Unraveling the role of podocyte turnover in glomerular aging and injury

Abstract

Podocyte loss is a major determinant of progressive CKD. Although recent studies showed that a subset of parietal epithelial cells can serve as podocyte progenitors, the role of podocyte turnover and regeneration in repair, aging, and nephron loss remains unclear. Here, we combined genetic fate mapping with highly efficient podocyte isolation protocols to precisely quantify podocyte turnover and regeneration. We demonstrate that parietal epithelial cells can give rise to fully differentiated visceral epithelial cells indistinguishable from resident podocytes and that limited podocyte renewal occurs in a diphtheria toxin model of acute podocyte ablation. In contrast, the compensatory programs initiated in response to nephron loss evoke glomerular hypertrophy, but not de novo podocyte generation. In addition, no turnover of podocytes could be detected in aging mice under physiologic conditions. In the absence of podocyte replacement, characteristic features of aging mouse kidneys included progressive accumulation of oxidized proteins, deposits of protein aggregates, loss of podocytes, and glomerulosclerosis. In summary, quantitative investigation of podocyte regeneration in vivo provides novel insights into the mechanism and capacity of podocyte turnover and regeneration in mice. Our data reveal that podocyte generation is mainly confined to glomerular development and may occur after acute glomerular injury, but it fails to regenerate podocytes in aging kidneys or in response to nephron loss.

Figure 1.

PECs are populating the glomerular tuft during kidney development and give rise to fully differentiated podocytes. (A) Schematic of inducible hPODXL.rtTA;tetO.Cre;mT/mG transgenic mice. (B) Labeling pattern in the glomerulus and possible ways of repopulation. (C) Expression of mG in PECs at P1 (arrow, labeled podocyte). Scale bar, 50 µm. (D) Podocyte foot process morphology in mG-labeled podocytes from hPODXL.rtTA;tetO.Cre;mT/mG and (E) hNPHS2.rtTA;tetO.Cre;mT/mG mice. Scale bars, 10 µm overview, 1 µm inlays. EGFP, enhanced green fluorescent protein; PCA, core promoter of chicken β-actin.

Figure 2.

Highly specific FACS analysis of podocyte populations. (A) Glomeruli were isolated by perfusion of the renal arteries with magnetic beads, followed by enzymatic digestion of the tissue and isolation of bead-filled glomeruli with a magnet. After digestion to single cells, podocytes were stained with a directly labeled antipodocin antibody. (B) mT/mG image of an hNPHS2.rtTA;TetO.Cre;mT/mG kidney previously induced with low efficiency. Scale bar, 10 µm. The same kidney was used for (C) analysis by flow cytometry. The podocyte population can be distinguished by antipodocin staining and can be split into mT- and mG-positive cells. (D) mT/mG image of an hNPHS2.Cre;mT/mG and (E) hNPHS2.rtTA;mT/mG glomerulus. Scale bar, 10 µm. (F) Validation of the flow cytometry–based method. Constitutively-labeled hNPHS2.Cre;mT/mG mice showed high specificity with very few false-positive cells, whereas no false-positive cells could be found in hNPHS2.rtTA;mT/mG mice. Error bars, SEM; n=3 per group. Scale bars, 10 µm. EGFP, enhanced green fluorescent protein.

Figure 3.

Acute podocyte loss in a DT model leads to limited podocyte renewal. (A) Schematic of transgenic podocyte-specific iDTR mice. (B) Albuminuria is dose-dependent after DT injection. (C) mT/mG fluorescent images of iDTR animals treated with a high (50 ng/g body weight) or low (5 ng and 2 ng/g body weight) dose of DT. (D) Nephrin and active caspase-3 staining 10 days after injection of DT. Scale bar, 10 µm. (E) Number of podocytes per glomerular cross-section after injection of DT, 2 ng/g body weight; >30 glomeruli, ≥10 animals per group; error bars, SEM; **P<0.01. (F) Albumin-to-creatinine ratio after injection of DT, 2 ng/g body weight. Error bars, SEM; n.s., not significant; *P<0.05; **P<0.01; ***P<0.001. (G) Flow cytometric analysis 4 weeks after injection of DT, 2 ng/g body weight; n≥10; error bars, SEM; *P<0.05. (H) Proportion of mT-positive cell increase in relation to ablated cells. EGFP, enhanced green fluorescent protein; PCA, core promoter of chicken β-actin.

Figure 4.

Nephron loss is not associated with podocyte renewal from bone marrow or PEC compartment. (A) Schematic of transgenic hNPHS2.rtTA;tetO.Cre;mT/mG mice and labeling pattern. (B) Schematic of the UNx protocol. (C) Kidney weight 12 weeks after UNx compared with control animals; n≥9; ***P<0.001. (D) Mean glomerular volume (*P<0.05) and (E) mean podocyte volume. n.s., not significant (P=0.06). (F) Number of podocytes per glomerulus. n.s., not significant (P=0.171). Error bars, SEM. (G) Flow cytometry analysis of labeled podocytes (n=9), (H) PECs (P=1, n=4) and (I) bone marrow cells (P=0.30, n=6) 12 weeks after nephrectomy compared with baseline values. Co, control; EGFP, enhanced green fluorescent protein; PCA, core promoter of chicken β-actin.

Figure 5.

Podocytes are not renewed during aging. (A) Schematic of hNPHS2.rtTA;tetO.Cre;mT/mG mice. (B) Induction protocol. (C) Flow cytometry analysis at 4 weeks and 12 months. Error bars, SEM; n.s., not significant (P=0.55); n>12. mT/mG fluorescent images at (D) 4 weeks and (E) 12 months (arrow, mT-labeled podocyte). Scale bar, 10 µm. (F) Schematic of hPODXL.rtTA;tetO.Cre;mT/mG mice. (G) Flow cytometry analysis at 4 weeks and 12 months. Error bars, SEM; n>6 per group; n.s., not significant (P=1). mT/mG fluorescent images at (H) 4 weeks and (I) 12 months (arrows, mG-labeled podocytes). Scale bar, 10 µm. (J) mG-labeled bone marrow was transplanted into irradiated mT mice. (K) Timeline of procedure. (L) Flow cytometry analysis 4 and 7 months after bone marrow transplant (BMT). Error bar, SEM; n=6 per group; n.s., not significant; P=0.054. mT/mG fluorescent images (M) 4 months and (N) 7 months after BMT (arrows, mG-positive cells). Scale bar, 10 µm. EGFP, enhanced greent fluorescent protein; PCA, core promoter of chicken β-actin.