Decreased miR-26a expression correlates with the progression of podocyte injury in autoimmune glomerulonephritis

Abstract

MicroRNAs contribute to the pathogenesis of certain diseases and may serve as biomarkers. We analyzed glomerular microRNA expression in B6.MRLc1, which serve as a mouse model of autoimmune glomerulonephritis. We found that miR-26a was the most abundantly expressed microRNA in the glomerulus of normal C57BL/6 and that its glomerular expression in B6.MRLc1 was significantly lower than that in C57BL/6. In mouse kidneys, podocytes mainly expressed miR-26a, and glomerular miR-26a expression in B6.MRLc1 mice correlated negatively with the urinary albumin levels and podocyte-specific gene expression. Puromycin-induced injury of immortalized mouse podocytes decreased miR-26a expression, perturbed the actin cytoskeleton, and increased the release of exosomes containing miR-26a. Although miR-26a expression increased with differentiation of immortalized mouse podocytes, silencing miR-26a decreased the expression of genes associated with the podocyte differentiation and formation of the cytoskeleton. In particular, the levels of vimentin and actin significantly decreased. In patients with lupus nephritis and IgA nephropathy, glomerular miR-26a levels were significantly lower than those of healthy controls. In B6.MRLc1 and patients with lupus nephritis, miR-26a levels in urinary exosomes were significantly higher compared with those for the respective healthy control. These data indicate that miR-26a regulates podocyte differentiation and cytoskeletal integrity, and its altered levels in glomerulus and urine may serve as a marker of injured podocytes in autoimmune glomerulonephritis.

Figure 1. Glomerular pathology of GN-model mice.

(a) Glomerular histology of C57BL/6 control mice and early- and late-stage B6.MRLc1 GN-model mice. PAM staining (upper panels), PAS staining (lower panels). Bars = 20 µm. GN mice developed membranous and proliferative lesions. (b) Histological scores of glomerular area, glomerular cell number, and sclerosis score in glomerulus. Values = mean ± SE. A significant difference from the control value is indicated by *(P<0.05) or **(P<0.01). (c) Level of uACR. Values = mean ± SE. A significant difference from the control value is indicated by *(P<0.05). (d) Early-stage glomerular ultrastructures of C57BL/6 control mice and B6.MRLc1 GN-model mice. Bars = 1 µm. Cap, capillary; Cl, capsular lumen; Pod, podocyte; GBM, glomerular basement membrane; Par, glomerular parietal epithelial cells. Arrowheads, podocyte foot process effacement. Asterisk, immune-complex deposition. Arrows, Bowman’s capsule.

Figure 2. Altered miRNA expression in the glomerulus of GN-model mice.

(a) Scatter plot of glomerular miRNA expression detected using microarray analysis to compare C57BL/6 control with early-stage B6.MRLc1 GN-model mice. (b) Scatter plot of glomerular miRNA microarray expression data to compare C57BL/6 control with late-stage B6.MRLc1 GN-model mice. Values = average of global normalized values. Red, black, and green lines indicate the boundaries of miRNAs expressed at 1.5-fold higher, equal, and 1.5-fold lower levels, respectively, in GN compared with control mice. (c) Heat map of glomerular miRNA expression determined using microarray analysis to compare C57BL/6 control mice and early- and late-stage B6.MRLc1 GN-model mice. Only those miRNAs with significant differential expression (P<0.05) in early and late GN compared with controls are listed. Red and green indicate increased and decreased expression, respectively. Heat-map scale bars indicate the relative changes. The relative expression and P values of each miRNA are shown in the table on the right. A significant difference from the control is indicated by *(P<0.005) or **(P<0.001).

Figure 3. miR-26a expression in the glomerulus and tubulointerstitium of GN-model mice and healthy humans.

(a) Localization of miR-26a expression using in situ hybridization with negative control and LNA probes for miR-26a. Female C57BL/6 mice at 9 months of age. Glo, glomerulus; Pod, podocyte; Cl, capsular lumen. Black arrows, miR-26a-positive signals in podocyte cytoplasm. Bars = 50 µm (low maginification) and 5 µm (high maginification). (b) miR-26a expression in isolated glomeruli and tubulointerstitium of C57BL/6 control mice and early-stage B6.MRLc1 GN-model mice. Bead perfusion method. PCR analysis. Values = mean ± SE. A significant difference from the tubulointerstitium is indicated by **(P<0.01). A significant difference from the control is indicated by ††(P<0.01). (c) miR-26a expression in isolated glomeruli of BXSB/MpJ control mice (males, 4 months of age) and BXSB/MpJ^Yaa lupus nephritis-model mice (males, 4 months of age). Bead perfusion method. PCR analysis. Values = mean ± SE. A significant difference from control is indicated by *(P<0.05). (d) miR-26a levels in urinary exosomes from C57BL/6 control mice and early-stage B6.MRLc1 GN-model mice. Values represent the proportional increases relative to controls. Values = mean ± SE. A significant difference from the control is indicated by *(P<0.05). (e) miR-26a expression in isolated glomeruli and tubulointerstitium of a healthy human. LMD method. PCR analysis. Values are expressed as the proportional increase compared with the tubulointerstitium. Values = mean ± SE. Glo, glomerulus. TI, tubulointerstitium.

Figure 4. Podocyte injuries in GN-model mice.

(a) Glomerular localization of podocin, NMIIA, synaptopodin, vimentin, and WT1 in C57BL/6 control mice and early- and late-stage B6.MRLc1 GN-model mice. (b) The glomerular area showing podocin, NMIIA, synaptopodin, vimentin, and WT1-positive cells. Values = mean ± SE. A significant difference from the control is indicated by *(P<0.05) or **(P<0.01).

Figure 5. Correlation between glomerular miR-26a expression and indices for podocyte injuries in GN-model mice.

(a and b) Expression of mRNAs encoding podocyte proteins and GN-associated genes in isolated glomeruli. Bead perfusion method. Values are expressed as the proportional increase relative to control glomeruli. Values = means ± SE. A significant difference from control is indicated by *(P<0.05) or **(P<0.01). (c) Correlation between glomerular miR-26a expression, uACR, and glomerular mRNA expression of podocyte proteins and GN-associated genes (n = 10, mice 9 months of age). Expression values are expressed as the fold-increase relative to those of the C57BL/6 control. Spearman's correlation test was used to analyze the correlation between two parameters, and a significant correlation is indicated by *(P<0.05) or **(P<0.01).

Figure 6. miR-26a expression in injured mouse podocytes.

(a) miR-26a expression in mouse renal cells. Values are expressed as the proportional increase relative to that of mouse collecting-duct cells. Values = means ± SE. A significant difference from collecting-duct cells is indicated by *(P<0.05) or **(P<0.01). A significant difference from mesangial cells is indicated by ††(P<0.01). (b) Viability of LPS- or PAN-treated mouse podocytes. Values = mean ± SE. Significant difference from untreated cells is indicated by *(P<0.05) or **(P<0.01). (c) Mouse podocytes treated with PBS, LPS, or PAN on day 12 after differentiation. LPS- or PAN-treated podocytes show cytoskeletal changes characterized by decreased cell size, altered shape, and weak actin fibers. Detection of actin and nuclei using phalloidin and Hoechst 33342, respectively. Bars = 20 µm. (d) miR-26a expression in PBS-, LPS-, or PAN-treated mouse podocytes. Values are expressed as the proportional increase relative to that of the PBS control. Values = mean ± SE. A significant difference from the PBS control is indicated by **(P<0.01). (e) miR-26a levels in culture medium and exosomes collected from culture medium after PAN treatment for 48 h. Mouse podocytes. Values are expressed as the proportional increase relative to that of the PBS control. Values = mean ± SE. Significant difference from the PBS control is indicated by **(P<0.01).

Figure 7. Correlation between miR-26a expression and podocyte differentiation.

(a) Differentiation of immortalized mouse podocytes. Podocytes at days 0 and 12 after differentiation. Bars = 5 µm. (b) Time-course of mRNA expression of podocyte and actin family proteins in mouse podocytes. Values are expressed as the proportional increase relative to that of day 0. Values = mean ± SE. (c) miR-26a expression in mouse podocytes. Values are expressed as the proportional increase relative to that of day 0. Values = means ± SE. A significant difference from day 0 is indicated by **(P<0.01). (d) Correlation between expression of miR-26a and mRNAs encoding podocyte and actin family proteins. Samples were collected on days 0, 4, 8, 10, 12, and 14 (n = 3 for each time point). Values are expressed as the proportional increase relative to that of day 0. Spearman's correlation test was used to analyze the correlation between two parameters, and a significant correlation is indicated by *(P<0.05) or **(P<0.01). (e) Expression of miR-26a and mRNAs encoding podocyte and actin family member proteins in mouse podocytes after silencing miR-26a expression on days 4 and 12. Values are expressed as the proportional increase relative to the siRNA control. Values = mean ± SE. A significant difference from the siRNA control is indicated by *(P<0.05) or **(P<0.01). (f) Immunoblotting analysis of pan-actin and vimentin expression in mouse podocytes with transient silencing of miR-26a. Values = mean ± SE. A significant difference from the control is indicated by *(P<0.05).

Figure 8. miR-26a levels in urine and glomerulus in patients with autoimmune GN.

(a) miR-26a levels in urinary exosomes of from healthy controls and patients with lupus nephritis. Values represent the proportional increases relative to controls. Values = mean ± SE. A significant difference from the control is indicated by *(P<0.05). (b) Correlation between miR-26a levels in urinary exosomes and urinary protein levels in patients with lupus nephritis. Spearman's correlation test was used to analyze the correlation between two parameters. (c) The ratio of miR-26a expression in isolated glomeruli (Glo) to tubulointerstitium (TI) of healthy controls to that of patients with lupus nephritis or IgA nephropathy. LMD method. PCR analysis. Values are expressed as the proportional increase relative to the control. Values = mean ± SE. A significant difference from the control is indicated by *(P<0.05).