Transforming growth factor-β-induced cross talk between p53 and a microRNA in the pathogenesis of diabetic nephropathy

Abstract

Elevated p53 expression is associated with several kidney diseases including diabetic nephropathy (DN). However, the mechanisms are unclear. We report that expression levels of transforming growth factor-β1 (TGF-β), p53, and microRNA-192 (miR-192) are increased in the renal cortex of diabetic mice, and this is associated with enhanced glomerular expansion and fibrosis relative to nondiabetic mice. Targeting miR-192 with locked nucleic acid-modified inhibitors in vivo decreases expression of p53 in the renal cortex of control and streptozotocin-injected diabetic mice. Furthermore, mice with genetic deletion of miR-192 in vivo display attenuated renal cortical TGF-β and p53 expression when made diabetic, and have reduced renal fibrosis, hypertrophy, proteinuria, and albuminuria relative to diabetic wild-type mice. In vitro promoter regulation studies show that TGF-β induces reciprocal activation of miR-192 and p53, via the miR-192 target Zeb2, leading to augmentation of downstream events related to DN. Inverse correlation between miR-192 and Zeb2 was observed in glomeruli of human subjects with early DN, consistent with the mechanism seen in mice. Our results demonstrate for the first time a TGF-β-induced feedback amplification circuit between p53 and miR-192 related to the pathogenesis of DN, and that miR-192-knockout mice are protected from key features of DN.

FIG. 1.

Increased p53 expression in the renal cortex of diabetic db/db mice relative to control db/+ mice. A: Western blot and quantification of p53 expression from renal cortical lysates of 10–12-week-old db/+ (n = 5) and db/db mice (n = 4). Representative immunostaining and quantification for p53 (B) and PAS (C) staining from cortical sections of db/+ and db/db mice; n = 4 per group. *P < 0.05; ***P < 0.001. Error bars, SEM. Scale bar, 20 μm.

FIG. 2.

miR-192 regulates p53 expression in the renal cortex of control and STZ-diabetic mice. A: Representative Western blot and quantification of p53 expression in cortical lysates of nondiabetic mice injected with normal saline (C-NS), negative control oligos LNA-anti-miR-239b (C-NC), or LNA-anti-miR-192 oligos (C-LNA) (2 mg/kg) for 6 h; n = 3 per group. B: Representative Western blot and quantification of p53 expression in cortical lysates of control mice injected with normal saline (C-NS, n = 2) and STZ-diabetic for 2 weeks injected with normal saline (STZ-NS, n = 3), negative control oligos LNA-anti-miR-239b (STZ-NC, n = 3), or LNA-anti-miR-192 (STZ-LNA, n = 4). C: Representative immunostaining and quantification of glomerular p53 from cortical sections of C-NS and STZ-NS mice (n = 5 per group) and STZ-NC (n = 6) and STZ-LNA mice (n = 7). *P < 0.05; ***P < 0.001. Error bars, SEM. Scale bar, 20 μm.

FIG. 3.

Generation of miR-192–KO mice. A: Schematic diagram showing design of the targeting vector used to generate miR-192-KO mice. B: Gel picture showing PCR screening for targeted miR-192–KO alleles from Neo-resistant ES clones. Arrows, positive clones identified by PCR screening. C: Southern blot confirming recombinant clones. D: Schematic diagram showing genomic structures of WT and KO alleles and PCR approach. E: Gel picture confirming germline transmission of the miR-192–deleted allele. F: Schematic genomic structures of Neo-deleted allele and PCR approach for identification of the deletion. G: Gel picture showing PCR analysis for genotyping performed using tail DNA. +/+, WT mice; +/−, heterozygous mice; −/−, homozygous mice. H and I: qRT-PCR analysis of miR-192 and Zeb2 in +/+ or −/− glomeruli (H) and MMCs (I); n = 3. **P < 0.01; ***P < 0.001. J: Scatter plot showing miR-192 and Zeb2 expression in glomeruli of type 2 diabetic human patients; n = 46. Patient information is provided in the RESULTS. R = −0.35; P value = 0.02.

FIG. 4.

Reduced p53 expression in kidneys of miR-192–KO diabetic mice. A: Western blot and quantification of p53 expression in cortical lysates of WT or miR-192-KO mice (control or STZ-diabetic for 2 weeks). Control mice, n = 3 per group; diabetic mice, n = 5 per group. Representative immunostaining and quantification of TGF-β and p53 (B) and PAS (C) staining from cortical sections of WT or miR-192-KO mice (control or STZ-diabetic). Control mice, n = 3–5 per group; diabetic mice, n = 3–6 per group. Bar graph adjacent to C shows glomerular area and mesangial matrix expansion. *P < 0.05; **P < 0.01; ***P < 0.001. Error bars, SEM. Scale bar, 20 μm.

FIG. 5.

Attenuation of key features of DN at 22 weeks in diabetic miR-192–KO mice. A: Western blot and quantification of p53 expression in cortical lysates of WT or miR-192-KO mice (control or STZ-diabetic for 22 weeks). Control mice, n = 3 per group; diabetic mice, n = 4 per group. Representative immunostaining and quantification of TGF-β and p53(B), PAS staining (C), and Masson’s trichrome staining (D) from cortical sections of WT and miR-192–KO control or STZ-diabetic mice; n = 3–6 mice per group. E: Urine protein levels at 19 weeks post–diabetes induction from WT and miR-192–KO control or STZ-injected mice (WT-C, n = 6; WT STZ-diabetic, n = 6; KO-C, n = 3; KO-STZ, n = 6). F: Bar graph showing relative ACR levels at 19 weeks post–diabetes induction from urine samples of WT or miR-192-KO mice (control or STZ-diabetic) (WT-C, n = 5; WT STZ-diabetic, n = 5; KO-C, n = 3; KO-STZ, n = 5). *P < 0.05; **P < 0.01; ***P < 0.001. Error bars, SEM. Scale bar, 20 μm.

FIG. 6.

TGF-β induces transcriptional activation of p53 through miR-192 in MMCs. qRT-PCR analysis of p53 in WT and miR-192-KO MMCs under basal conditions (n = 4) (A), after transfection of miR-192–KO MMCs with control oligos (NC, 10 nmol/L) or miR-192 mimic oligos (192-M, 10 nmol/L, n = 3) (B), and in WT and miR-192–KO MMCs after TGF-β treatment (10 ng/mL, n = 3) (C). D: Schematic genome structure of the p53 promoter region and the p53 promoter luciferase reporter construct. Luciferase assay results showing p53 promoter activity ± TGF-β (5 ng/mL) in WT and miR-192–KO MMCs (n = 4) (E), in miR-192–KO MMCs after transfection with negative control (NC, 10 nmol/L) or miR-192 mimic oligos (192-M, 10 nmol/L, n = 3) (F), and in miR-192–KO MMCs after transfection with control siRNA pool (Ctrl pool, 10 nmol/L) or Zeb2 siRNA pool (si-Zeb2, 10 nmol/L, n = 3) (G). H: Schematic genome structure of the WT and E-box mutant p53 promoter constructs. I: Luciferase assay results showing WT and E-box mutant p53 promoter activity in WT MMCs; n = 3. SD, serum depletion. *P < 0.05; **P < 0.01; ***P < 0.001. Error bars, SEM.

FIG. 7.

TGF-β induces transcriptional activation of miR-192 through p53 in MMCs. qRT-PCR analysis of miR-192 in WT and p53-KO MMCs, under basal and TGF-β–treated (10 ng/mL) conditions; n = 3 (A). miR-192 expression after exogenous p53 expression in WT MMCs (n = 3) (B) and p53-KO MMCs (n = 3) (C). Control, control vector; p53, p53 expression vector (100–200 ng/mL). D: Schematic genome structure of the miR-192 promoter region and the three miR-192 promoter reporter constructs used: P1 (1,871 bp) and P2 (245 bp), miR-192 promoter constructs with a p53-RE, and P3 (125 bp), a p53-RE deletion mutant miR-192 promoter construct. Luciferase assay results showing miR-192 promoter activity ± TGF-β (5 ng/mL) in WT MMCs (n = 4) (E) and after transfection of p53-KO MMCs with a control vector or p53 expression vector (100–200 ng/mL, n = 4) (F). SD, serum depletion. *P < 0.05; **P < 0.01; ***P < 0.001. Error bars, SEM.

FIG. 8.

Attenuation in key features associated with DN in MMCs from miR-192–KO and p53-KO mice. qRT-PCR analysis of Col1a2 and Col4a1 in WT or miR-192-KO MMCs treated ± TGF-β (10 ng/mL, n = 3) (A and B) and in WT or p53-KO MMCs treated ± TGF-β (10 ng/mL, n = 3) (C and D). E: Cellular hypertrophy levels ± TGF-β treatment (10 ng/mL for 24h) of WT and miR-192–KO MMCs (n = 4). F: A schematic model of the mechanism of reciprocal regulation of miR-192 and p53 under diabetic conditions, which can lead to hypertrophy and ECM accumulation associated with the pathogenesis of DN. *P < 0.05; **P < 0.01; ***P < 0.001. Error bars, SEM.