Figure 1.
Expression and co-localization of polarity proteins to slit diaphragms.
(A) In mouse kidneys at postnatal day 10 (P10), aPKCλ/ζ, phospho-aPKCλ/ζ T403/410, and PAR6β localize to the glomerular capillary tufts in a continuous pattern. Bars, 20 µm. (B) Double immunofluorescence for nephrin (green) and PAR3 (magenta). White regions show where PAR3 co-localizes with nephrin in rat podocytes. (C) Immunogold particles for PAR3 localize at the cytoplasmic face of the slit diaphragms. FP, foot process.
Figure 2.
Selective depletion of aPKCλ from mouse podocytes results in renal failure.
(A) Double immunofluorescence for aPKCλ (green) and WT1 (magenta) shows that the signals for aPKCλ are below the detectable level in mutant (aPkcλΔE5/floxE5;Nphs1-Cre, cKO) podocytes at P0 (arrowheads), whereas the tubular epithelial cells retained aPKCλ in the mutant (arrows). Bar, 20 µm. (B) No significant difference in the gross appearance of mutant (cKO) and control kidneys at P0. (C) One microliter of urine from each mouse at P0 was analyzed along with bovine serum albumin (BSA) by SDS-PAGE and CBB staining. (D) Serum creatinine concentration in mutant mice compared with controls. Triangles, controls carrying the Nphs1-Cre transgene; diamonds, mutants; bars, medians. The p values were determined by the two-tailed Mann–Whitney U-test. (E) Mutant mice (cKO) show growth retardation by the age of 4 weeks. Values are mean±S.E.M. (F) Kaplan–Meier survival curve for mutant (cKO) and control mice.
Figure 3.
Selective depletion of aPKCλ from mouse podocytes causes FSGS.
Periodic acid-Schiff staining (PAS) and immunohistochemistry for nephrin and synaptopodin in mutant (cKO) and control kidneys at P0, P10 and P21 show that mutant mice develop segmental to global glomerulosclerosis. Boxed regions are enlarged to show an irregular pattern of nephrin staining in mutant podocytes. Loss of podocytes (arrows) and occasional adhesion of glomeruli to Bowman's capsules (arrowheads) reveal the development of focal segmental glomerulosclerosis. Consistent with massive proteinuria in mutant mice at birth, PAS staining reveals occasional hyaline droplets, representing reabsorbed urinary protein, in the proximal renal tubules at P0. Bars, 50 µm.
Figure 4.
Ultrastructure of mutant podocytes.
(A) Although fine foot processes and slit diaphragms are observed at P0 (arrowheads), the foot processes form irregular adhesions and slit diaphragms are apically mislocalized at P7 (arrows). Mutant podocytes at P10 demonstrates effacement of foot processes and irregular adhesions between foot processes (asteriscs). Apparently, these adhesions did not form tight junctions. The glomerular basement membrane (GBM) is not significantly affected. US, urinary space; EnC, endothelial cell; CL, capillary lumen. Bars, 200 nm. (B) Immunogold electron microscopy in mutant podocytes at P7 reveals that podocalyxin (5-nm gold particles) loses its apically restricted localization, whereas ZO-1 (10-nm gold particles) localizes at the cell–cell junctions of podocytes (arrows). US, urinary space; CL, capillary lumen. Bar, 200 nm.
Figure 5.
PAR3 forms a complex with aPKC, nephrin, and podocin in vivo.
(A) The distribution of PAR3, aPKCλ, nephrin, and podocin was analyzed in differentially extracted fractions of rat renal tubules or glomeruli (Tub, Gl). Tub-S1, Gl-S1, highly soluble fractions extracted with 0.1% TritonX-100; Tub-S2, Gl-S2, relatively soluble fractions extracted with 1.0% TritonX-100 and 20 mM CHAPS; Tub-P2, Gl-P2, insoluble fractions. The blotted filter with anti-nephrin antibody was stripped and reprobed with anti-PAR3 antibody. (B) Immunoprecipitation for PAR3 was carried out on both glomerular fractions (Gl-S1, Gl-S2) using two independent anti-PAR3 antibodies (UBI, C2-2AP) and the immunoprecipitates were analyzed by immunoblotting. The blotted filter with anti-nephrin antibody was stripped and reprobed with anti-PAR3 antibody.
Figure 6.
PAR3 binds directly to nephrin in vitro.
(A) Domain structure, deletion constructs, and point mutants of PAR3 (left) and nephrin (right). The numbers refer to amino acids. PDZ, PDZ domain; aPKC-BR, aPKC-binding region; ECD, extracellular domain; TMD, transmembrane domain; ICD, intracellular domain. The C-terminal amino acids of nephrin show similarities to the type II PDZ-binding motif. The four Ys on nephrin represent the tyrosine residues responsible for binding to phosphatidylinositol-3 kinase p85 or Nck. (B) Mapping of the region in PAR3 that is required for binding to nephrin. Glutathione-S-transferase (GST)–nephrinICD-immobilized beads were incubated with 293T cell lysate expressing various T7-tagged PAR3 constructs as indicated. The bound proteins were analyzed by immunoblotting and CBB staining. (C) PAR3 directly binds to nephrin through the PDZ domains of PAR3. Purified GST–nephrin ICD2 was incubated with various PAR3 constructs fused with maltose-binding protein (MBP). The resulting complexes were immobilized on glutathione-Sepharose beads and analyzed by immunoblotting. (D) Mapping of the region in nephrin that is required for binding to the PDZ domains of PAR3. Purified MBP–PDZ123 was incubated with various nephrin constructs fused with GST, each at a final concentration of 50 nM. The resulting complexes were analyzed by immunoblotting and CBB staining. Numbers show relative intensities of the bands.
Figure 7.
aPKC activity is required for the appropriate distribution of nephrin and podocin.
(A) The aPKC inhibitor, myr-ζPS peptide (ζPS), significantly disturbs the appropriate distribution of nephrin and podocin. Differentially extracted proteins (S1, S2, P2; see Methods) from rat glomeruli incubated for 30 minutes with or without 20 µM ζPS were analyzed by immunoblotting (left). Quantitative analysis from five to seven independent experiments is shown (right). Values shown are mean±S.E.M. The p values were determined by two-tailed Student's t-test. (B) No significant difference in the formation of the nephrin–podocin complex by aPKC inhibitor, myr-ζPS peptide (ζPS). Glomeruli isolated from rat kidneys were incubated for 30 minutes with or without 20 µM of ζPS. The differentially extracted proteins (S1, S2) were incubated with normal rabbit IgG (NRb) or anti-podocin antibody (Pod). Immunoprecipitates (IP-S1, IP-S2) were analyzed by immunoblotting. Numbers show relative intensities of the bands. PAR3 and aPKC were consistently co-immunoprecipitated with podocin irrespective of ζPS treatment.