This article is mentioned in:

Introduction

Diabetes mellitus is the leading cause of end-stage renal disease (ESRD), based on the United States renal data system (1). Glycemic control and currently available pharmacotherapies delay, but cannot prevent, the progression of diabetic nephropathy (DN) towards ESRD (2,3). However, in order to develop novel therapies, a full understanding of the etiology of DN is required. Proteinuria, specifically microalbuminuria, is currently one of the earliest clinically identifiable markers of diabetes-induced renal damage, and the likelihood of progression to ESRD is significantly correlated with the level of albuminuria (4–7). Proteinuria not only predicts the speed of development in ESRD, but also correlates with renal decline. In addition, patients with DN with low levels of proteinuria have a markedly slower rate of disease progression, compared with those with a high rate of urinary protein excretion (8–11). The selection of targets and the timing of intervention are, thus critical for effectively preventing DN. Therefore, identifying the appropriate cellular targets for therapeutic intervention is crucial to enable the prevention of proteinuria. The progression of DN frequently begins with injury to podocytes, and there is a close association between the onset of albuminuria and podocytopathies (12,13), including foot process effacement, podocyte hypertrophy, detachment, apoptosis and epithelial-to-mesenchymal transition (EMT) (14). Based on various human and experimental models, DN is associated with a decreased number of podocytes per glomerulus and foot process effacement upon biopsy (15). Previous studies in OVE26 transgenic mice, a model of type-1 DN, demonstrated that reduced podocyte numbers and density frequently follow the onset of micro-albuminuria and more subtle podocyte injuries (16,17). Therefore, podocyte effacement and loss are critical events in the early progression of DN.

The EMT can also result in podocyte loss, during which epithelial cells undergo morphological changes. Podocyte EMT can be caused by the loss of epithelial P-cadherin, zonula occludens-1 or nephrin, and additional stimuli include the acquisition of mesenchymal FspI, desmin, collagen I and fibronectin (18). Podocytes can also undergo EMT under conditions of high glucose (19), which is associated with increased podocyte detachment, microalbuminuria and more severe glomerular pathology. Several intracellular signaling pathways regulate EMT, including transforming growth factor-β/small mothers against decapentaplegic, Wnt/β-catenin and integrin-linked kinase (ILK) (20). Glycogen synthase kinase 3β (GSK-3β) is involved in all these pathways, and thus may be pivotal in podocyte EMT (20). Therefore, GSK-3β inhibitors have been investigated in mesangial proliferative glomerulonephritis, crescent glomerulonephritis and lupus nephritis (21). In these studies, the GSK-3β inhibitor, (2′Z, 3′E)-6-bromoindirubin-3′-oxime (BIO), was found to inhibit high glucose-stimulated apoptosis in mesangial cells.

The mechanisms involved in glomerular injury and proteinuria during diabetes mellitus require additional investigation. The present study investigated the association between high glucose and EMT in podocytes and in the kidneys of db/db diabetic mice. In addition, the present study investigated the therapeutic potential of modulating GSK-3β and EMT, in the presence of high glucose, for DN.

Materials and methods

Animals

The Committee for the Care and Use of Laboratory Animals of Xinxiang Medical College (Xinxiang, China) approved all animal experiments. A total of 48 male db/db mice and 24 age-matched db/+ control mice were selected (Model Animal Research Center of Nanjing University, Nanjing, China). At an age of 12 weeks, the db/+ mice (normal control group; NC) and 24 db/db mice (diabetic nephropathy group; DN) were subcutaneously injected with dimethyl sulfoxide used as a diluent (Sigma-Aldrich, St. Louis, MO, USA), whereas the remaining 24 db/db mice (BIO intervention group; BIO) were injected with 320 µg/kg per day BIO (Sigma-Aldrich). All mice were housed in temperature- and humidity-controlled IVC-II independent supply isolation cages (Jinan Aonuo CNC Equipment Co., Ltd., Jinan, China), with free access to pure water and standard laboratory chow. When the mice were 12, 15 and 18 weeks old, eight mice in each group were sacrificed. On the day prior to sacrifice, the mice were individually housed in metabolic cages for the collection of urine over 24 h. The levels of urinary albumin excretion (UAE) were quantified using an albumin ELISA kit (Abcam, Cambridge, MA, USA). Heart blood (1 ml) was drawn and centrifuged at 625 × g at 4°C for 4 min. Serum glucose was measured at the time of sacrifice using an automatic analyzer (SPOTCHEM EZ SP-4430; ARKRAY, Inc., Kyoto, Japan). Following sacrifice of the mice, the kidneys were rapidly dissected and weighed, and the cortices were separated. Renal tissue sections (4 µm) were prepared and stained with hematoxylin and eosin (HE; Fuzhou Maixin Biotechnology Development Co., Ltd., Fuzhou, China), followed by assessment using light microscopy (Leica Microsystems GmbH, Wetzlar, Germany). In addition, electron microscopy was used to observe the micro-morphological changes in the renal tissues. A total of 20 glomeruli were evaluated for each mouse. The present study was approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University (Zhengzhou, China).

Cell culture and treatment

Cells of the MPC-5 clonal cell line of conditionally immortalized mouse podocytes, cultured in vitro from H-2Kb-tsA58 mice, were provided by American Mount Sinai Medical College, and were cultured, as previously described (22). For propagation, the cells were cultured at 33°C in RPMI 1640 medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10 U/ml mouse recombinant interferon-γ (Shanghai Sangon Biological Engineering Technology and Services, Co., Ltd., Shanghai, China) and 10% fetal bovine serum (Invitrogen; Thermo Fisher Scientific, Inc.) to enhance the expression of a thermosensitive T antigen. The cells were then grown under non-permissive conditions at 37°C, without interferon-γ, for 14 days to induce differentiation. The podocytes were then cultured with or without high glucose (25 mmol/l; Sigma-Aldrich), in the presence or absence of BIO (10 µmol/l) for 36 h.

Electron microscopy (EM)

Samples from 24 mice were used for the EM. Following anesthesia with ketamine (100 mg/kg; Sigma-Aldrich), the mice were sacrificed by cervical dislocation and 2% glutaraldehyde was perfused into the kidneys, which were then excised and immersed in the same fixative overnight. The tissue blocks were then fixed in 2% osmium tetroxide (Sigma-Aldrich) for 2 h at 4°C, dehydrated using an ethanol gradient and Epon-embedded (Fuzhou Maixin Biotechnology Development Co., Ltd.). Ultrathin sections (80 nm) stained with 4% uranyl acetate (Sigma-Aldrich) and with 1% lead citrate (Sigma-Aldrich), were cut using an ultra-microtome (EM UC7; Leica Microsystems GmbH) and examined by EM (JEM-100SX; JEOL, Ltd., Tokyo, Japan).

Western blotting

Western blotting was performed using established protocols to assess specific protein expression levels, as previously described (23). Radioimmunoprecipitation assay lysis buffer (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) according to the manufacturer's protocols, and protein concentration was determined using the Coomassie brilliant blue method. Proteins (30 µg) were separated on 10% SDS-PAGE gels for 3 h, then transferred to polyvinylidene fluoride membranes. Following washing with phosphate-buffered saline three times for 14 min, 1% bovine serum albumin (Fuzhou Maixin Biotechnology Development Co., Ltd.) was used to block the membrane for 2 h. Following blocking, the membranes were incubated with primary antibodies overnight at 4°C as follows: Rabbit polyclonal anti-podocin (dilution, 1:1,000; Bioss, Inc., Woburn, MA, USA; cat. no. bs6597R) rabbit polyclonal anti-nephrin (dilution, 1:1,000; cat. no. ab58968), rabbit polyclonal anti-synaptopodin (dilution, 1:200; cat. no. ab109560), rabbit polyclonal anti-α-SMA (dilution, 1:400; cat. no. ab5694), rabbit polyclonal anti-fibronectin (dilution, 1:2,000; cat. no. ab2413), rabbit polyclonal anti-phosphorylated-Tyr216-GSK-3β (dilution, 1:1,000; cat. no. ab75745), mouse polyclonal anti-vitamin D receptor (VDR; dilution, 1:1,000; cat. no. ab3508), and mouse monoclonal anti-β-actin (dilution, 1:1,000; cat. no. ab6276), mouse monoclonal anti-GSK-3β (dilution, 1:1,000; cat. no. ab75745; all from Abcam), rabbit monoclonal anti-phosphorylated-Ser9-GSK-3β (dilution, 1:1,000; cat. no. 5558), rabbit monoclonal anti-β-catenin (dilution, 1:1,000; cat. no. 8480), rabbit monoclonal anti-Snail (dilution, 1:1,000; cat. no. 3879; all from Cell Signaling Technology, Inc., Danvers, MA, USA). The secondary antibodies used incubated with the membrane for 2 h at room temperature and were as follows: alkaline phosphatase-labeled goat anti-rabbit IgG (dilution l:200; Abcam; cat. no ab97048) and alkaline phosphatase-labeled horse anti-mouse IgG (dilution l:200; Vector Laboratories, Inc., Burlingame, CA, USA; cat. no. AP-2000), goat anti-rabbit IgG (dilution l:200; Abcam; cat. no. ab6721) and donkey anti-mouse IgG (dilution l:300; Abcam; cat. no. ab150105). Visualization was performed using nitro-blue tetrazolium/5-bromo-4-chloro-3′-indolyphosphate coloration (Wuhan Boster Biological Technology, Ltd., Wuhan, China). The intensities of the bands were measured and quantified using ImageJ analysis software (version 1.46; imagej.nih.gov/ij/).

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis

Total RNA was isolated using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Synthesis of the first strand of cDNA was performed in 20 µl reaction buffer with 2 µg (13 µl) RNA, 4 µl 5X PrimeScript Buffer (Takara Biotechnology, Co., Ltd., Dalian, China), 1 µl PrimeScript RT Enzyme mix (Takara Biotechnology, Co., Ltd.), 1 µl Oligo-dT primer (50 µmol/l) and 1 µl random hexamers (100 µmol/l) at 37°C for 15 min, followed by 85°C for 5 sec. qPCR was performed, according to standard protocols with 3 µl aliquots of cDNA (5 ng/µl) using specific primer pairs, the sequences of which are shown in Table I. The reaction mixture (15 µl) contained 7.5 µl 2X Premix Ex Taq (Takara Biotechnology, Co., Ltd.), 0.25 µl (10 µmol/l) forward primer, 0.25 µl (10 µmol/l) reverse primer, 3 µl cDNA and 4 µl distilled water and the PCR reaction was performed in a Veriti® 96-Well Thermal Cycler (Applied Biosystems; Thermo Fisher Scientific, Inc.). Following an initial denaturation at 95°C for 3 min, the thermocycling conditions were as follows: For nephrin, 40 cycles of 95°C for 30 sec, 56°C for 30 sec and 72°C for 1 min, followed by 72°C for 5 min; for α-SMA, 40 cycles of 95°C for 30 sec, 52°C for 30 sec and 72°C for 1 min, followed by 72°C for 5 min; for GSK-3β, 40 cycles of 95°C for 30 sec, 59°C for 30 sec and 72°C for 1 min, followed by 72°C for 5 min; for Snail, 40 cycles of 95°C for 30 sec, 55°C for 30 sec and 72°C for 1 min, followed by 72°C for 5 min; for β-catenin, 45 cycles of 95°C for 30 sec, 51°C for 30 sec and 72°C for 1 min, followed by 72°C for 5 min; and for GAPDH, 40 cycles of 95°C for 30 sec, 60°C for 30 sec and 72°C for 1 min, followed by 72°C for 5 min. ImageJ software was used to visualize the qPCR product sizes fractionated on a 1.0% agarose gel (Sigma-Aldrich).

Table I Primer sequences, product sizes and annealing temperatures/durations for reverse transcription-quantitative polymerase chain reaction analysis.

Table I

Primer sequences, product sizes and annealing temperatures/durations for reverse transcription-quantitative polymerase chain reaction analysis.

Gene	Primer sequence	Product (bp)	Duration (sec)	Temp (°C)
GAPDH	F: 5′-CCTGCACCACCAACTGCTTAGC
	R: 5′-CCAGTGAGCTTCCCGTTCAGC	238	30	56
Nephrin	F: 5′-CCCAGGTACACAGAGCACAA
	R: 5′-CTCACGCTCACAACCTTCAG	200	30	55
α-SMA	F: 5′-TACTGCCGAGCGTGAGA
	R: 5′-GCTTCGTCGTATTCCTGTTT	489	30	51
GSK-3β	F: 5′-TTCAGGCCGCTGTTCACCG
	R: 5′-GTGCTGGTCTTTCCCGCGCA	152	30	60
β-catenin	F: 5′-CTCATTCCACCAACTGCTTGGC
	R: 5′-TAAGTGAGCTTTCGGTTCAGC	298	30	51
Snail	F: 5′-CCCTAGGTACACAGAGCACG
	R: 5′-GCCACGCTCACAACCTTACG	408	30	54

[i] α-SMA, α-smooth muscle actin; GSK-3β, glycogen synthase kinase 3β; F, forward; R, reverse.

GSK-3β kinase activity

The activity of GSK-3β was assayed using a GSK-3β Activity assay kit (Sigma-Aldrich), according to the manufacturer's protocol and as previously described (24). Activity was calculated following determining optical densities (OD) using the following formula: GSK-3β activity = [(OD sample − OD blank) × 0.1 × gradient concentration] / 0.005 × 6.22 × 5.

Statistical analysis

All data are presented as the means ± standard deviation of three independent experiments and statistical analysis was conducted using SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA). The physiological variables were analyzed using one-way analysis of variance followed by least significant difference multiple comparison post-hoc analysis. Other variables were compared using a Kruskal-Wallis non-parametric test followed by multiple comparisons using Duncan's method. For all comparisons, P<0.05 was considered to indicate a statistically significant difference.

Results

Effects of BIO on blood glucose, UAE, body weight and kidney weight

As summarized in Table II, blood glucose remained significantly elevated in the DN and BIO groups, compared with the NC group throughout the experiment. No significant differences were found between the DN and BIO groups, suggesting that BIO had no effect on blood glucose. The levels of UAE increased from the age of 12 weeks, and continued to increase until the age of 18 weeks. The increases in UAE were predominantly attenuated in the BIO-treated mice, compared with the DN mice, suggesting that BIO inhibited damage to the glomerular filtration membrane. The body and kidney weights were also increased by 12 weeks of age, and continued to increase at 15 and 18 weeks. The increases in body weight and kidney weight were markedly lower in the BIO group, compared with the DN group. These data suggested that BIO significantly attenuated DN.

Table II Effects of BIO on blood glucose, UAE, bodyweight, and kidney weight.

Table II

Effects of BIO on blood glucose, UAE, bodyweight, and kidney weight.

	Time point (age of mice)			F-value	P-value
	12 weeks	15 weeks	18 weeks
Blood glucose (mmol/l)
NC group	6.11±1.02	6.18±0.99	6.29±0.54	0.129	0.880
DN group	27.50±7.78a	28.03±7.91a	28.90±7.18a	0.222	0.804
BIO group	27.24±7.08a	28.29±4.57a	28.77±2.71a	0.502	0.523
F value	32.368	45.797	68.606	–
P-value	<0.05	<0.05	<0.05	–	–
Urinary albumin excretion (µg/24 h)
NC group	36.15±7.18	38.39±7.24	39.39±4.46	1.466	0.264
DN group	239.24±37.80a	244.66±35.30a,c	285.41±28.41a,c	44.826	<0.05
BIO group	239.54±24.08a	182.51±19.05a–c	135.65±18.96a–d	182.992	<0.05
F value	160.420	161.764	310.919	–	–
P-value	<0.05	<0.05	<0.05	–	–
Body weight (g)
NC group	11.96±1.02	14.54±1.19	16.06±1.27	62.437	<0.05
DN group	29.91±4.71a	34.24±4.08a	36.95±5.23a	30.326	<0.05
BIO group	29.93±4.85a	32.09±4.93a,b	32.49±4.77a,b	3.664	0.052
F value	55.200	66.077	56.106	–	–
P-value	<0.05	<0.05	<0.05	–	–
Kidney weight (mg)
NC group	107.32±8.77	116.02±8.82	123.92±7.24	96.946	<0.05
DN group	180.00±25.66a	189.95±19.27a,c	197.41±17.53a,c	8.318	<0.05
BIO group	180.78±14.97a	174.58±16.28a,b,c	159.90±22.53a–d	16.383	<0.05
F value	44.514	51.124	37.360	–	–
P-value	<0.05	<0.05	<0.05	–	–

a P<0.05, vs. NC group at same time point;

b P<0.05, vs. DN group at same time point.

c P<0.05, vs. in same treatment group at 12 weeks.

d P<0.05, vs. same treatment group at 15 weeks (n=8). NC, normal control (db/+); DN, diabetic neuropathy (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime-treated.

Kidney morphology

Following staining with HE, the renal cortices were observed under a light microscope. Compared with the NC group (Fig. 1A–C), increases in the mesangial matrix, occlusion of kidney tubules, sclerosis of the glomerulus, adhesion of balloon loops and Baumann's capsule walls were observed when the DN mice were 12 weeks old, and became more prominent by 18 weeks (Fig. 1D–F). These changes were significantly attenuated in the BIO group, compared with the DN group (Fig. 1G–I). Compared with the NC group (Fig. 2A–D), fusion and loss of foot processes, increased thickness of the glomerular basement membrane (GBM), increased mesangial matrix and collagen fiber, and hyaline degeneration of afferent and efferent vessels were detected in the DN mice (Fig. 2E–H). These changes were also significantly attenuated in the BIO group, compared with the DN group, suggesting that BIO slowed down the pathological changes in the glomerulus stimulated by DN (Fig. 2I–L).

Figure 1 HE staining of mouse kidney sections in the NC, DN and BIO groups at 12, 15 and 18 weeks of age. Kidney sections were harvested from mice in the NC, DN and BIO groups at 12, 15 and 18 weeks of age, and were stained with HE. The DN group exhibited increased mesangial matrix, adhesion of balloon loops and Baumann's capsule walls at 12 weeks, and the effects were more prominent by 18 weeks-of-age, compared with the mice in NC group. Following treatment with BIO for 6 weeks (from 12 weeks of age), there was a significant decrease in mesangial matrix in and the number and extent of glomerular sclerosis, compared with the untreated DN mice. Arrows indicate Kimmelstiel-Wilson nodules. Triangles indicate adhesion of balloon loops and Baumann's capsules. Scale bar=40 µm (n=8). HE, hematoxylin and eosin; NC, normal control (db/+); DN, diabetic neuropathy (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime-treated.

Figure 2 Ultrastructural examination using transmission electron microscopy in the NC, DN and BIO groups at 12, 15 and 18 weeks-of-age. Ultrathin sections (80 nm) were harvested from mice in the NC, DN and BIO groups at 12, 15 and 18 weeks of age, and were stained with 4% uranyl acetate and 1% lead citrate. Fusion and loss of foot processes, and increase in the thickness of the GBM were found in the DN group, compared with the NC group. Following treatment with BIO for 6 weeks (from 12 weeks of age), there was a significant decrease in foot process fusion and GBM thickness, compared with the DN group. The black boxes on the images at 18 weeks show the area of the image, which was magnified by 200% and shown to its left. White arrowheads indicate foot process effacement of podocytes; triangle indicate thickening of the GBM (n=8). NC, normal control group (db/+); DN, diabetic neuropathy group (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime-treated DN group; GBM, glomerular basement membrane.

Effects of BIO on the expression of epithelial cell markers

The expression levels of phenotypic markers for epithelial cells in mouse kidney tissues were compared, with β-actin as a loading control, using Western blotting and RT-qPCR analysis. With the progression of DN, the expression of nephrin decreased significantly in the DN group (Fig. 3). However, nephrin levels were significantly attenuated at 15 and 18 weeks of age in the BIO group, compared with the DN group (Fig. 3A). The results of the RT-qPCR analysis of mRNA levels of nephrin in mouse kidney tissue were analyzed using ImageJ software, which confirmed these observations (Fig. 3B). Similar expression patterns were observed in the protein expression levels of podocin and synaptopodin, two other epithelial markers (Fig. 3C and D).

Figure 3 Effect of BIO on the expression of epithelial cell markers. (A) Western blotting for the expression of nephrin in kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). (B) Reverse transcription-quantitative polymerase chain reaction analysis for the mRNA expression of nephrin in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). (C and D) Western blotting for the protein expression of podocin in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=3). 1, NC group; 2, DN group; 3, 15 week DN group; 4, 18 week DN group; 5, 12 week BIO group; 6, 15 week BIO group; 7, 18 week BIO group. (E) Western blotting for the protein expression of nephrin in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=4). 1–4: 1, normal glucose without BIO; 2, normal glucose with BIO; 3, high glucose without BIO; 4, high glucose with BIO. *P<0.05. vs. mice in NC group, or podocytes in the normal glucose group without BIO; #P<0.05, vs. 12-week-old mice in the DN group, or podocytes in the high glucose group without BIO treatment; ΔP<0.05, vs. 12-week-old mice in the DN+BIO group. NC, normal control group (db/+); DN, diabetic neuropathy group (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime.

The expression levels of nephrin in podocytes exposed to high or normal glucose, treated with or without BIO were compared. BIO had no effect under normal glucose conditions (Fig. 3A; P>0.05). Under conditions of high glucose, the expression of nephrin in the podocytes was significantly reduced, compared with the normal glucose group (P<0.05). However, treatment with BIO partially alleviated the high glucose-mediated reduction in the expression of nephrin (P<0.05), but did not restore expression levels to normal (P<0.05).

Effects of BIO on the expression of mesenchymal cell markers

The protein expression of mesenchymal cell phenotypic markers in the mouse kidney were normalized to β-actin, as an internal reference, for RT-qPCR analysis. As DN progressed, the expression of α-SMA in the DN group increased significantly in a time-dependent matter. However, these effects were attenuated at 15 and 18 weeks of age in the BIO group, compared with the DN group (Fig. 4A). RT-qPCR analysis of α-SMA in the kidneys of 15-week-old mice quantitatively confirmed these observations, as analyzed using ImageJ software (Fig. 4B). Similar expression patterns of fibronectin, another mesenchymal cell marker, were observed (Fig. 4C).

Figure 4 Effect of BIO on the expression levels of mesenchymal cell markers. (A) Western blotting for the expression of α-SMA in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). (B) Reverse transcription-quantitative polymerase chain reaction analysis for the mRNA expression of α-SMA in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). (C) Western blotting for the expression of fibronectin in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=4). 1, NC group; 2, DN group; 3, 15 week DN group; 4, 18 week DN group; 5, 12 week BIO group; 6, 15 week BIO group; 7, 18 week BIO group. (D) Western blotting for the expression of β-catenin in podocytes in high or normal glucose, with and without BIO treatment. Data are presented as the mean ± standard error of the mean (n=5). 1–4: 1, normal glucose without BIO; 2, normal glucose with BIO; 3, high glucose without BIO; 4, high glucose with BIO. *P<0.05, vs. mice or podocytes in NC group. #P<0.05, vs. 12-week-old mice in the DN group or podocytes in the high glucose group without BIO treatment; ΔP<0.05, vs. 12-week-old mice in the BIO group. α-SMA, α-smooth muscle actin; NC, normal control group (db/+); DN, diabetic neuropathy group (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime.

The expression levels of α-SMA in podocytes under conditions of high or normal glucose, with or without BIO treatment were then compared (Fig. 4D). Treatment with BIO had no effect on the expression of α-SMA under normal glucose concentrations (P>0.05). However, the expression of α-SMA was increased significantly under high-glucose conditions, compared with normal glucose (P<0.05). Treatment with BIO partially alleviated the high glucose-induced increase in the expression of α-SMA (P<0.05), but did not restore expression levels to normal (P<0.05).

Effects of BIO on the expression and activity of GSK-3β

GSK-3β, which is constitutively expressed in all eukaryotic cells, is a serine/threonine kinase. The important regulatory amino acids are Tyr-216 for activation and Ser-9 for inhibition. The expression levels of total-GSK-3β in the mouse kidney tissues were compared, with β-actin as a loading control, using RT-qPCR analysis. As DN progressed, the expression of total-GSK-3β in the DN group increased significantly in a time-dependent matter (P<0.05). However, total-GSK-3β levels were substantially attenuated at 15 and 18 weeks of age in the BIO group, compared with the DN group (P<0.05; Fig. 5A). ImageJ software was used to quantify the mRNA expression of GSK-3β in the mouse kidney following RT-qPCR analysis. The data from the animals at 15 weeks of age were consistent with the observations in the protein levels (P<0.05; Fig. 5B). The expression levels of pSer-9-GSK-3β increased significantly at 15 and 18 weeks in the BIO group (P<0.05; Fig. 5C), whereas the patterns observed in the expression of pTyr-216-GSK-3β were similar to those of total-GSK-3β (P<0.05; Fig. 5D). To confirm these observations, the kinase activity of GSK-3β was assessed using a kinase assay to determine the glomerular activity of GSK-3β (Fig. 5E). The activity of GSK-3β in the DN group increased with disease progression, in a time-dependent manner (P<0.05). Following treatment with BIO, the activity of GSK-3β decreased by 15 weeks, and continued to decline until of 18 weeks of age (P<0.05).

Figure 5 Effects of BIO on the expression and activity of GSK-3β. (A) Western blotting for the protein expression of total-GSK-3β in the kidneys of NC, DN and DN+BIO mice (n=4). (B) Reverse transcription-quantitative polymerase chain reaction analysis of mRNA expression of GSK-3β in the kidneys of NC, DN and DN+BIO mice (n=6). (C) Western blotting for pSer9-GSK-3β expression in the kidneys of NC, DN and DN+BIO mice (n=4). (D) Western blotting for the expression of pTyr216-GSK-3β in the kidneys of NC, DN and DN+BIO mice (n=4). 1, NC group; 2, DN group; 3, 15 week DN group; 4, 18 week DN group; 5, 12 week BIO group; 6, 15 week BIO group; 7, 18 week BIO group. (E) Kinase activity of GSK-3β in the kidneys of NC, DN and DN+BIO mice. (F) Western blotting for the protein expression of total-GSK-3β in podocytes in high or normal glucose, with and without BIO treatment (n=5). (G) Western blotting for the protein expression of pSer9-GSK-3β in podocytes in high or normal glucose, with and without BIO treatment (n=5). (H) Western blotting for the protein expression of pTyr216-GSK-3β in podocytes in high or normal glucose, with and without BIO treatment. Each bar represents mean ± SEM (n=5). 1–4: 1, normal glucose without BIO; 2, normal glucose with BIO; 3, high glucose without BIO; 4, high glucose with BIO. *P<0.05 vs. NC group or podocytes in normal glucose group without BIO treatment; #P<0.05, vs. 12-week-old mice in DN group or podocytes in high glucose group without BIO treatment; ΔP<0.05, vs. 12-week-old mice in DN+BIO group. All data are presented as the mean ± standard error of the mean. GSK-3β, glycogen synthase kinase 3β; NC, normal control group (db/+); DN, diabetic neuropathy group (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime.

The expression of total-GSK-3β in podocytes was also assessed, using RT-qPCR with β-actin as an internal reference (Fig. 5F). Under conditions of high glucose, the expression of total-GSK-3β increased significantly, compared with normal glucose (P<0.05). Treatment with BIO partially alleviated the high glucose-mediated increases in total-GSK-3β (P<0.05). Similar patterns of expression were observed for pTyr216-GSK-3β in the podocytes, whereas, consistent with previous observations, the protein expression patterns of pSer9-GSK-3β were reversed (P<0.05; Figs. 5G and H).

Effects of BIO on the expression levels of β-catenin, Snail and VDR

The expression levels of β-catenin in the mouse kidney tissues were compared, with β-actin as a loading control, using RT-qPCR (Fig. 6A). As DN progressed, the expression of β-catenin in the DN group increased significantly, in a time-dependent matter (Fig. 6A). However, the levels were substantially attenuated by BIO treatment at 15 and 18 weeks of age, compared with the DN group. RT-qPCR analysis of β-catenin in the kidney tissues of the 15-week-old mice confirmed these observations (Fig. 6B). Similar expression patterns for Snail were observed (Fig. 6C). However, the opposite patterns of expression were observed for VDR (P<0.05; Fig. 5D).

Figure 6 Effects of BIO on the expression levels of β-catenin, Snail and VDR. (A) Western blotting for the protein expression of β-catenin in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). (B) Reverse transcription-quantitative polymerase chain reaction analysis of the mRNA expression of β-catenin in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). (C) Western blotting for the protein expression of Snail expression in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=4). (D) Western blotting for the protein expression of VDR in the kidneys of NC, DN and DN+BIO mice. Data are presented as the mean ± standard error of the mean (n=6). 1, NC group; 2, DN group; 3, 15 week DN group; 4, 18 week DN group; 5, 12 week BIO group; 6, 15 week BIO group; 7, 18 week BIO group. (E) Western blotting for the protein expression of β-catenin in podocytes in high or normal glucose, with and without BIO treatment. Data are presented as the mean ± standard error of the mean (n=5). (F) Western blotting for the protein expression of VDR in podocytes in high or normal glucose, with and without BIO treatment. 1–4: 1, normal glucose without BIO; 2, normal glucose with BIO; 3, high glucose without BIO; 4, high glucose with BIO. *P<0.05, vs. mice in NC group or podocytes in normal glucose group without BIO treatment; #P<0.05, vs. 12-week-old mice in DN group, or podocytes in high glucose group without BIO treatment; ΔP<0.05, vs.12-week-old mice in DN+BIO group. VDR, vitamin D receptor; NC, normal control group (db/+); DN, diabetic neuropathy group (db/db); BIO, (2′Z, 3′E)-6-bromoindirubin-3′-oxime.

The expression of β-catenin in podocytes was also assessed and compared with β-actin, as a loading control, using RT-qPCR (Fig. 6E). Under normal glucose conditions, BIO treatment had no effect (P>0.05). However, under high-glucose conditions, the expression of β-catenin increased significantly, compared with the normal glucose group (P<0.05). Treatment with BIO partially attenuated the high glucose-induced increase in the expression of β-catenin (P<0.05). Consistent with the data from the mouse kidney tissues, the protein expression patterns of VDR in the podocytes were the reverse (Fig. 6F).

Discussion

Selection and establishment of animal models

db/db mice are one of the most commonly used animal models in studies investigating type II diabetes (25). The mouse contains a G>T point mutation in the leptin receptor (LepRdb/db), leading to its aberrant splicing derived by fat cells, and the formation of defective receptors (26,27). As a result, mice overeat, causing symptoms including obesity, hyperinsulinemia and insulin resistance. In addition, db/db mice develop progressive nephrotic disease, making the model suitable for long-term investigations.

The LepRdb/db mutation has been verified in C57BLKS/J (28), C57BL/6J (29) and FVB/NJ strains. The blood glucose levels of male C57BLKS/J db/db mice gradually increases from ~4 weeks of age, and hyperglycemia appears at ~8 weeks (>16 mmol/l). Microalbuminuria occurs at ~10–12 weeks (30–32), with morphological changes, including glomerular hypertrophy, basement membrane thickening, mesangial widening and foot process fusion (33) observed from 12–14 weeks of age. Finally, blood glucose levels peak at ~16 weeks (34–37). Therefore, the present study used male db/db C57BLKS/J mice to establish a model of DN.

The characteristics of the DN model used in the present study were as follows: i) at 12 weeks-of-age, mice exhibited significant polyuria, polydipsia and polyphagia, whereas body weight was 29.91±4.71 g and kidney weight was 180.00±25.66 mg, all of which were significantly different, compared with the NC group (P<0.05). ii) By 12 weeks of age, the blood glucose level in the DN mice was 27.5000±7.78 mmol/l and protein was detected in the urine. These observations were significantly different, compared with the NC group (P<0.05). iii) UAE over a 24 h period in the DN model mice was 239.24±37.80 µg at 12 weeks of age, which was also significantly different from that in the NC group (P<0.05). Finally, histopathological examination of the kidneys from the DN mice at 12 weeks of age revealed glomerular hypertrophy, basement membrane thickening and mesangial area widening, all of which increased in severity with disease progression. By 18 weeks of age, certain glomeruli exhibited lobulated sclerosis, a characteristic of DN mesangial cells, and capillary endothelial cells. Podocytes, which are glomerular epithelial cells attached to the outside of the GBM, constitute the glomerular filtration barrier between the GBM and capillary endothelium. Podocyte processes protrude from podocyte cell bodies, covering the outer surface of the GBM, and interacting with the GBM via proteoglycan adhesion molecules (38,39). The slit diaphragm, a zipper-like structure situated between two adjacent podocytes, is the final barrier of plasma protein filtration (35). Podocyte injury is involved in the progression of glomerulonephritis, DN, renal failure and other kidney diseases (40). Specifically, reductions in the number and density of podocytes have been linked to the pathological changes, which occur during DN (41). It has been suggested that podocyte phenotypic transformation, the process of its transdifferentiation into mesenchymal cells during cell damage, leads to the loss of specific podocyte protein markers, disrupted cell function and proteinuria (42).

Podocytes are highly differentiated epithelial cells. Several mature protein markers, including nephrin (43), podocin (44) and synaptopodin (45), are found on the surface and slit diaphragms of podocytes, and constitute a charge barrier of the GBM. They are also involved in cell-cell connections and signal transduction, and can maintain the integrity of podocyte foot processes and the normal function of the membrane hole (38,39). Phenotypic markers of mesenchymal cells include α-SMA and fibronectin, which are important for cellular activation and transdifferentiation.

A previous study has revealed that elevated blood glucose can cause mice to produce increased levels of fibronectin and laminin β1 in the extracellular matrix (46). In addition, Kang et al revealed that glucose levels can be raised following the induction of the expression of ILK, inducing podocyte transdifferentiation and injury (47).

Selection and application of treatments for DN

Clinically, DN treatment includes regulating blood glucose, controlling blood lipids and diet adjustment (48). Frequently used medications include angiotensin-converting enzyme inhibitors and angiotensin receptor blockers (49). However, these therapies only delay the development of ESRD, and do not effectively prevent the development of DN (50–52). GSK-3β is a serine/threonine kinase, which is expressed in all eukaryotic cells. The key amino acids for regulating GSK-3β activity are Tyr-216 for activation and Ser-9 for inhibition. Studies have shown that GSK-3β not only phosphorylates glycogen synthase to regulate the activity of glycolytic enzymes, but is also involved in signaling pathways, including insulin, Hedgehog, Notch and Wnt/β-catenin, to affect cellular differentiation, metabolism and apoptosis (53). During insulin signaling, the insulin receptor phosphorylates GSK-3β, leading to decreased GSK-3β activity in liver and muscle. This leads to reduced blood glucose levels and an increase in glycogen synthesis. In the Wnt pathway, GSK-3β, β-catenin, adenomatosis polyposis coli and Axin form a complex in the cytoplasm, leading to the degradation of β-catenin by the proteasome and the inhibition of Wnt gene transcription (54). Snail is a zinc finger transcription factor, which can transform epithelial cells into mesenchymal cells. GSK-3β can regulate the expression and activity of Snail, thereby regulating EMT (55,56). Therefore, the present study used BIO, an inhibitor of GSK-3β, to assess the effects of the expression and activity of GSK-3β on the progression of DN.

Effects of BIO on the development of diabetic nephropathy in db/db mice

In the present study, the db/db mice, as a model of DN, were treated with the GSK-3β inhibitor, BIO, and the effects on DN progression were assessed. At 18 weeks of age, the symptoms of the mice in BIO group, including polyuria, polydipsia and polyphagia, were significantly reduced. In addition, body weight was 32.49±4.77 g and kidney weight was 159.90±22.53 mg, which were significantly different, compared with those in the DN group (P<0.05). By 18 weeks of age, the blood glucose level in the BIO group was 28.7625±2.71 mmol/l, which was comparable to that in the DN group, suggesting that BIO had no effect on blood glucose (P>0.05). The UAE of mice in BIO group was 135.65±18.96 µg over 24 h, which was decreased significantly from the DN group (P<0.05). At ~18 weeks of age, renal pathological lesions, including mesangial cell proliferation and matrix, increased significantly in the DN group, but were alleviated by treatment with BIO. These data suggested that the GSK-3β inhibitor reduced proteinuria and slowed the progression of early DN. Of note, the mechanism underlying this protection of the kidney may involve adjusting EMT, but not lowering blood glucose.

Effects of BIO on GSK-3β in the db/db mouse kidney

Normal kidney tissue expresses small quantities of total-GSK-3β, however protein and mRNA expression levels were observed to increase with the progression of DN, in a time-dependent manner (P<0.05). Following treatment with BIO, the upregulation of total-GSK-3β was lower, compared with the control group (P<0.05). Similar expression patterns were observed for Tyr216-phosphorylated GSK-3β (pTyr216-GSK-3β) were observed. By contrast, Ser-9 phosphorylated GSK-3β (pSer9-GSK-3β) was expressed at a high level in normal renal tissues, but decreased with the development of DN, in a time-dependent manner (P<0.05). Following treatment with BIO, the levels of pSer9-GSK-3β were reduced, compared with the DN group (P<0.05). With the progression of DN, GSK-3β kinase activity in the DN group increased in a time-dependent matter, whereas treatment with BIO for 3 weeks led to a partial decrease in GSK-3β kinase activity, compared with the DN group. Following 6 weeks of BIO therapy, the activity of GSK-3β was further reduced (P<0.05). These results suggested that BIO, an inhibitor of GSK-3β, effectively affected the activity of GSK-3β, and may be useful for the treatment of DN.

Effects of BIO on EMT in the db/db mouse kidney

In the present study, as EMT developed in the renal cortex of mice in the DN group, the expression levels of total GSK-3β and pTyr216-GSK-3β, and the activity of GSK-3β all increased, whereas the levels of pSer9-GSK-3β declined. Therefore, it was hypothesized that the activation of GSK-3β is important in EMT, and that inhibiting the activity of GSK-3β may delay the development of EMT.

BIO was used to inhibit the activation of GSK-3β in the DN mice. When the mice were treated with BIO, the protein and mRNA expression levels of epithelial phenotype markers, including podocin, podocin and synaptopodin, increased in a time-dependent manner, which was significantly different, compared with the DN group (P<0.05). In addition, the expression levels of mesenchymal markers, including fibronectin and α-SMA, were decreased significantly, compared with those in the DN group (P<0.05). These results suggested that BIO, effectively reduced EMT in the db/db mice with DN.

During Wnt signaling, the upregulation of GSK-3β activity can lead to degradation of β-catenin (54) and decreased expression of Snail (57,58). Previous studies have revealed that using small interfering (si)RNA against GSK-3β can cause transdifferentiation of mammary gland tissue and skin epithelial cells (59). In addition, the activity of GSK-3β may maintain the structure of epithelial cells and their surface role in the normal podocyte phenotype (60,61). In contrast with these data, the present study revealed that normal kidney tissue expressed low levels of β-catenin and Snail. During the progression of DN, the expression levels of β-catenin and Snail increased, and the activity of GSK-3β was also upregulated in a time-dependent manner (P<0.05). Following treatment with BIO, the upregulation in the levels of β-catenin and Snail were inhibited significantly, compared with those in the DN group (P<0.05). These observations contradict previous observations in traditional Wnt signaling (62). It has been suggested that the activation of GSK-3β may directly stimulate transdifferentiation by triggering the transcription of relevant genes, and inhibiting the expression of epithelial cell markers (63). In the resting state, β-catenin is bound to fibronectin, which anchors in the cell membrane. This causes transdifferentiation of epithelial cells, following which β-catenin dissociates and migrates to the cytoplasm. Intracellular β-catenin then tends to migrate towards the nucleus (64), further promoting the transcription of transdifferentiation-associated genes.

Abnormalities in the expression of vitamin D are frequently observed in patients with chronic kidney disease (65). VDR, a receptor specific for vitamin D, has a protective effect in the kidneys, and its expression in the renal tissues of patients with DN is decreased (66). Once activated, VDR specifically combines with the DNA structural domain of vitamin D, and undergoes conformational change allowing it to combine with specific DNA regions. This leads to novel gene expression, resulting in a series of biological effects. In a previous study on human bone marrow stromal cells, the 1, 25 (OH)-2-vitamin D3 system activated VDR, which then combined with β-catenin in the nucleus, leading to the transport of β-catenin out of the nucleus (67). When activated by its ligand, VDR competitively combines with T cell transcription factor-4, which is normally bound to β-catenin, thus inhibiting the activity of β-catenin in colon cancer (68). In a mouse model of unilateral ureteral obstruction, it was observed that VDR was associated with the extent and development of renal fibrosis. Loss of the expression of VDR caused epithelial cell transdifferentiation, which may be mediated by β-catenin. Specifically, when the cells were treated with VDR siRNA, the expression of β-catenin was increased and transported to the nucleus, whereas the over expression of VDR inhibited the expression of β-catenin (69). The data presented in the present study demonstrated that the expression of VDR was decreased in the DN group, and the levels of β-catenin and Snail were increased. As expected, these changes were ameliorated following treatment with BIO, suggesting that GSK-3β may modulate the expression levels of β-catenin and Snail by regulating the expression of VDR. Therefore, the present study hypothesized that GSK-3β uses pathways in addition to the classical Wnt pathway to regulate and control EMT. This may explain why the activation of GSK-3β increased, and treatment with the GSK-3β inhibitor decreased, the expression levels of β-catenin and Snail, which were observations contradictory to the classical Wnt pathway.

Advantages and disadvantages of the clinical application of GSK-3β inhibitors

GSK-3β is a multifunctional protein, which is involved in several biological processes, including the transdifferentiation, proliferation, apoptosis and migration of several different cell types. During the transdifferentiation of renal tubular epithelial cells induced by high glucose, reports on GSK-3β activity are conflicting. It has been suggested that GSK-3β activity is reduced during the transdifferentiation of renal tubular epithelial cells induced by high glucose (70), whereas other have suggested that the activity of GSK-3β increases during high-glucose-mediated apoptosis in mesangial cells (71). In addition, studies have reported that the expression of GSK-3β increases in patients with type 2 diabetes and in rodent models of obesity and insulin resistance (72,73).

Further investigations are required to clarify the role of GSK-3β in high-glucose-stimulated DN, and specifically to clarify the association between GSK-3β, EMT and DN proteinuria. The data in the present study suggested that BIO, as an inhibitor of GSK-3β, can cause transdifferentiation of podocytes. Additional data suggested that LiCl, another inhibitor of GSK-3β has a similar role in EMT (data not shown) (74). Furthermore, the effects of GSK-3β in podocyte EMT were verified using GSK-3β RNA interference in podocytes, the results of which supported the conclusions of the present study. However, additional investigations are required to define the mechanisms underlying the effects of GSK-3β inhibitors in the treatment of DN, and the downstream signal transduction pathways. In addition, investigations on the long-term treatment with BIO is essential prior to its consideration as a therapeutic agent for the treatment of DN.

In conclusion, the present study demonstrated that, during the development of diabetic nephropathy, the expression levels of glomerular epithelial cell markers decreased, and the levels of mesenchymal-like markers increased in a time-dependent manner, suggesting the occurrence of EMT. In addition, the expression and activity of total-GSK-3β increased, the expression of pSer9-GSK-3β (the inhibitory site of GSK-3β) decreased and the expression of pTyr216-GSK-3β (the active site of GSK-3β) increased. This suggested that the activity of GSK-3β is closely associated with the development of DN in db/db mice. Following treatment with BIO, the expression of kidney epithelial cell markers in mice with DN increased, whereas the expression of mesenchymal-like markers decreased, compared with the untreated DN mice, suggesting that the inhibitor of GSK-3β partially reversed EMT. The present study also demonstrated that treatment with BIO reduced proteinuria leakage, delayed the deterioration of renal function and prevented structural changes in the kidneys of the mice with DN. This suggested that GSK-3β may regulate podocyte EMT, and that the GSK-3β inhibitor, BIO, may protect the filtration barrier of the glomerulus, reversing EMT and delaying the development of DN.

The present study also observed that the activation of GSK-3β increased the expression levels of β-catenin and Snail. Following GSK-3β inhibition by BIO, the expression levels of β-catenin and snail decreased. These effects may be due to the fact that EMT is directly associated with GSK-3β activity, which may inhibit the gene transcription of epithelial cell marker proteins. When EMT occurs, β-catenin dissociates from the cytomembrane and translocates to the cytoplasm, increased concentrations of intracellular β-catenin migrate and accumulate in the nucleus, accelerating the transcription of transdifferentiation-associated genes. In addition, it is possible that GSK-3β regulated the expression of VDR, and was consequently involved in regulating the expression of β-catenin and Snail. Therefore, the present study hypothesized that, rather than relying on the classical Wnt pathway to control EMT, GSK-3β can also regulate podocytes by modifying the levels of VDR, or by stimulating the nuclear accumulation of β-catenin.

Acknowledgments

The authors would like to thank Dr Li-Bin Zhang (Basic Medical College, Xinxiang Medical University) for discussion, and Dr Xin-Lai Qian (The Third Affiliated Hospital of Xinxiang Medical University, Xinxiang, China) and Dr Song-Xia Quan (The First Affiliated Hospital of Zhengzhou University) for their invaluable assistance in histological analyses. The authors would also like to thank Dr Ying Zhao (Department of Foreign Language, Henan Institute of Science and Technology, Zhengzhou, China) and Dr Peng Wan (Henan Time Culture Media Company, Zhengzhou, China) for their editorial assistance. This study was supported by grants from the National Basic Research Program of China 973 Program (grant nos, 2012CB517600 and 2012CB517606) and the National Natural Science Foundation of China (grant nos. 81070574 and 81270807).

References