Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2007 Jul;171(1):139-52.
doi: 10.2353/ajpath.2007.061116.

Disruption of glomerular basement membrane charge through podocyte-specific mutation of agrin does not alter glomerular permselectivity

Scott J Harvey  1 , George Jarad, Jeanette Cunningham, Angelique L Rops, Johan van der Vlag, Jo H Berden, Marcus J Moeller, Lawrence B Holzman, Robert W Burgess, Jeffrey H Miner
Affiliations

Disruption of glomerular basement membrane charge through podocyte-specific mutation of agrin does not alter glomerular permselectivity

Scott J Harvey et al. Am J Pathol. 2007 Jul.

Abstract

Glomerular charge selectivity has been attributed to anionic heparan sulfate proteoglycans (HSPGs) in the glomerular basement membrane (GBM). Agrin is the predominant GBM-HSPG, but evidence that it contributes to the charge barrier is lacking, because newborn agrin-deficient mice die from neuromuscular defects. To study agrin in adult kidney, a new conditional allele was used to generate podocyte-specific knockouts. Mutants were viable and displayed no renal histopathology up to 9 months of age. Perlecan, a HSPG normally confined to the mesangium in mature glomeruli, did not appear in the mutant GBM, which lacked heparan sulfate. Moreover, GBM agrin was found to be derived primarily from podocytes. Polyethyleneimine labeling of fetal kidneys revealed anionic sites along both laminae rarae of the GBM that became most prominent along the subepithelial aspect at maturity; labeling was greatly reduced along the subepithelial aspect in agrin-deficient and conditional knockout mice. Despite this severe charge disruption, the glomerular filtration barrier was not compromised, even when challenged with bovine serum albumin overload. We conclude that agrin is not required for establishment or maintenance of GBM architecture. Although agrin contributes significantly to the anionic charge to the GBM, both it and its charge are not needed for glomerular permselectivity. This calls into question whether charge selectivity is a feature of the GBM.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Structure of mouse agrin and properties of the mutant Agrn alleles. The N-terminal domain of agrin (NtA) that binds laminin γ1 is followed by follistatin-like (F), laminin EGF-like (L), serine/threonine-rich (ST), sperm protein/enterokinase/agrin (SEA), and EGF-like (E) domains. Laminin globular (G) domains at the C terminus bind integrin and dystroglycan receptors. Two putative attachment sites for GAGs are shown. A: A conditional allele (Agrnfl) was generated by introducing loxP sites within introns 6 and 33. Cre-mediated excision deletes most coding exons (light gray, numbered according to Rupp et al32) but leaves flanking exons (dark gray) and the untranslated regions (black) intact. B: In a knockout allele (Agrndel), sequence from within exon 6 through intron 33 is replaced by a loxP-flanked neo cassette. Transcripts derived from the Agrndel and Agrnfl alleles encode truncated proteins that lack the C terminus due to frameshifts. C: A gene trap allele (Agrnβ-geo) intercepts the Agrn gene 3′ of the last NtA-encoding exon. The positions of primers used for genotyping are indicated (arrows).
Figure 2
Figure 2
Podocytes are responsible for deposition of agrin in the GBM, and mutants synthesize N-terminal truncated forms of the protein. C-terminal agrin antibodies label the BM of pre-capillary loop-stage glomeruli (A) and mature GBM (B) in normal fetal kidney. BMs of collecting ducts, peritubular capillaries, and vascular smooth muscle cells are also stained. All BMs in Agrndel/del (C) and Agrnβ-geo/β-geo (D) mutants are negative. N-terminal antibodies label tubular, vascular, and glomerular BMs in normal fetal mice (E and F) and Agrndel/del mutants (G), whereas Agrnβ-geo/β-geo mutants are negative (H). C-terminal antibodies give the same staining pattern in normal adult kidney (I and J) noted in fetal mice but fail to label the GBM of adult podocyte-specific knockouts (K and L), indicating endothelial cells are not a significant source of agrin in mature GBMs. The N terminus of agrin is detected in most BMs of normal adults (M and N), and there is comparable labeling in conditional mutants (O and P). Thus, the Agrndel and Agrnfl alleles encode N-terminal truncated forms of agrin that localize to BMs. Scale bar: 150 μm in I, K, M, and O; 100 μm in F--H; 50 μm all other panels.
Figure 3
Figure 3
Truncated forms of agrin synthesized in mutants do not carry GAG chains. A: The Agrnfl and Agrndel alleles encode fragments of agrin with potential GAG attachment sites. COS-7 cells were transfected with a Myc-tagged construct (Agrin1–602) that represents the protein derived from a Cre-recombined Agrnfl allele to assess whether these sites (arrows) carry GAG chains. Agrin1–1171 includes known attachment sites and served as a control. B: Western blotting of medium and cell lysates for the Myc epitope detected bands of the expected size in samples from Agrin1–1171- and Agrin1–602-transfected cells whereas mock-transfected controls were negative. C: Agrin1–1171 was detected in the medium as a smear that shifted to a more discrete band on chondroitinase (cABC) but not heparinase (Hep) digestion, indicating that it carries predominantly CS-GAGs. Agrin1–602 was insensitive to digestion with cABC and Hep (not shown), indicating that it lacks GAG chains. D: Blotting of medium for perlecan revealed a high-molecular weight species that shifted to a discrete band consistent with the size of the perlecan core protein after Hep digestion, demonstrating that COS-7 cells possess the biosynthetic pathways necessary for synthesis of HSPGs.
Figure 4
Figure 4
Agrin mutants lose GBM-HS in the absence of compensation by perlecan. Perlecan is detected in all BMs of normal fetal kidney (A) and in agrin-deficient mice (B; E18.5 Agrndel/del shown). In podocyte-specific knockouts up to ∼7 months of age (D), staining for perlecan is properly localized to the mesangium and indistinguishable from controls (C). Antibody JM-403 recognizing native HS labels the GBM in normal adults (E) but not conditional mutants (F). Dual-labeling with mAb MI-91 to the N terminus of agrin is shown to localize glomeruli in control and mutant sections (G and H, respectively). Scale bar: 50 μm in A–D; 75 μm in E–H.
Figure 5
Figure 5
Expression of BM components and their receptors in mutant kidney. The laminin β2 chain is detected in normal fetal GBM (A), whereas Agrnβ-geo/β-geo mutants (B) show both BM and punctate labeling, suggesting β2-containing trimers were co-trapped by the NtA/β-geo fusion protein. Nidogen is detected in all BMs of normal fetal mice (C), and labeling is comparable in Agrnβ-geo/β-geo mutants (D). Glomerular staining for integrin α3 (E) and dystroglycan (G) in normal fetal mice shows a localization similar to that observed in Agrndel/del glomeruli (F and H, respectively). Scale bar = 50 μm.
Figure 6
Figure 6
Agrin mutants show no significant renal histological or ultrastructural abnormalities. Shown are representative H&E sections from E18.5 control (A and B) and Agrndel/del mutant kidneys (D and E). Glomerular ultrastructure in E18 Agrndel/del mutants (F) was indistinguishable from controls (C). Incomplete fusion of the GBM evident in the mutant is normal at late stages of glomerulogenesis. At 7 weeks of age, glomeruli of conditional mutants (J and K) display no histological or ultrastructural defects compared with controls (G and H). Mutant kidneys were normal by light microscopy up to ∼8 months of age, the latest time point examined. By EM, subepithelial irregularities of the GBM (ie, “bumps”) were noted in mutants beyond 4 months of age (L) that were more prominent than in controls (I). Scale bars: 500 μm in A and D; 50 μm in B, E, G, and J; 1 μm in all other panels.
Figure 7
Figure 7
Agrin mutants have a severe GBM charge defect. Anionic sites in the glomerular capillary wall were detected with the cationic probe PEI. In normal fetal mice (A, E17.5; C, E19), there is punctate labeling of the GBM distributed in a regular pattern along both laminae rarae. Subendothelial labeling was comparable in littermate Agrndel/del mutants, but subepithelial labeling was markedly decreased (B and D). Quantitative analysis revealed a significant reduction in the number of anionic sites in mutants (E). In normal adults, subendothelial labeling is diminished, but discrete subepithelial labeling remains (F, 4 months; G, 11 months). In adult conditional mutants, PEI labeling is markedly diminished (H, 4 months; I, 9 months), and quantitative analysis revealed a significant reduction in subepithelial anionic sites (J). Scale bars: 200 nm in C and D; 500 nm in all other panels.
Figure 8
Figure 8
The glomerular filtration barrier is not compromised in conditional agrin knockout mice. A: Urinary protein-to-creatinine ratios of adult conditional knockouts and controls were not significantly different (P = 0.34). Mutants do not show elevated urinary excretion of albumin (B) or immunoglobulin (C) compared with controls (P = 0.99 and 0.70, respectively). D: Mice were administered daily injections of BSA to test the integrity of the glomerular filtration barrier in the setting of a challenge. There was no significant difference in urinary protein-to-creatinine ratios between conditional knockout and control mice over time (P = 0.99). E: Renal excretion of the anionic tracer carboxymethyl Ficoll-70 was equivalent in conditional mutants and controls (P = 0.83), supporting the lack of a contribution to the charge barrier by the GBM. Values are mean ± SEM.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Chang RL, Deen WM, Robertson CR, Brenner BM. Permselectivity of the glomerular capillary wall: III. Restricted transport of polyanions. Kidney Int. 1975;8:212–218. - PubMed
    1. Bohrer MP, Baylis C, Humes HD, Glassock RJ, Robertson CR, Brenner BM. Permselectivity of the glomerular capillary wall: facilitated filtration of circulating polycations. J Clin Invest. 1978;61:72–78. - PMC - PubMed
    1. Rennke HG, Patel Y, Venkatachalam MA. Glomerular filtration of proteins: clearance of anionic, neutral and cationic horseradish peroxidase in the rat. Kidney Int. 1978;13:278–288. - PubMed
    1. Comper WD, Glascow EF. Charge selectivity in kidney ultrafiltration. Kidney Int. 1995;47:1242–1251. - PubMed
    1. Kerjaschki D, Sharkey DJ, Farquhar MG. Identification and characterization of podocalyxin: the major sialoprotein of the renal glomerular epithelial cell. J Cell Biol. 1984;98:1591–1596. - PMC - PubMed

Publication types