p66Shc mediates high-glucose and angiotensin II-induced oxidative stress renal tubular injury via mitochondrial-dependent apoptotic pathway

Abstract

p66Shc, a promoter of apoptosis, modulates oxidative stress response and cellular survival, but its role in the progression of diabetic nephropathy is relatively unknown. In this study, mechanisms by which p66Shc modulates high-glucose (HG)- or angiotensin (ANG) II-induced mitochondrial dysfunction were investigated in renal proximal tubular cells (HK-2 cells). Expression of p66Shc and its phosphorylated form (p-p66Shc, serine residue 36) and apoptosis were notably increased in renal tubules of diabetic mice, suggesting an increased reactive oxygen species production. In vitro, HG and ANG II led to an increased expression of total and p-p66Shc in HK-2 cells. These changes were accompanied with increased production of mitochondrial H(2)O(2), reduced mitochondrial membrane potential, increased translocation of mitochondrial cytochrome c from mitochondria into cytosol, upregulation of the expression of caspase-9, and ultimately reduced cell survival. Overexpression of a dominant-negative Ser36 mutant p66Shc (p66ShcS36A) or treatment of p66Shc- or PKC-β-short interfering RNAs partially reversed these changes. Treatment of HK-2 cells with HG and ANG II also increased the protein-protein association between p-p66Shc and Pin1, an isomerase, in the cytosol, and with cytochrome c in the mitochondria. These interactions were partially disrupted with the treatment of PKC-β inhibitor or Pin1-short interfering RNA. These data suggest that p66Shc mediates HG- and ANG II-induced mitochondrial dysfunctions via PKC-β and Pin1-dependent pathways in renal tubular cells.

Fig. 1.

A: expression of p66Shc and phosphorylated p66Shc (at Ser36 residue) (p-p66Shc; a–h), status of apoptosis (i–l), and oxidant stress (m–p) in kidneys of control CD1 (ICR) mice (column 1: a, e, i, and m) and those with streptozotocin (STZ)-induced diabetes mice (column 2: b, f, j, and n), db/m mice (column 3: c, g, k, and o), and db/db (column 4: d, h, l, and p) mice. In control mice (ICR and db/m), p66Shc and p-p66Shc expression mainly localized in the proximal tubules, and it was increased in mice with STZ-induced diabetes and db/db mice. The terminal deoxynucleotidyl transferase dUTP-mediated nick-end labeling (TUNEL) staining to mark the apoptotic nuclei also notably increased in kidney sections of diabetic mice, compared with the controls (i–l). Similarly, dihydroethidine (DHE) staining to assess the ROS generation showed increased staining in the cortical tubular cells of STZ-induced and db/db mice (m–p). Magnifications: ×200. B and C: Western blot (WB) analyses showed that both p66Shc and p-p66Shc expression were increased in the cortical kidney homogenates of diabetic mice. D and E: densitometry of the autoradiographic bands were depicted in B and C, respectively. Each bar graph represents a relative density ratio between p66Shc or p-p66Shc to β-actin. Values are means ± SE; N = 20 in each group. *P < 0.01 vs. control (Con).

Fig. 2.

Expression of p66Shc and p-p66Shc in HK-2 cells following exposure to high glucose (HG; 5–45 mM) and angiotensin II (ANG II; 10⁻⁹-10⁻⁵ M), as assessed by WB analysis (A1–A3, B1–B3, C1–C3, D1–D3) and real-time PCR (A4, B4, C4, D4). Expression of both p66Shc and p-p66Shc increased in a time- and dose-dependent manner in HK-2 cells treated with d-glucose (A and B) and ANG II (C and D). Expression of β-actin was unchanged. The bar graphs (columns 2, 3, and 4) represent the expression of p66Shc and p-p66Shc relative to β-actin. Values are means ± SE; N = 6. *P < 0.05, #P < 0.01 vs. control.

Fig. 3.

A1–A3: expression of p66Shc and p-p66Shc in HK-2 cells subjected to HG and ANG II treatments or individually transfected with wild-type (WT) p66Shc, p66ShcS36A (mutant), p66Shc-short interfering RNA (siRNA), or PKC-β-siRNA. The HG and ANG II treatment or overexpression of p66Shc significantly increased the expression of both p66Shc and p-p66Shc. The overexpression of p66ShcS36A mutant increased the expression of p66Shc, but not of its phosphorylated form, p-p66Shc. Transfection of empty vector or p66Shc-siRNA or PKC-β-siRNA had no significant effect on both p66Shc and p-p66Shc expression. B1–B3: effect of p66Shc, p66ShcS36A, p66Shc siRNA, or of PKC-β-siRNA on the expression of p66Shc and p-p66Shc in HK-2 cell that were treated with HG or ANG II. The p66Shc and p-p66Shc protein expression was increased in HK-2 cell treated with HG and ANG II, and this effect was abolished following transfection with p66Shc-siRNA or PKC-β-siRNA. The transfection of p66ShcS36A had no effect on the expression of p66Shc compared with HG or ANG II-treated groups, but it reduced the expression of p-p66Shc. Overexpression of p66Shc significantly increased the p66Shc and p-p66Shc protein expression, which was reduced following PKC-β-siRNA transfection. The bar graphs represent the densitometric measurements of the protein expression. Values are means ± SE; N = 6. *P < 0.01 vs. 5 mM d-glucose [low glucose (LG)]. @ P < 0.05 vs. 30 mM d-glucose. #P < 0.01 vs. 30 mM d-glucose (HG). & P < 0.01 vs. ANG II (10⁻⁷ M). $ P < 0.01 vs. p66Shc WT.

Fig. 4.

A, rows 1–3: confocal microscopy images of cells subjected to dichlorofluorescein-diacetate (DCFH-DA), MitoSOX, and TUNEL staining, and quantification of apoptosis and mitochondrial membrane potential (Δψm). By DCFH-DA staining, overexpression of p66ShcS36A, or treatment of PKC-β-siRNA attenuated intracellular reactive oxygen species (ROS) generation in HK-2 induced by HG (30 mM) or ANG II (10⁻⁷ M), while treatment of PKC-β-siRNA reduced ROS production in HK-2 induced by p66Shc serving as positive control (row 1). Similar results were seen upon staining with MitoSOX; however, there was no change observed in cells transfected with p66ShcS36A or p66Shc siRNA (row 2). TUNEL staining showed that HG or ANG II increased the degree of apoptosis, which was reduced with the transfection of p66ShcS36A or treatment of PKC-β-siRNA (row 3). B and C: the bar graphs represent a summary of flow cytometry analyses of cells stained with DCFH-DA and MitoSOX. D: the bar graph represents the quantification of apoptosis in HK-2 cells by TUNEL assay. E: Δψm as measured by tetramethylrhodamine ethylester perchlorate (TMRE) staining revealed a decrease of Δψm in HK-2 cells following HG and ANG II treatments, which was restored with the overexpression of p66ShcS36A or treatment of PKC-β-siRNA. Values are means ± SE; N = 6. *P < 0.01 vs. 5 mM d-glucose. #P < 0.01 vs. 30 mM d-glucose. & P < 0.01 vs. ANG II (10⁻⁷ M). $ P < 0.01 vs. p66Shc WT.

Fig. 5.

Effect of p66Shc on the HG- or ANG II-induced cytochrome c (Cyt.C) release. A: confocal microscopy revealed that, at a basal low concentration of d-glucose (LG, 5 mM), the Cyt.C is localized in the mitochondria (mCyt.C), as detected by anti-mCyt.C antibody (green, column 2) or Mitotracker dye (red, column 3). With the treatment of 30 mM d-glucose (HG) or ANG II (10⁻⁷ M), the Cyt.C was released into cytosol, and the release was inhibited by transfection of p66ShcS36A or PKC-β-siRNA. B: distribution of p-p66Shc, Cyt.C, and caspase-9 in mitochondrial vs. cytoplasmic compartments following various treatments, as assessed by WB analysis, with mitochondrial heat shock protein 70 (mHSP70) or GAPDH served as loading controls. C–F: the bar graphs represent a summary of WBs reflecting expression of mCyt.C (C) cytosolic Cyt.C (E), p-p66Shc (D), and caspase-9 (F). The HG or ANG II induced a decrease of mCyt.C and increase of p-p66Shc, cytosolic Cyt.C, and caspase-9, while overexpression of p66ShcS36A mutant or treatment of PKC-β-siRNA reversed these changes. Transfection of p66Shc WT had similar effects, as induced by HG or ANG II treatments on the expression of Cyt.C, p-p66Shc, and caspase-9, and they were inhibited by PKC-β-siRNA. DAPI, 4,6-diamidino-2-phenylindole. Values are means ± SE; N = 6. *P < 0.01 vs. control of 5 mM d-glucose. #P < 0.01 vs. 30 mM d-glucose. & P < 0.01 vs. ANG II (10⁻⁷ M). $ P < 0.01 vs. p66SShc WT.

Fig. 6.

A: status of mitochondrial DNA (mtDNA) in HK-2 cells exposed to HG or ANG II. HG or ANG II treatment led to a reduction in the expression of high-molecular-weight DNA (8,636 bp), while low-molecular-weight DNA (420 bp) was unaffected. These changes were normalized by the overexpression of p66ShcS36A or treatment of PKC-β-siRNA. The PKC-β-siRNA partially reversed the p66Shc-induced changes as well, which served as a positive control. B: the bar graph represents the densitometric analyses of the PCR product bands, followed by calculation of the high mtDNA-to-low mtDNA ratio. C and D: bar graphs depicting malondialdehyde (MDA) and lactate dehydrogenase (LDH) levels. Overexpression of p66ShcS36A or treatment of PKC-β-siRNA restored the increased MDA and LDH levels caused by HG or ANG II treatments. Values are means ± SE; N = 6. *P < 0.01 vs. 5 mM d-glucose. #P < 0.01 vs. 30 mM d-glucose. & P < 0.01 vs. ANG II (10⁻⁷ M). $ P < 0.01 vs. p66Shc WT. @ P < 0.01 vs. p66Shc WT + HG.

Fig. 7.

Expression of cytosolic p-p66Shc and mCyt.C in HK-2 cells exposed to HG and ANG II and treated with PKC-β inhibitor (LY333531) and Pin1 siRNA. A1, B1, C1, and D1: WB profiles of p-p66Shc and mCyt.C following various treatments. The β-actin and mHSP70 served as respective cytosolic and mitochondrial loading controls. A2, B2, C2, D2, A3, B3, C3, and D3: the bar graphs represent the density of the p-p66Shc (column 2) and mCyt.C (column 3) bands detected by WB procedures relative to that of β-actin and mHSP70. The PKC-β inhibitor partially normalized the HG or ANG II-induced changes in a dose-dependent manner in both cytosolic (p-p66Shc) and mitochondrial (mCyt.C) fractions. The Pin1-siRNA treatment also partially normalized the HG or ANG II-induced changes in the mitochondrial (mCyt.C), while they were accentuated in cytosolic (p-p66Shc) fractions. Values are means ± SE; N = 6. *P < 0.01 compared with 5 mM of LG. & P < 0.01 compared with HG. $P < 0.01 compared with 0 μM ANG II. @ P < 0.05 vs. 0 μM ANG II. #P < 0.01 vs. 10⁻⁷ M ANG II.

Fig. 8.

Protein-protein interactions between p-p66Shc and Pin1 in the cytosol (A and B), and between Cyt.C and p-p66Shc (C–F) in the mitochondria following HG, ANG II, PKC-β inhibitor (LY333531), and Pin1-siRNA treatments. The interactions were assessed by immunoprecipitation (IP)/WB procedures. A and B: the LY333531 inhibits the protein-protein interactions between p-p66Shc and Pin1 in the cytosolic fraction of HK-2 cells treated with HG or ANG II in a dose-dependent manner. A1 and B1 are autoradiograms, while A2 and B2 represent the densitometric analysis of density of the respective bands. C–F: both LY333531 and Pin1-siRNA inhibit the protein-protein interactions between p-p66Shc and mCyt.C in the mitochondrial fraction of HK-2 cells treated with HG or ANG II in a dose-dependent manner. C1, D1, E1, and F1 are autoradiograms, while C2, D2, E2, and F2 represent the densitometric analysis of density of the respective bands. The data included in A2, B2, C2, D2, E2, and F2 are representative of 3 experiments.

Fig. 9.

Schematic drawing depicting conceivable events as a consequence of HG and ANG II treatment with activation of PKC-β-p66Shc-Pin1-Cyt.C pathway and leading to oxidant stress and apoptosis of renal tubular epithelial cells. DAG, diacylglycerol.