- Email: [email protected]
The effect of an aldose reductase inhibitor (Epalrestat) on diabetic nephropathy in rats
ELSEVIER
Diabetes Research and Clinical Practice 25 (1994) 147-154
The effect of an aldose reductase inhibitor (Epalrestat) on diabetic nephropathy in rats Iori Itagaki* a, Kiyoshi Shimizu”, Yoshihisa Kamanakaa, Kazuhiko Ebatab, Ryuichi Kikkawab, Masakazu Hanedab, Yukio’ Shigetab aFukui Institute for Safety Research, One Pharmaceutical Co. Ltd., 50-10 Yamagishi, Mikuni-rho, Sakai-gun, Fukui 913, Japan bThird Department of Medicine, Shiga University of Medical Science, Seta, Otsu, Shiga 520-21, Japan
Received 10 January 1994; revision received 11 May 1994;accepted 9 June 1994
Abstract
In order to clarify the possible contribution of the abnormal polyol pathway to the development of diabetic nephropathy, the effect of aldose reductase inhibitor on renal function and morphology was examined in streptozotocin (STZ)-induced diabetic rats. Six months after STZ injection, glomerular filtration rate and renal plasma flow showed marked decline with significant increase in nuclear-free mesangial area (MA) and relative mesangial area (RMA; MA per glomerular area) in diabetic rats. Oral administration of an aldose reductase inhibitor, Epalrestat, prevented renal hypofunction and mesangial expansion in diabetic rats without influencing the levels of blood glucose. These results suggest that the abnormal polyol pathway in diabetic rats is closely related to the development of mesangial expansion, a morphologic representative of diabetic glomerulopathy, and renal hypofunction. Keywords: Aldose reductase inhibitor; Diabetic nephropathy; Glomerular morphometry; Mesangial expansion; Polyol pathway
1. Introduction In spite of the progress in the therapeutic treatment of diabetes mellitus, diabetic complications such as nephropathy remain unsolved at present. Histologically, diabetic nephropathy is characterized by the expansion of glomerular mesangium [ 1,2] which is believed to be a central mechanism for the loss of kidney function [3-61. However, the *Corresponding 82 6353.
author, Tel.: 0776 82 6161; Fax: 0776
pathogenesis of mesangial expansion due to diabetes is still unclear. Recently, the abnormal polyol pathway of glucose metabolism has been proposed to play an important role in the pathogenesis of diabetic complications including diabetic nephropathy 17-91. The aldose reductase, a rate-limiting enzyme of the polyol pathway, was found in both glomeruli and cultured mesangial cells and the accumulation of sorbitol was observed both in diabetic glomeruli [7] and in the mesa&al cells cultured under high glucose conditions [9]. If the
0168-8227/94/$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0168-8227(94)00969-2
148
I. Itagaki et al. /Diabeies
Res. Clin. Pratt.
abnormal polyol pathway causes diabetic nephropathy, the mesa&al expansion and renal hypofunction in diabetes might be prevented by the administration of aldose reductase inhibitor (ARI). To prove this hypothesis, we have examined the functional and morphologic changes in diabetic kidneys and the effect of aldose reductase inhibition on them. 2. Materials ad methods 2.1. Experimental protocol Male Sprague-Dawley rats (179.8-248.8 g, 6 weeks) were made diabetic by a single injection of 55 mg/kg streptozotocin (STZ; Sigma Chemical Co., St. Louis, MO) in 0.1 M citrate buffer, pH 4.5. The rats were housed individually in stainless steel wire mesh cages in a room with controlled temperature (23 i: 2°C) and humidity (55 f 10%) in a barrier system. They were allowed free access to commercial chow (CE-2, CLEA Japan Inc., Tokyo, Japan) and ultrafiltrated tap water. The rats given an injection of citrate buffer were used as a non-diabetic control group (C group). Diabetic rats were divided into three groups as follows: (1) an untreated-diabetic group (D group); (2) an AR&treated diabetic group (DA group): rats were treated with Epalrestat (5[( 12,2E)-2-methyl-3-phenylpropenylidenl-4-0x0-2thoxo-3-thiazolidine acetic acid, Ono Pharmaceutical Co. Ltd., Osaka, Japan) mixed in the above mentioned diet in 600 ppm; (3) an insulintreated diabetic group (DI group): rats were treated with daily subcutaneous injection of insulin (ULTRALENTE ISZILIN, Shimizu Pharmaceutical Co. Ltd., Shimizu, Japan) except Sunday. Administration of Epalrestat or insulin was started on day 3 of the experiment after conlirming the development of diabetes. The dose of insulin was established for individual animals within the range of 2- 16 units/rat so that urinary glucose of animals became negative and blood glucose levels became 80-150 mg/dl. Those animals in D or DA groups with marked emaciation together with ketonuria were also treated with 2 units/rat of insulin to produce levels of 400-700 mg/dl of blood glucose. All animals were kept for 6 months after induction of diabetes. The levels of blood glucose
25 (1994) 147-154
were determined on days 3,6, 13,20,27,55,83 and 174 by the glucose oxidase method using a Spectrophotometer 150-20 (Hitachi, Hitachi, Japan). 2.2. Examination of renal function Six months after induction of diabetes, rats were anesthetized by intraperitoneal injection of pentobarbital sodium (30 mgfkg). A catheter was inserted in the left carotid artery for blood sampling and another into the jugular vein for the infusion of inulin and p-aminohippurate (PAH) solution. The bladder was catheterized for urine sampling. A solution of inulin (1%) and PAH (2.5%) in normal saline was infused at a rate of 60 pl/min for 20 min as a priming load followed by constant infusion at a rate of 20 &nin during the study period. After a 60-min equilibration period, urine was collected every 30 min and blood samples were taken 15 min after every urine sampling period. Inulin and PAH were measured using the cysteine/tryptophan reaction and the method of Brun, with slight modification [lo], respectively. Urine samples were treated with concentrated HCl for 10 min at 70°C to hydrolyze glycated PAH. The average of three clearance periods was used for the calculation of the values of glomerular filtration rate (GFR) and renal plasma flow (RPF). 2.3. Morphometric analysis of the glomeruli The kidneys obtained from the rats in each group were fixed in 10% buffered formalin and embedded in paraffin. Each kidney was sectioned three times in 2 pm thickness (they were nonserially sectioned more than 20 pm apart from one another, to ensure that each glomerulus located in separate sections was not counted twice) and was stained with periodic acid-Stiff (PAS) to use for the morphometric analysis of glomeruli. The glomeruli of which the cut surface followed its vascular pole were measured. Glomeruli which were severely distorted were excluded from the count. A computer-assisted color image analyzer SP- 1000 (OLYMPUS-Avionix, Tokyo, Japan) was used for the glomerular morphometric analysis. The system consisted of a light microscope, a high resolution video camera, a computer connected to an image processor and a TV monitor. Image enhancements and measurements were made using IC 7098 ver-
I. Itagaki et al. /Diabetes Rex Clin. Pratt. 25 (1994) 147-154
Fig. 1. Making of glomerular images for morphometric analysis. A microscopic image of a glomerulus in the PAS stained kidney section was inputted in the image analyzer. The image of the glomerular tuft was electronically painted (dotted area) except the PAS-positive mesangial area. The measurement was automatically performed on the painted image of the glomerular tuft by the analyzer.
149
sion 1.50, which is the system software. A raw image of a glomerulus in the microscope was inputted to the analyzer through the camera and displayed on the TV screen in x 840 magnification. The whole image of the glomerular tuft in the analyzer was electronically enhanced by computer operation. It was painted, except for the PASpositive mesangial matrix (nuclear-free mesangial area), as shown in Fig. 1. The PAS-positive part other than the mesangium, such as the basement membrane, was carefully excluded from the mesangial area. The following measurements were automatically performed on the enhanced image by pixel counting; (1) glomerular area (GA, pm2), which was the total area of the painted or unpainted pixels on the glomerular tuft, indicating whole glomerular area, (2) nuclear-free mesangial area (MA, pm*), which was the total area of the unpainted pixels on the image of glomerular tuft and (3) relative mesangial area (RMA, %) which was the portion of MA in GA. Morphometric analysis was performed on 15 glomeruli per rat, and these data were averaged. 2.4. Statistical analysis The group comparison of each parameter was
Table 1 Clinical data of the rats Blood glucose
Blood glucose
Body weight
Kidney weight
[PI (mg/dl)
WI (m&W
(a)
k)
Control
Mean f SD. n
116.20 + 7.59 8
116.20 f 7.56 7
605.53 f 62.14 8
1.81 f 0.17 8
Diabetic
Mean f S.D. n Significance
462.44 f 48.05 8 *‘*###
462.44 f 48.05 7 l**###
353.11 + 50.29 9 ***###
2.32 f 0.49 8
Diabetic + ARI
Mean f S.D. n Significance
432.11 f 59.82 7 l**###
432.11 f 59.82 7 ***###
375.36 f 58.36 8 ***###
2.38 zt 0.59 8
Diabetic + insulin
Mean f SD. n
152.88 f 24.43 7
142.18 f 18.49 7
591.59 f 76.92 8
1.96 f 0.29 I
Values are means f SD. and the number. Blood glucose data of every determination point are averaged. Body weight and kidney weight are estimated after the maintenance period. Blood glucoce [F]: the value of the rats prepared for the kidney functional study. Blood glucoce [hi]: the value of the rats prepared for the glomerular morphometry. ARI: aldose reductase inhibitor. *** vs. C group P < 0.001; ### vs. DI group P < 0.001; - , no significance.
150
I. Itagaki et al. /Diabetes Res. Clin. Pratt. 25 (1994) 147-154
carried out by Scheffe’s test. On the analysis of the blood glucose, two-way analysis of variance (ANOVA) was performed before the group comparison. Rank correlation coefficient between RMA and reciprocal of GFR/rat, RPF/rat, GFRikidney wt. and RPFikidney wt. (l/GFR per rat, l/RPF per rat, l/GFR per kidney wt. and l/RPF per kidney wt.) was analyzed to discuss the relationship between renal function and mesangial expansion. P < 0.05 was considered statistically significant in the tests. 3. Results 3.1. Metabolic parameters The levels of blood glucose of the rats in the D and DA groups were elevated to markedly high levels in comparison with those in the C group throughout the 6 month maintenance period. The averages of every determination point were in-
T
d
r”l
T
dicated in Table 1 (the data of the rats prepared for functional study: [F], of morphometric analysis: [Ml). Significant differences (P < 0.001) were found between the C or DI group and the D or DA group, but no significance was found between the latter two groups. Therefore, Epalrestat did not influence blood glucose in the diabetic rats and the animals of the DA group were regarded as being diabetic to a similar level as that of the D group over the 6 months. After 6 months, the body weight of the rats in the D and DA groups was significantly (P < 0.001) smaller than that in the C or DI groups (Table 1). Kidney weight of the rats in the D and DA groups was slightly increased in contrast to the C or DI group, although the difference was not significant (Table 1). 3.2. Renal function GFR/rat, RPF/rat, GFR/kidney wt. and RPFI kidney wt. of the rats in the D group all showed
‘or
RPF/rat
GFR/rat
8003.$
2-
.9 P 6 E .4-
l-
T2-
. 3
O-
O-
GFR/kid wt
RPF/kid wt
FF
Fig. 2. Renal functions of rats 6 months after induction of diabetes. 0, control group (GFR/rat, GFRIkidney wt.: n = 8, RPF/rat, RPF/kidney wt., FF: n = 6); a, untreated diabetic group (GFRkat, GFRIkidney wt., RPF/rat, RPFkidney wt., FF: n = 8); Pp, diabetic group with Epalrestat treatment (GFRkat, GFRkidney wt.: n = 7, RPF/rat, RPFikidney wt., FF: n = 6); 0, insulin treated diabetic group (GFR/rat, GFRkidney wt.: n = 7, RPF/rat, RPFIkidney wt., FF: n = 6). *P < 0.05.
I. Itagaki
et al. /Diabetes
X 10 pm2
Rex
Clin. Pratt.
25 (1994)
1200
1500
1000
1200
12
-5 I*
800
8
900
600
6
600
400
300
200
-I-
0
15
%
Pm2
1800
147-154
I
***
0
Mesangial
Glomerular area
area
Relative mesangial area
Fig. 3. Results of the glomerular morphometric analysis 6 months after induction of diabetes. 0, control group (n = 7); q, untreated diabetic group (n = 7); 0, diabetic group with Epalrestat treatment (n = 7); 9 insulin treated diabetic group (n = 7). *P < 0.05; **P < 0.01; ***p < 0.001.
L+ +++ + ++ + ++ L
251 +
15
p=O.Ol
T .3 $ c9
II n
”
+ I
I
0
;
1
I
21
B er; -
‘5
++
+
10 +
++
p=o.o5
++
+ +++ +
n~
+
+
+
5
5 0
+
+
+
+
++
$
+
+b +
p=O.Ol
+
+++ +
0
+
5
10
15
20
25
RMA Rank
20
+ &
10
I&ALlk20
20
+b
20
l--t---cfi +c + 24
++
++ ++ +
+
+
p&O1
+
Fig. 4. The rank correlation between RMA and each reciprocal of the renal functions. RMA and GFR/rat: n = 22, ys = 0.545 (P < 0.01); RMA and RPF/rat: n = 19, ys = 0.472 (P < 0.05); RMA and GFRkidney wt.: n = 22, ys = 0.605 (P < 0.01); RMA and RPF/kidney wt.: n = 19, ys = 0.600 (P < 0.01).
152
I. Itagaki et al. /Diabetes Res. Ch. Pratt, 25 (1994) 147-1.54
marked declines in comparison with the C group 6 months after induction of diabetes. In comparison with the D group, these parameters in the DA group were maintained in similar levels to those in the C group, indicating that renal hypofunction due to diabetes was suppressed in the DA group. In the DI group, parameters of kidney function were also similar to those of the C group (Fig. 2). Filtration fraction (FF) was similar in every group (Fig. 2). Urinary albumin excretion measured by ELISA method was increased (P < 0.05) both in the DA group (3.16 f 1.71 mg/lOO g body wt. per 24 h) and in the D group (3.24 f 2.34) as compared with the C group (0.44 f 0.30), although the data are not shown in the Table. The value of the DI group (0.72 f 0.49) was similar to that of the C group. 3.3. Morphometric analysis Glomerular area (GA) showed no difference between the groups. However, mesangial expansion was demonstrated both absolutely and relatively in the D group, as the MA (RMA) value shown in Fig. 3. In the DA group, the MA (RMA) value was smaller than in the D group (P < O.OOl), indicating that mesangial expansion was apparently suppressed in this group. In the DI group, the MA (RMA) value was similar to that in the C group. In order to investigate whether mesangial expansion influences the decline of renal function or not, the rank correlation coefficient between RMA and each of l/GFR per rat, l/RPF per rat, l/GFR per kidney wt. and l/RPF per kidney wt. was analyzed in all groups of the rats. All parameters of the kidney function were significantly related with RMA (Fig. 4). 4. Discussion The present study indicates that both functional and histological abnormalities of the kidneys of diabetic rats are prevented by the administration of an ARI, Epalrestat. Six months after STZ injection, RPF and GFR were decreased in untreated diabetic rats and both were prevented by either the administration of Epalrestat or the treatment with insulin. Untreated diabetic rats exhibited significant mesangial expansion, which was also
prevented by either Epalrestat or insulin treatment. Although the image analyzer used in this study is not commonly used for research in diabetic nephropathy, a similar computer-assisted image analyzer is used in other fields of medical science [ 11,121. It is quite appropriate to analyze glomerular changes by the analyzer, since the data for GA was similar to that of Orloff et al. [13] measured by Camera lucida tracing (9924 f 510 pm2 in control rats). From the results of glomerular morphometry in the present study, it was demonstrated that the mesangial matrix of the untreated diabetic rats showed a marked expansion both absolutely and relatively 6 months after the induction of diabetes. The lesion was regarded as a diabetic change because the normalization of blood glucose by insulin could prevent this mesangial expansion. This mesangial expansion was significantly suppressed in diabetic rats treated with Epalrestat without improving the glycemic control. These results are in agreement with the results reported by Mauer et al. [ 141who showed that Sorbinil, another ARI, was able to prevent the mesangial expansion of diabetic rats. However, Mauer et al. found the mesangial expansion only in diabetic rats on a high protein diet (50% protein) and they could not find a significant mesangial expansion in diabetic rats on a 20% protein diet. In the present study, both control and diabetic rats received the standard chow containing 25% protein. Although a high protein diet may accelerate the diabetic changes of the kidneys through the hemodynamic abnormalities, mesa&al expansion has been reported in diabetic rats on the usual protein diet [5,13,14]. Glomerular sorbitol content in the DA group (0.35 * 0.42 nmol/mg protein) was less than in the D group (0.93 f 0.40). The decline of sorbitol content without influencing blood glucose due to Epalrestat treatment suggests an inhibition of aldose reductase in the glomeruli. However, the incomplete suppression by Epalrestat in comparison with insulin may suggest the presence of other factors involved in mesangial expansion which could not be identified by this study. On the other hand, GFR and RPF showed a marked decline in the D group. This result& sup-
I. ltagaki et al. /Diabetes Res. Clin. Pratt. 25 (1994) 147-154
ported by other reports [5,14-171 indicating renal hypofunction in severely diabetic animals, although many reports demonstrate hypertiltration in diabetic animals with moderate glycemic control [ 16,18-241. In comparison with the D group, GFR and RPF in the DA group was kept at the same levels as those in the C or DI groups in accordance with the report of Craven and DeRubertis [ 161. The suppressive effect of AR1 on renal hypofunction suggests a relationship between abnormal polyol metabolism and renal hypofunction in diabetes. Rank correlation coefficients between RMA and functional studies were calculated to confirm the idea that renal hypofunction in diabetes and its reversal by Epalrestat were caused through morphologic changes of the mesangium. Each parameter significantly correlated with RMA, suggesting the influence of mesa&al change on renal function. Albuminuria, however, was not suppressed by the AR1 Epalrestat, in contrast to the results of other researches using various ARIs [ 19,24-281. The results in this study may be due to severe inhibition of albumin reabsorption in the renal proximal tubules caused by a mechanism independent of polyol metabolism [29] or glycogen deposition [30,31]. In summary, the present study suggests that abnormal polyol metabolism in diabetes might be related to both mesangial expansion and renal hypofunction and thus Epalrestat, which prevented these changes in diabetes, might be considered to be a drug worthy of investigation in future clinical studies. Furthermore, the mesangial cells have been reported to produce type IV collagen [32], a major component of the mesangial matrix and the production of type IV collagen was found to be enhanced in the mesangial cells cultured under high glucose conditions [33]. It remains to be clarified, however, whether the enhanced activity of the polyol pathway plays a role in abnormal collagen production of mesangial cells in diabetes. Referems
111Murphy,G.F., Kwan,T.H. and Mihm, Jr., M.C. (1984) The Kidney. In: S.L. Robbins, R.S. Cotran and V.
153
Kumar, (Ed.), Pathologic Bases of Disease. Third Edition. WB Saunders, Philadelphia, PA, pp. 991-1061. 121@sterby, R. (1992) Glomerular structural changes in type 1 (insulin-dependent) diabetes mellitus: causes, consequences, and prevention. Diabetologia 35, 803-812. 131Deckert, T., Feldt-Rasmussen, B., Borch-Johnsen, K., Jensen, T. and Kofoed-Enevoldsen, A. (1989) Albuminuria reflects widespread vascular damage. The Steno hypothesis. Diabetologia 32, 219-226. 141 Mauer, SM., Steffes, M.W., Ellis, E.N., Sutherland, D.E.R., Brown, D.M. and Goetx, F.C. (1984) Structuralfunctional relationship in diabetic nephropathy. J. Clin. Invest. 74, 1l43- 1155. [51 Michels, L.D., O’Donnel, M.P. and Keane, W.F. (1984) Glomerular hemodynamic and structural correlations in long-term experimental diabetic rats. J. Lab. Clin. Med. 103, 840-847. WI Steffes, M.W., @sterby, R., Chaves, B. and Mauer, S.M. (1989) Mesangial expansion as a central mechanism for loss of kidney function in diabetic patients. Diabetes 38, 1077-1081. 171 Corder, C.N., Braughler, J.M. and Culp, P.A. (1979) Quantitative histochemistry of the sorbitol pathway in glomeruli and small arteries of human diabetic kidney. Folia His&hem. Cytochem. 17, 137-146. PI Kaneko, M., Carper, D., Nishimura, C., Millen, J., Bock, M. and Hohman, T.C. (1990) Induction of aldose reductase expression in rat kidney mesangial cells and Chinese hamster ovary cells under hypertonic condition. Exp. Cell Res. 188, 135-140. 191 Kikkawa, R., Umemura, K., Haneda, M., Arimura, T., Ebata, K. and Shigeta, Y. (1987) Evidence for existence of polyol pathway in cultured rat mesangial cells. Diabetes 36, 240-243. 1101 Dalton, R.N., Wiseman, M.J., Turner, C. and Viberti, G.T.I. (1988) Measurement of urinary paraaminohippuric acid in glycosuric diabetics. Kidney Int. 34, 117-120. 1111 Arden, KC., Bucana, C.D., Johnston, D.A. and Pahtak, S. (1989) Computer-assisted image analysis of silver staining in normal and leukemic bone marrow. Int. J. Cancer 43, 395-398. [I21 Derenzini, M., Nardi, F., Farabegoli, F., Ottinetti, A., Roncaroli, F. and Bussolati, G. (1989) Distribution of silver-stained interphase nucleolar organizer regions as a parameter to distinguish neoplastic from nonneoplastic reactive cells in human effusions. Acta Cytol. 33, 491-498. 1131 Orloff, M.J., Yamanaka, N., Greenleaf, G.E., Huang, Y., Huang, D. and Leng, X. (1986) Reversal of mesangial enlargement in rats with long-standing diabetes by whole pancreas transplantation. Diabetes 35. 347-354. [14] Mauer, S.M., Steffes, M.W., Azar, S. and Brown, D.M. (1989) Effects of Sorbinil on glomerular structure and function in long-term-diabetic rats. Diabetes 38, 839-846. [I51 Michels, L.D., Davidman, M. and Keane, W.F. (1981)
154
I. Itagaki et al. /Diabetes Res. Clin. Pratt. 25 (1994)
Determinants of glomerular filtration and plasma flow in experimental diabetic rats. J. Lab. Chn. Med. 98, 869-885. ]16J Craven, P.A. and DeRubertis, F.R. (1989) Sorbinil suppresses glomerular prostaglandin production in the streptozotocin diabetic rat. Metabolism 38, 649-654. [I71 Hostetter, T.H., Yroy, J.L. and Brenner, B.M. (1981) Glomerular hemodynamics in experimental diabetes mellitus. Kidney Int. 19, 410-415. [IS] Bank, N., Mower, P., Aynedjian, H.S., Wilkes, B.M. and Silverman, S. (1989) Sorbinil prevents glomerular hyperfusion in diabetic rats. Am. J. Physiol. 256, FIOOO-F1006. [19] Chang, W.P., Dimitriadis, E., Allen, T., Dunlop, M.E., Cooper, M. and Larkins, R.G. (1991) The effect of aldose reductase inhibitors on glomerular prostaglandin production and urinary albumin excretion in experimental diabetes mellitus. Diabetologia 34, 225-231. [20] Daniels, B.S. and Hostetter, T.H. (1989) Aldose reductase inhibition and glomerular abnormalities in diabetic rats. Diabetes 38, 981-986. [21] Goldfarb, S., Ziyadeh, F.N., Kern, E.F.O. and Simmons, D. (1991) Effects of polyol-pathway inhibition and dietary myo-inositol on glomerular hemodinamic function in experimental diabetes mellitus in rats. Diabetes 40, 465-471. [22] Green, D. (1988) The pathogenesis and prevention of diabetic neuropathy and nephropathy. Metabolism 37, 25-29. 1231 Pedersen, M.M., Christiansen, J.S. and Mogensen, C.E. (1991) Reduction of glomerular hyperfiltration in normoalbumic IDDM patients by 6 mo of aldose reductase inhibition. Diabetes 40, 27-53 1. [24] Tilton, R.G., Chang, K., Pughese, G. et al. Prevention of diabetic rats by aldose reductase inhibitors. Diabetes 38, 1258-1270. [25] Beyer-Mears, A., Cruz, E., Edelist, T. and Varagiannis,
147-154
E. (1986) Diminished proteinuria in diabetes mellitus by Sorbinil, an aldose reductase inhibitor. Pharmacology 32, 52-60. (261 Beyer-Mears, A., Murray, F.T., Del Val M., Cruz, E. and Sciadini, M. (1988) Reversal of proteinuria by Sorbinil, an aldose reductase inhibitor in spontaneously diabetic (BB) rats. Pharmacology 36, 112-120. [27] Jenings, P.E., Nightingale, S., Le Guen, C. et al. (1990) Prolonged aldose reductase inhibition in chronic peripheral diabetic neuropathy: effects on microangiopathy. Diabetic Med. I, 63-68. [28] McCaleb, M.L., Serdy, J., Ackerman, D.M. and Dvornik, D. (1988) Prevention of urinary albumin excretion in 6 month streptozotocin-diabetic rats with the aldose reductase. inhibitor Tolrestat. J. Diab. Complications 2, 16-18. [29] Kaneda, K., Iwao, J., Sakata, N. and Takebayashi, S. (1992) Correlation between mitochondrial enlargement in renal proximal tubules and microalbuminuria in rats with early stmptozotocin-induced diabetes. Acta Pathol. Jpn. 42, 855-860. [30] Kato, C., Abe, T., Murakami, T. and Kobayashi, A. (1987) Glycogen deposition of renal tubular ceils in strep tozotocin induced diabetic rats. J. Clin. Electron Microsc. 20, 5-6. 1311 Kato, C., Abe, T., Sawada, S. and Sugiura, Y. (1986) An electron microscopic study of the tubular lesions in experimental diabetic rats. J. Clin. Electron Microsc. 19, 5-6. 1321 Ayo, S.H., Radnic, R.A., Garoni, J.A., Glass, W.F. and Kreisberg, J.I. (1990) High glucose. causes an increse in extracellular matrix proteins in cultured mesangial cells. Am. J. Pathol. 136, 1339-1348. [33] Haneda, M., Kikkawa, R., Horide, N. et al. (1991) Glucose enhances type IV collagen production in cultured rat glomerular mesangial cells. Diabetlogia 34, 198-200.
Diabetes Research and Clinical Practice 25 (1994) 147-154
The effect of an aldose reductase inhibitor (Epalrestat) on diabetic nephropathy in rats Iori Itagaki* a, Kiyoshi Shimizu”, Yoshihisa Kamanakaa, Kazuhiko Ebatab, Ryuichi Kikkawab, Masakazu Hanedab, Yukio’ Shigetab aFukui Institute for Safety Research, One Pharmaceutical Co. Ltd., 50-10 Yamagishi, Mikuni-rho, Sakai-gun, Fukui 913, Japan bThird Department of Medicine, Shiga University of Medical Science, Seta, Otsu, Shiga 520-21, Japan
Received 10 January 1994; revision received 11 May 1994;accepted 9 June 1994
Abstract
In order to clarify the possible contribution of the abnormal polyol pathway to the development of diabetic nephropathy, the effect of aldose reductase inhibitor on renal function and morphology was examined in streptozotocin (STZ)-induced diabetic rats. Six months after STZ injection, glomerular filtration rate and renal plasma flow showed marked decline with significant increase in nuclear-free mesangial area (MA) and relative mesangial area (RMA; MA per glomerular area) in diabetic rats. Oral administration of an aldose reductase inhibitor, Epalrestat, prevented renal hypofunction and mesangial expansion in diabetic rats without influencing the levels of blood glucose. These results suggest that the abnormal polyol pathway in diabetic rats is closely related to the development of mesangial expansion, a morphologic representative of diabetic glomerulopathy, and renal hypofunction. Keywords: Aldose reductase inhibitor; Diabetic nephropathy; Glomerular morphometry; Mesangial expansion; Polyol pathway
1. Introduction In spite of the progress in the therapeutic treatment of diabetes mellitus, diabetic complications such as nephropathy remain unsolved at present. Histologically, diabetic nephropathy is characterized by the expansion of glomerular mesangium [ 1,2] which is believed to be a central mechanism for the loss of kidney function [3-61. However, the *Corresponding 82 6353.
author, Tel.: 0776 82 6161; Fax: 0776
pathogenesis of mesangial expansion due to diabetes is still unclear. Recently, the abnormal polyol pathway of glucose metabolism has been proposed to play an important role in the pathogenesis of diabetic complications including diabetic nephropathy 17-91. The aldose reductase, a rate-limiting enzyme of the polyol pathway, was found in both glomeruli and cultured mesangial cells and the accumulation of sorbitol was observed both in diabetic glomeruli [7] and in the mesa&al cells cultured under high glucose conditions [9]. If the
0168-8227/94/$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0168-8227(94)00969-2
148
I. Itagaki et al. /Diabeies
Res. Clin. Pratt.
abnormal polyol pathway causes diabetic nephropathy, the mesa&al expansion and renal hypofunction in diabetes might be prevented by the administration of aldose reductase inhibitor (ARI). To prove this hypothesis, we have examined the functional and morphologic changes in diabetic kidneys and the effect of aldose reductase inhibition on them. 2. Materials ad methods 2.1. Experimental protocol Male Sprague-Dawley rats (179.8-248.8 g, 6 weeks) were made diabetic by a single injection of 55 mg/kg streptozotocin (STZ; Sigma Chemical Co., St. Louis, MO) in 0.1 M citrate buffer, pH 4.5. The rats were housed individually in stainless steel wire mesh cages in a room with controlled temperature (23 i: 2°C) and humidity (55 f 10%) in a barrier system. They were allowed free access to commercial chow (CE-2, CLEA Japan Inc., Tokyo, Japan) and ultrafiltrated tap water. The rats given an injection of citrate buffer were used as a non-diabetic control group (C group). Diabetic rats were divided into three groups as follows: (1) an untreated-diabetic group (D group); (2) an AR&treated diabetic group (DA group): rats were treated with Epalrestat (5[( 12,2E)-2-methyl-3-phenylpropenylidenl-4-0x0-2thoxo-3-thiazolidine acetic acid, Ono Pharmaceutical Co. Ltd., Osaka, Japan) mixed in the above mentioned diet in 600 ppm; (3) an insulintreated diabetic group (DI group): rats were treated with daily subcutaneous injection of insulin (ULTRALENTE ISZILIN, Shimizu Pharmaceutical Co. Ltd., Shimizu, Japan) except Sunday. Administration of Epalrestat or insulin was started on day 3 of the experiment after conlirming the development of diabetes. The dose of insulin was established for individual animals within the range of 2- 16 units/rat so that urinary glucose of animals became negative and blood glucose levels became 80-150 mg/dl. Those animals in D or DA groups with marked emaciation together with ketonuria were also treated with 2 units/rat of insulin to produce levels of 400-700 mg/dl of blood glucose. All animals were kept for 6 months after induction of diabetes. The levels of blood glucose
25 (1994) 147-154
were determined on days 3,6, 13,20,27,55,83 and 174 by the glucose oxidase method using a Spectrophotometer 150-20 (Hitachi, Hitachi, Japan). 2.2. Examination of renal function Six months after induction of diabetes, rats were anesthetized by intraperitoneal injection of pentobarbital sodium (30 mgfkg). A catheter was inserted in the left carotid artery for blood sampling and another into the jugular vein for the infusion of inulin and p-aminohippurate (PAH) solution. The bladder was catheterized for urine sampling. A solution of inulin (1%) and PAH (2.5%) in normal saline was infused at a rate of 60 pl/min for 20 min as a priming load followed by constant infusion at a rate of 20 &nin during the study period. After a 60-min equilibration period, urine was collected every 30 min and blood samples were taken 15 min after every urine sampling period. Inulin and PAH were measured using the cysteine/tryptophan reaction and the method of Brun, with slight modification [lo], respectively. Urine samples were treated with concentrated HCl for 10 min at 70°C to hydrolyze glycated PAH. The average of three clearance periods was used for the calculation of the values of glomerular filtration rate (GFR) and renal plasma flow (RPF). 2.3. Morphometric analysis of the glomeruli The kidneys obtained from the rats in each group were fixed in 10% buffered formalin and embedded in paraffin. Each kidney was sectioned three times in 2 pm thickness (they were nonserially sectioned more than 20 pm apart from one another, to ensure that each glomerulus located in separate sections was not counted twice) and was stained with periodic acid-Stiff (PAS) to use for the morphometric analysis of glomeruli. The glomeruli of which the cut surface followed its vascular pole were measured. Glomeruli which were severely distorted were excluded from the count. A computer-assisted color image analyzer SP- 1000 (OLYMPUS-Avionix, Tokyo, Japan) was used for the glomerular morphometric analysis. The system consisted of a light microscope, a high resolution video camera, a computer connected to an image processor and a TV monitor. Image enhancements and measurements were made using IC 7098 ver-
I. Itagaki et al. /Diabetes Rex Clin. Pratt. 25 (1994) 147-154
Fig. 1. Making of glomerular images for morphometric analysis. A microscopic image of a glomerulus in the PAS stained kidney section was inputted in the image analyzer. The image of the glomerular tuft was electronically painted (dotted area) except the PAS-positive mesangial area. The measurement was automatically performed on the painted image of the glomerular tuft by the analyzer.
149
sion 1.50, which is the system software. A raw image of a glomerulus in the microscope was inputted to the analyzer through the camera and displayed on the TV screen in x 840 magnification. The whole image of the glomerular tuft in the analyzer was electronically enhanced by computer operation. It was painted, except for the PASpositive mesangial matrix (nuclear-free mesangial area), as shown in Fig. 1. The PAS-positive part other than the mesangium, such as the basement membrane, was carefully excluded from the mesangial area. The following measurements were automatically performed on the enhanced image by pixel counting; (1) glomerular area (GA, pm2), which was the total area of the painted or unpainted pixels on the glomerular tuft, indicating whole glomerular area, (2) nuclear-free mesangial area (MA, pm*), which was the total area of the unpainted pixels on the image of glomerular tuft and (3) relative mesangial area (RMA, %) which was the portion of MA in GA. Morphometric analysis was performed on 15 glomeruli per rat, and these data were averaged. 2.4. Statistical analysis The group comparison of each parameter was
Table 1 Clinical data of the rats Blood glucose
Blood glucose
Body weight
Kidney weight
[PI (mg/dl)
WI (m&W
(a)
k)
Control
Mean f SD. n
116.20 + 7.59 8
116.20 f 7.56 7
605.53 f 62.14 8
1.81 f 0.17 8
Diabetic
Mean f S.D. n Significance
462.44 f 48.05 8 *‘*###
462.44 f 48.05 7 l**###
353.11 + 50.29 9 ***###
2.32 f 0.49 8
Diabetic + ARI
Mean f S.D. n Significance
432.11 f 59.82 7 l**###
432.11 f 59.82 7 ***###
375.36 f 58.36 8 ***###
2.38 zt 0.59 8
Diabetic + insulin
Mean f SD. n
152.88 f 24.43 7
142.18 f 18.49 7
591.59 f 76.92 8
1.96 f 0.29 I
Values are means f SD. and the number. Blood glucose data of every determination point are averaged. Body weight and kidney weight are estimated after the maintenance period. Blood glucoce [F]: the value of the rats prepared for the kidney functional study. Blood glucoce [hi]: the value of the rats prepared for the glomerular morphometry. ARI: aldose reductase inhibitor. *** vs. C group P < 0.001; ### vs. DI group P < 0.001; - , no significance.
150
I. Itagaki et al. /Diabetes Res. Clin. Pratt. 25 (1994) 147-154
carried out by Scheffe’s test. On the analysis of the blood glucose, two-way analysis of variance (ANOVA) was performed before the group comparison. Rank correlation coefficient between RMA and reciprocal of GFR/rat, RPF/rat, GFRikidney wt. and RPFikidney wt. (l/GFR per rat, l/RPF per rat, l/GFR per kidney wt. and l/RPF per kidney wt.) was analyzed to discuss the relationship between renal function and mesangial expansion. P < 0.05 was considered statistically significant in the tests. 3. Results 3.1. Metabolic parameters The levels of blood glucose of the rats in the D and DA groups were elevated to markedly high levels in comparison with those in the C group throughout the 6 month maintenance period. The averages of every determination point were in-
T
d
r”l
T
dicated in Table 1 (the data of the rats prepared for functional study: [F], of morphometric analysis: [Ml). Significant differences (P < 0.001) were found between the C or DI group and the D or DA group, but no significance was found between the latter two groups. Therefore, Epalrestat did not influence blood glucose in the diabetic rats and the animals of the DA group were regarded as being diabetic to a similar level as that of the D group over the 6 months. After 6 months, the body weight of the rats in the D and DA groups was significantly (P < 0.001) smaller than that in the C or DI groups (Table 1). Kidney weight of the rats in the D and DA groups was slightly increased in contrast to the C or DI group, although the difference was not significant (Table 1). 3.2. Renal function GFR/rat, RPF/rat, GFR/kidney wt. and RPFI kidney wt. of the rats in the D group all showed
‘or
RPF/rat
GFR/rat
8003.$
2-
.9 P 6 E .4-
l-
T2-
. 3
O-
O-
GFR/kid wt
RPF/kid wt
FF
Fig. 2. Renal functions of rats 6 months after induction of diabetes. 0, control group (GFR/rat, GFRIkidney wt.: n = 8, RPF/rat, RPF/kidney wt., FF: n = 6); a, untreated diabetic group (GFRkat, GFRIkidney wt., RPF/rat, RPFkidney wt., FF: n = 8); Pp, diabetic group with Epalrestat treatment (GFRkat, GFRkidney wt.: n = 7, RPF/rat, RPFikidney wt., FF: n = 6); 0, insulin treated diabetic group (GFR/rat, GFRkidney wt.: n = 7, RPF/rat, RPFIkidney wt., FF: n = 6). *P < 0.05.
I. Itagaki
et al. /Diabetes
X 10 pm2
Rex
Clin. Pratt.
25 (1994)
1200
1500
1000
1200
12
-5 I*
800
8
900
600
6
600
400
300
200
-I-
0
15
%
Pm2
1800
147-154
I
***
0
Mesangial
Glomerular area
area
Relative mesangial area
Fig. 3. Results of the glomerular morphometric analysis 6 months after induction of diabetes. 0, control group (n = 7); q, untreated diabetic group (n = 7); 0, diabetic group with Epalrestat treatment (n = 7); 9 insulin treated diabetic group (n = 7). *P < 0.05; **P < 0.01; ***p < 0.001.
L+ +++ + ++ + ++ L
251 +
15
p=O.Ol
T .3 $ c9
II n
”
+ I
I
0
;
1
I
21
B er; -
‘5
++
+
10 +
++
p=o.o5
++
+ +++ +
n~
+
+
+
5
5 0
+
+
+
+
++
$
+
+b +
p=O.Ol
+
+++ +
0
+
5
10
15
20
25
RMA Rank
20
+ &
10
I&ALlk20
20
+b
20
l--t---cfi +c + 24
++
++ ++ +
+
+
p&O1
+
Fig. 4. The rank correlation between RMA and each reciprocal of the renal functions. RMA and GFR/rat: n = 22, ys = 0.545 (P < 0.01); RMA and RPF/rat: n = 19, ys = 0.472 (P < 0.05); RMA and GFRkidney wt.: n = 22, ys = 0.605 (P < 0.01); RMA and RPF/kidney wt.: n = 19, ys = 0.600 (P < 0.01).
152
I. Itagaki et al. /Diabetes Res. Ch. Pratt, 25 (1994) 147-1.54
marked declines in comparison with the C group 6 months after induction of diabetes. In comparison with the D group, these parameters in the DA group were maintained in similar levels to those in the C group, indicating that renal hypofunction due to diabetes was suppressed in the DA group. In the DI group, parameters of kidney function were also similar to those of the C group (Fig. 2). Filtration fraction (FF) was similar in every group (Fig. 2). Urinary albumin excretion measured by ELISA method was increased (P < 0.05) both in the DA group (3.16 f 1.71 mg/lOO g body wt. per 24 h) and in the D group (3.24 f 2.34) as compared with the C group (0.44 f 0.30), although the data are not shown in the Table. The value of the DI group (0.72 f 0.49) was similar to that of the C group. 3.3. Morphometric analysis Glomerular area (GA) showed no difference between the groups. However, mesangial expansion was demonstrated both absolutely and relatively in the D group, as the MA (RMA) value shown in Fig. 3. In the DA group, the MA (RMA) value was smaller than in the D group (P < O.OOl), indicating that mesangial expansion was apparently suppressed in this group. In the DI group, the MA (RMA) value was similar to that in the C group. In order to investigate whether mesangial expansion influences the decline of renal function or not, the rank correlation coefficient between RMA and each of l/GFR per rat, l/RPF per rat, l/GFR per kidney wt. and l/RPF per kidney wt. was analyzed in all groups of the rats. All parameters of the kidney function were significantly related with RMA (Fig. 4). 4. Discussion The present study indicates that both functional and histological abnormalities of the kidneys of diabetic rats are prevented by the administration of an ARI, Epalrestat. Six months after STZ injection, RPF and GFR were decreased in untreated diabetic rats and both were prevented by either the administration of Epalrestat or the treatment with insulin. Untreated diabetic rats exhibited significant mesangial expansion, which was also
prevented by either Epalrestat or insulin treatment. Although the image analyzer used in this study is not commonly used for research in diabetic nephropathy, a similar computer-assisted image analyzer is used in other fields of medical science [ 11,121. It is quite appropriate to analyze glomerular changes by the analyzer, since the data for GA was similar to that of Orloff et al. [13] measured by Camera lucida tracing (9924 f 510 pm2 in control rats). From the results of glomerular morphometry in the present study, it was demonstrated that the mesangial matrix of the untreated diabetic rats showed a marked expansion both absolutely and relatively 6 months after the induction of diabetes. The lesion was regarded as a diabetic change because the normalization of blood glucose by insulin could prevent this mesangial expansion. This mesangial expansion was significantly suppressed in diabetic rats treated with Epalrestat without improving the glycemic control. These results are in agreement with the results reported by Mauer et al. [ 141who showed that Sorbinil, another ARI, was able to prevent the mesangial expansion of diabetic rats. However, Mauer et al. found the mesangial expansion only in diabetic rats on a high protein diet (50% protein) and they could not find a significant mesangial expansion in diabetic rats on a 20% protein diet. In the present study, both control and diabetic rats received the standard chow containing 25% protein. Although a high protein diet may accelerate the diabetic changes of the kidneys through the hemodynamic abnormalities, mesa&al expansion has been reported in diabetic rats on the usual protein diet [5,13,14]. Glomerular sorbitol content in the DA group (0.35 * 0.42 nmol/mg protein) was less than in the D group (0.93 f 0.40). The decline of sorbitol content without influencing blood glucose due to Epalrestat treatment suggests an inhibition of aldose reductase in the glomeruli. However, the incomplete suppression by Epalrestat in comparison with insulin may suggest the presence of other factors involved in mesangial expansion which could not be identified by this study. On the other hand, GFR and RPF showed a marked decline in the D group. This result& sup-
I. ltagaki et al. /Diabetes Res. Clin. Pratt. 25 (1994) 147-154
ported by other reports [5,14-171 indicating renal hypofunction in severely diabetic animals, although many reports demonstrate hypertiltration in diabetic animals with moderate glycemic control [ 16,18-241. In comparison with the D group, GFR and RPF in the DA group was kept at the same levels as those in the C or DI groups in accordance with the report of Craven and DeRubertis [ 161. The suppressive effect of AR1 on renal hypofunction suggests a relationship between abnormal polyol metabolism and renal hypofunction in diabetes. Rank correlation coefficients between RMA and functional studies were calculated to confirm the idea that renal hypofunction in diabetes and its reversal by Epalrestat were caused through morphologic changes of the mesangium. Each parameter significantly correlated with RMA, suggesting the influence of mesa&al change on renal function. Albuminuria, however, was not suppressed by the AR1 Epalrestat, in contrast to the results of other researches using various ARIs [ 19,24-281. The results in this study may be due to severe inhibition of albumin reabsorption in the renal proximal tubules caused by a mechanism independent of polyol metabolism [29] or glycogen deposition [30,31]. In summary, the present study suggests that abnormal polyol metabolism in diabetes might be related to both mesangial expansion and renal hypofunction and thus Epalrestat, which prevented these changes in diabetes, might be considered to be a drug worthy of investigation in future clinical studies. Furthermore, the mesangial cells have been reported to produce type IV collagen [32], a major component of the mesangial matrix and the production of type IV collagen was found to be enhanced in the mesangial cells cultured under high glucose conditions [33]. It remains to be clarified, however, whether the enhanced activity of the polyol pathway plays a role in abnormal collagen production of mesangial cells in diabetes. Referems
111Murphy,G.F., Kwan,T.H. and Mihm, Jr., M.C. (1984) The Kidney. In: S.L. Robbins, R.S. Cotran and V.
153
Kumar, (Ed.), Pathologic Bases of Disease. Third Edition. WB Saunders, Philadelphia, PA, pp. 991-1061. 121@sterby, R. (1992) Glomerular structural changes in type 1 (insulin-dependent) diabetes mellitus: causes, consequences, and prevention. Diabetologia 35, 803-812. 131Deckert, T., Feldt-Rasmussen, B., Borch-Johnsen, K., Jensen, T. and Kofoed-Enevoldsen, A. (1989) Albuminuria reflects widespread vascular damage. The Steno hypothesis. Diabetologia 32, 219-226. 141 Mauer, SM., Steffes, M.W., Ellis, E.N., Sutherland, D.E.R., Brown, D.M. and Goetx, F.C. (1984) Structuralfunctional relationship in diabetic nephropathy. J. Clin. Invest. 74, 1l43- 1155. [51 Michels, L.D., O’Donnel, M.P. and Keane, W.F. (1984) Glomerular hemodynamic and structural correlations in long-term experimental diabetic rats. J. Lab. Clin. Med. 103, 840-847. WI Steffes, M.W., @sterby, R., Chaves, B. and Mauer, S.M. (1989) Mesangial expansion as a central mechanism for loss of kidney function in diabetic patients. Diabetes 38, 1077-1081. 171 Corder, C.N., Braughler, J.M. and Culp, P.A. (1979) Quantitative histochemistry of the sorbitol pathway in glomeruli and small arteries of human diabetic kidney. Folia His&hem. Cytochem. 17, 137-146. PI Kaneko, M., Carper, D., Nishimura, C., Millen, J., Bock, M. and Hohman, T.C. (1990) Induction of aldose reductase expression in rat kidney mesangial cells and Chinese hamster ovary cells under hypertonic condition. Exp. Cell Res. 188, 135-140. 191 Kikkawa, R., Umemura, K., Haneda, M., Arimura, T., Ebata, K. and Shigeta, Y. (1987) Evidence for existence of polyol pathway in cultured rat mesangial cells. Diabetes 36, 240-243. 1101 Dalton, R.N., Wiseman, M.J., Turner, C. and Viberti, G.T.I. (1988) Measurement of urinary paraaminohippuric acid in glycosuric diabetics. Kidney Int. 34, 117-120. 1111 Arden, KC., Bucana, C.D., Johnston, D.A. and Pahtak, S. (1989) Computer-assisted image analysis of silver staining in normal and leukemic bone marrow. Int. J. Cancer 43, 395-398. [I21 Derenzini, M., Nardi, F., Farabegoli, F., Ottinetti, A., Roncaroli, F. and Bussolati, G. (1989) Distribution of silver-stained interphase nucleolar organizer regions as a parameter to distinguish neoplastic from nonneoplastic reactive cells in human effusions. Acta Cytol. 33, 491-498. 1131 Orloff, M.J., Yamanaka, N., Greenleaf, G.E., Huang, Y., Huang, D. and Leng, X. (1986) Reversal of mesangial enlargement in rats with long-standing diabetes by whole pancreas transplantation. Diabetes 35. 347-354. [14] Mauer, S.M., Steffes, M.W., Azar, S. and Brown, D.M. (1989) Effects of Sorbinil on glomerular structure and function in long-term-diabetic rats. Diabetes 38, 839-846. [I51 Michels, L.D., Davidman, M. and Keane, W.F. (1981)
154
I. Itagaki et al. /Diabetes Res. Clin. Pratt. 25 (1994)
Determinants of glomerular filtration and plasma flow in experimental diabetic rats. J. Lab. Chn. Med. 98, 869-885. ]16J Craven, P.A. and DeRubertis, F.R. (1989) Sorbinil suppresses glomerular prostaglandin production in the streptozotocin diabetic rat. Metabolism 38, 649-654. [I71 Hostetter, T.H., Yroy, J.L. and Brenner, B.M. (1981) Glomerular hemodynamics in experimental diabetes mellitus. Kidney Int. 19, 410-415. [IS] Bank, N., Mower, P., Aynedjian, H.S., Wilkes, B.M. and Silverman, S. (1989) Sorbinil prevents glomerular hyperfusion in diabetic rats. Am. J. Physiol. 256, FIOOO-F1006. [19] Chang, W.P., Dimitriadis, E., Allen, T., Dunlop, M.E., Cooper, M. and Larkins, R.G. (1991) The effect of aldose reductase inhibitors on glomerular prostaglandin production and urinary albumin excretion in experimental diabetes mellitus. Diabetologia 34, 225-231. [20] Daniels, B.S. and Hostetter, T.H. (1989) Aldose reductase inhibition and glomerular abnormalities in diabetic rats. Diabetes 38, 981-986. [21] Goldfarb, S., Ziyadeh, F.N., Kern, E.F.O. and Simmons, D. (1991) Effects of polyol-pathway inhibition and dietary myo-inositol on glomerular hemodinamic function in experimental diabetes mellitus in rats. Diabetes 40, 465-471. [22] Green, D. (1988) The pathogenesis and prevention of diabetic neuropathy and nephropathy. Metabolism 37, 25-29. 1231 Pedersen, M.M., Christiansen, J.S. and Mogensen, C.E. (1991) Reduction of glomerular hyperfiltration in normoalbumic IDDM patients by 6 mo of aldose reductase inhibition. Diabetes 40, 27-53 1. [24] Tilton, R.G., Chang, K., Pughese, G. et al. Prevention of diabetic rats by aldose reductase inhibitors. Diabetes 38, 1258-1270. [25] Beyer-Mears, A., Cruz, E., Edelist, T. and Varagiannis,
147-154
E. (1986) Diminished proteinuria in diabetes mellitus by Sorbinil, an aldose reductase inhibitor. Pharmacology 32, 52-60. (261 Beyer-Mears, A., Murray, F.T., Del Val M., Cruz, E. and Sciadini, M. (1988) Reversal of proteinuria by Sorbinil, an aldose reductase inhibitor in spontaneously diabetic (BB) rats. Pharmacology 36, 112-120. [27] Jenings, P.E., Nightingale, S., Le Guen, C. et al. (1990) Prolonged aldose reductase inhibition in chronic peripheral diabetic neuropathy: effects on microangiopathy. Diabetic Med. I, 63-68. [28] McCaleb, M.L., Serdy, J., Ackerman, D.M. and Dvornik, D. (1988) Prevention of urinary albumin excretion in 6 month streptozotocin-diabetic rats with the aldose reductase. inhibitor Tolrestat. J. Diab. Complications 2, 16-18. [29] Kaneda, K., Iwao, J., Sakata, N. and Takebayashi, S. (1992) Correlation between mitochondrial enlargement in renal proximal tubules and microalbuminuria in rats with early stmptozotocin-induced diabetes. Acta Pathol. Jpn. 42, 855-860. [30] Kato, C., Abe, T., Murakami, T. and Kobayashi, A. (1987) Glycogen deposition of renal tubular ceils in strep tozotocin induced diabetic rats. J. Clin. Electron Microsc. 20, 5-6. 1311 Kato, C., Abe, T., Sawada, S. and Sugiura, Y. (1986) An electron microscopic study of the tubular lesions in experimental diabetic rats. J. Clin. Electron Microsc. 19, 5-6. 1321 Ayo, S.H., Radnic, R.A., Garoni, J.A., Glass, W.F. and Kreisberg, J.I. (1990) High glucose. causes an increse in extracellular matrix proteins in cultured mesangial cells. Am. J. Pathol. 136, 1339-1348. [33] Haneda, M., Kikkawa, R., Horide, N. et al. (1991) Glucose enhances type IV collagen production in cultured rat glomerular mesangial cells. Diabetlogia 34, 198-200.