High glucose-induced thioredoxin-interacting protein in renal proximal tubule cells is independent of transforming growth factor-beta1

Abstract

Hyperglycemia is a causative factor in the pathogenesis of diabetic nephropathy. Here, we demonstrate the transcriptional profiles of the human proximal tubule cell line (HK-2 cells) exposed to high glucose using cDNA microarray analysis. Thioredoxin-interacting protein (Txnip) was the gene most significantly increased among 10 strongly up-regulated and 15 down-regulated genes. Txnip, heat shock proteins 70 and 90, chemokine (C-C motif) ligand 20, and matrix metalloproteinase-7 were chosen for verification of gene expression. Real-time reverse transcriptase-polymerase chain reaction confirmed the mRNA expression levels of these five genes, consistent with microarray analysis. The increased protein expression of Txnip, CCL20, and MMP7 were also verified by Western blotting and enzyme-linked immunosorbent assay. Increased expression of Txnip and of nitrotyrosine, as a marker of oxidative stress, were confirmed in vivo in diabetic Ren-2 rats. Subsequent studies focused on the dependence of Txnip expression on up-regulation of transforming growth factor (TGF)-beta1 under high-glucose conditions. Overexpression of Txnip and up-regulation of Txnip promoter activity were observed in cells in which the TGF-beta1 gene was silenced in HK-2 cells using short interfering RNA technology. High glucose further increased both Txnip expression and its promoter activity in TGF-beta1 silenced cells compared with wild-type cells exposed to high glucose, suggesting that high glucose induced Txnip through a TGF-beta1-indepen-dent pathway.

Figure 1

High glucose up-regulated Txnip expression in HK-2 cells. HK-2 cells were exposed to 5 mmol/L (normal glucose) and 30 mmol/L (high glucose) d-glucose for 11 days. Thirty mmol/L l-glucose acts as osmotic control. A: Real-time RT-PCR was performed, and Txnip mRNA expression was normalized to the housekeeping gene β-actin. Txnip protein expression was measured by Western blotting, and Coomassie Brilliant Blue staining was used as equal loading control. The experiments were repeated three times, with a representative blot shown (B), and the densitometry of Western blotting is quantified (C). D: HK-2 cells were exposed to 5 mmol/L (normal glucose) and 30 mmol/L (high glucose) d-glucose for 1, 2, 4, and 11 days. Txnip and β-actin real-time RT-PCR was performed. Results are mean ± SEM and shown as fold change compared with control. ***P < 0.0005, **P < 0.005, *P < 0.05; n = 3.

Figure 2

High glucose significantly regulated mRNA and protein expression of HSP70, HSP90, CCL20, and MMP7 after HK-2 cells were exposed to high glucose for 11 days. HK-2 cells were exposed to 5 mmol/L (normal glucose) and 30 mmol/L (high glucose) d-glucose for 11 days. Thirty mmol/L l-glucose acts as osmotic control. Real-time RT-PCR was performed, and HSP70 (A), HSP90 (B), CCL20 (C), and MMP7 (E) mRNA expressions were normalized to the housekeeping gene β-actin. Supernatant was collected as described in Materials and Methods, and CCL20 (D) and MMP7 (F) proteins were measured using enzyme-linked immunosorbent assay. Results are mean ± SEM and shown as fold change compared with control. ***P < 0.0005, **P < 0.005; n = 3.

Figure 3

Nitrotyrosine and Txnip expression in nondiabetic and diabetic Ren-2 rats. Diabetic Ren-2 rats (II) were associated with an increase in nitrotyrosine (A) and Txnip (B) immunostaining within the tubules when compared with nondiabetic Ren-2 rats (I). Positive staining of nitrotyrosine (ROS) (C) and Txnip (D) in the dilated tubules in nondiabetic and diabetic Ren-2 rats was counted separately and normalized to the number of tubules counted in each chosen field. The data were expressed as percentage of positive staining of ROS or Txnip in dilated tubules per total tubules. Txnip protein in whole rat kidney tissue was measured by Western blotting, and Coomassie Brilliant Blue staining was used as equal loading control (E and F). Results are mean ± SEM and shown as fold change compared with control. ***P < 0.0005, **P < 0.005, *P < 0.05; n = 3. Original magnifications, ×340 (A and B).

Figure 4

The effect of TGF-β1 on Txnip expression in HK-2 cells. HK-2 cells were exposed to 10 ng/ml TGF-β1 for 0, 1, 2, 4, and 6 days. A: RNA was collected for measurement of Txnip by real-time RT-PCR and normalized to the housekeeping gene β-actin. B and C: Txnip protein was measured by Western blotting, and Coomassie Brilliant Blue staining was used as equal loading control. Results are mean ± SEM and shown as fold change compared with control. ***P < 0.0005; n = 3.

Figure 5

The role of TGF-β1 on Txnip expression under high-glucose conditions. Thirty nmol/L of nonspecific or TGF-β1 siRNAs were introduced into HK-2 cells, respectively, at subconfluence using Lipofectamine 2000. Cells were then treated with or without 30 mmol/L d-glucose (HG) for 72 hours. A and D: RNA was collected for measurement of Txnip mRNA expression. TGF-β1-silenced cells were either exposed to 30 μg/ml of pan-specific TGF-β antibody (aT) or 30 μg/ml rabbit IgG (negative control) or treated with or without HG. B, C, E, and F: Txnip protein was measured by Western blotting, and Coomassie Brilliant Blue staining was used as equal loading control. Results are mean ± SEM and are shown as fold change. ***P < 0.0005, **P < 0.005, *P < 0.05, ^###P < 0.0005, ^##P < 0.005, ^#P < 0.05; n = 3.

Figure 6

The role of TGF-β1 on Txnip promoter activity under high-glucose conditions. HK-2 cells were 85% subconfluent when seeded on a 24-well plate and transfected with 2 μg of Txnip promoter pGL3 (Firefly) and 1 ng of pRL-SV40 (Renilla) using Lipofectamine 2000. A: Cells were maintained in medium containing 5 and 30 mmol/L d-glucose for 24, 48, and 72 hours after 6 hours of transfection. Cell lysate was then collected for Txnip promoter assay. B: HK-2 cells were exposed to 10 ng/ml TGF-β1 for 0, 4, 8, 16, 24, and 48 hours, and cell lysate was then collected for Txnip promoter assay. Thirty nmol/L of nonspecific or TGF-β1 siRNAs were introduced into HK-2 cells at subconfluence using Lipofectamine 2000. C: Cells were then treated with or without 30 mmol/L d-glucose (HG) for 24 hours, and cell lysate was then collected for Txnip promoter assay. Firefly luciferase activity was normalized to Renilla, which serves as internal control. All results are mean ± SEM and are shown as fold change. ***P < 0.0005, **P < 0.005, *P < 0.05, ^#P < 0.05; n = 3.