Combined effect of probucol and insulin on renal damage in diabetic rats fed a high cholesterol diet

gy 548 (2006) 174–180www.elsevier.com/locate/ejphar

European Journal of Pharmacolo

Combined effect of probucol and insulin on renal damagein diabetic rats fed a high cholesterol diet

Masumi Yoshida a, Hitoshi Kimura a, Kouhei Kyuki a, Mikio Ito b,⁎

a Pharmacology Division, Nihon Bioresearch Inc.; 6-104 Mazima, Hukuju-cho, Hashima, Gifu 501-6251, Japanb Laboratory of Analytical Pharmacology, Faculty of Pharmacy, Meijo University; 150 Yagotoyama, Tenpaku-ku, Nagoya 468-8503, Japan

Received 17 November 2005; received in revised form 7 July 2006; accepted 11 July 2006Available online 22 August 2006

Abstract

We investigated the effects of long-term treatment with probucol, a hypolipidemic agent with antioxidative action, insulin, or their combination onrenal damage in streptozotocin-induced diabetic rats fed a high cholesterol diet. Increases in urinary albumin and lipid peroxide excretions wereobserved in these diabetic rats, when both urinary parameters were measured at 8 and 15 weeks after streptozotocin administration. Daily treatmentwith probucol, insulin, or their combinationmarkedly suppressed the increase in the 24 h urinary albumin and lipid peroxide excretions. Furthermore,glycogen degeneration of distal tubules, fatty degeneration of glomerular endothelium, and hypertrophy of glomeruli and mesangium were observedin the kidneys of the diabetic animals, when histopathological evaluation was performed at 4, 8 and 15 weeks (glomerular andmesangial hypertrophywas observed only at 15 weeks). Combined probucol and insulin treatment was the most effective in suppressing these renal histopathologicalchanges. These results indicate that combined treatment with probucol and insulin is useful in preventing the progression of renal damage in diabeticrats. The possible mechanisms for the preventive effect of this combined treatment will be discussed.© 2006 Elsevier B.V. All rights reserved.

Keywords: Diabetes; Renal injury; Hyperglycemia; Hyperlipidemia; Lipid peroxide

1. Introduction

Diabetic nephropathy is characterized by persistent protein-uria, hypertension and declining renal function. Microalbumi-nuria is detected at the early stage of diabetic nephropathy.Approximately 30–40% of type 1 diabetic patients experiencerenal complications which are a major cause of mortality (Greenand Hougaard, 1984; Rossing et al., 1995). Hyperlipidemia inaddition to hyperglycemia has been frequently observed ininsulin-dependent(Nikkila and Kekki, 1973; Nikkila andHormila, 1978) and non-insulin-dependent diabetic patients(Reaven and Greenfield, 1981). Likewise, hyperlipidemia wasalso observed in experimental insulin-dependent (Ito et al.,2001) and non-insulin-dependent diabetic animal models (Manet al., 1977). Hyperlipidemia, in addition to hyperglycemia, hasbeen thought to be a major risk factor for the development and

⁎ Corresponding author. Tel.: +81 52 832 1781; fax: +81 52 834 8090.E-mail address: [email protected] (M. Ito).

0014-2999/$ - see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.ejphar.2006.07.053

progression of diabetic complications such as nephropathy(Shestakova et al., 1993; Hadjadj et al., 2004).

Recently, oxidative stress has been showed to play an im-portant role in the development and progression of diabeticnephropathy (Ha and Lee, 2001). Under hyperglycemicconditions, oxygen-derived free radicals are produced mainlythrough a glycation reaction (Sakurai and Tsuchiya, 1988; Huntet al., 1991). Scheuer et al. (2000) have shown that hyperlip-idemia can aggravate glomerulosclerosis and chronic tubuloin-terstinal damage in kidneys, and that oxidative stress contributesto the deleterious effects of hyperlipidemia on renal damage.

Probucol has been widely used clinically for the prevention ofthe progression of atherosclerosis, because this agent acts as apotent antioxidant (Regnstrom et al., 1990; Baumastark et al.,1992) in addition to having a lipid-lowering action (Schwarts,1988). Pretreatment with probucol, for example, reportedlyprevents the in vivo diabetogenic actions of alloxan in rats,probably by means of its strong free radical scavenger action(Matsushita et al., 1989). Probucol has also been shown toprevent the beta-cell function in diabetic C57BL/KsJ-db/db

mailto:[email protected]

http://dx.doi.org/10.1016/j.ejphar.2006.07.053

175M. Yoshida et al. / European Journal of Pharmacology 548 (2006) 174–180

mice through reduction of oxidative stress (Gorogawa et al.,2002). On the other hand, insulin has been widely used for thetreatment of type 1 (insulin-dependent) diabetes mellitus as ahormone supplement therapy. From these findings, it seemsdistinctly possible that insulin in combination with probucolmay exert a more beneficial effect than insulin alone on diabeticnephropathy.

Recently, we demonstrated that a long-term high cholesteroldiet for streptozotocin-induced diabetic rats not only precipi-tated the onset of cataracts but also increased the incidence ofthis complication, with along the mortality rate (Yoshida et al.,2005). Furthermore, in this study, we demonstrated that long-term combined treatment with probucol and insulin synergis-tically prevents the development and progression of cataracts instreptozotocin-induced diabetic rats fed a high cholesterol diet.

In the present preliminary experiment, we clarified that along-term high cholesterol diet aggravated streptozotocin-in-duced renal damage in rats with increased urinary lipid peroxideexcretion, an index of lipid peroxidation in kidneys. Therefore,in the present study, in order to clarify whether or not probucol, alipid-lowering agent with an antioxidant action, acts as a sup-plement to insulin therapy for diabetic nephropathy, we inves-tigated the effects of long-term treatment with probucol, insulin,or a combination of the two, on renal damage in streptozotocin-induced diabetic rats fed a high cholesterol diet.

2. Materials and Methods

2.1. Animals

Seven-week-old male Sprague–Dawley rats (Charles River,Hino) were housed in an isolator caging system in an air-conditioned animal room at 23±1 °C. All experimental pro-cedures described were approved by the Committee for Ethicsand Animal Experimentation of Nihon Bioresearch Inc.

2.2. Drugs

The drugs used were probucol (Sinlestal®, Daiichi Pharma-ceutical Co. Ltd., Tokyo, Japan) and insulin Isophan Insulin(Aqueous Suspension), NPHiszilin®, Takeda PharmaceuticalCo., Ltd., Osaka).

2.3. Induction of diabetes and experimental procedure

Rats were fasted for 20 h before the experiment and thendivided into 5 groups. These rats were fed a standard diet beforethe experiment and a standard diet, high cholesterol diet, orprobucol-supplemented high cholesterol diet after starting theexperiment. The standard diet (pellets) consisted of 20% casein,63.2% sucrose, 10% corn oil, 2% agar, 0.8% vitamin mixtureand 4% salt mixture. The high cholesterol diet (pellets) consistedof the standard diet with 1% cholesterol (wt/wt) and 0.5% cholicacid (wt/wt) in place of an equal amount of sucrose. Further-more, the probucol-supplemented diet contained 1% probucol(wt/wt) in the high cholesterol diet. These diets were producedby Oriental Yeast Co., Ltd. (Tokyo). To induce insulin-deficient

diabetes, the fasted rats in four groups were injected intrave-nously with streptozotocin (Sigma Chemical Co., Ltd., St Louis,MO, U.S.A.) dissolved in a citrate buffer (0.1Mwith pH 4.5) at adose of 40mg/kg. The fasted rats in one group were injected withthe equivalent volume of citrate buffer as the non-diabetic group.The experimental groups were as follows (Table 1): (1) Standarddiet-fed normal (SD-fedNor) group, (2) high cholesterol diet-feddiabetic (HCD-fed Diab) group, (3) probucol-supplementedhigh cholesterol diet-fed diabetic (Prob-suppl HCD-fed Diab)group, (4) high cholesterol diet-fed diabetic+ insulin (HCD-fedDiab+Ins) group and (5) probucol-supplemented high choles-terol diet-fed diabetic+ insulin (Prob-suppl HCD-fed Diab+ Ins)group. The rats in the HCD-fed Diab+Ins and Prob-suppl HCD-fed Diab+Ins groups received a daily s.c. injection (2 U/rat) ofinsulin at 10:00 a.m. from the day after the start of the exper-iment. Blood samples for the determination of serum glucose,total cholesterol and triglyceride contents were taken from thecavernous sinus with a capillary under light ether anesthesia at9:00 a.m. 0 (before the injection of streptozotocin or citratebuffer), and 1, 2, 4, 8, 12 and 15 weeks after the start of theexperiment (Yoshida et al., 2005). The serum biochemicalparameters were determined using an Automated ChemistryAnalyzer (AU 400, Olympus Optical Co., Ltd., Tokyo, Japan).At 0, 8 and 15 weeks after the start of the experiment, theseanimals were kept in individual metabolic cages, and the 24 hurine samples were collected for the determination of urinaryalbumin and lipid peroxide contents. The urinary albumin wasdetermined by immunoturbidity method using a kit for thedetermination of microalbumin (Bayer Diagnostics, Germany).Urinary lipid peroxide was determined as the concentration ofmalondialdehyde, a secondary product of lipid peroxidation, byadding thiobarbituric acid in accordance with the method of Yagi(1976). At 4, 8 and 15 weeks, animals picked from each groupwere sacrificed by decapitation under light ether anesthesia andtheir kidneys removed for histopathological evaluation.

2.4. Histopathological evaluation of kidneys

Kidneys removed for histopathological evaluation were fixedin 10% neutral buffered formalin. Paraffin sections (2 μm) werestained with hematoxylin and eosin (HE) or periodic acid-Schiffreaction (PAS) for light microscopy. Glycogen degeneration ofdistal tubules (HE stain), fatty degeneration of glomerular endo-thelium (HE stain) and glomerular and mesangial hypertrophy(PAS stain) were observed in 10 distal tubules or glomeruli persection for histopathological evaluation of the kidneys. Thedegree of the degeneration of the tubules and glomerular endo-thelium was semiquantitatively scored from 0 to 3 as follows:Normal, score 0; mild, score 1; moderate, score 2; severe, score3. Representativemicrographs of the degrees of the degenerationare shown in Fig. 1. The average score for each group wascalculated for the histopathological evaluation. Furthermore,glomerular and mesangial areas were measured using imageanalysis software (Win ROOF) (Mitani Corporation, Fukui).The PAS-positive stained portion of each glomerulus wasmeasured as the mesangial area. Histopathological analysis wasperformed by two independent observers.

Table 1Effects of a long-term treatment with probucol, insulin or their combination on serum glucose, total cholesterol and triglyceride levels in streptozotocin-induceddiabetic rats fed a high cholesterol diet

Group A. Serum glucose (mg/dl)

0 1 2 4 8 12 15 (weeks)

SD-fed Nor 140±5 133±2 122±2 129±4 138±3 131±3 153±3(n=10) (n=10) (n=10) (n=10) (n=10) (n=10) (n=10)

HCD-fed Diab 131±2 584±18aa 663±25aa 570±107aa 634±42aa 540±51aa 491±27aa

(n=10) (n=10) (n=10) (n=8) (n=6) (n=5) (n=3)Prob-suppl HCD-fed Diab 130±4 582±28 595±24b 566±32 624±33 569±30 474±20

(n=10) (n=10) (n=10) (n=10) (n=8) (n=8) (n=7)HCD-fed Diab+Ins 124±2 546±21 456±21bb 406±17 308±21bb 277±14bb 298±19bb

(n=10) (n=10) (n=10) (n=10) (n=10) (n=8) (n=8)Prob-suppl HCD-fed Diab + Ins 133±4 534±18 429±21bb,cc 381±23b,c 272±19bb,cc 305±22bb,cc 278±22bb,cc

(n=10) (n=10) (n=10) (n=10) (n=10) (n=10) (n=9)

B. Total cholesterol (mg/dl)



(n=10) (n=10) (n=10) (n=8) (n=6) (n=5) (n=3)Prob-suppl HCD-fed Diab 64±4 1571±359bb 2684±641bb 4012±780bb 3684±1019bb 3466±932bb 2387±925b

(n=10) (n=10) (n=10) (n=10) (n=8) (n=8) (n=7)HCD-fed Diab+Ins 60±3 559±69bb 634±77bb 1148±190bb 807±163bb 944±121bb 1084±186bb

(n=10) (n=10) (n=10) (n=10) (n=10) (n=8) (n=8)Prob-suppl HCD-fed Diab + Ins 66±4 584±56bb,c 886±145bb,cc 1431±153bb,cc 1189±202bb,cc 1758±284bb,c 1306±249bb

(n=10) (n=10) (n=10) (n=10) (n=10) (n=10) (n=9)

C. Triglyceride levels (mg/dl)



(n=10) (n=10) (n=10) (n=8) (n=6) (n=5) (n=3)Prob-suppl HCD-fed Diab 95±15 606±210 484±129bb 704±225bb 969±490bb 1850±290bb 1149±687

(n=10) (n=10) (n=10) (n=10) (n=8) (n=8) (n=7)HCD-fed Diab+Ins 92±13 531±105 353±73bb 311±57bb 131±19bb 218±40bb 248±38

(n=10) (n=10) (n=10) (n=10) (n=10) (n=8) (n=8)Prob-suppl HCD-fed Diab + Ins 82±11 663±127 389±60bb 360±56bb 144±21bb 234±33bb,cc 220±32

(n=10) (n=10) (n=10) (n=10) (n=10) (n=10) (n=9)

SD-fed Nor: Standard diet-fed normal group, HCD-fed Diab: High cholesterol diet-fed diabetic group, Prob-suppl HCD-fed Diab: Probucol-supplemented highcholesterol diet-fed diabetic group, HCD-fed Diab+Ins: High cholesterol diet-fed diabetic+ insulin group and Prob-suppl HCD-fed Diab+Ins: Probucol-supplemented high cholesterol diet-fed diabetic+ insulin group. Each value represents the mean±S.E.M. n indicates the number of rats used. aaPb0.01, as comparedwith the SD-fed Nor group (Student's t-test). bPb0.05, bbPb0.01, as compared with the HCD-fed Diab group (Duncan's test). cPb0.05, ccPb0.01, as compared withthe Prob-suppl HCD-fed Diab group (Duncan's test).

176 M. Yoshida et al. / European Journal of Pharmacology 548 (2006) 174–180

2.5. Statistical analysis

The data are expressed as the mean±S.E.M. The data wereanalyzed for statistical significance by Duncan's test for mul-tiple comparison or Student's t-test for comparison betweentwo groups.

3. Results

3.1. Non-fasting serum glucose, total cholesterol and triglycer-ide levels

In our previous study (Yoshida et al., 2005), the serumglucose levels in the Prob-suppl HCD-fed Diab were notdifferent from those in the HCD-fed Diab (control) group for theexperimental period (Table 1). These levels in the HCD-fed

Diab+Ins and Prob-suppl HCD-fed Diab+Ins groups weresignificantly lower than those in the HCD-fed Diab group buthigher than those in the SD-fedNor group throughout 2–15weeks.Thus, blood glucose control by the daily treatment with in-sulin (2U/rat, s.c.) was insufficient. In addition, the lowering effectof combined treatment with probucol and insulin on the glucoselevels was not significantly different from those of insulin alone.

The number of rats in all groups was 10 (n=10) at 0–2 weeks(Table 1). However, the number of rats in the HCD-fed Diab(control) group at 4, 8, 12 and 15 weeks was 8, 6, 5 and 3,respectively. The time course-related decreases in the number ofthe animals were due to increases in the number of deadanimals. Specifically, the mortality rate of rats in the controlgroup at 4, 8, 12 and 15 weeks was 20%, 40%, 50%, and 70%,respectively. On the other hand, the rate of rats in the Prob-supplHCD-fed Diab, HCD-fed Diab+Ins, and Prob-suppl HCD-fed

Fig. 1. Representative photomicrographs of kidneys from normal andstreptozotocin-induced diabetic rats with nephropathy. A. Glycogen degener-ation of distal tubules (arrow). HE stain×200. B. Fatty degeneration ofendothelium (arrow). HE stain×400. The degree of histopathological change inA and B in kidneys was scored from 0 to 3. A. 0: Normal; 1: Mild (vacuolardegeneration in some distal tubules); 2: Moderate (distal tubules with sporadicvacuolar degeneration); 3: Severe (Many distal tubules with vacuolardegeneration). B. 0: Normal; 1: Mild (a few small-sized lipid droplets inglomerulus); 2: Moderate (sporadic small- or middle-sized lipid droplets inglomerulus); 3: Severe (many large-sized lipid droplets in glomerulus).

Fig. 2. Effects of long-term treatment with probucol, insulin, or their combinationon (A) urinary albumin and (B) lipid peroxide excretions in streptozotocin-induced diabetic rats fed a high cholesterol diet. SD-fed Nor: Standard diet-fednormal group, HCD-fed Diab: high cholesterol diet-fed diabetic group, prob-suppl HCD-fed Diab: probucol-supplemented high cholesterol diet-fed diabeticgroup, HCD-fed Diab+Ins: high cholesterol diet-fed diabetic+ insulin group andProb-suppl HCD-fed Diab+Ins: probucol-supplemented high cholesterol diet-fed diabetic+ insulin group. Each value represents the mean±S.E.M. of 5 rats.**Pb0.01, as compared with the SD-fed Nor group (Student's t-test). ##Pb0.01,as compared with the HCD-fed Diab group (Duncan's test). $Pb0.05, $$Pb0.01,as compared with the probucol-suppl HCD-fed Diab (Duncan's test). †Pb0.05 ascompared with the HCD-fed Diab+Ins group (Duncan's test).


Diab+Ins groups at 15 weeks was 30%, 20% and 10%,respectively. Thus, combined treatment with both agents wasthe most effective at decreasing the death rate.

The non-fasting serum total cholesterol levels in the Prob-suppl HCD-fed Diab, HCD-fed Diab+Ins, and Prob-suppl HCD-fed Diab+Ins groups were significantly lower than those of theHCD-fed Diab (control) group throughout the observation periodof 1–15 weeks (Table 1B) (Yoshida et al., 2005). Thus, it is clearthat daily treatment with probucol, insulin or the combination ofboth agents markedly suppressed hypercholesterolemia in theHCD-fed Diab rats. However, the lowering effect of combinedtreatment with both agents on the total cholesterol levels was notsignificantly different from those of either agent alone.

The non-fasting serum triglyceride levels in the Prob-supplHCD-fed Diab, HCD-fed Diab+Ins and Prob-suppl HCD-fedDiab+ Ins groups were significantly lower than those in theHCD-fed Diab (control) group throughout 2–12 weeks afterstreptozotocin administration. However, the lowering effect ofcombined treatment with both agents on the triglyceride levelswas not significantly different from those of either agent alone.

3.2. Urinary albumin and lipid peroxide excretions

The 24 h urinary albumin excretion in the HCD-fed Diabgroup at 8 and 15 weeks was 13.1 and 15.2 times, respectively,higher than that in the SD-fed Nor group (both weeks: 0.10±0.05 mg/24 h) (Fig. 2A). The urinary albumin excretion in theProb-suppl HCD-fed Diab, HCD-fed Diab+Ins and Prob HCD-fed Diab+Ins groups was markedly decreased, comparedwith that of the respective HCD-fed Diab (control) group at8 (% decrease: 69%, 76%, and 91%, respectively) and 15 weeks(% decrease: 61%, 82%, and 88%). Combined probucol andinsulin treatment was much more effective in decreasing urinaryalbumin excretion than probucol alone at 8 and 15 weeks.

The 24 h urinary lipid peroxide excretion in the HCD-fed Diabgroup at 8 and 15 weeks was 4.9 and 6.8 times, respectively,higher than that in the SD-fed Nor group (8 weeks: 79.4±4.0 nmol/24 h; 15 weeks: 97.4±4.5 nmol/24 h) (Fig. 2B). The24 h urinary lipid peroxide excretion in the Prob-suppl HCD-fedDiab, HCD-fed Diab+Ins, and Prob-suppl HCD-fed Diab+Insgroups was significantly decreased, compared with that in therespective HCD-fed Diab (control) group at 8 (% decrease: 46%,61%, and 74%, respectively) and 15 weeks (% decrease: 54%,60%, and 77%, respectively). The combined probucol and insulintreatment was significantly more effective at decreasing theurinary lipid peroxide excretion than probucol alone at 8 weeksand treatment with either agent alone at 15 weeks.

3.3. Histopathology of kidneys

When histopathological changes in the kidneys wereevaluated at 4, 8 and 15 weeks after streptozotocin administra-tion, glycogen degeneration in the distal tubules (Fig. 3A) andfatty degeneration in the glomerular endothelium (Fig. 3B) wereobserved at all weeks in the kidneys of animals evaluated in theHCD-fed Diab group. The distal tubule glycogen degenerationscores for the Prob-suppl HCD-fed Diab, HCD-fed Diab+Insand Prob-suppl HCD-fed Diab+Ins groups were significantly


decreased, compared with those for the HCD-fed Diab (control)group at 4 (% decrease: 64%, 79%, and 86%, respectively),8 (% decrease: 47%, 73%, and 87%, respectively) and 15 weeks(% decrease: 47%, 80% and 90%, respectively). The combinedprobucol and insulin treatment was significantly more effectiveat suppressing glycogen degeneration in distal tubules thaneither probucol or insulin alone at 8 and 15 weeks.

Glomerular endothelium fatty degeneration scores for theHCD-fed Diab+Ins and Prob-suppl HCD-fed Diab+Ins groupswere significantly decreased, compared with those in the HCD-fed Diab (control) group at all weeks measured (% decrease: 67%and 67%, respectively, at 4 weeks; 62% and 92%, respectively, at8 weeks; 86% and 93%, respectively, at 15 weeks). The fattydegeneration score for the Prob-suppl HCD-fed Diab group wassignificantly decreased, compared with that in the control group,at only 8 weeks (% decrease: 46%).

The combined probucol and insulin treatment was significant-ly more effective at suppressing fatty degeneration in theglomerular endothelium than either agent alone at 4 and 8 weeks.

When the glomerular and mesangial areas were measured at15 weeks after streptozotocin administration, the glomerular(Fig. 4A) and mesangial areas (Fig. 4B) in the HCD-fed Diab

Fig. 3. Effects of long-term treatment with probucol, insulin, or theircombination on (A) glycogen degeneration in distal tubules and (B) fattydegeneration in glomerular endothelium in streptozotocin-induced diabetic ratsfed a high cholesterol diet. Designation of groups as in Fig. 2. Each valuerepresents the mean±S.E.M. of 5 rats. The score of the SD-fed Nor group (@)was 0. ⁎⁎Pb0.01, as compared with the SD-fed Nor group (Student's t-test).#Pb0.05, ##Pb0.01, as compared with the HCD-fed Diab group (Duncan'stest). $$Pb0.01, as compared with the probucol-suppl HCD-fed Diab group(Duncan's test). †Pb0.05, as compared with the HCD-fed Diab+Ins group(Duncan's test).

Fig. 4. Effects of long-term treatment with probucol, insulin, or their combinationon (A) glomerular and (B)mesangial areas in streptozotocin-induced diabetic ratsfed a high cholesterol diet. Designation of groups as in Fig. 2. Each valuerepresents the mean±S.E.M. of 5 rats. ⁎⁎Pb0.01, as compared with the SD-fedNor group (Student's t-test). ##Pb0.01, as compared with the HCD-fed Diabgroup (Duncan's test). $$Pb0.01, as comparedwith the probucol-suppl HCD-fedDiab group (Duncan's test). ††Pb0.01, as compared with the HCD-fed Diab+Insgroup (Duncan's test).

group were 2.0 and 4.4 times greater, respectively, than those inthe SD-fed Nor group. The glomerular area in the Prob-supplHCD-fed Diab, HCD-fed Diab+Ins and Prob-suppl HCD-fedDiab+Ins groupswas significantly decreased, comparedwith thatin the HCD-fed Diab (control) group (% decrease: 19%, 19%, and35%, respectively). The combined probucol and insulin treatmentthus was significantly more effective at suppressing the increasein glomerular area than either agent alone.

The mesangial area in the Prob-suppl HCD-fed Diab, HCD-fed Diab+Ins, and Prob-suppl HCD-fed Diab+Ins groups wassignificantly decreased, compared with that in the HCD-fedDiab (control) group (% decrease: 29%, 25%, and 50%, res-pectively). Likewise, the combined treatment with both agentswas significantly more effective at suppressing the increase inmesangial area than either agent alone. Thus, the combinedtreatment with both agents can be said to be the most effective insuppressing renal pathological changes.

4. Discussion

The present study demonstrates that long-term combinedtreatment with probucol, a hypolipidemic agent with antioxida-tive action, and insulin synergistically prevents the developmentand progression of renal damage in streptozotocin-induceddiabetic rats fed a high cholesterol diet.


In the present experiment, an increase in urinary albuminexcretion was observed in streptozotocin-induced diabetic ratsfed a high cholesterol diet, compared with normal rats fed astandard diet at 8 and 15 weeks after streptozotocin administra-tion. Furthermore, glycogen degeneration in the distal tubules,fatty degeneration in the glomerular endothelium, and glomer-ular and mesangial hypertrophy were observed in the diabeticrats fed a high cholesterol diet at 4, 8 and 15 weeks (glomerularand mesangial hypertrophy were observed only at 15 weeks).The severity of the urinary albumin excretion and these histopa-thological changes in the kidneys was greater in the diabetic ratsfed a high cholesterol diet than in the diabetic rats fed a standarddiet (data not shown). However, massive albuminuria and severehistopathological changes such as glomerulosclerosis were notobserved in streptozotocin-induced diabetic rats fed a highcholesterol diet even at 15 weeks after streptozotocin adminis-tration. Thus, the degree of the changes in urinary albuminexcretion and histopathology in the kidneys of diabetic rats fed ahigh cholesterol diet at 15 weeks resembled nephropathy at anearly stage in diabetic patients. It has been reported that, withstreptozotocin-induced diabetic spontaneously-hypertensiverats, glomerulosclerosis with amarked decrease in renal functionis observed at 10 weeks after streptozotocin injection (Fabriset al., 1999). Therefore, it is conceivable from our results and theabove finding that hypertension may be the major risk factor forthe development and progression of diabetic nephropathy.

In the present experiment, we used doses of insulin (2 U/rat/day, s.c.) and probucol (a high cholesterol diet supplementedwith 1% W/W probucol) that were insufficient for controllingblood glucose or lipids, because it would be difficult to evaluateany synergistic effect of combined probucol and insulin admin-istration, if their respective doses were both sufficient (Table 1).As mentioned above, we administered probucol mixed with diet.

The average daily food intake in the Prob-suppl HCD-fedDiab and Prob-suppl HCD-fed Diab+Ins rats was 39–61 gand 30–41 g, respectively, throughout the observation period of1–15 weeks after streptozotocin administration (Yoshida et al.,2005). If a diabetic rat weighing 270 g intakes 40 g of the foodpellets containing 1% W/W probucol per 24 h, the oral dosageof this agent is 1480 mg/kg/day. As a result, treatment withprobucol, insulin or their combination markedly decreased uri-nary albumin excretion and suppressed all histopathologicalchanges in the kidneys mentioned above in the diabetic rats feda high cholesterol diet. Combined probucol and insulin treat-ment was the most effective at suppressing urinary albuminexcretion and renal pathological changes. Recently, Agardhet al. (2002) reported that long-term (6 months) treatment withprobucol (1% w/w in food pellets) prevented changes in renalmorphology in streptozotocin-induced diabetic rats, but in-creased urinary albumin excretion in contrast with our results. Itseems natural to conclude that the preventive effect of probucolon renal damage in diabetic rats results in the reduction inurinary albumin excretion. Therefore, the discrepant result isdifficult to understand.

In our previous study (Yoshida et al., 2005), treatment withinsulin and the combination of probucol and insulin markedlyimproved hyperglycemia, although probucol alone had no

effect. In addition, treatment with probucol, insulin, and theircombination markedly improved hypercholesterolemia andhypertriglyceridemia. In this experiment, we determined serumtotal cholesterol content, but did not identify separate amounts oflow-density lipoprotein (LDL) cholesterol and high-densitylipoprotein (HDL) cholesterol. Tsutsumi et al. (1999) reportedthat a marked increase in serum total cholesterol in streptozo-tocin-induced diabetic rats fed a cholesterol-rich diet was due tothe marked increase in non-HDL cholesterol such as LDL,because non-HDL cholesterol:HDL cholesterol was 93:7. It isknown that probucol reduces both LDL and HDL cholesterollevels. From the above report, the effect of probucol on serumHDL cholesterol seems not to be significant. Probucol has beenshown to increase cholesteryl ester transfer protein activity inhypercholesteroalemic patients (Franceschini et al., 1991).However, the lowering effects of combined treatment withprobucol and insulin on serum glucose, total cholesterol andtriglyceride levels in diabetic rats fed a high cholesterol diet werenot significantly different from those of insulin alone. Theseresults suggest that a beneficial effect of combined treatmentwith both agents on diabetic renal damage may not be due totheir direct blood glucose or lipid-lowering actions.

As described in the Introduction, oxygen-derived free rad-icals and lipid peroxide, which are easily formed in the diabeticstate, are thought to play important roles in the development ofdiabetic complications.

In the present experiment, we measured the 24 h urinary lipidperoxide content as an index of lipid peroxidation in vivo, inkidneys, especially at 8 and 15 weeks after streptozotocin ad-ministration.Marked increase in urinary lipid peroxide excretionwas recognized in the diabetic rats fed a high cholesterol diet atboth weeks. Treatment with probucol, insulin or the combinationof both agents markedly suppressed the increase in urinary lipidperoxide excretion. The combined treatment with both agentswas the most effective at decreasing the lipid peroxide excretion.In addition, there was a clear relationship between the decreasein urinary lipid peroxide excretion and the prevention of urinaryalbumin excretion or renal damage by the combination of pro-bucol and insulin (data not shown). These results stronglysuggest that the preventive effect of the combined treatment withprobucol and insulin on diabetic renal damage may be mainlydue to free radical scavenging actions. Insulin, unlike probucol,does not possess a direct antioxidative action. Therefore, thesuppressive action of insulin on the increase in urinary lipidperoxide excretion of the diabetic rats may be a secondary actiondue to its hypoglycemic action.

In summary, it is concluded from these results that the com-bined treatment with probucol and insulin is useful in preventingthe development and progression of diabetic renal damage. Inaddition, it is valuable to use antioxidants such as probucol as asupplement of insulin therapy for diabetic nephropathy.

References

Agardh, C.-D., Stenram, U., Torffvit, O., Agardh, E., 2002. Effects of inhibitionof glycation and oxidative stress on the development of diabetic nephropathyin rats. J. Diabetes Its Complicat. 16, 395–400.


Baumastark, M.W., Aristegui, R., Zoller, T., Frey, I., Berg, A., Keul, J., 1992.Probucol incorporated into LDL particles in vivo, inhibits generation of lipidperoxides more effectively than endogenous antioxidants alone. Clin.Biochem. 25, 395–397.

Fabris, B., Candido, R., Armini, L., Fischetti, F., Calci, M., Bardelli, M., Fazio,M., Campanacci, L., Carretta, R., 1999. Control of glomerular hyperfiltra-tion and renal hypertrophy by an angiotensin converting enzyme inhibitorprevents the progression of renal damage in hypertensive diabetic rats.J. Hypertens. 17, 1925–1931.

Franceschini, G., Chiesa, G., Sirtori, C.R., 1991. Probucol increases cholesterylester transfer protein activity in hypercholesterolaemic patients. Eur. J. Clin.Invest. 21, 384–388.

Gorogawa, S., Kajimoto, Y., Umayahara, Y., Kaneto, H., Watanabe, H., Kuroda,A., Kawamori, D., Yasuda, T., Matsuhisa, M., Yamasaki, Y., Hori, M., 2002.Probucol preserve pancreatic beta-cell function through reduction ofoxidative stress in type 2 diabetes. Diabetes Res. Clin. Prac. 57, 1–10.

Green, A., Hougaard, P., 1984. Epidemiological studies of diabetes mellitus inDenmark. Diabetologia 26, 190–194.

Hadjadj, S., Duly-Bouhanick, B., Bekherraz, A., Brldoux, F., Gallois, Y.,Mauco, G., Ebran, J., Marre, M., 2004. Serum triglycerides are a predictivefactor for the development and the progression of renal and retinalcomplications in patients with type 1 diabetes. Diabetes Metab. 230, 43–51.

Ha, H., Lee, H.B., 2001. Oxidative stress in diabetic nephropathy: basic andclinical information. Curr. Diabetes Rep. 1, 282–287.

Hunt, J.V., Smith, C.C., Wolff, S.P., 1991. Autoxidative glycosylation andpossible involvement of peroxides and free radicals in LDL modification byglucose. Diabetes 39, 1420–1424.

Ito, M., Kondo, Y., Kakatani, A., Hayashi, K., Naruse, A., 2001. Character-ization of low dose streptozotocin-induced progressive diabetes in mice.Environ. Toxicol. Pharmacol. 9, 71–78.

Matsushita, M., Yoshino, G., Iwai, M., Matsuwa, K., Morita, M., Iwatani, I.,Yoshida, M., Kazumi, T., Baba, S., 1989. Protective effect of probucol onalloxan diabetes in rats. Diabetes Res. Clin. 7, 313–316.

Man, Z.-W., Zhu, M., Noma, Y., Toide, K., Sato, T., Asahi, Y., Hirashima, T.,Mori, S., Kawano, K., Mizuno, A., Sano, T., Shima, K., 1977. Impaired β-cellfunction and deposition of fat droplets in the pancreas as a consequence of

hypertriglyceridemia inOLETF rat, a model of spontaneousNIDDM.Diabetes46, 1718–1724.

Nikkila, E.A., Kekki, M., 1973. Plasma triglyceride transport kinetics indiabetes mellitus. Metabolism 22, 1–22.

Nikkila, E.A., Hormila, P., 1978. Serum lipid and lipoproteins in insulin-treateddiabetes. Demonstration of increased high density lipoprotein concentra-tions. Diabetes 27, 1078–1085.

Reaven, G.M., Greenfield, M.S., 1981. Diabetic hypertriglyceridemia. Evidencefor three clinical syndrome. Diabetes 30 (Suppl. 2), 66–75.

Regnstrom, J., Walldius, D., Carlson, L.A., Nilsson, J., 1990. Effect of probucoltreatment on susceptibility of low density lipoprotein isolated fromhypercholesterolemic patients to became oxidatively modified in vivo.Atherosclerosis 82, 43–51.

Rossing, P., Rossing, K., Jacobsen, P., Parving, H.-H., 1995. Unchangedincidence of diabetic nephropathy in IDDM patients. Diabetes 44, 739–743.

Sakurai, T., Tsuchiya, S., 1988. Superoxide production from nonenzymaticallyglycation protein. FEBS Lett. 236, 406–410.

Scheuer, H., Gwinner, W., Hohbach, J., Grone, E.F., Brandes, R.P., Malle, E.,Olbricht, C.J., Walli, A.K., Grone, H.J., 2000. Oxidative stress inhyperlipidemia-induced renal damage. Am. J. Physiol., Renal Physiol.278, F63–F74.

Schwarts, C.J., 1988. The probucol experience: a review of the past and a look atthe future. Am. J. Cardiol. 62, 1B–5B.

Shestakova, M.V., Dedov, I.I., Neverov, N.I., Severgina, E.S., Diuzhev, T.G.,Ponomarev, A.B., 1993. Hyperlipidemia as a factor in the development andprogress of diabetic nephropathy. Probl. Endokrinol. (Mosk) 39, 7–9.

Tsutsumi, K., Inoue, Y., Yoshida, C., 1999T. Acceleration of development ofdiabetic cataract by hyperlipidemia and low high-density lipoprotein in rats.Biol. Pharm. Bull. 22, 37–41.

Yagi, K., 1976. A simple fluorometric assay for lipoperoxide in blood plasma.Biochem. Med. 15, 212–216.

Yoshida, M., Kimura, H., Kyuki, K., Ito, M., 2005. Combined effect of probucoland insulin on cataracts of diabetic rats fed a high cholesterol diet. Eur.J. Pharmacol. 513, 159–168.

Documents

Combined effect of probucol and insulin on renal damage in diabetic rats fed a high cholesterol diet