Reversibility of structural and functional damage in a model of advanced diabetic nephropathy

Abstract

The reversibility of diabetic nephropathy remains controversial. Here, we tested whether replacing leptin could reverse the advanced diabetic nephropathy modeled by the leptin-deficient BTBR ob/ob mouse. Leptin replacement, but not inhibition of the renin-angiotensin-aldosterone system (RAAS), resulted in near-complete reversal of both structural (mesangial matrix expansion, mesangiolysis, basement membrane thickening, podocyte loss) and functional (proteinuria, accumulation of reactive oxygen species) measures of advanced diabetic nephropathy. Immunohistochemical labeling with the podocyte markers Wilms tumor 1 and p57 identified parietal epithelial cells as a possible source of regenerating podocytes. Thus, the leptin-deficient BTBR ob/ob mouse provides a model of advanced but reversible diabetic nephropathy for further study. These results also suggest that restoration of lost podocytes is possible but is not induced by RAAS inhibition, possibly explaining the limited efficacy of RAAS inhibitors in promoting repair of diabetic nephropathy.

Figure 1.

Leptin replacement results in reduction of collagen IV positive mesangial matrix. (A) Representative glomeruli showing mesangial matrix accumulation in BTBR ob/ob mice with leptin replacement was significantly decreased compared with 24-week-old BTBR ob/ob control mice. BTBR ob/ob mice treated with hydralazine, losartan, or enalapril showed no difference in mesangial matrix collagen IV immunostaining. Original magnification, ×400. (B) Morphometric analysis of glomerular tuft area (GTA) occupied by type IV collagen immunostaining of mesangial matrix. +++ P<0.001 versus 18- and 24-week-old BTBR ob/ob mice, *** P<0.001 versus 24-week-old BTBR WT mice.

Figure 2.

Representative silver methenamine–stained glomeruli demonstrate reduction of mesangial matrix and mesangiolysis following leptin replacement for 6 weeks. (A) Silver methenamine staining of 24-week-old BTBR ob/ob mice shows increased mesangial matrix compared with WT mice, similar to the results with collagen IV immunostaining. BTBR ob/ob mice with leptin replacement show dramatic reduction in mesangial matrix, equivalent to BTBR WT mice. BTBR ob/ob mice treated with losartan or enalapril treatment demonstrate stabilization of the expanded mesangial matrix at the level of mice at the start of treatment (e.g., equivalent to 18-week-old untreated BTBR ob/ob mice) but show no regression of matrix. Original magnification, ×400. (B) Morphometric analysis of silver methenamine–positive matrix and mesangiolysis was assessed. BTBR ob/ob mice have significantly increased matrix and numbers of glomeruli with mesangiolysis compared with WT mice. Leptin replacement significantly decreased matrix compared with age-matched untreated BTBR ob/ob mice. Treatment with hydralazine, losartan, and enalapril also significantly reduced mesangial matrix, but matrix did not regress to WT levels as seen with leptin replacement. Leptin replacement significantly decreased mesangiolysis compared with age-matched untreated BTBR ob/ob mice, whereas the other treatment modalities had no effect on the degree of mesangiolysis seen. GTA, glomerular tuft area. ***P<0.001, **P<0.01, *P<0.05 versus 24-week-old BTBR WT mice; +++P<0.001, ++P<0.01, +P<0.05 versus 24-week-old BTBR ob/ob mice.

Figure 3.

Leptin replacement reduces basement membrane thickening and mesangial matrix accumulation. (A) Electron micrographs of representative glomeruli shows normal architectural organization and preservation of podocyte foot processes in BTBR WT mice. BTBR ob/ob mice have expanded mesangial areas (M) with matrix accumulation and lucencies indicating mesangial lysis (*). The inset shows areas of lysis that contain red blood cells and red blood cell fragments in some foci (arrow). Podocyte foot processes are incompletely preserved. Leptin replacement results in restoration of normal mesangial and podocyte structure. Original magnification, ×2900. (B) BTBR ob/ob mice develop thickened basement membranes (BMs), which are reduced with leptin replacement. (C) High-power electron micrographs of representative glomerular basement membranes. Podocyte foot processes are normal in BTBR WT mice but are focally completely effaced in BTBR ob/ob mice. Leptin replacement reduces basement membrane thickening and improves podocyte foot process architecture. Original magnification, ×13,000.

Figure 4.

BTBR ob/ob mice with leptin replacement demonstrate replenished podocyte density and decreased glomerular volume. Podocyte density is increased, as assessed by WT-1–expressing cells in BTBR ob/ob mice with leptin replacement. (A) Representative glomeruli immunostained with WT-1. BTBR ob/ob mice have decreased density of WT-1–positive podocytes and increased glomerular cross-sectional area. Treatment with hydralazine, losartan, or enalapril has no effect on podocyte density. BTBR ob/ob mice with leptin replacement have restoration of podocyte density compared with untreated BTBR ob/ob mice. In BTBR ob/ob mice with leptin replacement, there was increased expression of WT-1 by PECs lining the Bowman capsule (arrows). Original magnification, ×400. (B) Decreased podocyte density and increased glomerular volume in untreated BTBR ob/ob mice. Podocyte density and glomerular volume were normalized to values near those of BTBR WT mice with leptin treatment, whereas treatment with hydralazine, losartan, and enalapril had little effect. ***P<0.001, **P<0.01, *P<0.05 versus 24-week-old BTBR WT and BTBR ob/ob mice treated with leptin; ++P<0.01 versus 24-week-old BTBR ob/ob and BTBR ob/ob mice treated with hydralazine, losartan, and enalapril.

Figure 5.

p57-positive podocytes were counted in serial sections using the fractionator/dissector method. One hundred serial sections were cut and stained with an antibody specific for p57, a second specific marker of podocytes. p57-positive podocytes seen in both sections of the dissector pair were not counted, ensuring that each podocyte was counted only once. Use of this method confirmed that BTBR ob/ob mice had significantly fewer total podocytes per glomerulus and that podocyte number was restored with 6 weeks of leptin treatment. p57-positive PECs were seen only in leptin treated mice (arrow). Original magnification, ×400.

Figure 6.

Expression of WT-1 by PECs in BTBR ob/ob mice with leptin replacement. (A) Double immunostaining with WT-1 (gray/black), which stains nuclei of podocytes, and claudin-1 (red/brown), which stains cytoplasm and cell junctions of PECs, is shown in low- and high-power views of representative glomeruli. BTBR ob/ob mice with leptin replacement demonstrate de novo expression of WT-1 by PECs lining the Bowman capsule. Arrows point to cells that express both WT-1, a podocyte marker, and the PEC marker claudin-1. Original magnifications, ×400 and ×1000. (B) Graphic representation of quantitation of WT-1–positive PECs in study groups. **P<0.01 versus BTBR WT mice, ++ P<0.01 versus 18- and 24-week-old BTBR ob/ob mice treated with hydralazine, losartan, and enalapril; δδ P<0.01 versus 18- and 24-week-old BTBR ob/ob mice treated with hydralazine, losartan, and enalapril.

Figure 7.

Superoxides are upregulated in BTBR ob/ob mice. (A) Representative glomeruli stained with DHE show little to no superoxide in BTBR WT mice. Superoxide (red stain) is markedly increased in glomeruli of BTBR ob/ob mice at 28 weeks. Leptin replacement for 8 weeks abolished the glomerular superoxide. (B) Glomerulus from 24 week BTBR ob/ob mouse stained with DHE (red) and WT-1 (green). Merged pictures show WT-1–positive podocytes that are also positive for superoxide (arrows). Original magnification, ×400.

Figure 8.

ObRa, the short form of the leptin receptor, is expressed in glomeruli of BTBR WT and BTBR ob/ob mice. RT-PCR was performed using RNA from whole cortex and from isolated glomeruli using primers specific for ObRa and ObRb, the long form of the leptin receptor. Levels of ObRa, but not ObRb, were detectable in glomeruli of BTBR WT and BTBR ob/ob mice (upper panel). Immunohistochemistry performed using an antibody specific for phosphorylated-Stat3 shows upregulated expression in BTBR ob/ob mice treated with leptin within intrinsic renal cells, within glomeruli and tubular epithelium (bottom panel), indicating signaling either directly through leptin or through some other intermediate mediator. Original magnification, ×400.