Pioglitazone attenuates diabetic nephropathy through an anti-inflammatory mechanism in type 2 diabetic rats

Abstract

Background. Peroxisome proliferator-activated receptors (PPARs) are nuclear transcription factors that play a role in insulin sensitivity, lipid metabolism and inflammation. However, the effects of PPARγ agonist on renal inflammation have not been fully examined in type 2 diabetic nephropathy.

Methods. In the present study, we investigated the effect and molecular mechanism of the PPARγ agonist, pioglitazone, on the progression of diabetic nephropathy in type 2 diabetic rats. Inflammatory markers including NF-κB, MCP-1 and pro-fibrotic cytokines were determined by RT-PCR, western blot, immunohistochemical staining and EMSA. In addition, to evaluate the direct anti-inflammatory effect of PPARγ agonist, we performed an in vitro study using mesangial cells.

Results. Treatment of OLETF rats with pioglitazone improved insulin sensitivity and kidney/body weight, but had a little effect on blood pressure. Pioglitazone treatment markedly reduced urinary albumin and MCP-1 excretion, and ameliorated glomerulosclerosis. In cDNA microarray analysis using renal cortical tissues, several inflammatory and profibrotic genes were significantly down-regulated by pioglitazone including NF-κB, CCL2, TGFβ1, PAI-1 and VEGF. In renal tissues, pioglitazone treatment significantly reduced macrophage infiltration and NF-κB activation in association with a decrease in type IV collagen, PAI-1, and TGFβ1 expression. In cultured mesangial cells, pioglitazone-activated endogenous PPARγ transcriptional activity and abolished high glucose-induced collagen production. In addition, pioglitazone treatment also markedly suppressed high glucose-induced MCP-1 synthesis and NF-κB activation.

Conclusions. These data suggest that pioglitazone not only improves insulin resistance, glycaemic control and lipid profile, but also ameliorates renal injury through an anti-inflammatory mechanism in type 2 diabetic rats.

diabetic nephropathy, inflammation, nuclear factor-kappa B, PPARγ agonist, type 2 diabetes

Introduction

The prevalence of type 2 diabetes is rapidly increasing and diabetic nephropathy is a major cause of end-stage renal disease in many developed countries [ 1 ]. Recent studies have suggested the emerging role of inflammatory processes in the pathogenesis of diabetic nephropathy [ 2–7 ]. Moreover, some studies showed that various anti-inflammatory agents prevented the development of glomerular injury in streptozotocin-induced diabetic rats [ 8 and 9 ]. Taken together, these findings suggest that anti-inflammatory agents might prevent the occurrence and progression of diabetic nephropathy.

Peroxisome proliferator-activated receptors (PPARs) have been shown to be critical factors in regulating diverse biologic process, including lipid metabolism, adipogenesis, insulin sensitivity, immune response, cell growth and differentiation [ 10–13 ]. Ligands for PPAR-γ, such as thiazolidinediones (TZD), have been known to have renoprotective actions in various renal diseases [ 14,15 ], and a number of PPAR-γ agonists have been developed and are currently under evaluation for their efficacy in animal models of diabetic nephropathy [ 16–18 ]. However, most studies have focused on the lipid-lowering effect and metabolic alterations, while little in vivo evidence is currently available about the anti-inflammatory effect of PPARγ agonist in diabetic nephropathy.

In this study, we investigated the effect of the PPAR-γ agonist, pioglitazone, on renal function and renal inflammation during the progression of diabetic nephropathy in type 2 diabetic rats. Additionally, in order to define the anti-inflammatory action of pioglitazone, we conducted an in vitro study examining the effects of pioglitazone on high glucose-induced MCP-1 synthesis, NF-κB activation and collagen synthesis to determine whether MCP-1 synthesis was regulated by pioglitazone in cultured mesangial cells.

Subjects and methods

Animal studies

We used male Otsuka Long-Evans Tokushima Fatty (OLETF) rats supplied by the Tokushima Research Institute (Otsuka Pharmaceutical Co., Tokyo, Japan) as type 2 diabetic models. Age-matched male Long-Evans Tokushima Fatty (LETO) rats served as the genetic control for OLETF rats. The rats at 20 weeks of age were divided into three groups ( n = 6/group). Group 1 consisted of LETO control rats. Group 2 animals were OLETF type 2 diabetes rats and Group 3 consisted of OLETF rats treated with 10 mg/kg of pioglitazone (Takeda Pharmaceutical Co., Osaka, Japan) mixed with rat chow for 6 months. Rats had free access to rat chow and tap water, and were caged individually under a controlled temperature (23 ± 2°C) and humidity (55 ± 5%) environment with an artificial light cycle. Daily amounts of food intake were checked at regular intervals to affirm the dose of the administered drug. At the end of the study period, systolic blood pressure was measured using tail-cuff plethysmography (LE 5001-Pressure Meter, Letica SA, Barcelona, Spain). Plasma glucose levels were measured by a glucose oxidase-based method and creatinine levels were determined by a modified Jaffe method. Plasma insulin levels, serum and urine adiponectin levels were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Linco Research, St. Charles, MO, USA). Serum and urine MCP-1 concentrations were measured by quantitative sandwich ELISA using a commercial kit (Biosource Inc., Camarillo, CA, USA). Serum and urinary βig-h3 levels were measured by ELISA as described previously [ 19 ]. Serum triglyceride and cholesterol analyses were also performed using a GPO-Trinder kit (Sigma, St. Louis, MO, USA). To determine the urinary protein and albumin excretion, individual rats were caged in the metabolic cage and 24-h urine was collected at the end of the study. The urinary protein was measured by nephelometry, and urinary albumin was determined by a competitive ELISA (Shibayagi, Shibukawa, Japan). We sacrificed rats under anaesthesia by intraperitoneal injection of sodium pentobarbital (50 mg/kg). Experiments were conducted in accordance with the Korea University Guide for Laboratory Animals.

cDNA microarray analysis

To identify and study the inflammatory molecules that were changed by pioglitazone treatment, cDNA microarray analysis was performed at the end of the study period. Total RNAs isolated using Trizol reagent from renal cortical tissue were used to synthesize ³² P-labelled cDNAs by reverse transcription. The general array methodology used was based on the procedures of DeRisi et al . [ 20 ]. The clustering of changes in gene expression was determined using a public domain cluster based on a pairwise complete-linkage cluster analysis.

Analysis of gene expression by real-time quantitative PCR in tissues and cells

Total RNA was extracted from renal cortical tissues or experimental cells with the Trizol reagent and further purified using an RNeasy Mini kit (Qiagen, Valencia, CA, USA). Table 1 shows the nucleotide sequence of primers. Quantitative gene expression was performed on a Bio-Rad iCycler system (Bio-Rad, Hercules, CA, USA) using SYBR Green technology. Real-time RT-PCR was performed for 10 min at 50°C and 5 min at 95°C. Subsequently, 45 cycles were applied, consisting of denaturation for 10 s at 95°C and annealing with extension for 30 s at 60°C. The ratio of each gene and β-actin level (relative gene expression number) was calculated by subtracting the threshold cycle number (Ct) of the target gene from that of β-actin and raising 2 to the power of this difference.

Table 1

Biochemical parameters in experimental animals

	LETO	OLETF	OLETF + pioglitazone
Body weight (g)	590 ± 54	595 ± 56	718 ± 52 ‡
Kidney WT/BW (mg/g)	0.26 ± 0.0	0.42 ± 0.05 *	0.25 ± 0.02 §
SBP (mmHg)	103 ± 5	151 ± 6 *	145 ± 4 ‡
Urine volume (mL/day)	11 ± 4	23 ± 3 **	8 ± 2 §
Glucose (mg/dL)	166 ± 3	402 ± 46 *	189 ± 11 §
Insulin (ng/mL)	1.01 ± 0.11	3.10 ± 0.43 *	0.78 ± 0.15 §
HOMA-IR	0.41 ± 0.05	3.06 ± 0.48 *	0.36 ± 0.06 §
Cholesterol (mg/dL)	113 ± 5	147 ± 23	99 ± 5
Triglyceride (mg/dL)	39 ± 6	234 ± 93 **	23 ± 2 §§
Creatinine (mg/dL)	0.65 ± 0.02	0.53 ± 0.04	0.62 ± 0.05
U-protein (mg/mg Cr)	6.0 ± 3.6	16.5 ± 4.2 **	4.1 ± 1.5 §§
U-MA (μg/mg Cr)	30.9 ± 4.0	495.7 ± 8.2 *	132.6 ± 23.2 §

Values are expressed as mean ± SEM. LETO, Long-Evans Tokushima Fatty; OLETF, Otsuka Long-Evans Tokushima Fatty; WT, weight; BW, body weight; SBP, systolic blood pressure; HOMA-IR, homeostasis model assessment-insulin resistance score; U, urine; MA, microalbuminuria. ^*P < 0.01, ^**P < 0.05, LETO versus OLETF; ^‡P < 0.05, LETO versus OLETF + pioglitazone; ^§P < 0.01, ^§§P < 0.05, OLETF versus OLETF + pioglitazone.

Table 2

Genes up- or down-regulated by pioglitazone treatment in renal cortical tissues of OLETF rats

Name of gene (fold-change)	Name of gene (fold-change)
Increased genes
Caspase 3 (+3.33)	Epithelial sodium channel gamma subunit (ENACγ) (+3.16)
Myo-inositol 1 phosphate synthase A1 (+3.16)	Pyrimidinergic receptor P2Y (+3.02)
Solute carrier family 12, member 3 (+2.84)	Ribosomal protein L7 (+2.65)
PPAR gamma (+2.61)	Inducible nitric oxide synthase (+2.52)
Protein tyrosine phosphatase (+2.30)	Stearoyl-coenzyme A desaturase 1 (SCD1) (+2.30)
Tumour protein p53 (+2.23)	Serum/glucocorticoid-regulated kinase (SGK) (2.15)
Insulin-like growth factor 1 (IGF1) (+2.14)
Decreased genes
Complement component 3 receptor 1 (−5.50)	Macrophage migration inhibitory factor (−4.41)
Nuclear factor of kappa light chain gene enhancer 1 (−4.36)	MCP-1 (−4.10)
Transforming growth factor beta 1 (TGFβ1) (−4.10)	Integrin, alpha 2 (−3.96)
Colony stimulating factor 1 (−3.94)	Tumour necrosis receptor superfamily, member 5 (−3.91)
Matrix metalloproteinase-9 (MMP-9) (−3.88)	Fibroblast growth factor 10 (FGF 10) (−3.86)
Smad nuclear interacting protein 1 (−3.70)	Plasminogen activator inhibitor 1 (−3.61)
Mitogen activated protein kinase 11 (−3.58)	Thymosin beta 4 (−3.54)
Caspase recruitment domain family, member 9 (−3.55)	Growth differentiation factor 6 (−3.51)
Insulin-like growth factor 2 receptor (IGF2R) (−3.48)	Chemokine (C-C motif) ligand 4 (−3.47)
Integrin, beta 6 (−3.41)	Epidermal growth factor receptor (EGFR) (−3.33)
Prostaglandin E receptor 1 (−3.30)	Bone morphogenetic protein 15 (BMP 15) (−3.21)
Fibronectin receptor beta (−3.11)	Connective tissue growth factor (CTGF) (−3.0)
TNF receptor-associated protein 1 (−2.97)	Transforming growth factor beta receptor 2 (−2.94)
Protein kinase C, beta 1 (−2.43)	Vascular endothelial growth factor (VEGF) (−2.04)

cDNA microarray analysis was performed at the end of the study period. Total RNAs that were isolated using Trizol reagent from renal cortical tissue were then used to synthesize ³² P-labelled cDNAs by reverse transcription. Cluster analysis was performed on Z-transformed microarray data. The values represent the average fold change compared with control diabetic OLETF rats.

Table 3

Primer sequences for real-time quantitative PCR

Target gene	Primer sequence (5′ to 3′)	Amplicon length (bp)
MCP-1, forward	CACCTGCTGCTACTCATTCACT	415
MCP-1, reverse	GTTCTCTGTCATACTGGTCACTTCT
TGF-β1, forward	ATACAGGGCTTTCGATCCAGG	360
TGF-β1, reverse	GTCCAGGCTCCAAATATAGG
PAI-1, forward	ATGAGATCAGTACTGCGGACGCCATCTTTG	333
PAI-1, reverse	GCACGGAGATGGTGCTACCATCAGACTTGT
Procollagen α1(I), forward	TGGTCCCTCTGGAAATGCTGGACC	297
Procollagen α1(I), reverse	CAGGAGAACCAGGAGAACCAGG
Procollagen α1(IV), forward	TAGGTGTCAGCAATTAGGCAGG	484
Procollagen α1(IV), reverse	CGGACCACTATGCTTGAAGTGA
β-actin, forward	TCATGAGGTAGTCCGTCAGG	460
β-actin, reverse	TCTAGGCACCAAGGTGTG

MCP-1, monocyte chemotactic peptide-1; TGF-β1, transforming growth factor-β1; PAI-1, plasminogen activator inhibitor 1. In this experiment, each sample was run in triplicate and the corresponding non-reverse-transcribed mRNA sample was used as a negative control. The mRNA level of each sample was normalized to that of β-actin mRNA.

Histologic examination and immunohistochemical staining for ED-1, TGF-β ₁ , PAI-1 and type IV collagen

The kidney tissues embedded in paraffin were cut into 4-μm-thick slices, and were stained with periodic acid Schiff and Masson's trichrome. A semi-quantitative score sclerosis index (SI) was used to evaluate the degree of glomerulosclerosis according to the method described by Ma et al . [ 21 ]. Severity of sclerosis for each glomerulus was graded from 0 to 4± as follows: 0, no lesion; 1±, sclerosis of <25% of the glomerulus; 2±, 3± and 4±, sclerosis of 25–50%, >50–75% and >75% of the glomerulus, respectively. All histologic examinations were carried out by renal pathologist in a blinded manner, and more than 80 glomeruli were analysed in kidney sections from each rats. For immunohistochemistry, kidney sections were transferred to a 10 mmol/L citrate buffer solution at a pH of 6.0, and then heated at 80°C for 30 min to retrieve antigens for ED-1 and TGF-β ₁ staining. Alternatively, the sections were transferred to a 1 M EDTA buffer solution (pH 8.0) for type IV collagen, Biogenex Retrievit (pH 8.0) (InnoGenex, San Ramon, CA, USA) for PAI-1 and microwaved for 10–20 min for antigen retrieval prior to type IV collagen and PAI-1 staining. After washing in water, 3.0% H ₂ O ₂ in methanol was applied for 20 min in order to block endogenous peroxidase activity, and the slides were incubated at room temperature for 20 min with normal goat serum (ED-1, TGF-β ₁ and type IV collagen) or 10% powerblock (PAI-1) to prevent nonspecific detection. The slides then were incubated for 1 h with primary antibody against mouse monoclonal anti-monocyte/macrophage (ED-1) (1:100; Serotec, Oxford, UK) and rabbit polyclonal anti-type IV collagen (1:200; Santa Cruz Biotechnology, Santa Cruz, CA, USA), or for 2 h with a rabbit polyclonal anti-TGF-β ₁ antibody (1:200; Santa Cruz Biotechnology). For PAI-1 staining, slides were incubated at 4°C overnight with a rabbit anti-PAI-1 antibody (1:50; American Diagnostica, Stamford, CT, USA). The slides were then incubated with a secondary antibody for 30 min. For coloration, slides were incubated at room temperature with a mixture of 0.05% 3,3′-diaminobenzidine containing 0.01% H ₂ O ₂ , and then counterstained with Mayer's haematoxylin. Negative control sections were stained under identical conditions with a buffer solution substituting for the primary antibody. For evaluation of immunohistochemical staining results, glomerular fields were graded semi-quantitatively as described previously [ 22 ].

Mesangial cell culture

A part of the renal cortex from Sprague–Dawley rats was obtained immediately after surgical nephrectomy, and this was grown in Dulbecco's modified Eagle's medium supplemented with 17% foetal calf serum, 100 μg/ml of penicillin/streptomycin, 1% HEPES, 2 g sodium bicarbonate and 2 mM l -glutamine. To evaluate the effect of pioglitazone on MCP-1 and collagen synthesis under a high glucose condition (30 mM), sub-confluent MCs were serum-starved for 24 h, and then pioglitazone was administered at a final concentration of 10 μM to the culture media containing 30 mM glucose. Total soluble collagen was measured in culture supernatants by the Sircol™ soluble collagen assay kit (Biocolor, Belfast, UK). All experimental groups were cultured in triplicate and harvested at 24 h for extraction of total RNA and protein.

Transient transfection and luciferase reporter activity assay

To determine the endogenous PPAR transcriptional activity and the effect of pioglitazone on its activity, PPAR reporter activity was measured in cultured MCs. A PPAR reporter plasmid containing firefly luciferase was linked to three PPRE consensus sequences, which was provided by Dr Raymond N Dubois (Vanderbilt University, Nashville, TN, USA) [ 23 ]. Cells were transfected with 1 μg of PPRE reporter plasmid and 1 μg of plasmid containing Renilla luciferase driven by a TK promoter for 24 h using Superfect (Qiagen, Valencia, CA, USA). Different concentrations of pioglitazone were added at final concentrations of 0.1 μM, 1.0 μM and 10 μM. After 24 h, the luciferase activity was determined using the dual luciferase assay system (Promega Corp., Madison, WI, USA). To control for differences in transfection efficiency from well to well, a plasmid containing Renilla luciferase driven by a TK promoter was included in each transfection.

Extraction of nuclear proteins and western blotting

Nuclear protein from tissue and cells were extracted using a commercial nuclear extraction kit (Active Motif, Carlsbad, CA, USA). Twenty micrograms of protein was electrophoresed on a 10% SDS–PAGE mini-gel under denaturing conditions. The proteins were transferred onto a polyvinylidene difluoride membrane (Immobilon-P; Millipore, Bedford, MA, USA), and the membranes were hybridized with a rabbit polyclonal anti-p65 antibody (Santa Cruz Biotechnology) diluted 1:1000 in blocking buffer overnight at 4°C. The filter was then incubated with a horseradish peroxidase-conjugated secondary antibody, diluted 1:1000, for 60 min at room temperature. The detection of specific signals was performed using an ECL method.

Electrophoretic mobility shift assay (EMSA)

Nuclear proteins were extracted from renal cortical tissues or experimental cells using a commercial nuclear extraction kit (Active Motif, Carlsbad, CA, USA). Briefly, nuclear extracts (5 μg of protein) were preincubated with poly(dl-dC)zpoly(dl-dC) (2 μg), dithiothreitol (0.3 mM) and a reaction buffer (12 mM Tris, pH 7.9, 2 mM MgCl ₂ , 60 mM KCl, 0.12 mM EDTA, and 12% glycerol) for 30 min at 4°C. ³² P-labelled double-stranded oligonucleotide probes for NF-κB containing a consensus NF-κB sequence (5′-AGTTGAGGGGACTTTCCCAGGC-3′, Promega, WI, USA) were then added, and the reaction mixtures were incubated for 10 min at 37°C. The reaction mixtures were then separated on a 4% non-denaturing polyacrylamide gel at 200 V for 2 h. The gel was dried and autoradiographed. The specificity of the complex was checked using an anti-p65 antibody (Santa Cruz Biotechnology). NF-κB activity was quantified by densitometric analysis with NIH image-analysis software (Version 1.61).

Statistical analysis

We used non-parametric analysis due to the relatively few samples present. Results were expressed as mean ± SEM. A Kruskall–Wallis test was used for comparison of more than two groups, followed by a Mann–Whitney U -test for comparison using a microcomputer-assisted program with SPSS for Windows 10.0 (Spss Inc., Chicago, IL, USA). A value of P < 0.05 was considered to be statistically significant.

Results

Biochemical parameters in experimental animals

Table 2 shows the various biochemical results for each group. Treatment of OLETF rats with pioglitazone for 6 months induced an increase of body weight. Systolic blood pressure was greater in OLETF rats than in LETO rats, and pioglitazone treatment had no significant effect on systolic blood pressure. Plasma insulin, triglyceride levels and HOMA-IR were markedly decreased in the pioglitazone group compared with the diabetic group. Urinary protein and albumin excretion in the OLETF rats was significantly higher than in the LETO rats. Lastly, pioglitazone treatment significantly reduced urinary protein and albumin excretion.

Gene expression profiles in OLETF rats and pioglitazone-treated rats

The expression levels of 41 cDNAs, which represent 3.55% of the total DNA elements on the array, were changed by more than twofold in the pioglitazone treatment group. The changes ranged from −5.50-fold to +3.33-fold. As shown in Table 3 , pioglitazone treatment up-regulated 13 genes, including PPARγ and regulatory genes for apoptosis and solute handling. More genes (28 genes) were down-regulated with pioglitazone treatment. Interestingly, genes related to inflammation such as NF-κB, MCP-1, TNF-α and MAK kinase were down-regulated after pioglitazone treatment.

Renal histologic change in experimental animals

Figure 1 shows the representative glomerular pathology in the experimental groups at the end of the study period. Diabetic OLETF rats showed more severe glomerulosclerosis when compared with LETO rats. Pioglitazone treatment significantly reduced the glomerulosclerosis.

Fig. 1

Representative renal histologic findings in experimental animals. ( A ) Long-Evans Tokushima Fatty (LETO) rat at 46 weeks of age. ( B ) Otsuka Long-Evans Tokushima Fatty (OLETF) rat at 46 weeks of age. ( C ) OLETF rat treated with pioglitazone at 10 mg/kg per day for 6 months at 46 weeks of age. ( D ) Glomerulosclerosis index in each group. Data are shown as mean ± SEM. Pio, pioglitazone; ^*P < 0.05 LETO versus OLETF, ^#P < 0.05 OLETF versus OLETF with pioglitazone; original magnification ×400; PAS stain.

Effect of pioglitazone on MCP-1 and adiponectin levels in experimental animals

Although the serum level of MCP-1 did not differ among the three groups, urinary MCP-1 excretion was of significantly higher level in OLETF rats than LETO rats, and treatment with pioglitazone markedly reduced urinary MCP-1 excretion. The serum adiponectin level, a major anti-inflammatory adipocytokine, was decreased in OLETF rats compared to LETO rats, and recovered upon pioglitazone treatment (Figure 2 ).

Fig. 2

Effect of pioglitazone on MCP-1 and adiponectin levels of experimental animals. MCP-1 ( A ) and adiponectin ( B ) concentrations in serum and urine collected for 24 h. The urinary concentration was adjusted with creatinine concentration. Data are shown as mean ± SEM. Pio, pioglitazone; ^*P < 0.05 LETO versus OLETF, ^#P < 0.05 OLETF versus OLETF with pioglitazone.

Effect of pioglitazone on macrophage infiltration in the kidney

Only modest macrophage infiltration, as assessed by ED-1 staining, was observed in the glomerulus of LETO rats. In contrast, increased numbers of ED-1 positive macrophages were detected in the diabetic kidney of untreated OLETF rats. Pioglitazone treatment induced a significant reduction in renal ED-1 positive macrophage infiltration (Figure 3 ).

Fig. 3

Immunohistochemistry of ED-1 in experimental animals. ( A ) LETO rat at 46 weeks of age. ( B ) OLETF rat at 46 weeks of age. (C) OLETF rat treated with pioglitazone (10 mg/kg per day) for 6 months at 46 weeks of age. Positive staining for ED-1 is shown by short arrows. ( D ) Quantitative immunostaining score for ED-1 positive staining cells. Data are shown as mean ± SEM. Pio, pioglitazone; ^*P < 0.01 LETO versus OLETF, ^#P < 0.01 OLETF versus OLETF with pioglitazone; original magnification ×400.

Effect of pioglitazone on NF-κB activation in the renal cortex

To confirm that the pioglitazone-induced down-regulation of MCP-1 is associated with the NF-κB pathway, gel mobility-shift assays were performed using renal cortical tissues. As shown in Figure 4 , the NF-κB binding activity of nuclear protein was significantly increased in untreated OLETF rats compared to LETO rats and pioglitazone treatment markedly suppressed NF-κB activation. In addition, western blot analysis of the p65 subunit of NF-κB also showed a similar result with that of EMSA (Figure 4 ).

Fig. 4

Effect of pioglitazone on NF-κB activation in diabetic rats. ( A ) Electrophoretic mobility shift assay (EMSA) for NF-κB. Long-Evans Tokushima Fatty (LETO) rat at 46 weeks of age. Otsuka Long-Evans Tokushima Fatty (OLETF) rat at 46 weeks of age. OLETF rat treated with pioglitazone at 10 mg/kg per day for 6 months at 46 weeks of age. ( B ) Quantitative analysis for NF-κB activation. ( C ) Representative immunoblot for NF-κB p65 using nuclear protein from renal cortical tissues. Data are shown as mean ± SEM. Pio, pioglitazone; ^*P < 0.01 LETO versus OLETF, ^#P < 0.01 OLETF versus OLETF with pioglitazone.

Effect of pioglitazone on the expression of TGF-β1, PAI-1, type IV collagen and TGFβ-inducible gene h3 (βig-h3)

Gene expression for TGF-β1, PAI-1 and type IV collagen was markedly increased in OLETF rats, and dramatically suppressed by pioglitazone treatment. In accordance with TGF-β1 mRNA expression, serum and urinary levels of βig-h3, which is a biological activity marker of TGF-β1 in the kidney, were dramatically increased in OLETF rats. Pioglitazone treatment significantly decreased serum and urinary levels of βig-h3 to near control rats’ levels (Figure 5 ). Figure 6 shows representative immunostaining results for TGF-β1 (Figures 6 A–C), PAI-1 (Figures 6 D–F) and type IV collagen (Figures 6 G–I). In the diabetic kidney, TGF-β1, PAI-1 and type IV collagen immunoreactivity was primarily increased in the glomerular mesangium, and pioglitazone treatment induced significant reductions in renal TGF-β1, PAI-1 and type IV collagen protein expression (Figure 7 ).

Fig. 5

mRNA expression of TGF-β1 ( A ), PAI-1 ( B ) and type IV collagen ( C ) in renal cortical tissues measured by RT-PCR. ( D ) βig-h3 levels in serum and 24-h urine measured by ELISA. Urinary concentration of βig-h3 was adjusted by creatinine concentration data are shown as mean ± SEM. PAI-1, plasminogen activator inhibitor; βig-h3, TGFβ-inducible gene h3; Pio, pioglitazone; ^*P < 0.01 LETO versus OLETF, ^#P < 0.01 OLETF versus OLETF with pioglitazone.

Fig. 6

Immunohistochemistry for TGFβ1, PAI-1 and type IV collagen in experimental animals. Representative immunostained findings at 46 weeks of age for TGFβ1 ( A – C ), PAI-1 ( D – F ) and type IV collagen ( G – I ). LETO rat at 46 weeks of age ( A , D and G ); OLETF rat at 46 weeks of age ( B , E and H ); OLETF rat treated with pioglitazone (10 mg/kg per day) for 6 months at 46 weeks of age ( C , F and I ). Original magnification ×400

Fig. 7

Immunostaining score for TGFβ1 ( A ), PAI-1 ( B ) and type IV collagen ( C ) in glomeruli. Data are shown as mean ± SEM. Pio, pioglitazone; ^*P < 0.01 versus LETO, ^#P < 0.01 OLETF versus OLETF with pioglitazone.

Effect of pioglitazone on endogenous PPARγ transcriptional activity, MCP-1, collagen production and NF-κB activity in cultured MCs

Mesangial cells exhibited a significant induction of luciferase activity after stimulation with pioglitazone in a dose-dependent manner. The high glucose condition induced a significant up-regulation in MCP-1 and collagen gene transcription and protein excretion, and pioglitazone treatment abolished this high glucose-induced MCP-1 and collagen expression and secretion. In addition, high glucose stimuli increased NF-κB activity measured by p65 western blotting using extracted nuclear protein, and NF-κB activity decreased upon pioglitazone treatment (Figures 8 and 9 ).

Fig. 8

Effect of different concentrations of pioglitazone on PPRE transcriptional activity ( A ) NF-κB activity ( B ) and MCP-1 production in cultured mesangial cells ( C , D ) in cultured mesangial cells. ( A ) PPRE-luciferase activity was normalized to Renilla luciferase activity ( n = 12 in each condition). Data are shown as mean ± SEM. ^*P < 0.01 compared to control. ( B ) Representative immunoblot for NF-κB p65 using nuclear protein form cultured cells. Cells were treated with high glucose stimuli (30 mM of glucose) with or without pioglitazone at a concentration of 10 μM for 24 h. ( C , D ) mRNA expression for MCP-1 was measured by RT-PCR and the concentration of MCP-1 in culture supernatant was measured by ELISA. Data are shown as mean ± SEM. NG, normal glucose condition; HG, high glucose condition (30 mM); Pio, pioglitazone; ^*P < 0.01 versus normal glucose condition, ^#P < 0.01 versus high glucose condition.

Fig. 9

Effect of pioglitazone on procollagen α1 chain mRNA expression in type I collagen, procollagen α1 chain in type IV collagen ( A , B ), and collagen protein secretion ( C ) in cultured mesangial cells. Cells were cultured under a high glucose medium (30 mM of d -glucose) with or without pioglitazone at a concentration of 10 μM. Total collagen secreted in the supernatant was measured by a collagen assay kit. The collagen concentration was corrected using the protein concentration of the cells. Data are shown as mean ± SEM. NG, normal glucose (5 mM); HG, high glucose (30 mM); Pio, pioglitazone; ^*P < 0.01 versus a normal glucose condition, ^#P < 0.01 versus a high glucose condition.

Discussion

In the present study, we demonstrated that treatment with the PPARγ agonist, pioglitazone, resulted in a marked improvement of urinary albumin excretion and ameliorated glomerulosclerosis in these animals. The renoprotective effect of pioglitazone was associated with attenuation of the renal inflammatory process including NF-κB activation, MCP-1 synthesis, suppression of profibrotic molecules and reduction of collagen synthesis in the diabetic kidney. We also provided evidence that in addition to metabolic improvement, pioglitazone could directly activate endogenous renal PPARγ, resulting in the suppression of inflammatory molecules and down regulation of collagen synthesis in cultured mesangial cells.

PPARγ agonists have been considered to have direct beneficial effects on the diabetic kidney disease. They have been shown to reduce proteinuria and improve glomerulosclerosis, both in animal and human diabetic nephropathy studies [ 16–18 , 24,25 ]. Multiple mechanisms have been suggested for the renoprotective effect of PPARγ including inhibition of reactive oxygen species, suppression of PAI-1, improvement of glomerular hyperfiltration and renal endothelial dysfunction [ 26–28 ]. Based on the physiologic action of PPARγ agonist, the anti-inflammatory effect may contribute to the renoprotective effect in the diabetic kidney disease. Recently, Ohga et al . reported that pioglitazone ameliorates renal injury through the inhibition of NF-κB activation, ICAM-1 expression and macrophage infiltration in streptozotocin-induced diabetic rats [ 29 ]. However, it is still unknown whether the renal protective effect of PPARγ agonist in type 2 diabetic nephropathy is related to an anti-inflammatory mechanism.

In the present study, we tested the hypothesis that pioglitazone prevents diabetic renal injury through an anti-inflammatory mechanism and examined its effect on metabolic parameters in type 2 diabetic rats. Pioglitazone treatment induced a down regulation of gene expression and urinary excretion of MCP-1, and a reduction of macrophage accumulation in the diabetic kidney. Furthermore, we noted that pioglitazone treatment markedly suppressed the gene expression and activation of NF-κB in the diabetic kidney. Although we cannot provide direct evidence of MCP-1 overproduction through NF-κB activation in this study, NF-κB has been considered as an important upstream regulator of MCP-1 synthesis [ 30 ]. These results are consistent with other studies about the anti-inflammatory effect of PPARγ agonist on diabetic nephropathy [ 31,32 ]. Furthermore, these findings support the possibility that pioglitazone reduces the transcriptional activation of pro-inflammatory mediators and subsequent inflammation by blocking NF-κB pathway activation in the diabetic kidney.

The importance of NF-κB in the pathogenesis of diabetic nephropathy was reported in other studies [ 33,34 ], but the regulatory effect of PPARγ agonist on NF-κB has not been reported in the type 2 diabetic rat model. In our study, pioglitazone-induced down-regulation of NF-κB activity may be partially mediated by the hypoglycaemic effect of pioglitazone in diabetic rats. To exclude the confounding effect of hyperglycaemia on NF-κB activation, we performed in vitro experiments. In cultured mesangial cells, high glucose-induced NF-κB activation and MCP-1 overproduction was dramatically suppressed by pioglitazone treatment under a high glucose condition. These results suggest that pioglitazone treatment directly inhibited activation of NF-κB and MCP-1 synthesis in the kidney.

We also found that pioglitazone treatment down-regulated many genes that are involved in fibrosis and matrix accumulation. Furthermore, we confirmed that renal synthesis of TGFβ1, PAI-1 and type IV collagen, and urinary excretion of βig-h3 reflecting in vivo TGFβ activity [ 19 ], were dramatically decreased by pioglitazone treatment, which were consistent with the microarray data.

The above-mentioned pioglitazone-induced improvement in renal function is further supported by the in vitro experiments. Pioglitazone treatment directly reduced NF-κB activity, and down-regulated MCP-1 and collagen synthesis. These in vitro findings suggest that pioglitazone directly contributes to the observed renoprotective effects, as well as its metabolic effects, in type 2 diabetic rats via PPARγ activation.

The limitation of this study is that glucose levels were not the same among experimental groups, and additional metabolic factors, such as decreased triglyceride levels, may be operative in producing a beneficial effect of pioglitazone treatment [ 35 ].

In conclusion, the present study shows that the PPARγ agonist decreased urinary albumin excretion and ameliorates glomerulosclerosis through inhibition of NF-κB activation. Moreover, subsequent decreased inflammation, fibrosis and matrix accumulation, including TGFβ1 and collagen synthesis, were observed in the diabetic kidney upon pioglitazone treatment. These findings suggest that PPARγ agonist ameliorates diabetic nephropathy through anti-inflammatory mechanisms in type 2 diabetic rats.

We thank the Tokushima Research Institute, Otsuka Pharmaceutical Co., Ltd, for provision of Otsuka Long-Evans Tokushima Fatty (OLETF) rats. This work was supported in part by a Korea University grant.

Conflict of interest statement . None declared.

© Oxford University Press 2008