High glucose induces phosphorylation and oxidation of mitochondrial proteins in renal tubular cells: A proteomics approach

Abstract

Mitochondrial dysfunction has been thought to play roles in the pathogenesis of diabetic nephropathy (DN). However, precise mechanisms underlying mitochondrial dysfunction in DN remained unclear. Herein, mitochondria were isolated from renal tubular cells after exposure to normal glucose (5.5 mM glucose), high glucose (25 mM glucose), or osmotic control (5.5 mM glucose + 19.5 mM mannitol) for 96 h. Comparative proteomic analysis revealed six differentially expressed proteins among groups that were subsequently identified by tandem mass spectrometry (nanoLC-ESI-ETD MS/MS) and confirmed by Western blotting. Several various types of post-translational modifications (PTMs) were identified in all of these identified proteins. Interestingly, phosphorylation and oxidation were most abundant in mitochondrial proteins whose levels were exclusively increased in high glucose condition. The high glucose-induced increases in phosphorylation and oxidation of mitochondrial proteins were successfully confirmed by various assays including MS/MS analyses. Moreover, high glucose also increased levels of phosphorylated ezrin, intracellular ATP and ROS, all of which could be abolished by a p38 MAPK inhibitor (SB239063), implicating a role of p38 MAPK-mediated phosphorylation in high glucose-induced mitochondrial dysfunction. These data indicate that phosphorylation and oxidation of mitochondrial proteins are, at least in part, involved in mitochondrial dysfunction in renal tubular cells during DN.

Figure 1

Purity of mitochondrial isolation. Immunoreactive bands of markers for mitochondria (COX4), cytoplasm (carbonic anhydrase II), and endoplasmic reticulum (GRP94) in whole cell lysate and mitochondrial fraction are shown. The full-membrane images of these blots are provided in Supplementary Fig. S1.

Figure 2

2-D proteome map of differentially expressed mitochondrial proteins. (A–C) Representative 2-D gels, stained with Deep Purple fluorescence dye, of mitochondrial proteins under normal glucose, high glucose, and osmotic control, respectively (n = 5 gels were derived from 5 individual culture flasks for each condition; a total of 15 gels were subjected to comparative analysis). Differentially expressed proteins are labelled with numbers, which correspond to those reported in Tables 1 and 2.

Figure 3

Intensity levels of differentially expressed protein spots. (A–C) Intensity levels of spots no. 181, 610, and 661, respectively, which were increased exclusively in high glucose condition. (D–E) Intensity levels of spots no. 296, 655, and 799, respectively, which were decreased in both high glucose and osmotic control conditions. Each bar represents mean ± SEM of intensity data obtained from five gels per group. *p < 0.05 vs. normal glucose and osmotic control; **p < 0.01 vs. normal glucose and osmotic control; #p < 0.05 vs. normal glucose; ## p < 0.01 vs. normal glucose.

Figure 4

Validation of the proteomic data by Western blot analysis. (A) Immunoreactive bands of tubulin, lamin A/C and annexin A2 in whole cell and mitochondrial fraction. The full-membrane images of these blots are provided in Supplementary Fig. S3. (B) Ponceau S-stained membranes were used to verify equal protein loading in each sample. (C) Band intensity of tubulin normalized with total protein. (D) Band intensity of lamin A/C normalized with total protein. (E) Band intensity of annexin A2 normalized with total protein. Each bar represents mean ± SEM of intensity data obtained from three independent samples. *p < 0.05 vs. normal glucose and osmotic control.

Figure 5

Potential PTMs identified from all differentially expressed mitochondrial proteins. (A) Number of potentially modified residues by individual PTMs in “each protein” identified by nanoLC-ESI-ETD MS/MS. (B) Summation of all the potentially modified residues by individual PTMs in “all proteins” identified. Insets demonstrate the zoom-in details of phosphorylation and oxidation, respectively. See more details in Supplementary Table S2.

Figure 6

Detection and quantitative analysis of mitochondrial phosphoproteins. (A) Zoom-in images of representative 2-D gels, stained with Pro-Q Diamond phosphoprotein gel stain followed by SYPRO Ruby total protein gel stain, of mitochondrial proteins under normal glucose, high glucose, and osmotic control, respectively (n = 3 gels were derived from 3 individual culture flasks for each condition; a total of 9 gels were analyzed). The full-length gels of these zoom-in images are shown in Supplementary Fig. S4. (B) Demonstrates the whole 2-D gel, of which 2-D spot pattern was consistent with that of the Deep Purple stained gels used for initial comparative analysis (shown in Fig. 1). (C) Quantitative analysis of levels of phosphoproteins (Pro-Q Diamond) in relative to (normalized with) total protein (SYPRO Ruby) detected in each group. Each bar represents mean ± SEM of intensity data obtained from three individual gels per group. *p < 0.05 vs. normal glucose and osmotic control.

Figure 7

Oxidative stress assays. (A) Immunoreactive bands of oxidized mitochondrial proteins as detected by Oxyblot assay. (B) Quantitative analysis of oxidized mitochondrial proteins. (C) Histograms of DCF fluorescence intensity representing ROS level measured by flow cytometry. (D) Quantitative analysis of intracellular ROS level. Each bar represents mean ± SEM of intensity data obtained from three independent samples per group. *p < 0.05 vs. normal glucose and osmotic control.

Figure 8

MS/MS analyses of phosphorylated and oxidatively modified proteins. Pie charts and bar graphs summarize all of the identified peptides from all differentially expressed mitochondrial proteins. The data include the proportion of non-modified and modified peptides, percentage and number of the modified residues, specifically with phosphorylation at serine (S), threonine (T) or tyrosine (Y), oxidation or dioxidation at methionine (M), and oxidation, dioxidation or trioxidation at cysteine (C). See illustrative MS/MS spectra with phosphorylated or oxidatively modified residue(s) in Supplementary Figs. S5 and S6.

Figure 9

Levels of phosphorylated ezrin (p-ezrin) and intracellular ATP. (A) Immunoreactive bands of p-ezrin and total ezrin in the cells. (B) Band intensity of p-ezrin normalized with that of total ezrin. (C) Intracellular ATP level measured by luciferin-luciferase ATP assay. Each bar represents mean ± SEM of the data obtained from three independent samples. *p < 0.05 vs. normal glucose and osmotic control.

Figure 10

Effects of p38 MAPK inhibitor (SB239063) on levels of p-ezrin, intracellular ATP and ROS. (A) Immunoreactive bands of p-ezrin and total ezrin in high glucose-treated cells without or with SB239063 cotreatment. (B) Band intensity of p-ezrin normalized with total ezrin in high glucose-treated cells without or with SB239063 cotreatment. (C) Intracellular ATP level detected by luciferin-luciferase ATP assay. (D) Intracellular ROS level measured by DCFH-DA assay and flow cytometry. Each bar represents mean ± SEM of the data obtained from three independent samples. *p < 0.05 vs. high glucose.