Ligation of α-Dystroglycan on Podocytes Induces Intracellular Signaling: A New Mechanism for Podocyte Effacement? Authors: Nils P. J. Vogtländer, Henk Jan Visch, Marinka A. H. Bakker, Jo H. M. Berden, Johan van der Vlag Published: 2009, PLOS ONE DOI: 10.1371/journal.pone.0005979 ABSTRACT α-Dystroglycan is a negatively charged glycoprotein that covers the apical and basolateral membrane of the podocyte. Its transmembrane binding to the cytoskeleton is regulated via tyrosine phosphorylation (pY892) of β-dystroglycan. At the basolateral side α-dystroglycan binds the glomerular basement membrane. At the apical membrane, it plays a role in the maintenance of the filtration slit. In this study, we evaluated whether ligation of α-dystroglycan with specific antibodies or natural ligands induces intracellular signaling, and whether there is an effect on podocyte architecture. Conditionally immortalized podocytes were exposed in vitro to antibodies to α-dystroglycan, and to fibronectin, biglycan, laminin and agrin. Intracellular calcium fluxes, phosphorylation of β-dystroglycan and podocyte architecture were studied. Antibodies to α-dystroglycan could specifically induce calcium signaling. Fibronectin also induced calcium signaling, and led to dephosphorylation of pY892 in β-dystroglycan. Ligation of α-dystroglycan resulted in an altered actin architecture, a decreased number of podocyte pedicles and a more flattened appearance of the podocyte. We conclude that ligation of α-dystroglycan on podocytes induces intracellular calcium signaling, which leads to an altered cytoskeleton architecture akin to the situation of foot process effacement. In particular the ability of fibronectin to induce intracellular signaling events is of interest, since the expression and excretion of this protein is upregulated in several proteinuric diseases. Therefore, fibronectin-induced signaling via dystroglycan may be a novel mechanism for foot process effacement in proteinuric diseases. INTRODUCTION Podocytes are highly specialized glomerular visceral epithelial cells that cover the outside of the glomerular basement membrane [1] . Podocytes together with the glomerular basement membrane and the glomerular endothelium constitute the size and charge selective filter of the glomerulus. The podocyte cell body forms primary and secondary extensions, which form foot processes that interdigitate with those of neighboring podocytes and that are attached to each other by the slit diaphragm. In many glomerular diseases with proteinuria, podocyte foot processes are effaced, as their extensions are lost. Dystroglycan was identified as a component of the dystrophin glycoprotein complex, which is involved in muscular dystrophies [2] , [3] . Thereafter, dystroglycan was described as a component in epithelial cells [4] , [5] , and we have shown that dystroglycan is expressed by podocytes, both at the basolateral and the apical cell membrane [6] , [7] . The dystroglycan encoding gene DAG1 was first cloned and sequenced in 1992. Dystroglycan is post-translationally cleaved into the extracellular α subunit and the integral membrane β subunit, which non-covalently interact [2] . Dystroglycan provides a link between the extracellular matrix like the glomerular basement membrane to the actin cytoskeleton of the podocyte. α- Dystroglycan binds to laminin G modules that are present in laminin and agrin, whereas β-dystroglycan is connected to the actin cytoskeleton via dystrophin or its ubiquitous homologue utrophin in the kidney [6] , [8] , [9] . α-Dystroglycan undergoes massive O-mannosyl glycosilation with terminal sialic acids, which serve by virtue of their negative charge as an anti-adhesion layer on the apical podocyte membrane, analogous to podocalyxin, a sialic acid-rich glycoprotein that also covers the apical cell membrane of the podocyte [7] . The glycosilation of α-dystroglycan is necessary for its binding to laminin G modules [8] . We have shown previously that deglycosilation of α-dystroglycan by reactive oxygen species results in a loss of binding to agrin and laminin [10] . The small leucine-rich repeat at the C-terminal tail of α-dystroglycan has been shown to bind to chondroitin sulfate A and/or C on biglycan [11] . Biglycan is a small leucine-rich proteoglycan with chondroitin sulfate or dermatan sulfate side chains. It has been shown to be predominantly localized in the tubular interstitium and intrarenal arteries. Minimal expression of biglycan in the mesangial matrix has been described for normal human kidney and various glomerular diseases [12] – [16] . TGF-β-mediated mesangial upregulation of biglycan has been described in Thy-1 mediated membranoproliferative glomerulonephritis in the rat [17] . TGF-β induces fibrosis and as a consequence it is especially co-localized with collagen type I in fibrosis [13] .The assembly of chondroitin sulfate side chains decreases after exposure to puromycin-aminonucleoside and dexamethasone [18] , [19] . Biglycan is not detectable in the urine of patients with glomerular diseases [15] . It was recently shown that the C-terminal domain of muscle α-dystroglycan can also bind to fibronectin and fibrinogen [20] . Fibronectin is a 500 kDa protein that consists of two similar 250 kDa subunits linked via a disulfide bond. By post-translational processing plasma and tissue isoforms are formed. An increased expression of both plasma and tissue fibronectin has been documented in various glomerular diseases like diabetic nephropathy, puromycin-aminonucleoside and adriamycin nephrosis [21] – [24] . Furthermore, urinary excretion of fibronectin has been shown to correlate with proteinuria in IgA-nephropathy, membranous glomerulopathy, and diabetic nephropathy and with glomerulosclerosis in lupus nephritis and chronic glomerulonephritis [25] – [27] . The extracellular N-terminal domain of β-dystroglycan interacts with α-dystroglycan, whereas the intracellular C-terminal domain interacts with utrophin [28] – [30] . In particular, a PPxY motif including a tyrosine residue at position 892 in β-dystroglycan in the C-terminal domain is necessary for interaction with WW domains in utrophin [28] – [30] . WW domains are 30–40 amino acid motifs containing two conserved tryptophan residues, that bind to proline-rich sequences [31] . The interaction of β-dystroglycan with dystrophin/utrophin is stabilized by a helix-turn-helix motif, i.e. EF hand, close to the PPxY motif in dystrophin/utrophin [29] , [32] . Furthermore, dephosphorylation of Y892 (pY892) in β-dystroglycan enables binding to utrophin/dystrophin [30] , [33] . Phosphorylation of β-dystroglycan enhances recruitment of SH2/SH3 domain containing proteins like c-Src, Fyn, Csk, NCK, SHC and Grb-2, which all are involved in signaling events [34] , [35] . This suggests a role for dystroglycan in signaling. In this study, we investigated whether ligation of α-dystroglycan by monoclonal antibodies against α-dystroglycan and natural ligands for α-dystroglycan, such as laminin, agrin, biglycan and fibronectin, leads to intracellular signaling as assessed by measuring calcium fluxes and to changes in the phosphorylation status of β-dystroglycan at position 892 (Y892). In addition, we investigated whether α-dystroglycan ligation had an impact on actin cytoskeleton architecture and podocyte morphology. RESULTS The binding of cultured mouse podocytes to EHS laminin, agrin, collagen A and Matrigel coated plates was quantified with the hexosaminidase assay. Podocytes were pre-incubated with α-dystroglycan specific antibodies, IIH6 or VIA4.1, directed to carbohydrate epitopes on α-dystroglycan, prior to seeding on the coatings. It is known that monoclonal antibody IIH6 inhibits the binding of α-dystroglycan to laminin G modules, whereas monoclonal antibody VIA 4.1 does not [4] , [36] . Indeed, the number of podocytes binding to laminin decreased after pre-incubation with IIH6, but remained unaltered after pre-incubation with VIA4.1 ( Figure 1A ). Pre-incubation with isotype controls for either of the α-dystroglycan antibodies (TEPC and MOPC respectively) did not affect the binding. As expected, similar results were obtained with agrin as coated ligand (data not shown). However, the binding of podocytes to collagen A, which is comprised of type I collagen that does not contain laminin G modules, was unaltered by the addition of either of the two α-dystroglycan specific antibodies ( Figure 1B ). Furthermore, the binding of podocytes to Matrigel© which is considered as a surrogate basement membrane preparation containing a mixture of several extracellular matrix molecules, could not be prevented by MoAb IIH6 (data not shown). Therefore, monoclonal antibody IIH6 specifically inhibits the binding of mouse podocytes to laminin EHS and agrin, indicating an important role for α-dystroglycan for the binding of podocytes to laminin G modules containing proteins such as laminin and agrin in the glomerular basement membrane. Podocytes were seeded on laminin (A) or collagen A coatings (B) after pre-incubation with either monoclonal antibody IIH6 using TEPC as an isotype control, or with VIA 4.1 using MOPC 21 as an isotype control. Monoclonal antibody IIH6 specifically inhibits podocyte binding to laminin, but not to collagen A. Next, we analyzed whether α-dystroglycan on mouse podocytes could mediate outside-in signaling. The ability of monoclonal antibodies IIH6 and VIA 4.1 and their respective isotype controls to induce calcium fluxes was assessed by the fura 2 method, which measures the cytosolic calcium concentration. Both monoclonal antibodies could specifically induce calcium signaling after ligation of α-dystroglycan, whereas the isotype controls did not ( Figure 2A, B ). The peak of the cytosolic calcium concentration was similar in the presence or absence of extracellular calcium, indicating that the cytosolic calcium influx was derived from the intracellular calcium stores. Although these experiments indicate that ligation of α-dystroglycan mobilizes intracellular calcium stores, this induction by specific antibodies is however, not a physiological event. Therefore, the ability of natural ligands of α-dystroglycan to induce a similar rise in cytosolic calcium was investigated. Both laminin and agrin, which both can bind to glomerular α-dystroglycan [10] , were not able to induce a rise in cytosolic calcium ( Figure 2C, D ). Biglycan induced a modest calcium influx ( Figure 2F ). Since the interaction of biglycan with muscle α- dystroglycan depends on the chondroitin sulfate side chains of biglycan [11] , we also tested the effect of adding chondroitin sulfate A, B and C, and heparan sulfate. However, both chondroitin sulfate preparations as well as heparan sulfate did not induce calcium mobilization in the cytosol (data not shown). Since fibronectin is a known ligand of muscle α-dystroglycan, and a protein which is involved in several glomerular diseases [21] – [27] , we evaluated the effect of fibronectin on calcium fluxes. Since we could not rule out that RGD motives in fibronectin could induce signaling via α 3 β 1 integrin, these RGD fibronectin receptor recognition sequences in integrins were blocked by pre-incubation with RGD peptide. We included RGD blocking, although α3β1integrins are mainly expressed at the basal and not at the apical side of podocytes. Addition of RGD peptide did not result in calcium fluxes. Fibronectin, after blocking of α 3 β 1 integrin with RGD peptides, induced the strongest cytosolic calcium influx of all 4 endogenous ligands of α-dystroglycan tested ( Figure 2E ). This finding indicates that fibronectin can induce calcium fluxes in podocytes, via other molecules than integrins. To further investigate whether the dystroglycan complex is involved in fibronectin-induced calcium fluxes in podocytes, its effect on the phosphorylation state of Y892 in β-dystroglycan was investigated. It appeared that stimulation of podocytes by fibronectin (concomitantly with blocking with RGD peptide) led to a decrease in phosphorylation of Y892, while the total amount of β-dystroglycan was unaltered ( Figure 3 ). This latter finding suggests that α-dystroglycan is involved in fibronectin-induced calcium influxes in podocytes. We have also studied the expression level of pY892 in renal sections of control rats and rats with adriamycin nephropathy, with normal and effaced podocytes, respectively. We applied several antigen retrieval techniques, and an epitope preservation approach by inclusion of phosphatase inhibitors. However, despite an additional signal amplification, the detectable expression levels of pY892 in both control and adriamycin rats were extremely low and, therefore, not conclusive (data not shown), which seems in contrast to our clear findings with Western blot analysis ( Figure 3 ). However, a similar discrepancy between immunofluorescence and Western blot analysis has been observed on skeletal muscle by others [37] . It has been described that differentiation of conditionally immortalized podocytes induces process (filopodia) formation [38] . In figure 4 we show the effect of the anti-α-dystroglycan monoclonal antibodies and fibronectin on the morphology of podocytes and actin cytoskeleton. All ligands induced a retraction of filopodia and a reorganization of the cytoskeleton. Statistical analysis of these effects revealed a significant effect for the anti-α-dystroglycan monoclonals IIH6 and VIA 4.1, but only a trend for fibronectin ( Table 1 ). However, in several fibronectin-stimulated cells retraction of filopodia was clearly visible as shown in the lower panels of figure 4 . This reorganization of the actin cytoskeleton may be a situation reminiscent of foot process effacement. These data suggest that fibronectin-induced signaling events, mediated by the dystroglycan complex leads to an altered podocyte architecture. Mouse podocytes were differentiated on collagen A coatings for 2 weeks. Intracellular calcium concentrations were monitored by the fura-2 method, with and without extracellular calcium. Both α-dystroglycan specific monoclonal antibodies induced calcium fluxes, most likely from intracellular stores. Isotype controls were negative (A, B). The laminin G modules-containing proteins agrin and laminin were unable to induce a calcium influx in podocytes (C, D). From the natural ligands of α-dystroglycan tested, fibronectin induced the most pronounced calcium influx. Note that blocking of potential integrin binding sites with a RGD peptide did not influence the fibronectin effect (E). The proteoglycan biglycan induced a modest calcium influx (F). Cultured mouse podocytes were incubated for 3 minutes with fibronectin after which a decrease of pY892 was found in Western blot analysis of whole podocyte cell lysates (right panel), whereas the total amount of β-dystroglycan was not affected (left panel). Total β-dystroglycan was probed with NCL-BDG and pY892 was probed with anti-pY892. Differentiated podocytes on collagen coatings were incubated with either monoclonal antibodies IIH6 or VIA 4.1, their respective isotype controls (TEPC or MOPC) or fibronectin. The podocyte pedicles (arrow) are probed with antibodies against Mena (green, which also localizes along stress fibers and in focal contacts at the tips of stress fibers [58] , [65] ). The pedicles were lost after 6 hours of incubation with the antibodies. As these pedicles may be the in vitro manifestation of foot processes, their retraction may imply foot process effacement. Percentage of podocytes with smooth surfaces were counted, which yielded for IIH6 60% versus TEPC 30% (χ 2  = 20.8, p<0.001), for VIA4.1 51% versus MOPC 28% (χ 2  = 9.01, p<0.01) and fibronectin 38% versus control 30% (χ 2  = 1.3, p = 0.25). (Red = phalloidin (actin); blue = DAPI (DNA)). Smooth 0–0.2 filopodia/5 µm, intermediate 0.2–1 filopodia/5 µm or filopodia rich ≥1 filopodia/5 µm cell border. Significantly different when comparing frequencies in both columns; IIH6 vs. TEPC183 χ 2  = 20.8, p<0.001. VIA4.1 vs. MOPC21 χ 2  = 9.01, p<0.01 Fibronectin+(FN+) vs. control (FN-) χ 2  = 1.3, p = 0.25. TABLE: Percentages of podocytes with smooth, intermediate or filopodia rich surfaces after in vitro ligation with α-dystroglycan specific antibodies, fibronectin and their respective controls. IIH6 ** TEPC 183 ** VIA4.1 *** MOPC21 *** FN+ FN- Filopodia rich * 12 25 17 28 18 23 Intermediate * 28 45 32 44 44 46 Smooth * 60 30 51 28 38 30 DISCUSSION α-Dystroglycan is a negatively charged glycoprotein, which covers the apical and basolateral membrane of the podocyte. At the basolateral side α-dystroglycan binds laminin G modules in agrin and laminin. The α 3 β 1 integrin is also expressed at the basolateral side of podocytes, where it extracellulary binds to laminin, fibronectin and collagen. In this study we investigated the importance of α-dystroglycan for the binding of podocytes to laminin, agrin and several other extracellular matrix molecules with a new binding assay. Binding of monoclonal antibody IIH6 but not monoclonal antibody VIA4.1 inhibited the binding of podocytes to laminin and agrin. Our findings with podocytes are similar to the data for the binding of Schwannoma cells to laminin as previously described [36] . This is also underlined by the observation that monoclonal antibody VIA 4.1 did not inhibit the binding/growth of podocytes to collagen nor to laminin and agrin, since the carbohydrates recognized by VIA4.1 are not involved in the binding to laminin G moieties [4] , [36] . However, no impairment of binding of podocytes to collagen type I coated plates was observed after incubation with monoclonal antibody IIH6. Furthermore, binding of podocytes to Matrigel-coated plates also could not be inhibited by IIH6, which not unexpectedly suggests that podocytes also can bind to additional molecules without laminin G modules. As over 50% of the binding of podocytes to laminin was lost by the addition of monoclonal antibody IIH6, it can be concluded that for binding to laminin α-dystroglycan on podocytes is of equal or even greater importance than α 3 β 1 integrins. Therefore, our in vitro findings suggest that α-dystroglycan represents an important mechanism for podocytes to anchor to the glomerular basement membrane. At the apical domain of the podocyte the assumed function of the dystroglycan complex is maintenance of the filtration slit by its negative charge [7] . In this study, we investigated whether dystroglycan plays a role in outside-inside signaling events. Indeed, we observed a specific cytosolic calcium influx in cultured mouse podocytes after ligation of α-dystroglycan by two different anti α-dystroglycan antibodies. This rise in the cytosolic calcium concentration comes mainly from the intracellular calcium stores, since it was not dependent on the extracellular calcium concentration. This increase in cytosolic calcium is presumably mediated via inositol-1,4,5 tri-phosphate in the endoplasmatic reticulum [39] . Since induction of calcium fluxes by these anti α-dystroglycan specific antibodies is not a physiological event, we were interested in the effects of natural ligands of dystroglycan like, laminin, agrin, biglycan and fibronectin. From these ligands fibronectin induced the most pronounced signal, biglycan induced only a weak increase in cytosolic calcium, while agrin and laminin were not able to induce such an effect. This difference may be related to the fact that the function of α-dystroglycan on the basal side of the podocyte is different from the function at the apical side. Furthermore, clustering of α-dystroglycan can be induced by the antibodies, fibronectin and biglycan, as they may all serve as bivalent ligands [20] . In contrast, agrin and laminin probably serve as monovalent ligands. Tandem LG modules are necessary for α-dystroglycan binding. Since agrin contains only 3 LG modules, only one tandem is available [40] . Laminin possesses 5 LG modules which enable the formation of maximally two LG tandems. However, their spatial relationship within the molecule may be to close to cross link α-dystroglycan [41] , for review see [42] . The fibronectin induced rise in cytosolic calcium can in theory be mediated via many complexes. The integrin α 3 β 1 can bind fibronectin via a RGD module, and is intracellulary linked to integrin linked kinase (ILK) which is involved in regulation of the actin cytoskeleton, morphology and cell survival via signaling events [43] , [44] . However, the involvement of integrins in the fibronectin-induced rise in cytosolic calcium was excluded by blocking with an RGD peptide. Furthermore, probing of podocytes with fibronectin induced a dephosphorylation of β-dystroglycan, which makes a role for α-dystroglycan in fibronectin induced signaling likely. Immunoprecipitation studies in our laboratory have indeed showed that glomerular α-dystroglycan can bind fibronectin (data not shown), confirming previously published observations [20] . Intact fibronectin chains (∼250 kDa) can only pass the glomerular basement membrane if proteinuria is present, and indeed urinary fibronectin correlates with proteinuria [25] – [27] . Importantly, tissue fibronectin is upregulated during several glomerular diseases [21] – [24] . The unraveling of the downstream signaling pathways after dystroglycan ligation in podocytes remains challenging. The PPxY domain of β-dystroglycan can recruit SH2/SH3 domain-containing proteins like c-Src, Fyn, Csk, NCK, SHC and Grb-2. These interactions are enhanced by phosphorylation of pY892 [34] . From these, c-Src has been shown to be able to phosphorylate Y892 in c-Src overexpressing COS7 cells [37] . Also the expression of pY892 β-dystroglycan increases in COS7 cells when cultured on coatings of laminin, agrin or fibronectin [37] . Caveolin 1 and 3 co-express with β-dystroglycan in lipid raft domains, and bind when Y892 is phosphorylated. Caveolin 3 has been shown to inhibit the c-Src induced phosphorylation of Y892. [45] . Furthermore, the proline-rich SH3 region of Grb-2 binds at the C terminal side of β-dystroglycan [35] . p125 FAK has been shown to be linked to Grb-2, but not directly to β-dystroglycan [46] . Taken together, the dynamic role of β-dystroglycan in signaling seems to provide docking sites for multiple kinases when phosphorylated, or to provide a stable anchorage to the actin cytoskeleton when dephosphorylated. A similar phenomenon was illustrated by a study in which the Src family kinases were shown to be important for the stability of dystroglycan containing clusters induced by agrin [47] . β-dystroglycan also is able to bind ezrin, which also affects its interaction with the actin cytoskeleton [48] . Dystroglycan suppresses laminin 1 and -10/11 induced ERK activation via α 6 β 1 integrin and direct linking of β-dystroglycan to the ERK-MAP kinase pathway has also been described in COS 7 cells [49] , [50] . Recently, genetic interaction screens on presumed modifier genes for dystroglycan and dystrophin in many drosophila melanogaster mutants, added even more putative signaling pathways, like the Notch, TGF-β, EGFR, Semaphorin-Plexin, Frazzled-Netrin and Slit-Robo pathways [51] . Unfortunately, in co-immunoprecipitation experiments we could not find any binding of β-dystroglycan to Src, Grb2, p125 FAK and ezrin in rat kidney lysates (data not shown). So the signaling pathways linked to β-dystroglycan in podocytes remains subject for further investigation. Alterations in podocyte shape and function are calcium dependent. This has been shown in the calcium switch model in podocytes, in which 90 minutes removal of extracellular calcium results in an increased paracellular flux of dextrans [52] . Mutations in the canonical transient receptor potential 6 (TRPC6) result in an increased amplitude of the calcium current from the extracellular calcium stores, and eventually leads to a form of autosomal dominant focal segmental glomerulosclerosis [53] , [54] . However, the calcium flux induced via α-dystroglycan is derived from the intracellular calcium stores, which suggests that TRPC6 is not involved in these events. Furthermore, podocyte shape alterations can be induced by sialidase or puromycin aminonucleoside in rats; dephosphorylation of ezrin linked to podocalyxin via NHERF2 results in dissociation from the actin cytoskeleton and concomitant foot process effacement [55] . The involvement of β-dystroglycan in fibronectin-induced signaling is shown by the decrease in phosphorylation of Y892, which however enhances binding of the dystroglycan complex to the actin cytoskeleton via utrophin. The impact of this enhanced binding to actin was visualized by changes in podocyte morphology. We observed a reorganization of the actin cytoskeleton after ligation of α-dystroglycan, with monoclonal antibodies to α-dystroglycan and in a subset of podocytes with fibronectin. This was accompanied by a retraction of branches of cultured mouse podocytes. Similar observations have been made by Rüdiger et al. after exposure of mouse podocytes to protamin sulfate, which resulted in instant calcium influxes, and podocyte foot process retraction after 2 hrs [56] . Protamin sulfate inhibits the binding of α-dystroglycan to laminin and can also bind α-dystroglycan, in vivo injection results in immediate foot process effacement and endocytosis of α-dystroglycan [57] . Thus, the in vitro retraction of foot processes may reflect in vivo podocyte foot process effacement. In summary, the binding of fibronectin to apical free accessible α-dystroglycan leads to signaling events in the podocytes, as indicated by a rise in cytosolic calcium and a dephosphorylation of β-dystroglycan, enabling utrophin and the actin cytoskeleton to bind and retract the podocyte foot process. This may represent a novel mechanism of podocyte foot process effacement in proteinuric disease. METHODS Conditionally immortalized mouse podocytes (MPC-5, generously provided by Peter Mundel, Division of Nephrology, Albert Einstein College of Medicine, Bronx, NY, USA) were cultured essentially as described before [7] , [38] . MPC-5 express α-dystroglycan at their apical and basal cell membrane [7] . Passages 12–17 were used in these experiments. Indirect immunofluorescence was performed as described previously [7] . Anti-Mena antibody (Becton Dickinson, Transduction Laboratories, San Jose, CA, USA) dilution 1∶40 and goat anti-mouse IgA FITC (Southern Biotech, Birmingham, Alabama, USA) dilution 1∶50 were used to probe the filopodia. [58] . Phalloidin-TRITC (Molecular Probes, Leiden, The Netherlands) dilution 1∶150, and DAPI (Molecular Probes) dilution 3*10 −5 was used to visualize the actin cytoskeleton and DNA, respectively. Sections were investigated by a Leica DM 4000 B microscope (Leica Microsystems, Rijswijk, the Netherlands). Filopodia count on the cell border was studied as follows: The number of filopodia of 2 µm or longer was counted per 5 µm cell border. Cells were divided into 3 categories: smooth 0–0.2 filopodia/5 µm, intermediate 0.2–1 filopodia/5 µm or filopodia rich ≥1 filopodia/5 µm cell border. The frequencies in the groups were compared using χ 2 . For the investigation of binding, propagated podocytes were seeded overnight on laminin EHS (Campro Scientific; Veenendaal, The Netherlands) or collagen A (Biochrom, Berlin, Germany) coated 96 microwell plates (Costar, Corning Life Sciences, Corning, NY) using 3000 podocytes/well with the addition of 1 µM α-dystroglycan specific monoclonal antibodies, IIH6 or VIA4.1 [59] , or their respective isotype controls, MOPC 21 or TEPC 183 (Sigma, St Louis, MO). Notably, the glycoepitopes on glomerular α-dystroglycan are recognized by the antibodies IIH6 and VIA4.1 and expressed by cultured mouse podocytes [7] , [60] . Podocytes were inversely centrifuged at 200 g for 1 minute and carefully rinsed three times with phosphate buffered saline (PBS, Merck, Darmstadt, Germany). To evaluate the number of podocytes a hexosaminidase essay was performed in which the amount of the house keeping enzyme hexosaminidase is quantified by measuring the digestion of p-nitrophenyl- N- acetyl-β-D-glucosamidine [61] . In short: podocytes were incubated overnight at 37°C with 100 µl 50 mM citrate (Merck), 3.75 mM p-nitrophenyl- N- acetyl-β-D-glucosamidine (Sigma) and 0.25% Triton ×100 (Sigma). Thereafter, the reaction was blocked with a solution containing 50 mM glycine (Sigma) and 5 mM EDTA, pH 10.4, 150 µl/well. Optical density was measured at 405 nm. Two-tailed Student- t test was performed and P ≤0.05 was considered significant. MEASUREMENT OF CALCIUM FLUXES Propagated podocytes were seeded on collagen A coated glass cover slips and differentiated for 3 weeks. Subsequently, cells were loaded with 5 mM fura-2-AM (Molecular Probes) and 0.025% pluronic-F127 (Molecular Probes), added to the medium of differentiated podocytes for 20 minutes at 37°C. Subsequently, they were washed and mounted in a bath-chamber in 137 mM NaCl, 3.6 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 5,5 mM D-glucose and 10 mM Hepes-Tris buffer pH 7.4 (Sigma) at 37°C and placed under an inverted microscope (Nikon) attached to a videorate confocal microscope (Noran Instruments, Middleton, WI). Calcium-free measurements were performed in the same buffer without calcium and with 2 mM ethylene glycol tetra-acetic acid (EGTA, Sigma) added. The podocytes were excited 15 times/min at 340 and 380 nm, with an exposure time of 100 ms and single cell emission at 492 nm was monitored as a measure of the cytosolic free calcium concentration. Dynamic video imaging was carried out using the MagiCal hardware and Tardis software [62] . The effects of the addition of 1 µM biglycan (Sigma), 1 µM laminin, 1 µM IIH6, 1 µM VIA4.1, 1 µM MOPC 21, 1 µM TEPC 183 (Sigma), 1 µM heparan sulfate (Seikagaku), a mixture of 2 mg/ml chondroitin sulfate A, 1 µM B and C (Sigma), or 0.025 mg/ml rat fibronectin (250 kDa chains, Sigma) on calcium fluxes were evaluated. Only for the measurement of the effect of fibronectin on calcium fluxes, podocytes were prior to the addition of fibronectin maintained in serum-free medium for 24 hrs. Possible fibronectin binding sites on integrins were blocked with GRGDS peptide (160 µM, Sigma). WESTERN BLOT ANALYSIS Western blot analysis was performed according to the methods of Laemmli and as described recently [10] , [63] . Whole podocyte cell lysates were blotted onto nitrocellulose and blocked with 5% non-fat dry milk (Biorad, Veenendaal, the Netherlands) in PBS. Phosphatase inhibitors, 10 mM NaF (Sigma) and 1 mM sodium orthovanadate (Sigma), were added to the buffers used during the whole procedure to prevent dephosphorylation [64] . Anti-pY892 (1∶50 BD Biosciences, San Jose, USA) and NCL-BDG (1∶10 Novocastra, Newcastle upon Tyne, United Kingdom) were used as primary antibodies. Peroxidase labeled goat anti mouse IgG H+L 1∶5000, was used as secondary antibody. REFERENCES [1] Pavenstadt H Kriz W Kretzler M 2003 Cell biology of the glomerular podocyte. Physiol Rev 83: 1 253 307 [2] Ibraghimov-Beskrovnaya O Ervasti JM Leveille CJ Slaughter CA Sernett SW 1992 Primary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix. Nature 355: 6362 696 702 [3] Hoffman EP Fischbeck KH Brown RH Johnson M Medori R 1988 Characterization of dystrophin in muscle-biopsy specimens from patients with Duchenne's or Becker's muscular dystrophy. N Engl J Med 318: 21 1363 1368 [4] Durbeej M Larsson E Ibraghimov-Beskrovnaya O Roberds SL Campbell KP 1995 Non-muscle alpha-dystroglycan is involved in epithelial development. J Cell Biol 130: 1 79 91 [5] Durbeej M Henry MD Ferletta M Campbell KP Ekblom P 1998 Distribution of dystroglycan in normal adult mouse tissues. J Histochem Cytochem 46: 4 449 457 [6] Raats CJ van den Born J Bakker MA Oppers-Walgreen B Pisa BJ 2000 Expression of agrin, dystroglycan, and utrophin in normal renal tissue and in experimental glomerulopathies. Am J Pathol 156: 5 1749 1765 [7] Vogtlander NPJ Dijkman H Bakker MAH Campbell KP van der Vlag J 2005 Localization of alpha-dystroglycan on the podocyte: from top to toe. J Histochem Cytochem 53: 11 1345 1353 [8] Ervasti JM Campbell KP 1993 A role for the dystrophin-glycoprotein complex as a transmembrane linker between laminin and actin. J Cell Biol 122: 4 809 823 [9] Regele HM Fillipovic E Langer B Poczewki H Kraxberger I 2000 Glomerular expression of dystroglycans is reduced in minimal change nephrosis but not in focal segmental glomerulosclerosis. J Am Soc Nephrol 11: 3 403 412 [10] Vogtlander NPJ Tamboer WPM Bakker MAH van der Vlag J Berden JH 2006 Reactive oxygen species deglycosilate glomerular alpha-dystroglycan. Kidney Int 69: 9 1526 1534 [11] Bowe MA Mendis DB Fallon JR 2000 The small leucine-rich repeat proteoglycan biglycan binds to alpha- dystroglycan and is upregulated in dystrophic muscle. J Cell Biol 148: 4 801 810 [12] Schaefer L Grone HJ Raslik I Robenek H Ugorcakova J 2000 Small proteoglycans of normal adult human kidney: distinct expression patterns of decorin, biglycan, fibromodulin, and lumican. Kidney Int 58: 4 1557 1568 [13] Stokes MB Holler S Cui Y Hudkins KL Eitner F 2000 Expression of decorin, biglycan, and collagen type I in human renal fibrosing disease. Kidney Int 57: 2 487 498 [14] Moss J Shore I Woodrow D 1998 An ultrastructural study of the colocalization of biglycan and decorin with AA amyloid fibrils in human renal glomeruli. Amyloid 5: 1 43 48 [15] Kuroda M Sasamura H Kobayashi E Shimizu-Hirota R Nakazato Y 2004 Glomerular expression of biglycan and decorin and urinary levels of decorin in primary glomerular disease. Clin Nephrol 61: 1 7 16 [16] Schaefer L Raslik I Grone HJ Schonherr E Macakova K 2001 Small proteoglycans in human diabetic nephropathy: discrepancy between glomerular expression and protein accumulation of decorin, biglycan, lumican, and fibromodulin. Faseb J 15: 3 559 561 [17] Okuda S Languino LR Ruoslahti E Border WA 1990 Elevated expression of transforming growth factor-beta and proteoglycan production in experimental glomerulonephritis. Possible role in expansion of the mesangial extracellular matrix. J Clin Invest 86: 2 453 462 [18] Bjornson Granqvist A Ebefors K Saleem MA Mathieson PW Haraldsson B 2006 Podocyte proteoglycan synthesis is involved in the development of nephrotic syndrome. Am J Physiol Renal Physiol 291: 4 F722 730 [19] Kuroda M Sasamura H Shimizu-Hirota R Mifune M Nakaya H 2002 Glucocorticoid regulation of proteoglycan synthesis in mesangial cells. Kidney Int 62: 3 780 789 [20] Hall H Bozic D Michel K Hubbell JA 2003 N-terminal alpha-dystroglycan binds to different extracellular matrix molecules expressed in regenerating peripheral nerves in a protein-mediated manner and promotes neurite extension of PC12 cells. Mol Cell Neurosci 24: 4 1062 1073 [21] Kubosawa H Kondo Y 1998 Alterations in the distribution of plasma fibronectin and the ultrastructure of podocytes in the peripheral glomerular loops in nephrotic rats. Virchows Arch 433: 5 449 455 [22] Meier M Park JK Overheu D Kirsch T Lindschau C 2007 Deletion of protein kinase C-beta isoform in vivo reduces renal hypertrophy but not albuminuria in the streptozotocin-induced diabetic mouse model. Diabetes 56: 2 346 354 [23] Lee S Kim W Moon SO Sung MJ Kim DH 2007 Renoprotective effect of COMP-angiopoietin-1 in db/db mice with type 2 diabetes. Nephrol Dial Transplant 22: 2 396 408 [24] Cohen MP Lautenslager GT Hud E Shea E Wang A 2007 Inhibiting albumin glycation attenuates dysregulation of VEGFR-1 and collagen IV subchain production and the development of renal insufficiency. Am J Physiol Renal Physiol 292: 2 F789 F795 [25] Soylemezoglu O Wild G Dalley AJ MacNeil S Milford-Ward A 1997 Urinary and serum type III collagen: markers of renal fibrosis. Nephrol Dial Transplant 12: 9 1883 1889 [26] Senatorski G Paczek L Sulowicz W Gradowska L Bartlomiejczyk I 1998 Urine activity of cathepsin B, collagenase and urine excretion of TGF-beta 1 and fibronectin in membranous glomerulonephritis. Res Exp Med (Berl) 198: 4 199 206 [27] Kilis-Pstrusinska K Wikiera-Magott I Zwolinska D Kopec W Rzeszutko M 2002 Analysis of collagen IV and fibronectin in blood and urine in evaluation of nephrotic fibrosis in children with chronic glomerulonephritis. Med Sci Monit 8: 10 CR713 CR719 [28] Jung D Yang B Meyer J Chamberlain JS Campbell KP 1995 Identification and characterization of the dystrophin anchoring site on beta-dystroglycan. J Biol Chem 270: 45 27305 27310 [29] Rentschler S Linn H Deininger K Bedford MT Espanel X 1999 The WW domain of dystrophin requires EF-hands region to interact with beta-dystroglycan. Biol Chem 380: 4 431 442 [30] James M Nuttall A Ilsley JL Ottersbach K Tinsley JM 2000 Adhesion-dependent tyrosine phosphorylation of (beta)-dystroglycan regulates its interaction with utrophin. J Cell Sci 113(Pt 10) 1717 1726 [31] Macias MJ Wiesner S Sudol M 2002 WW and SH3 domains, two different scaffolds to recognize proline-rich ligands. FEBS Lett 513: 1 30 37 [32] Huang X Poy F Zhang R Joachimiak A Sudol M 2000 Structure of a WW domain containing fragment of dystrophin in complex with beta-dystroglycan. Nat Struct Biol 7: 8 634 638 [33] Ilsley JL Sudol M Winder SJ 2001 The interaction of dystrophin with beta-dystroglycan is regulated by tyrosine phosphorylation. Cell Signal 13: 9 625 632 [34] Sotgia F Lee H Bedford MT Petrucci T Sudol M 2001 Tyrosine phosphorylation of beta-dystroglycan at its WW domain binding motif, PPxY, recruits SH2 domain containing proteins. Biochemistry 40: 48 14585 14592 [35] Yang B Jung D Motto D Meyer J Koretzky G 1995 SH3 domain-mediated interaction of dystroglycan and Grb2. J Biol Chem 270: 20 11711 11714 [36] Matsumura K Chiba A Yamada H Fukuta-Ohi H Fujita S 1997 A role of dystroglycan in schwannoma cell adhesion to laminin. J Biol Chem 272: 21 13904 13910 [37] Sotgia F Bonuccelli G Bedford M Brancaccio A Mayer U 2003 Localization of phospho-beta-dystroglycan (pY892) to an intracellular vesicular compartment in cultured cells and skeletal muscle fibers in vivo. Biochemistry 42: 23 7110 7123 [38] Mundel P Reiser J Zuniga Mejia Borja A Pavenstadt H Davidson GR 1997 Rearrangements of the cytoskeleton and cell contacts induce process formation during differentiation of conditionally immortalized mouse podocyte cell lines. Exp Cell Res 236: 1 248 258 [39] Berridge MJ Bootman MD Roderick HL 2003 Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 4: 7 517 529 [40] Gesemann M Cavalli V Denzer AJ Brancaccio A Schumacher B 1996 Alternative splicing of agrin alters its binding to heparin, dystroglycan, and the putative agrin receptor. Neuron 16: 4 755 767 [41] Tisi D Talts JF Timpl R Hohenester E 2000 Structure of the C-terminal laminin G-like domain pair of the laminin alpha2 chain harbouring binding sites for alpha-dystroglycan and heparin. Embo J 19: 7 1432 1440 [42] Timpl R Tisi D Talts JF Andac Z Sasaki T 2000 Structure and function of laminin LG modules. Matrix Biol 19: 4 309 317 [43] Blattner SM Kretzler M 2005 Integrin-linked kinase in renal disease: connecting cell-matrix interaction to the cytoskeleton. Curr Opin Nephrol Hypertens 14: 4 404 410 [44] Kretzler M 2002 Regulation of adhesive interaction between podocytes and glomerular basement membrane. Microsc Res Tech 57: 4 247 253 [45] Sotgia F Lee JK Das K Bedford M Petrucci TC 2000 Caveolin-3 directly interacts with the C-terminal tail of beta -dystroglycan. Identification of a central WW-like domain within caveolin family members. J Biol Chem 275: 48 38048 38058 [46] Cavaldesi M Macchia G Barca S Defilippi P Tarone G 1999 Association of the dystroglycan complex isolated from bovine brain synaptosomes with proteins involved in signal transduction. J Neurochem 72: 4 1648 1655 [47] Sadasivam G Willmann R Lin S Erb-Vogtli S Kong XC 2005 Src-family kinases stabilize the neuromuscular synapse in vivo via protein interactions, phosphorylation, and cytoskeletal linkage of acetylcholine receptors. J Neurosci 25: 45 10479 10493 [48] Spence HJ Chen YJ Batchelor CL Higginson JR Suila H 2004 Ezrin-dependent regulation of the actin cytoskeleton by beta-dystroglycan. Hum Mol Genet 13: 15 1657 1668 [49] Ferletta M Kikkawa Y Yu H Talts JF Durbeej M 2003 Opposing roles of integrin alpha6-beta1 and dystroglycan in laminin-mediated extracellular signal-regulated kinase activation. Mol Biol Cell 14: 5 2088 20103 [50] Spence HJ Dhillon AS James M Winder SJ 2004 Dystroglycan, a scaffold for the ERK-MAP kinase cascade. EMBO Rep 5: 5 484 489 [51] Kucherenko MM Pantoja M Yatsenko AS Shcherbata HR Fischer KA 2008 Genetic modifier screens reveal new components that interact with the Drosophila dystroglycan-dystrophin complex. PLoS ONE 3: 6 e2418 [52] Hunt JL Pollak MR Denker BM 2005 Cultured podocytes establish a size-selective barrier regulated by specific signaling pathways and demonstrate synchronized barrier assembly in a calcium switch model of junction formation. J Am Soc Nephrol 16: 6 1593 1602 [53] Reiser J Polu KR Moller CC Kenlan P Altintas MM 2005 TRPC6 is a glomerular slit diaphragm-associated channel required for normal renal function. Nat Genet 37: 7 739 744 [54] Winn MP Conlon PJ Lynn KL Farrington MK Creazzo T 2005 A mutation in the TRPC6 cation channel causes familial focal segmental glomerulosclerosis. Science 308: 5729 1801 1804 [55] Takeda T McQuistan T Orlando RA Farquhar MG 2001 Loss of glomerular foot processes is associated with uncoupling of podocalyxin from the actin cytoskeleton. J Clin Invest 108: 2 289 301 [56] Rudiger F Greger R Nitschke R Henger A Mundel P 1999 Polycations induce calcium signaling in glomerular podocytes. Kidney Int 56: 5 1700 1709 [57] Kojima K Davidovits A Poczewski H Langer B Uchida S 2004 Podocyte flattening and disorder of glomerular basement membrane are associated with splitting of dystroglycan-matrix interaction. J Am Soc Nephrol 15: 8 2079 2089 [58] Yanagida-Asanuma E Asanuma K Kim K Donnelly M Young Choi H 2007 Synaptopodin protects against proteinuria by disrupting Cdc42:IRSp53:Mena signaling complexes in kidney podocytes. Am J Pathol 171: 2 415 427 [59] Ohlendieck K Ervasti JM Snook JB Campbell KP 1991 Dystrophin-glycoprotein complex is highly enriched in isolated skeletal muscle sarcolemma. J Cell Biol 112: 1 135 148 [60] Michele DE Barresi R Kanagawa M Saito F Cohn RD 2002 Post-translational disruption of dystroglycan-ligand interactions in congenital muscular dystrophies. Nature 418: 6896 417 422 [61] Landegren U 1984 Measurement of cell numbers by means of the endogenous enzyme hexosaminidase. Applications to detection of lymphokines and cell surface antigens. J Immunol Methods 67: 2 379 388 [62] Visch HJ Rutter GA Koopman WJ Koenderink JB Verkaart S 2004 Inhibition of mitochondrial Na+-Ca2+ exchange restores agonist-induced ATP production and Ca2+ handling in human complex I deficiency. J Biol Chem 279: 39 40328 40336 [63] Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 259 680 685 [64] Sharma SK Carew TJ 2002 Inclusion of phosphatase inhibitors during Western blotting enhances signal detection with phospho-specific antibodies. Anal Biochem 307: 1 187 189 [65] Asanuma K Yanagida-Asanuma E Faul C Tomino Y Kim K 2006 Synaptopodin orchestrates actin organization and cell motility via regulation of RhoA signalling. Nat Cell Biol 8: 5 485 491