Suppression of renal fibrosis by galectin-1 in high glucose-treated renal epithelial cells

E XP E RI ME N T AL C E L L R E SE A RC H 31 6 ( 20 1 0) 3 2 82 – 3 29 1

available at www.sciencedirect.com

www.elsevier.com/locate/yexcr

Research Article

Suppression of renal fibrosis by galectin-1 in high glucose-treated renal epithelial cells Kazuhiro Okano⁎, Yuki Tsuruta, Tetsuri Yamashita, Mari Takano, Yoshihisa Echida, Kosaku Nitta Department of Medicine, Kidney Center, Tokyo Women's Medical University, Tokyo 162-8666, Japan

A R T I C L E I N F O R M A T I O N

A B S T R A C T

Article Chronology:

Diabetic nephropathy is the most common cause of chronic kidney disease. We investigated the

Received 20 April 2010

ability of intracellular galectin-1 (Gal-1), a prototype of endogenous lectin, to prevent renal

Revised version received

fibrosis by regulating cell signaling under a high glucose (HG) condition. We demonstrated that

24 August 2010

overexpression of Gal-1 reduces type I collagen (COL1) expression and transcription in human

Accepted 29 August 2010

renal epithelial cells under HG conditions and transforming growth factor-β1 (TGF-β1)

Available online 7 September 2010

stimulation. Matrix metalloproteinase 1 (MMP1) is stimulated by Gal-1. HG conditions and TGF-β1 treatment augment expression and nuclear translocation of Gal-1. In contrast, targeted

Keywords:

inhibition of Gal-1 expression reduces COL1 expression and increases MMP1 expression. The

Diabetes

Smad3 signaling pathway is inhibited, whereas two mitogen-activated protein kinase (MAPK)

Galectin-1

pathways, p38 and extracellular signal-regulated kinase (ERK), are activated by Gal-1, indicating

Renal fibrosis

that Gal-1 regulates these signaling pathways in COL1 production. Using specific inhibitors of

Smad3

Smad3, ERK, and p38 MAPK, we showed that ERK MAPK activated by Gal-1 plays an inhibitory role

Mitogen-activated protein kinase

in COL1 transcription and that activation of the p38 MAPK pathway by Gal-1 plays a negative role

TGF-β1

in MMP1 production. Taken together, two MAPK pathways are stimulated by increasing levels of Gal-1 in the HG condition, leading to suppression of COL1 expression and increase of MMP1 expression. © 2010 Elsevier Inc. All rights reserved.

Introduction Diabetic nephropathy is the most common cause of chronic kidney disease in many countries; it is pathologically characterized by cellular hypertrophy and increased extracellular matrix (ECM) accumulation [1,2]. A high glucose (HG) condition activates several growth factors, such as transforming growth factor (TGF)-β1, in a variety of cells [3]. An HG condition is also known to stimulate collagen synthesis, at least partially, through the TGFβ1/Smad signaling pathway in renal and vascular cells [4]. On the contrary, neutralizing antibody to TGF-β1 decreases collagen

mRNA in kidneys of streptozotocin-induced diabetic mice [5]. These data suggest an important contribution of TGF-β1/Smad signaling in ECM formation in diabetic nephropathy. Galectin-1 (Gal-1) is expressed in podocytes with diffuse mesangial proliferation and focal segmental glomerulosclerosis [6] and in cultured human tubular epithelial cells [7]. Peritoneal administration of recombinant Gal-1 prevents renal fibrosis of nephrotoxic serum nephritis in Wistar–Kyoto rats [8] and decreases the severity of nonobese diabetic mice [9]. These data suggest that exogenous Gal-1 may be beneficial for preventing glomerulonephritis. Gal-1 is expressed both in the cell nuclei and

⁎ Corresponding author. Department of Medicine, Kidney Center, Tokyo Women's Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan. Fax: +81 3 3356 0293. E-mail address: [email protected] (K. Okano). 0014-4827/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2010.08.015

E XP E RI ME N T AL C E L L R E S EA RC H 31 6 ( 20 1 0) 3 2 8 2– 3 29 1

the cytosol and is secreted into the extracellular space through cell membranes [10], suggesting that Gal-1 has a variety of intra- and extracellular functions. With respect to intracellular functions, Gal-1 is known to stimulate the extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase (MAPK) pathway [11,12]. T cell-mediated immune responses are inhibited by Gal-1, thus suggesting its potential use for immunosuppression [13]. The activity of Smad3 signaling is decreased by intracellular Gal-1 in TGF-β1-treated renal tubular cells, followed by suppression of type I collagen (COL1) transcription [14]. In general, the lectin activity of Gal-1 is observed when it acts extracellularly, whereas the protein–protein interactions of Gal-1 affect its intracellular functions [10]. In the present series of experiments, using cultured human renal epithelial cells, we examined whether a high ambient glucose concentration affects intracellular functions of Gal-1 and

3283

how Gal-1 modulates cross-talk between the Smad3 and MAPK pathways in regulation of COL1.

Materials and methods Reagents and antibodies Recombinant human TGF-β1 was obtained from R&D Systems (Minneapolis, MN, USA). The following antibodies were purchased from the suppliers: goat anti-Gal-1 monoclonal antibody (mAb) (R&D Systems), rabbit polyclonal antibodies to COL1 and rabbit mAb to matrix metalloproteinase 1 (MMP1) (Abcam, Cambridge, MA, USA), and Alexa Fluor 488-conjugated donkey anti-goat IgG and Alexa Fluor 488-conjugated goat anti-rabbit IgG (Molecular Probes, Eugene, OR, USA). PD98059, SB239063, and SIS3

Fig. 1 – Effect of high glucose (HG) on localization of galectin-1 (Gal-1). (A–D) HKC cells were seeded on gelatin-coated coverslips and treated with NG or HG (16.5 mM or 30 mM, respectively) for 72 h with or without transforming growth factor (TGF)-β1 (4 ng/ml) for 24 h. After fixing, the cells were blotted with antibodies against Gal-1 and visualized with appropriate Alexa Fluor 488-conjugated antibodies. (E) Cytoplasmic and nuclear fractions prepared from cells treated as described above were subjected to Western blotting. The graph shows results of densitometric analysis of nuclear Gal-1 levels from three separate experiments. HG significantly increased accumulation of Gal-1 in the nucleus compared to NG (*p < 0.05).

3284

E XP E RI ME N T AL C E L L R E SE A RC H 31 6 ( 20 1 0) 3 2 82 – 3 29 1

(Calbiochem, Darmstadt, Germany) were dissolved in dimethyl sulfoxide (DMSO) and used as stock solutions.

Cell culture and glucose condition Immortalized human renal epithelial cell lines (HKC cells), a kind gift from L. C. Racusen, were cultured as described previously. The cells were maintained with Dulbecco's modified Eagle's medium (DMEM)/F12 (glucose 16.5 mM) supplemented with 10% heatinactivated fetal bovine serum (FBS), penicillin/streptomycin, and amphotericin B [15]. After passage, the cells were cultured in DMEM/F12 medium at a normal glucose (NG, 16.5 mM) or HG concentration (30 mM). Mannitol was used as an osmotic control.

Plasmid constructs We generated a full-length human Gal-1/pcDNA3.1 plasmid construct as described previously [14]. The Smad-binding element-luciferase (SBE-LUC) and the MMP1-LUC reporter constructs were kindly provided by Dr. Vogelstein and Dr. Galloway, respectively [16,17]. The α2(I) collagen (COL1A2)-LUC construct, comprising 376 bp of COL1A2 promoter sequence and 58 bp of the transcribed sequence fused to the LUC reporter gene, was described previously [18]. pFA-Elk1 and pFR-LUC plasmids were purchased from Stratagene (La Jolla, CA, USA).

were transfected with the above-mentioned RNAi duplexes, according to the manufacturer's protocol for RiboJuice siRNA transfection reagent (Novagen). The final concentration of Stealth™ RNAi was 100 nM. A negative control was also purchased and transfected in the same way as described above [19].

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and Western blot Cells were treated as described above and then lysed on ice in RIPA buffer [50 mM Tris–HCl (pH 7.5), 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, and 0.1% SDS] containing protease and phosphatase inhibitors [20]. Cytoplasmic and nuclear extracts were prepared according to the method of Deng et al. [21]. The whole-cell lysate, nuclear, or cytoplasmic extract was resolved by SDS–PAGE, followed by immunoblotting with appropriate antibodies. Nuclear extracts were prepared as described previously [14]. The resulting bands were densitometrically analyzed using the ImageJ 1.42 program for Windows.

Immunofluorescence cytochemistry Cells were replated on coverslips coated with 1 mg/ml gelatin at a density of 5 × 104/well with HG medium containing 0.5% FBS. Twenty-four hours later, the cells were transfected with indicated plasmids and cultured with an HG medium containing 0.5% FBS for 48 h, then treated with or without TGF-β1 (4 ng/ml) for 24 h. Twenty-four hours after TGF-β1 treatment, the cells were fixed with 3.7% formaldehyde followed by 1% Triton-X. They were blotted for the indicated antibodies and were visualized with appropriate Alexa Fluor 488-conjugated antibodies.

Reporter assay The cells were seeded on six-well plates at a density of 5 × 104/well. Twenty-four hours later, transfection was performed using FuGene 6 (Roche, Indianapolis, IN, USA) with the indicated plasmid DNA, together with CMV-SPORT-β-galactosidase (Invitrogen) as a control for transfection efficiency. Five hours after transfection, the cells were cultured in a fresh HG or NG medium containing 0.5% FBS for another 48 h and then stimulated with 4 ng/ml TGF-β1 for an additional 24 h. They were harvested with reporter lysis buffer (Promega, Madison, WI, USA) and LUC activity was measured as described previously [8]. β-Galactosidase activity was assayed using a Promega kit and used to standardize for transfection efficiency.

RNAi transfection Stealth™ RNAi duplexes were synthesized by Invitrogen. RNAi 1 (5′-UUCGUAUCCAUCUGGCAGCUUGACG-3′) and RNAi 2 (5′UUGAAUUCGUAUCCAUCUGGCAGCU-3′) were designed to target different coding regions of the human Gal-1 mRNA sequence (GenBank accession no. NM_002305). Cells were seeded at 5 × 104/well in a six-well plate. Twenty-four hours later, they

Fig. 2 – Effect of Gal-1 on induction of α2(I) collagen (COL1A2) and matrix metalloproteinase 1 (MMP1) transcriptional activities. (A) HKC cells were conditioned for 72 h and then transfected with the COL1A2-luciferase (LUC) reporter construct; the activity induced by 24 h stimulation with 4 ng/ml TGF-β1 was evaluated. The graph shows the mean ± SD of LUC activity corrected for β-galactosidase expression from a representative experiment performed in triplicate. Similar results were obtained from three individual experiments. *p < 0.05 vs. control treated with NG and TGF-β1. (B) Cells were treated as described above and then transfected with the MMP1-LUC reporter construct. The graph shows the mean ± SD of LUC activities corrected for β-galactosidase expression. *p < 0.05 vs. control treated with NG and TGF-β1.

E XP E RI ME N T AL C E L L R E S EA RC H 31 6 ( 20 1 0) 3 2 8 2– 3 29 1

Statistical analyses Data are presented as mean± SD. Differences between the experimental and control groups were evaluated using Student's t test, and a value of p < 0.05 was considered statistically significant.

Results The expression level of Gal-1 increased, and there was a slight increase in the nuclear translocation of Gal-1 in TGF-β1-treated HKC cells (Figs. 1A and B). The HG condition strongly increased the expression levels of Gal-1 (Fig. 1C). The expression level and nuclear translocation of Gal-1 were highly stimulated in HG- and TGF-β1-treated HKC cells (Fig. 1D). To quantitatively evaluate these increases, we investigated the expression level of Gal-1 in the cytoplasm and the nucleus. While TGF-β1 alone has a potential to increase the Gal-1 expression both in the cytoplasm and the nucleus, the HG condition more strongly increases the level of Gal-1 than TGF-β1 alone (Fig. 1E). The HKC cells under HG and TGF-β1 treatment increase the Gal-1 expression 6 times higher in the cytoplasm and twice higher in the nucleus compared to the cells under the NG condition. Next, we investigated transcriptional activities related to renal fibrogenesis using COL1A2 and MMP1 promoter constructs. Overexpressed Gal-1 increases COL1A2 promoter activities in the cells

3285

treated HG alone, while significantly decreased COL1A2 promoter activity compared to control plasmids in both HG and TGF-β1treated cells in either the HG or the NG condition. In the control plasmid group, the HG milieu almost abolished MMP1 promoter activity compared to the NG condition with or without TGF-β1 (Fig. 2A). Overexpression of Gal-1 increased the transcriptional activity of MMP1 in the cells treated HG alone significantly higher than that observed in the control under the same condition. In both HG- and TGF-β1-stimulated cells, overexpression of Gal-1 also rescued MMP1 promoter activity compared to the control plasmid (Fig. 2B). However, the MMP1 reporter activities were decreased in NG- or HG-treated cells with TGF-β1 stimulation compared to those without TGF-β1. These suggest that Gal-1 has an inhibitory role on COL1 production in both HG- and TGF-β1treated cells and that Gal-1 has bifunctional activities on COL1 and MMP1 transcription with HG and/or TGF-β1 stimulation. Fig. 3A shows that overexpressed Gal-1 inhibited COL1 expression, whereas the MMP1 expression level was enhanced compared to the control. There was a significant difference in protein levels of COL1 in Gal-1-overexpressing HKC cells compared to the control (Fig. 3B). Overexpression of Gal-1 showed significant reduction of COL1 expression in no relation to culture media, NG, HG, or MN. The effects of overexpressed Gal-1 were examined by immunofluorescence cytochemistry. Consistent with COL1A2 reporter assay, Gal-1 increased expression levels of COL1 in HG-treated cells without TGF-β1 compared to control plasmid (Figs. 4A and C).

Fig. 3 – Effect of HG on type I collagen (COL1) and MMP1 expression. (A) COL1 and MMP1 expression was assayed by Western blot in HKC cells that were incubated with NG (16.5 mM), HG (30 mM), or mannitol (30 mM) for 72 h, followed by TGF-β1 (4 ng/ml) for 24 h. The bottom panel shows expression levels of Gal-1 with and without overexpression of Gal-1 (control and Gal-1, respectively). (B) The graph shows the relative densities of COL1 or MMP1 as mean ± SD of three independent experiments. *p < 0.05 vs. control treated with NG with or without TGF-β1.

3286

E XP E RI ME N T AL C E L L R E SE A RC H 31 6 ( 20 1 0) 3 2 82 – 3 29 1

Fig. 4 – Effect of Gal-1 on COL1 and MMP1 expressions. (A–D) Cells were plated on gelatin-coated coverslips, conditioned with HG medium for 72 h, and transfected with control or Gal-1 or control plasmid 48 h before fixation. TGF-β1 treatment was performed 24 h before fixation. They were then fixed and stained for COL1. (E–H) Cells were treated as above and stained for MMP1.

The expression levels of MMP1 were also enhanced by overexpression of Gal-1 compared to control plasmid in the HG condition without TGF-β1 (Figs. 4E and G). In both HG- and TGFβ1-treated cells, Gal-1 decreased the expression level of COL1 (Figs. 4B and D) but still increased that of MMP1 (Figs. 4F and H). These results indicate that overexpression of Gal-1 has a negative effect on COL1 production in HKC cells exposed to HG and TGF-β1 and that Gal-1 has a bifunctional role in COL1 production according to various stimulations. Next, we examined the role of endogenous Gal-1 using RNAi. The levels of COL1 were significantly increased by suppression of

Gal-1 expression by two independent Stealth™ RNAi duplexes, RNAi 1 and RNAi 2 (Fig. 5A, upper panel, and Fig. 5B). On the contrary, expression levels of MMP1 were significantly decreased by suppressing Gal-1 (Fig. 5A, middle panel). These data suggested that endogenous Gal-1 has a negative effect on COL1 production but increases MMP1 production in HG- and TGF-β1-treated HKC cells. We also investigated which signaling cascades are involved in the suppression of COL1 production by Gal-1. Overexpression of Gal-1 decreased the fold induction of Smad3-mediated reporter activity; however, there was no significant difference between the

E XP E RI ME N T AL C E L L R E S EA RC H 31 6 ( 20 1 0) 3 2 8 2– 3 29 1

3287

Fig. 5 – Role of endogenous Gal in COL1 and MMP1 expression. (A) Two different, double-stranded Gal-1 RNAi were transfected to the HKC cells that were seeded on six-well plates and conditioned with HG for 72 h and TGF-β1 for 24 h. Expression levels of Gal-1, COL1, and MMP1 were assayed by Western blotting. (B) Relative densities of COL1 and MMP1 are shown as mean ± SD of three independent experiments. *p < 0.01 vs. control RNAi.

control and the Gal-1-overexpressing group in HKC cells treated with HG and TGF-β1 (Fig. 6A). Elk1-LUC reporter activity was significantly increased by overexpression of Gal-1 compared to the control (Fig. 6B). Gal-1 has a potential to activate ERK MAPKmediated reporter activities in HG-stimulated cells with or without TGF-β1. These results indicate that Gal-1 plays an inhibitory role in Smad3 signaling but a stimulatory role in ERK MAPK signaling in HG- and TGF-β1-treated HKC cells. Further, it is suggested that MAPK pathways have a more important role on COL1 transcription in HG-stimulated renal cells. To investigate further mechanisms, we employed specific inhibitors against these signaling molecules. COL1A2-LUC activities were augmented by inhibitors of ERK or p38 MAPK (PD or SB, respectively) compared to the control (DMSO) in cells without Gal-1 overexpression (Fig. 7A). With Gal-1 overexpression, treatment with an inhibitor of ERK MAPK significantly increased COL1A2-LUC activity compared to the control plasmid group treated with the same inhibitor; however, there was no difference between the two groups treated with an inhibitor of p38 with or without Gal-1 overexpression. Treatment with SIS3 strongly inhibited COL1A2LUC activities, and there was no obvious difference between the two groups with and without Gal-1 overexpression. These results indicate that the inhibitory effect of Gal-1 on COL1A2 transcription is mainly mediated by the Smad3 and ERK MAPK pathways under HG and TGF-β1 stimulation (Fig. 8A). MMP1-LUC reporter activity was decreased by blockade of the ERK MAPK pathway but increased by blockade of the p38 MAPK pathway in cells with

controlled plasmid overexpression (Fig. 7B). Gal-1 overexpression increased these effects, particularly in the p38 MAPK pathway. The data suggest that activation of the p38 MAPK pathway by Gal-1 negatively regulates MMP1 transcription in cells under HG and TGF-β1 stimulation, although Gal-1 itself plays a positive role in MMP1 transcription (Fig. 8B). Taken together, we concluded that Gal-1 affects two MAPK pathways in HG-stimulated renal epithelial cells, resulting in a decrease in COL1 expression and an increase in MMP1 expression.

Discussion Although the Smad3 pathway plays a central role in ECM production by renal cells in the presence of HG and TGF-β1, a variety of signaling molecules, such as components of the ERK MAPK pathway, regulate the mechanism. Gal-1 has numerous antiinflammatory effects [22]. Recently, Shimizu et al. [23] reported that Gal-1 activates the ERK 1/2 MAPK pathway in podocytes, leading to alteration of the slit diaphragm. This suggests that Gal-1 regulates the ERK MAPK pathway in the progression of renal fibrosis. To our knowledge, this is the first report demonstrating the intracellular effect of Gal-1 on COL1 production by renal cells treated with HG and TGF-β1. Either HG or TGF-β1 alone can activate the Smad3 signaling pathway. Renal fibrogenesis in the diabetic kidney is chronic, and the alteration of renal tissues becomes apparent

3288

E XP E RI ME N T AL C E L L R E SE A RC H 31 6 ( 20 1 0) 3 2 82 – 3 29 1

Fig. 6 – Effect of Gal-1 on Smad3- and extracellular signal-regulated kinase (ERK)-mediated reporter activities. (A) Cells were conditioned with HG for 72 h with or without TGF-β1 for 24 h and then transfected with Smad-binding element-LUC promoter constructs. The graph shows fold induction of activity over control (mean ± SD, n = 3) in each condition. NS indicates no significant difference between control plasmids and Gal-1 with HG and TGF-β1 treatment. (B) Transfection was performed under the conditions described above, except that the reporter construct was Elk1. Data are mean ± SD from triplicate samples. *p < 0.01 vs. control with HG and TGF-β1.

approximately a decade after the onset of diabetes. Although the diabetic milieu could enhance Smad3 activity and partly increase TGF-β1 expression, the long-term effects of HG on the relatively transient activity of Smad3 are less clear. To address this issue, we examined Smad responses to a single stimulation with TGF-β1 after HG conditioning [20]. First, we found that total levels of Gal-1 expression were strongly enhanced and that the nuclear translocation of Gal-1 increased in renal epithelial cells treated with HG and TGF-β1. We also showed that Gal-1 activated the MAPK pathways in HG- and TGF-β1-stimulated HKC cells, resulting in a decrease in COL1 transcription and expression. Since TGF-β1 is an inducer of intracellular Gal-1 in mammary adenocarcinoma cells [24], the

Fig. 7 – COL1 and MMP1 transcriptional activities HG and specificity of mitogen-activated protein kinase (MAPK) or Smad3 inhibitors. (A) Cells were treated with HG (72 h) and TGF-β1 (24 h), followed by various inhibitors for 24 h. Transfection with COL1A2 promoter constructs was performed 5 h before adding TGF-β1. The graph shows fold induction of activity compared to control (mean ± SD, n = 3) in each condition. NS indicates no significant difference between the control and Gal-1 groups with the same treatment. *p < 0.01 vs. control plasmid group with the same treatment. (B) Transfection was performed under the conditions described above, except that the reporter construct was MMP1. Data are mean ± SD from triplicate samples. NS indicates no significant difference between the control and Gal-1 groups with the same treatment. *p < 0.01 vs. control group with the same treatment.

increase in Gal-1 expression was strongly enhanced by both HG and TGF-β1 stimulation. Several reports showed collaboration between the Smad3 and MAPK pathways in accumulation of ECM in the kidney [20,25,26]. Our results indicate that Gal-1 participates in the cross-talk between the Smad3 and MAPK pathways. One possible mechanism is that growth-inhibitory or antiinflammatory potential is dominant with a high concentration of Gal-1 [27]. A relatively high concentration of Gal-1, particularly in the nucleus, may inhibit the binding of Smad3 to COL1 promoter regions such as the SBE motif [14]. Another explanation is that activation of the MAPK pathway by Gal-1 itself prevents ECM production by Smad3 in HG- and TGF-β1-stimulated renal cells. In light of our results, we believe that the second scenario is more likely. In the HG condition, activation of MAPK pathways by Gal-1 seems more dominant than that of Smad3 by Gal-1 because addition of TGF-β1 showed no significant difference in SBE-LUC activities between Gal-1 and control plasmid in HG-treated renal cells. Blockade of the ERK MAPK pathway by a specific inhibitor increases COL1A2 promoter activities, suggesting that activated ERK MAPK itself plays a negative role in COL1A2 transcription by the TGF-β1/Smad3 pathway. However, activation of MAPK by

E XP E RI ME N T AL C E L L R E S EA RC H 31 6 ( 20 1 0) 3 2 8 2– 3 29 1

3289

Fig. 8 – Possible mechanisms underlying the protective effect of Gal-1 against HG- and TGF-β1-induced renal fibrogenesis. (A) Gal-1 inhibits the Smad3 pathway and activates the ERK MAPK pathway, resulting in a decreased COL1 transcription and expression in HG- and TGF-β1-treated renal epithelial cells. (B) Gal-1 inhibits the Smad3 pathway and activates the p38 MAPK pathway, resulting in an increased MMP1 transcription and expression in HG- and TGF-β1-treated renal epithelial cells.

Gal-1 in the HG condition augments COL1A2 transcription compared to NG, suggesting Gal-1 may have bifunctional activities on ECM regulation. Gal-1 is reported to have a similar character with TGF-β1 in having positive and negative effect on growthregulation [28]. Gal-1 is reported as an activator of MAPK pathway regardless of TGF-β1 or HG stimulation [29]. Gal-1 has strong immunoregulatory effects through its ability to inhibit T-cell effecter function [30]. These suggest that Gal-1 has a pivotal role in regulation of renal fibrogenesis by Gal-1. It is left for future study whether bifunctional role of Gal-1 in renal fibrosis is mediated mainly through MAPK pathways in the HG condition. Blockade of ERK or p38 MAPK pathways diminishes the inhibitory effect of Gal-1 in the HG condition, suggesting that activation of two MAPK pathways by Gal-1 has a critical role in COL1 transcription. On the other hand, the effect of Gal-1 on MMP1 transcriptional activities is somewhat different from COL1 transcription. MMP1 transcriptional activities are decreased by Gal-1 in the NG condition with or without TGF-β1, while increased in the HG condition. This raises a possibility that contribution of Gal-1 on COL1 and MMP1 expression is more dependent on glucose concentration than TGF-β1, while the effect of Gal-1 on COL1 expression seems dependent on TGF-β1. Further study is required to know precise mechanisms. Another possibility is that inhibitory effect of Gal-1 on renal fibrosis is more dependent on decrease of COL1 transcription than on increase of MMP1 transcription in NG and/or HG condition.

Another novel finding in the present study was that the ERK and p38 MAPK pathways made complementary contribution to ECM production by renal cells. In a unilateral ureteral obstruction model, treatment with another p38 MAPK inhibitor, FR167653, caused a marked decrease in renal fibrosis, accompanied with decrease in the COL1A2 mRNA level [31]. Our results show that the effect of blockade of the p38 MAPK pathway by SB239063 on COL1A2 promoter activities did not differ significantly between the control and Gal-1 plasmid groups. This can be explained by the difference between the experimental models. Furthermore, different fragments of α collagen may be regulated differently. However, COL1 expression is decreased by Gal-1, suggesting that the difference between the two studies results from the difference in the experiment models, such as ischemic vs. diabetic models. To our knowledge, there is no report demonstrating that the p38 MAPK pathway regulates MMP1 production in HG-treated renal cells. The p38 MAPK pathway contributes to MMP1 expression in primary human macrophages, rheumatoid arthritis synovial fibroblasts, decidual cells, and human neutrophils [32– 35]. In Alport model mice, other MMPs, such as MMP2 and MMP9, are regulated by the p38 MAPK but not by the ERK MAPK pathway [36]. This suggests that MMPs are mainly regulated by the p38 MAPK pathway. Our results suggest that p38 MAPK is a more important regulator of MMP1 in renal fibrosis than ERK MAPK;

3290

E XP E RI ME N T AL C E L L R E SE A RC H 31 6 ( 20 1 0) 3 2 82 – 3 29 1

however, ERK MAPK pathway plays a critical role in COL1 production in HG- and TGF-β1-treated HKC cells. In conclusion, we demonstrated that Gal-1 plays an important role in prevention of ECM production in HG-treated renal epithelial cells. Two MAPK pathways, p38 and ERK, are stimulated by increasing levels of intracellular Gal-1 in the HG condition, leading to decreasing levels of COL1 expression and increasing levels of MMP1 expression. Optimization of the variables for therapy with Gal-1 could give rise to new possibilities for the treatment of renal fibrosis.

[16]

[17]

[18]

REFERENCES [19] [1] US Renal Data System, USRDS 2004 annual data report, Am. J. Kidney Dis. 45 (2005) 8–280. [2] S. Adler, Structure-function relationships associated with extracellular matrix alteration in diabetic glomerulopathy, J. Am. Soc. Nephrol. 5 (1994) 1165–1172. [3] A. Nakao, T. Imamura, S. Souchelnyskyi, M. Kawabata, A. Ishisaki, E. Oeda, K. Tamaki, J. Hanai, C.H. Heldin, K. Miyazono, P. ten Dijke, TGF-beta receptor-mediated signaling through Smad2, Smad3 and Smad4, EMBO J. 16 (1997) 5353–5362. [4] J.H. Li, X.R. Huang, H.J. Zhu, R. Johnson, H.Y. Lan, Role of TGF-beta signaling in extracellular matrix production under high glucose conditions, Kidney Int. 63 (2003) 2010–2019. [5] K. Sharma, Y. Jin, J. Guo, F.N. Ziyadeh, Neutralization of TGF-b antibody attenuates kidney hypertrophy and the enhanced extracellular matrix gene expression in STZ-induced diabetic mice, Diabetes 45 (1996) 522–530. [6] D. Ostalska-Nowicka, J. Zachwieja, M. Nowicki, E. Kaczmarek, A. Siwinska, M. Witt, Immunohistochemical detection of galectin-1 in renal biopsy specimens of children and its possible role in proteinuric glomerulopathies, Histopathology 51 (2007) 468–476. [7] A. Burger, S. Filsinger, D.N. Cooper, G.M. Hansch, Expression of the 14 kDa galactose-binding protein, galectin-1, on human tubular epithelial cells, Kidney Int. 50 (1996) 754–759. [8] Y. Tsuchiyama, J. Wada, H. Zhang, Y. Morita, K. Hiragushi, K. Hida, K. Shikata, M. Yamamura, Y.S. Kanwar, H. Makino, Efficacy of galectins in the amelioration of nephrotoxic serum nephritis in Wister Kyoto rats, Kidney Int. 58 (2000) 1941–1952. [9] M.J. Perone, S. Bertera, W.J. Shufesky, S.J. Divito, A. Montecalvo, A.R. Mathers, A.T. Larregina, M. Pang, N. Seth, K.W. Wucherpfennig, M. Trucco, L.G. Baum, A.E. Morelli, Suppression of autoimmune diabetes by soluble galectin-1, J. Immunol. 182 (2009) 2641–2653. [10] I. Camby, M. Le Mercier, F. Lefranc, R. Kiss, Galectin-1: a small protein with major functions, Glycobiology 16 (2006) 137R–157R. [11] A. Paz, R. Haklai, G. Elad, E. Ballan, Y. Kloog, Galectin-1 binds oncogenic H-Ras to mediate Ras membrane anchorage and cell transformation, Oncogene 20 (2000) 7486–7493. [12] G. Elad-Sfadia, R. Haklai, E. Ballan, H.J. Gabius, Y. Kloog, Galectin-1 augments Ras activation and diverts Ras signals to Raf-1 at the expense of phosphoinositide 3-kinase, J. Biol. Chem. 277 (2002) 37169–37175. [13] L.G. Baum, D.P. Blackall, S. Arias-Magallano, D. Nanigian, S.Y. Uh, J.M. Browne, D. Hoffmann, C.E. Emmanouilides, M.C. Territo, G.C. Baldwin, Amelioration of graft versus host disease by galectin-1, Clin. Immunol. 109 (2003) 295–307. [14] K. Okano, K. Uchida, K. Nitta, T. Hayashida, Galectin-1 suppresses alpha2(I) collagen through Smad3 in renal epithelial cells, Cell. Mol. Life Sci. 65 (2008) 3304–3311. [15] L.C. Racusen, C. Monteil, A. Sgrignoli, M. Lucskay, S. Marouillat, J.G. Rhim, J.P. Morin, Cell lines with extended in vitro growth

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27] [28]

[29]

[30]

[31]

potential from human renal proximal tubule: characterization, response to inducers, and comparison with established cell lines, J. Lab. Clin. Med. 129 (1997) 318–329. L. Zawal, J.L. Dai, P. Buckhaults, S. Zhou, K.W. Kinzler, B. Vogelstein, S.E. Kern, Human Smad3 and Smad4 are sequence-specific transcription activators, Mol. Cell 1 (1998) 1611–1617. J.A. Benanti, D.K. Williams, K.L. Robinson, H.L. Ozer, D.A. Galloway, Induction of extracellular matrix-remodeling genes by the senescence-associated protein APA-1, Mol. Cell. Biol. 22 (2002) 7385–7397. A.C. Poncelet, M.P. De Caestecker, H.W. Schnaper, The transforming growth factor-beta/SMAD signaling pathway is present and functional in human mesangial cells, Kidney Int. 56 (1999) 1354–1365. H. Xiao, Z. Wu, H. Shen, A.L. Luo, Y.F. Yang, X.B. Li, D.Y. Zhu, In vivo reversal of P-glycoprotein-mediated multidrug resistance by efficient delivery of Stealth™ RNAi, Basic Clin. Pharmacol. Toxicol. 103 (2008) 342–348. T. Hayashida, H.W. Schnaper, High ambient glucose enhances sensitivity to TGF-β1 via extracellular signal-regulated kinase and protein kinase Cδ activates in human mesangial cells, J. Am. Soc. Nephrol. 15 (2004) 2032–2041. W.G. Deng, Y. Zhu, A. Montero, K.K. Wu, Quantitative analysis of binding of transcription factor complex to biotinylated DNA probe by a streptavidin–agarose pulldown assay, Anal. Biochem. 323 (2003) 12–18. M.J. Perone, S. Bertera, W.J. Shufesky, S.J. Divoto, A. Montecalvo, A.R. Mathers, A.T. Larregina, M. Pang, N. Seth, K.W. Wucherpfennig, M. Trucco, L.G. Baum, A.E. Morelli, Suppression of autoimmune diabetes by soluble galectin-1, J. Immunol. 182 (2009) 2641–2653. M. Shimizu, J. Khoshnoodi, Y. Akimoto, H. Kawakami, H. Hirano, E. Higashihara, M. Hosoyamada, Y. Sekine, R. Kurayama, K. Joh, J. Hirabayashi, K. Kasai, K. Tryggvason, N. Ito, K. Yan, Expression of galectin-1, a new component of slit diaphragm, is altered in minimal change nephritic syndrome, Lab. Invest. 89 (2009) 178–195. C.M. Daroqui, J.M. Ilarregui, N. Rubinstein, M. Salatino, M.A. Toscano, P. Vazquez, A. Bakin, L. Puricelli, E.B. de Kier Joffe, G.A. Rabinovich, Regulation of galectin-1 expression by transforming growth factor b1 in metastatic mammary adenocarcinoma cells: implications for tumor-immune escape, Cancer Immunol. Immunother. 56 (2007) 491–499. M.P. de Caestecker, E. Peik, A.B. Roberts, Role of transforming growth factor-β signaling in cancer, J. Natl Cancer Inst. 92 (2000) 1388–1402. M. Awazu, K. Ishikura, M. Hida, M. Hoshiya, Mechanisms of mitogen-activated protein kinase cascade is activated in glomeruli of diabetes, J. Am. Soc. Nephrol. 10 (1999) 738–745. K. Scott, C. Weinberg, Galectin-1: a bifunctional regulator of cellular proliferation, Glycoconj. J. 19 (2004) 467–477. H.L. Moses, E.Y. Yang, J.A. Pietenpol, TGF-beta stimulation and inhibition of cell proliferation: new mechanistic insights, Cell 63 (1990) 245–247. A. Matsuda, Y. Suzuki, G. Honda, S. Muramatsu, O. Matsuzaki, Y. Nagano, T. Doi, K. Shimotohno, T. Harada, E. Nishida, H. Hayashi, S. Sugano, Large-scale identification and characterization of human genes that activate NF-kB and MAPK signaling pathways, Oncogene 22 (2003) 3307–3318. J. van der Leij, A. van den Berg, T. Blokzijl, G. Harms, H. van Goor, P. Zwiers, R. van Weeghel, S. Poppema, L. Visser, Dimeric galectin-1 induces IL-10 production in T-lymphocytes; an important tool in the regulation of the immune response, J. Pathol. 204 (2004) 511–518. M. Nishida, Y. Okumura, H. Sato, K. Hamaoka, Delayed inhibition of p38 mitogen-activated protein kinase ameliorates renal fibrosis in obstructive nephropathy, Nephrol. Dial. Transplant. 23 (2008) 2520–2524.

E XP E RI ME N T AL C E L L R E S EA RC H 31 6 ( 20 1 0) 3 2 8 2– 3 29 1

[32] L. Rand, J.A. Green, L. Saraiva, J.S. Friedland, P.T. Elkington, Matrix metalloproteinase-1 is regulated in tuberculosis by a p38 MAPK-dependent, p-aminosalicylic acid-sensitive signaling cascade, J. Immunol. 182 (2009) 5865–5872. [33] E.M. Noh, J.S. Kim, H. Hur, B.H. Park, E.K. Song, M.K. Han, K.B. Kwon, W.H. Yoo, I.K. Shim, S.J. Lee, H.J. Youn, Y.R. Lee, Cordycepin inhibits IL-1beta-induced MMP-1 and MMP-3 expression in rheumatoid arthritis synovial fibroblasts, Rheumatology 48 (2009) 45–48. [34] C. Oner, F. Schatz, G. Kizilay, W. Murk, L.F. Buchwalder, U.A. Kayisli, A. Arici, C.J. Lockwood, Progression-inflammatory cytokine interactions affect matrix metalloproteinase-1 and -3

3291

expression in term decidual cells, J. Clin. Endocrinol. Metab. 93 (2008) 252–259. [35] S.M. Oh, D.H. Hahm, I.H. Kim, S.Y. Choi, Human neutrophil lactoferrin trans-activates the matrix metalloproteinase 1 gene through stress-activated MAPK signaling modules, J. Biol. Chem. 276 (2001) 42575–42579. [36] D. Cosqrove, D.T. Meehan, D. Delimont, A. Pozzi, X. Chen, K.D. Rodgers, R.M. Tempero, M. Zallocchi, V.H. Rao, Integrin alpha1beta1 regulates matrix metalloproteinases via P38 mitogen-activated protein kinase in mesangial cells: implications for Alport syndrome, Am. J. Pathol. 172 (2008) 761–773.