Glomerular clusterin expression is increased in diabetic nephropathy and protects against oxidative stress-induced apoptosis in podocytes

Abstract

Clusterin, a glycoprotein encoded by the CLU gene, is expressed in many tissues, including the kidney, and clusterin expression is upregulated in the glomeruli of patients with various forms of kidney disease. Here, we investigated the role of clusterin in diabetic nephropathy (DN). In this study, we found that glomerular clusterin expression was increased in both patients with DN and streptozotocin-induced diabetic mice and that it co-localised with the podocyte marker WT1, indicating clusterin is expressed in podocytes. In our in vitro analysis, we found no significant change in CLU mRNA expression in podocytes following stimulation with high glucose and angiotensin II; in contrast, CLU mRNA expression was significantly upregulated following methylglyoxal stimulation. Methylglyoxal treatment also significantly decreased the mRNA expression of the slit diaphragm markers ZO-1 and NEPH1 and significantly increased the mRNA expression of the oxidative stress marker HO-1. Lastly, we showed that pre-incubating podocytes with recombinant human clusterin protein increased podocyte survival, prevented slit diaphragm damage, and reduced oxidative stress‒induced apoptosis following methylglyoxal stimulation. Taken together, our results indicate that glomerular clusterin is upregulated in DN, and this increase in clusterin expression may protect against oxidative stress-induced apoptosis in podocytes, providing a possible new therapeutic target for DN and other kidney diseases.

Figure 1

Glomerular clusterin expression is increased in patients with DN. (a,b) Representative images of kidney sections obtained from a healthy subject (Control; a) and a patient with DN (DN; b) stained for clusterin; the scale bars represent 200 µm. (c,d) Representative high-magnification images of a and b; the scale bars represent 50 µm. (e,f) The high-magnification views of the rectangles in (c,d). Clusterin staining was present along the outer side of the GBM at the place where the podocytes are located (arrowheads; the square shows a zoomed in image of a clusterin-positive podocyte). (g) Summary of clusterin staining (semi-quantitative score) in the glomeruli of healthy subjects (Control; n = 10) and patients with DN (DN; n = 12); ***P < 0.001 vs. control (Student’s t-test). (h) Summary of CLU mRNA measured in isolated glomeruli from healthy subjects (Control; n = 11) and patients with DN (DN; n = 24); **P < 0.01 vs. control (Student’s t-test).

Figure 2

Clusterin expression is increased in the glomeruli of mice with type 1 diabetes and co-localises with podocytes. (a, b) Representative images of mouse kidney sections obtained from control mice (Control; a) and STZ-treated mice 15 weeks after diabetes was induced (STZ; b); the scale bars represent 100 µm. (c, d) The high-magnification views of the rectangles in (a, b); the scale bars represent 50 µm. (e, f) Representative images of the adjacent sections stained for clusterin (e) and WT1 (f), showing co-localisation of clusterin with WT1 in the glomeruli (arrowheads). (g) Averaged clusterin-positive area (measured by using ImageJ) in the glomeruli of control and STZ-treated mice 15 weeks after diabetes was induced; **P < 0.01 vs. control (Student’s t-test).

Figure 3

Clusterin expression in podocytes cultured under various diabetic conditions. (a–d) Human podocytes were cultured for the indicated times in PAN (30 µg/mL) (a), Glucose (5 mM or 25 mM) (b), Angiotensin II (1 µM) (c), or the indicated concentrations of MGO for 24 h (d). The CLU mRNA expression after these treatments was measured, the mRNA levels were normalised to the respective control. In (a–c), *P < 0.05 vs. the respective control at the same time point (Student’s t-test); in (d), *P < 0.05 vs. control, and †P < 0.05 vs. 0.5 mM MGO (one-way ANOVA). PAN puromycin aminonucleoside, MGO methylglyoxal.

Figure 4

Methylglyoxal reduces podocyte viability, down-regulates ZO-1 and NEPH1 expression, and up-regulates HO-1 expression. (a) Cell viability was measured in podocytes incubated with the indicated concentrations of MGO for 24 h; ***P < 0.001 vs. control, ^†††P < 0.001 vs. 0.5 mM MGO, and ^‡‡‡P < 0.001 vs. 1.0 mM MGO (one-way ANOVA). (b–d) Podocytes were incubated with the indicated concentrations of MGO for 24 h, followed by measuring ZO-1 (b), NEPH1 (c), and HO-1 (d) mRNA expression; the mRNA levels were normalised to the respective control; *P < 0.05 vs. control, **P < 0.01 vs. control, ^†P < 0.05 vs. 0.5 mM MGO, and ^††P < 0.01 vs. 0.5 mM MGO (one-way ANOVA). (e,f) Podocytes were incubated with MGO (1.5 mM) for 24 h, with or without pre-treatment with NAC (150 µmol/L) for 1 h, followed by measuring HO-1 (e) and CLU (f) mRNA expression; the mRNA levels were normalised to the respective control; **P < 0.01 vs. control, ***P < 0.001 vs. control, and ^†††P < 0.001 vs. MGO (one-way ANOVA). MGO methylglyoxal, NAC N-acetyl-l-cysteine.

Figure 5

Recombinant human clusterin protein protects podocytes against methylglyoxal-induced oxidative stress and apoptosis. (a) Cell viability was measured in podocytes incubated with the indicated concentrations of rhCLU for 24 h. (b) Cell viability was measured in podocytes pre-incubated for 4 h with the indicated concentrations of rhCLU, followed by 24 h in the presence or absence of 1.5 mM MGO. Cell viability in the presence of MGO was normalised to cells stimulated in the absence of rhCLU; *P < 0.05 vs. 0 µg/mL rhCLU + 1.5 mM MGO (one-way ANOVA). (c) Representative images of podocytes treated with Milli-Q water (Control) or podocytes pre-treated for 4 h with 2.0 µg/mL rhCLU, 2 h with 1.5 mM MGO, or both, followed by MitoSOX Red staining. The scale bars indicate 20 µm and the boxed areas are shown at higher magnification in the bottom row. (d) Representative images of podocytes treated as described in (c), followed by DCFDA staining. The scale bars indicate 100 µm. (e) Caspase3/7 activity in podocytes treated with Milli-Q water (Control) or MGO (1.5 mM) for 12 h with or without pre-treated with 2.0 µg/mL rhCLU for 4 h. The measurement was normalised to the control; **P < 0.01 vs. control, ^††P < 0.01 vs. rhCLU, and ^‡‡P < 0.01 vs. MGO (one-way ANOVA). (f) Podocytes treated with Milli-Q water (Control) or MGO (1.5 mM) for 2 h with or without pre-treatment with 2.0 µg/mL rhCLU for 4 h, followed by measuring BAX and BCL2 mRNA expression. The BAX/BCL2 ratio was calculated and normalised to the control; **P < 0.01 vs. control, ^††P < 0.01 vs. rhCLU, and ^‡P < 0.05 vs. MGO (one-way ANOVA). (g,h) Podocytes treated with Milli-Q water (Control) or MGO (1.5 mM) for 2 h with or without pre-treated with 2.0 µg/mL rhCLU for 4 h, followed by measuring ZO-1 (g) and NEPH1 (h) mRNA expression; the mRNA levels were normalised to the respective control; *P < 0.05 vs. control, **P < 0.01 vs. control ***P < 0.001 vs. control, ^†P < 0.05 vs. rhCLU, ^††P < 0.01 vs. rhCLU, and ^‡P < 0.05 vs. MGO (one-way ANOVA). MGO methylglyoxal, rhCLU recombinant human clusterin protein.