Lipid peroxidation regulates podocyte migration and cytoskeletal structure through redox sensitive RhoA signaling

Abstract

Early podocyte loss is characteristic of chronic kidney diseases (CKD) in obesity and diabetes. Since treatments for hyperglycemia and hypertension do not prevent podocyte loss, there must be additional factors causing podocyte depletion. The role of oxidative stress has been implicated in CKD but it is not known how exactly free radicals affect podocyte physiology. To assess this relationship, we investigated the effects of lipid radicals on podocytes, as lipid peroxidation is a major form of oxidative stress in diabetes. We found that lipid radicals govern changes in podocyte homeostasis through redox sensitive RhoA signaling: lipid radicals inhibit migration and cause loss of F-actin fibers. These effects were prevented by mutating the redox sensitive cysteines of RhoA. We therefore suggest that in diseases associated with increased lipid peroxidation, lipid radicals can determine podocyte function with potentially pathogenic consequences for kidney physiology.

Keywords: Chronic kidney disease; Cysteine; Lipid peroxidation; Podocyte; Reactive lipids; RhoA.

Fig. 1

Alkyl (lipid peroxyl) radical generation in cultured podocytes. (A) Conditionally immortalized mouse podocytes were cultured as described in Methods. Representative pictures show podocytes maintained at 33 °C with IFNγ present and for differentiation cells were switched to non-permissive conditions (37 °C, no IFNγ). Podocytes were differentiated for at least 11 days or until increased synaptopodin levels were evident as verified by Western blot analysis. (B) The free radical donor AAPH was used to generate alkyl radicals and mimic lipid peroxidation. Equations show the temperature sensitive decomposition of AAPH, the resultant radical formation in the cell membrane and the method for calculating free radical production for the incubation periods.

Fig. 2

Lipid peroxyl radicals influence podocyte migratory parameters. (A-B) “Wound” scratch assay shows impaired migration of podocytes exposed to lipid radicals generated at 2 µmol/min for 4 h. Representative pictures of duplicate experiments, number of cells migrating into the wound was counted after 72 h. (C) Track directions of untreated cells and podocytes exposed to lipid radicals (0.8 − 2 µmol/min) were followed in random individual cells (colored lines show representatives) for 72 h using an Image J manual cell tracking plug-in. (D-E) Total distance traveled and average velocity have been calculated from the tracks. Colored tracks show a more static phenotype in cells with 2 µmol/min R• treatment; with significantly less distance traveled, and lower velocity. Duplicate experiments, n = 12 viewing areas (4 per group), at least six cells tracked in each viewing area. *p < 0.05 vs. control (two-tailed Student's t-test).

Fig. 3

Lipid peroxyl radicals induce cytoskeletal rearrangements in podocytes and increase RhoA activation. (A) Upper panels: Representative images of untreated podocytes and podocytes incubated with AAPH to generate alkyl radicals (R^•). After incubation (4 h), cells were stained for F-actin as described in Materials and methods. F-actin orientation was observed under a fluorescent microscope (10 pictures/group at random viewing areas). Podocytes exposed to lipid radicals display cortical rearrangement and loss of F-actin transversal fibers. Lower panels: representative greyscale images of control and R^• treated podocytes for anisotropy and orientation analysis, showing the anisotropy vectors in red (Image J “FibrilTool” application). Yellow arrows show examples of actin-rich centers in podocytes exposed to lipid peroxidation. (B) Anisotropy analysis of control and lipid radical treated podocytes confirms significant loss of transversal fibers. (C) Orientation analysis of F-actin fibers, showing significant reorientation in podocytes exposed to lipid peroxidation. Orientation angles from − 90 to + 90 degrees were grouped into 20 degree intervals. Duplicate experiments in six well plates, n = at least 30 cells analyzed per group (3 cells each picture, 10 pictures each group), *p < 0.05 vs. control. (D) Lipid radicals generated from AAPH activate RhoA in a dose-dependent manner. RhoA activation was measured using an active (GTP-bound) RhoA specific “G-LISA” assay. Duplicate experiments, n = 4 per group, *p < 0.05 vs. control (one-way ANOVA for multiple comparisons).

Fig. 4

Podocytes sense lipid peroxyl radicals through the redox sensitive cysteine residues of RhoA. (A-B) Comparison of podocyte migration using the “wound” assay in normal cells, wild type RhoA transduced podocytes and in cells transduced with the mutant C16/20 A RhoA, after lipid peroxide radical exposure (0.8 − 2 µmol/min, 4 h). Number of cells migrating into the wound was counted in each group in duplicate experiments after 72 h. While normal or Wt podocytes have significantly impaired migration at 2 µmol/min radical exposure, C16/20 A RhoA mutant podocytes migrate normally. N= 4 viewing areas each group, * p < 0.05 vs. control. (C) Activation of RhoA by lipid radicals is blunted in podocytes bearing C16/20 A mutated RhoA as measure by a “G-LISA” assay (see also Fig. 2). (D) Track direction of C16/20 A RhoA transduced podocytes were followed in random individual cells (colored lines show representatives) for 72 h using an Image J manual cell tracking plug-in. (E-F) Total distance traveled and average velocity have been calculated from the tracks. Podocytes with mutant RhoA display normal migratory parameters. (G) F-actin fibers were visualized and (H-I) anisotropy and fiber orientation were measured similarly to as shown in Fig. 3, in wild type RhoA and C16/20 A RhoA transduced cells using the highest radical concentration. (One-way ANOVA analysis was used for multiple comparisons).