Src-Family Kinase Fyn Phosphorylates the Cytoplasmic Domain of Nephrin and Modulates Its Interaction with Podocin

Transcription

1 J Am Soc Nephrol 15: , 2004 Src-Family Kinase Fyn Phosphorylates the Cytoplasmic Domain of Nephrin and Modulates Its Interaction with Podocin HONGPING LI,* SERGE LEMAY,* LAMINE AOUDJIT,* HIROSHI KAWACHI, and TOMOKO TAKANO* *McGill University Health Centre, Montreal, Québec, Canada; and Niigata University School of Medicine, Niigata, Japan Abstract. Visceral glomerular epithelial cells (GEC) are critical for normal permselectivity of the kidney. Nephrin is a molecule that is expressed specifically in GEC in a structure called the slit diaphragm and is required for normal morphology and permselectivity of GEC. However, the mechanisms of action of nephrin are not understood precisely. The intracellular domain of nephrin has six conserved tyrosine residues. It was hypothesized that these tyrosine residues are phosphorylated by Srcfamily kinases and that this phosphorylation modulates the function of nephrin. A transient transfection system was used to study the role of tyrosine phosphorylation of the cytoplasmic domain of nephrin in its function. When nephrin was cotransfected with Src-family kinases Fyn or Src in Cos-1 cells, nephrin was strongly tyrosine phosphorylated by Fyn and less so by Src. The results with tyrosine-to-phenylalanine mutations suggested that multiple tyrosine residues contribute to phosphorylation mediated by Src-family kinases. The intracellular domain of nephrin is known to interact with another slit diaphragm protein, podocin. When nephrin and podocin were transfected with Fyn, the interaction between nephrin and podocin was augmented significantly. Podocin was not tyrosine phosphorylated by Fyn; thus, the increased interaction is likely to be secondary to tyrosine phosphorylation of nephrin. Fyn also significantly augmented the activation of the AP-1 promoter induced by nephrin and podocin. In summary, Fyn phosphorylates the cytoplasmic domain of nephrin on tyrosine, leading to enhanced association with podocin and downstream signaling of nephrin. Received December 4, Accepted September 15, Correspondence to Dr. Tomoko Takano, 3775 University Street, Room 236, Montreal, Quebec H3A 2B4. Phone: ; Fax: ; tomoko.takano@mcgill.ca / Journal of the American Society of Nephrology Copyright 2004 by the American Society of Nephrology DOI: /01.ASN Visceral glomerular epithelial cells (GEC) or podocytes are critical for structural integrity as well as permselectivity of the glomerulus. In 1998, Kestila et al. (1) identified a new molecule, nephrin, which localizes in the slit diaphragm of podocytes. Mutations of nephrin cause severe congenital proteinuria and renal failure (Finnish-type nephrotic syndrome). This breakthrough discovery was followed by identification of a number of other podocyte-associated molecules essential for glomerular permselectivity, such as CD2AP, podocin, and -actinin-4 (2). These molecules are believed to maintain the normal morphology and permselectivity of podocyte by interacting with the actin cytoskeleton, but their precise mechanisms of action are largely unknown. After the initial cloning of nephrin in 1998, substantial amounts of information were obtained about the expression level and distribution pattern of nephrin in various kidney diseases. However, information about the signal transduction of nephrin is only beginning to accumulate. Nephrin belongs to the Ig superfamily. It is a transmembrane protein, and its extracellular domain has eight Ig-like domains and a fibronectin-like domain. It is believed that nephrin molecules from adjacent foot processes bind to each other in a homophilic manner, which serves as a backbone for the slit diaphragm (3). When the homophilic binding of nephrin was disrupted, intracellular actin organization was disrupted (4). Thus, in addition to its function as an adhesion molecule, nephrin is likely to transmit extracellular signals into the cells (4). Podocin, which was identified in 2000, is a membrane-associated protein that belongs to the stomatin protein family. Mutations of podocin cause autosomal recessive steroid-resistant nephrotic syndrome in childhood (5). Podocin interacts with nephrin via its C- terminus (amino acids 125 to 385) (6,7). Nephrin-podocin interaction is believed to be important for the recruitment of nephrin to lipid raft (8). Also, podocin augments the activation of the AP-1 promoter by nephrin (6). Thus, it is likely that the nephrin podocin interaction has an important role in glomerular permselectivity (2). Nephrin has a short cytoplasmic domain (~150 amino acids), which contains six tyrosine residues conserved in human, mouse, and rat sequences (Figure 1A). Recently, several groups reported that nephrin is tyrosine phosphorylated in the cytoplasmic domain (9 11). The injection of a monoclonal antibody that causes morphologic changes of podocyte induced marked tyrosine phosphorylation of nephrin (10). Clustering of nephrin by antibodies caused strong tyrosine phosphorylation of nephrin in the cytoplasmic domain (9). Also, it was shown that the Src-family tyrosine kinase Fyn bound and phosphorylated nephrin (11). However, the significance of nephrin tyrosine phosphorylation has yet to be elucidated. It was re-

2 J Am Soc Nephrol 15: , 2004 Fyn Phosphorylates Nephrin 3007 Materials and Methods Tissue culture media, isopropylthio- -galactoside, and lipofectamine 2000 reagent were purchased from Invitrogen-Life Technologies (Burlington, ON, Canada). Sodium orthovanadate and other standard chemicals were from Sigma Chemical Co. (St. Louis, MO). Protease inhibitor cocktail, reagents for PCR, and calf intestinal alkaline phosphatase were from Roche Diagnostics (Laval, QC, Canada). Antiphosphotyrosine antibody Py20 was from BD Biosciences (San Jose, CA). Rabbit anti-cd2ap antibody, rabbit anti-fyn antibody, mouse anti-myc antibody, and protein A-Sepharose were from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-Src [py 418 ] phospho-specific antibody was from Biosource International (Camarillo, CA). Src phosphorylated at Y 418 is known to be active. Dual-Luciferase Reporter Assay System was from Promega (Madison, WI). cdna encoding Src, T lymphocyte Fyn (FynT), kinase-negative FynT, and C-terminal Src kinase (Csk; all in the mammalian expression vector pme18s) were gifts from Dr. Junichi Abe (University of Rochester, Rochester, NY) (17 19). cdna Tac/zeta/zeta (20) was a gift from Dr. Sylvain Latour (Hôpital Necker Enfants-Malades, Paris, France). Isolation of full-length rat nephrin was reported earlier (21), and the coding region was subcloned into the EcoRV/XbaI sites of pcdna3.1/hisa (Invitrogen-Life Technologies). Plasmids for AP-1 luciferase, nephrin-f, and F-podocin were gifts from Dr. Thomas Benzing (Freiburg, Germany) (6). Figure 1. Alignment of the cytoplasmic domains of human, rat, and mouse nephrin (A) and Tac/nephrin construct (B). (A) Six conserved tyrosine residues are indicated by arrows. Numbers are according to the rat sequence (accession no. NM022628). (B) The extracellular domain of the human IL-2 receptor (Tac) was connected to the transmembrane/cytoplasmic domain of rat nephrin to form a chimera Tac/nephrin. Full-length nephrin is shown for comparison. cently reported that the cytoplasmic domain of nephrin binds to the regulatory p85 subunit of phosphoinositide 3-kinase (PI3K) and together with CD2AP and podocin activates the Akt signaling pathway (12). It was suggested that this binding is dependent on tyrosine phosphorylation of nephrin (12). Src-family kinases are a large family of nonreceptor tyrosine kinases, which are known to localize in the lipid rafts and are involved in a variety of signal transduction pathways (13,14). For example, c-src phosphorylates intracellular molecules that interact with and transmit signals from adhesion molecules, such as integrins and cadherins, leading to rearrangement in the cytoskeleton (15). Yu et al. (16) reported that in Fyn-deficient mice (fyn / ), effacement of podocyte foot processes was observed, accompanied by significant proteinuria. Similar findings were reported in fyn / yes / double knockout mice (11). Thus, it is tempting to speculate that tyrosine phosphorylation by Src-family kinases may regulate protein protein interactions and signal transduction of nephrin. The current study demonstrates that the cytoplasmic domain of nephrin is a substrate for the Src-family kinase Fyn and that tyrosine phosphorylation of nephrin modulates its interaction with its molecular partner, podocin. Cells and Transfection Cos-1 cells and HEK293T cells were grown in DMEM supplemented with 10% FBS. Cells were transfected with the indicated amounts of plasmid using lipofectamine 2000 reagent (Invitrogen- Life Technologies) following the manufacturer s instructions. Induction of Passive Heymann Nephritis Passive Heymann nephritis (PHN) was induced in male Sprague- Dawley rats (150 to 175 g body wt; Charles River, St. Constant, QC, Canada) by intravenous injection (400 l/rat) of sheep anti-fx1a antiserum as described previously (22). Significant proteinuria was observed 14 d after injection (~160 mg/d; normal rats excrete 10 mg protein/d). Plasmid Preparations To construct Tac/nephrin (Figure 1B), we amplified cdna corresponding to the transmembrane and cytoplasmic domain of rat nephrin (21) by reverse transcription PCR (RT-PCR) from RNA prepared from rat glomeruli. After the sequence was confirmed, this fragment was subcloned into Tac/zeta/zeta to replace the zeta/zeta portion, to form a chimera of the extracellular domain of the human IL-2 receptor (Tac) and the transmembrane/cytoplasmic domains of rat nephrin. Finally, Tac/nephrin was subcloned into the mammalian expression vector pcdna3.1( )/hygro (Invitrogen-Life Technologies). cdna corresponding to the coding region of mouse podocin (5) was amplified by RT-PCR from mouse kidney RNA. After the sequence was confirmed, this fragment was cloned into pcdna3.1( )/Myc-HisB (Invitrogen-Life Technologies) to construct Myc (HIS)-tagged podocin (Myc/podocin). cdna corresponding to the coding region of mouse CD2AP (23) was amplified by RT-PCR from mouse liver RNA. After the sequence was confirmed, the fragment was subcloned into pcdna3.1( )/hygro (Invitrogen-Life Technologies). Tyrosine to phenylalanine mutations of nephrin were carried out by PCR-based mutagenesis. To construct mutants in full-length rat nephrin, we transferred the fragments that contained mutations from Tac/nephrin to full-length rat nephrin using BglII and XbaI sites. As part of the subcloning process, His-tag was removed from the original full-length rat nephrin and the expression vector was changed to pcdna3.1 (Invitrogen-Life Technologies). All sequences were confirmed by automated sequencing at Sheldon Biotechnology Centre (McGill University, Montreal, QC, Canada).

3 3008 Journal of the American Society of Nephrology J Am Soc Nephrol 15: , 2004 Preparation of Antinephrin Antiserum cdna corresponding to the cytoplasmic domain of rat nephrin was amplified by RT-PCR from rat glomerular RNA and was subcloned into pgex-5x-2 (Pharmacia) to construct a glutathione S transferase (GST)-fusion protein. The construct was expressed in bacteria XL- 10-Gold (Stratagene), and GST-fusion protein was induced by isopropylthio- -galactoside. The GST-fusion protein was separated by SDS-PAGE, purified, and used to immunize rabbits. Specificity of the antiserum was verified using cell lysates from Tac/nephrin-transfected Cos-1 cells and rat glomerular lysates. Immunoprecipitation and Immunoblotting Cells or glomeruli were lysed in ice-cold IP buffer (1% Triton X-100, 125 mm NaCl, 10 mm Tris [ph 7.4], 1 mm EDTA, 1 mm EGTA, 2 mm Na 3 VO 4, 10 mm sodium pyrophosphate, and 25 mm NaF) that contained a protease inhibitor cocktail (Roche Diagnostics). After insoluble components were removed by centrifugation (14,000 rpm, 5 min, 4 C), protein concentrations of supernatants were quantified using a commercial reagent (Bio-Rad). Immunoprecipitation was performed with 5 l of antiserum and 20 l of protein A- Sepharose per sample. Equal amounts of protein (25 to 50 g of total lysates) or immunoprecipitates were separated by 7.5% SDS-PAGE under reducing conditions. Proteins were electrophoretically transferred to nitrocellulose membrane, blocked with 5% dry milk (or 5% BSA for Py20), and incubated with first antibodies for 16 h at 4 C. After three washes, membranes were incubated with secondary antibodies conjugated with horseradish peroxidase, and horseradish peroxidase activity was detected by enhanced chemiluminescence (Amersham Pharmacia Biotech, Baie d Urfé, QC, Canada). Protein content was quantified using scanning densitometry (NIH Image software). Luciferase Assay AP-1 luciferase assay was performed as reported by Huber et al. (6) with minor modification. HEK293T cells were plated in 24-well plates and transfected with plasmids using lipofectamine 2000 reagent in Opti-MEM. Dual-Luciferase Reporter System was used to normalize transfection efficiencies. One day later, cells were harvested and cell lysates were subjected to luciferase assay following the manufacturer s instructions. Statistical Analyses Data are presented as mean SEM. The t statistic was used to determine significant differences between two groups. One-way ANOVA was used to determine significant differences among groups. When significant differences were found, individual comparisons were made between groups using the t statistic. Results Fyn Tyrosine Phosphorylates the Cytoplasmic Domain of Nephrin To test the hypothesis that the cytoplasmic domain of nephrin is a substrate for Src-family kinases, we first used a chimeric construct in which the transmembrane and cytoplasmic domain of rat nephrin is connected to the extracellular domain of the human IL-2 receptor Tac (Tac/nephrin; Figure 1B). Tac has been used in the past to study the function of the cytoplasmic domain of transmembrane molecules and the effect of cross-linking of the extracellular domain by the monoclonal anti-tac antibody 7G7 (20,24). When Tac/nephrin was transiently transfected in Cos-1 cells and immunoprecipitated by antinephrin antiserum, no tyrosine phosphorylation was observed (Figure 2A). However, when Tac/nephrin was co-transfected with the Src-family kinase Fyn, Tac/nephrin was strongly tyrosine phosphorylated (Figure 2A). This tyrosinephosphorylated band at ~90 kd was not observed in the absence of nephrin (Figure 2A). Tac/nephrin was expressed as multiple bands ranging from 70 to 90 kd presumably because of differential glycosylation in the extracellular Tac domain and/or by ubiquitination/degradation (20); however, tyrosinephosphorylated nephrin appeared as a single band of ~90 kd (Figure 2A). Similar tyrosine phosphorylation was observed when the cytoplasmic domain of nephrin (not shown) or the full-length rat nephrin (Figure 2E) was used instead of Tac/ nephrin. Another Src-family kinase Src also tyrosine phosphorylated Tac/nephrin; however, phosphorylation was significantly weaker as compared with Fyn (Figure 2B). Moreover, there was a second tyrosine-phosphorylated band (~80 kd) observed only with Src (Figure 2B). A similar difference in electrophoretic mobility between Fyn and Src was observed with full-length nephrin (Figure 2F). Although the precise nature of the lower band is not clear, it seems that the lower band represents hypophosphorylated nephrin. These results suggest that the cytoplasmic domain of nephrin is a substrate of Fyn and Src. Fyn phosphorylates nephrin more strongly, as compared with Src, and there may be a qualitative and/or quantitative difference between tyrosine phosphorylation by Fyn and Src. We next studied whether tyrosine phosphorylation of nephrin is dependent on the kinase activity of Src-family kinases. First, a kinase-negative mutant of Fyn (K299M) (19) failed to phosphorylate Tac/nephrin, indicating that tyrosine phosphorylation of nephrin is dependent on the kinase activity of Fyn (Figure 2C). It is known that phosphorylation of carboxyl terminal tyrosine (Tyr 527 in Src) by C-terminal Src kinase (Csk) inhibits activity of Src-family kinases. Thus, we also tested the impact of Csk on Fyn/Src-medicated nephrin phosphorylation. When the same amounts of Fyn or Src (0.2 g) were transfected into Cos-1 cells with Tac/nephrin (0.2 g), Tac/nephrin was tyrosine phosphorylated more strongly by Fyn, as compared with Src (Figure 2D, lanes 3 and 6), consistent with the results in Figure 2B. When Csk was co-transfected, tyrosine phosphorylation of Tac/nephrin was markedly attenuated (Figure 2D, lanes 8 and 9). In fact, tyrosine phosphorylation by Src was not apparent in the presence of Csk (Figure 2D, lane 9). These results support that tyrosine phosphorylation of nephrin is dependent on the kinase activity of the Src-family kinases. Although these results do not prove that nephrin is a substrate of Src-family kinases in vivo, Verma et al. (11) reported that nephrin phosphorylation is markedly reduced in fyn / mice. Together, it is reasonable to conclude that nephrin is tyrosine phosphorylated, at least in part, by Src-family kinases. Identification of Tyrosine Residues Phosphorylated by Fyn We next studied which among the six conserved tyrosine residues within the nephrin cytoplasmic domain are phosphorylated by Src-family kinases. Because Fyn phosphorylated

4 J Am Soc Nephrol 15: , 2004 Fyn Phosphorylates Nephrin 3009 Figure 2. Nephrin is tyrosine phosphorylated in Cos-1 cells. Cos-1 cells were transfected in 10-cm plates (A, B, C, E, and F) or in 35-mm plates (D) with the indicated amounts of plasmids using lipofectamine 2000 reagent. One to 2 d later, cell lysates were immunoprecipitated with rabbit antinephrin antiserum (A, B, C, E, and F) and immunoprecipitates (A, B, C, E, and F) or total cell lysates (25 g; D) were blotted for phosphotyrosine (Py20) and nephrin. Note that Tac/nephrin is detected as multiple bands ranging from 70 to 90 kd. Tyrosine-phosphorylated Tac/nephrin migrated at ~90 kd as a single band with Fyn (A through D) and at ~90 and ~80 kd as double bands with Src (B and D). Full-length nephrin (FL-nephrin) migrated at ~180 kd as a doublet (E and F). When FL-nephrin was phosphorylated by Fyn, phosphorylated FL-nephrin co-migrated predominantly with the top band of the doublet, whereas when it was phosphorylated by Src, it co-migrated equally with the two bands of the doublet (F). nephrin more strongly than Src, subsequent studies focused on Fyn. We first evaluated the possibility of tyrosine phosphorylation of the six conserved residues using NetPhos 2.0 prediction program provided by the Center for Biologic Sequence Analysis. ( (25). This program predicts tyrosine residues that are likely to be phosphorylated in eukaryotic cells. On the basis of the results shown in Figure 3A, we chose to focus on Y1152, Y1171, Y1204, and Y1238, which were predicted to be the most likely sites of phosphorylation. When these tyrosine residues were mutated in Tac/nephrin individually to phenylalanine (Figure 3B), tyrosine phosphorylation decreased as follows: wild type, 100%; Y1152F, 72 10%; Y1171F, 68 1%; Y1204F, 53 8%; Y1228F, 89 8%; wild type without Fyn, 0 0% (Figure 3C). When these tyrosine residues were mutated in combinations, tyrosine phosphorylation decreased more markedly (compared with wild type [100%]; double [Y1204/1228F], 43 11%; triple [Y1152/1204/1228F], 27 13%; and quadruple [Y1152/1171/1204/ 1228F], 8 5%). Analogous experiments were performed using full-length rat nephrin with similar results (Figure 3D). Two additional tyrosine residues, which had low probability of phosphorylation (Y1127 and Y1194), were also tested. As compared with wild type (100%), Fyn-mediated tyrosine phosphorylation of Y1127F and Y1194F were 100 6% (n 6; NS) and 79 6% (n 5; P 0.05), respectively. Phosphorylation of Y1127F and Y1228F were not significantly different from wild type, suggesting that these two residues do not contribute to phosphorylation by Fyn. All other mutants were phosphorylated significantly less than wild type, indicating that multiple tyrosine residues of nephrin contribute to phosphorylation by Fyn. These residues may be phosphorylated directly by Fyn or may facilitate the phosphorylation of the other residues. Tyrosine Phosphorylation of the Cytoplasmic Domain of Nephrin Facilitates Its Interaction with Podocin The intracellular domain of nephrin is known to interact with another slit diaphragm protein, podocin. We next studied

5 3010 Journal of the American Society of Nephrology J Am Soc Nephrol 15: , 2004 whether this interaction would be affected by tyrosine phosphorylation of nephrin. When Tac/nephrin and Myc/podocin were co-transfected in Cos-1 cells and nephrin was immunoprecipitated, podocin was co-immunoprecipitated with nephrin weakly/inconsistently (Figure 4A). When Fyn was transfected with Tac/nephrin and Myc/podocin, co-immunoprecipitation of nephrin and podocin increased significantly (Figure 4A). Similar association of nephrin and podocin was also observed when the full-length rat nephrin was used instead of Tac/ nephrin (Figure 4D). Podocin was not tyrosine phosphorylated by Fyn (Figure 4B); thus, the increased co-immunoprecipitation is likely to be secondary to tyrosine phosphorylation of nephrin. Nephrin is also known to interact with the adapter molecule CD2AP (15,16); however, we could not demonstrate co-immunoprecipitation of nephrin and CD2AP in this experimental system (data not shown). Moreover, the addition of CD2AP had no consistent impact on nephrin podocin interaction (Figure 4A, left). We next studied the dose-response effect of Fyn on nephrin podocin association (Figure 4C). When 0 to 1 g of Fyn was transfected with Tac/nephrin and Myc/ podocin, tyrosine phosphorylation of nephrin increased in parallel with the amount of Fyn. Nephrin podocin interaction was most evident with 1 g of Fyn at both 24 and 48 h after transfection. Smaller amounts of Fyn also mildly increased the interaction when tested at 48 h after transfection (Figure 4C). Thus, within the range tested, Fyn facilitates nephrin podocin interaction in a dose-dependent manner. Kinase negative mutant of Fyn did not augment nephrin podocin interaction, suggesting that the effect of Fyn is dependent on its kinase activity (Figure 4, E and F) Tyrosine Residues Are Important for Nephrin Podocin Interaction We next tested nephrin podocin co-immunoprecipitation using various Tac/nephrin mutants in the presence of Fyn. When normalized to the wild-type Tac/nephrin (100%), each mutant co-immunoprecipitated podocin as follows: Y1152F, 84 14%; Y1171F, 88 15%; Y1204F, 69 10%; Y1228F, 97 20%; double (Y1204/1228F), 67 11%; triple (Y1152/ 1204/1228F), 69 2%; quadruple (Y1152/1171/1204/1238F), 54 6%; wild-type without Fyn, 42 3% (Figure 5A). Y1204F and all of the multiple mutants showed significantly Figure 3. Identification of the tyrosine residues that are phosphorylated by Fyn. (A) Probability of tyrosine phosphorylation of the six conserved tyrosine residues were studied using the NetPhos 2.0 prediction program provided by the Center for Biologic Sequence Analysis ( NetPhos). This program predicts tyrosine residues that are likely to be phosphorylated in eukaryotic cells with sensitivity in the range of 69 to 96% (25). Numbers are in the range of 0 to 1, 1 being the highest possibility for phosphorylation. (B) Four conserved tyrosine residues in the cytoplasmic domain of nephrin are indicated (Y1152, Y1171, Y1204, and Y1228). In Tac/nephrin, each residue was mutated to phenylalanine individually (Y1152F, Y1171F, Y1204F, and Y1228F) or in combinations (double [Y1204/ 1228F], triple [Y1152/1204/1228F], and quadruple [Y1152/1171/ 1204/1228F]). (C) One microgram each of wild-type (WT) or mutant Tac/nephrin was transfected into Cos-1 cells with Fyn (1 g). One to 2 d later, cell lysates were immunoprecipitated with rabbit antinephrin antiserum and immunoprecipitates were blotted for phosphotyrosine (Py20). In control, wild-type Tac/nephrin was transfected without Fyn. (Top) Immunolot. (Bottom) densitometry. Values are normalized for expression of Tac/nephrin; n 4 to 7. For expression of Tac/ nephrin, all bands between 70 and 90 kd were quantified. All mutants except Y1228F are significantly different from wild type with Fyn (*P 0.05 versus wild type with Fyn). (D) The same experiments as in C were repeated with FL-nephrin and its mutants with similar results. *P 0.05 versus wild type with Fyn.

6 J Am Soc Nephrol 15: , 2004 Fyn Phosphorylates Nephrin 3011 Figure 4. Fyn augments nephrin podocin interaction. (A) Cos-1 cells were transfected with the indicated amounts of plasmids in 10-cm plates. One to 2 d later, cell lysates were immunoprecipitated with rabbit antinephrin antiserum. Immunoprecipitates were blotted for myc to study co-immunoprecipitation of nephrin with podocin. Total cell lysates (25 g) were analyzed for expression of Tac/nephrin and Myc/podocin. Note that different amounts of Myc/podocin were used in the left and right panels. In general, when 2 g of Myc/podocin was used, there was more baseline (without Fyn) co-immunoprecipitation of nephrin and podocin and co-immunoprecipitation was observed in a more consistent manner, as compared with when 1 g of Myc/podocin was used. (B) Cos-1 cells were transfected with Myc/podocin with or without Fyn. One day later, cell lysates were immunoprecipitated with mouse anti-myc antibody and immunoprecipitates were blotted for phosphotyrosine (Py20). Expression of Myc/podocin was confirmed in total cell lysates. Myc/podocin was not tyrosine phosphorylated by Fyn. (C) Cos-1 cells were transfected with different amounts of Fyn, and nephrin podocin interaction was studied 24 or 48 h later. (D) Full-length rat nephrin was transfected with Myc/podocin and Fyn and analyzed as in A. Effect of kinase negative Fyn (kn) was tested with Tac/nephrin (E) and FL-nephrin (F). less podocin co-immunoprecipitation, as compared with wild type. Similar results were obtained for Y1204F and quadruple mutant in full-length nephrin (compared with wild type with Fyn [100%]; Y1204F, 75 10%; quadruple, 41 4%; wild type without Fyn, 34 7%; Figure 5B). Of note, the wild-type nephrin in the absence of Fyn or the quadruple mutant in the presence of Fyn still demonstrated significant interaction with podocin under these experimental conditions (Figure 5), whereas tyrosine phosphorylation of these two nephrins was minimal (Figure 3). These results suggest that tyrosine phosphorylation of nephrin facilitates nephrin podocin interaction but is not a prerequisite for this interaction. Fyn Augments AP-1 Promoter Activation by Nephrin and Podocin It was reported previously that nephrin activates the AP-1 promoter, which is significantly augmented by podocin (6). Although the physiologic significance of this pathway is yet to be established, we used this system as a readout to address the functional impact of Fyn on nephrin/podocin-mediated

7 3012 Journal of the American Society of Nephrology J Am Soc Nephrol 15: , 2004 signaling (Figure 6). The AP-1 promoter was activated by 2.9-fold by nephrin and podocin. Addition of Fyn significantly augmented this activation to 4.4-fold. Fyn alone mildly activated the AP-1 promoter; however, when Fyn was added to nephrin alone or podocin alone, the increase of the AP-1 promoter activity was small. These results are consistent with the hypothesis that Fyn augments nephrin podocin interaction, thereby augmenting nephrin-mediated signaling. Nephrin Is Tyrosine Phosphorylated In Vivo To begin to address the role of tyrosine phosphorylation of nephrin by Src-family kinases in vivo, we first studied whether nephrin is tyrosine phosphorylated in normal glomeruli. We isolated glomeruli from normal rats and lysed in buffer that contained phosphatase inhibitors (see the Materials and Methods section). Glomerular lysates were immunoprecipitated for nephrin and blotted for phosphotyrosine and nephrin. Similar to the results in cultured cells, we identified tyrosine phosphorylation of nephrin (Figure 7A). Antinephrin immunoblot on glomerular lysates revealed two bands of ~180 and ~170 kd, consistent with previous reports (26,27). This doublet has been attributed to differential glycosylation. Tyrosine-phosphorylated nephrin always migrated at ~180 kd. When glomerular lysates were treated with alkaline phosphatase, the ~180 kd phosphorylated nephrin band disappeared (Figure 7C). However, when the same samples were blotted for nephrin, there was no visible decrease in the intensity of the ~180 kd band (Figure 7C), suggesting either that dephosphorylation does not change electrophoretic mobility or that only a small fraction of 180-kD nephrin is tyrosine phosphorylated. We also confirmed that rat glomeruli express Fyn and Src (Figure 7B). The phospho-specific antibody directed at the positive regulatory tyrosine of Src (Y418) recognized a phosphorylated band of ~60 kd even in normal rat glomeruli (Figure 7B). Because the region surrounding Y418 is highly conserved in all of the related Src-family kinases, the phospho-specific Src antibody is likely to cross-react with the other members of the family. Figure 5. Tyrosine mutations of the cytoplasmic domain of nephrin decrease nephrin podocin interaction. (A) Wild-type and mutant Tac/nephrin as described in Figure 3 were transfected with Myc/ podocin and Fyn in Cos-1 cells. One day later, cell lysates were immunoprecipitated with rabbit antinephrin antiserum and immunoprecipitates were blotted for Myc to study co-immunoprecipitation of nephrin with podocin. Total cell lysates (25 g) were analyzed for expression of Tac/nephrin and Myc/podocin. (Top) Immunoblots. (Bottom) Densitometry. Results were normalized for expression of Tac/nephrin as in Figure 3C. *P 0.05 versus wild type with Fyn; n 5. (B) The same experiments as in A were repeated for Y1204F and quadruple mutant prepared in full-length nephrin with similar results. *P 0.05 versus wild type with Fyn; n 5to8. Figure 6. Fyn augments nephrin-podocin mediated AP-1 activation. HEK293T cells were plated in 24-well plates and transfected with AP-1 luciferase (50 ng for all wells), nephrin-f (200 ng), F-podocin (200 ng), and Fyn (75 ng) as indicated, and luciferase activity was quantified as in the Materials and Methods section. *P 0.05 versus nephrin podocin; n 3.

8 J Am Soc Nephrol 15: , 2004 Fyn Phosphorylates Nephrin 3013 affected in the PHN model of membranous nephropathy. In PHN, it is known that complement C5b-9 mediated GEC injury leads to morphologic changes of GEC and proteinuria. Glomerular lysates from normal rats and rats with PHN were immunoprecipitated with antinephrin antiserum and blotted for phosphotyrosine. We anticipated that in glomeruli of rats with PHN, tyrosine phosphorylation of nephrin would be decreased, leading to diminished interaction of nephrin and podocin. However, to our surprise, tyrosine phosphorylation was markedly increased ( fold) in glomeruli from rats with PHN, as compared with normal rats (n 6; P 0.005; Figure 7A). The amount of active (phosphorylated) Src(-family kinases) was also increased in glomeruli from rats with PHN (Figure 7B). Tyrosine phosphorylation of nephrin was not increased on day 5, when proteinuria was not observed yet (data not shown). These results suggest that loss of glomerular permselectivity can be associated with both hypophosphorylation or hyperphosphorylation of nephrin, as occurs respectively in Fyn-deficient mice and in the PHN rat model (see the Discussion section). Figure 7. Nephrin is tyrosine phosphorylated in rat glomeruli. (A) Passive Heymann nephritis (PHN) was induced as in the Materials and Methods section. Glomerular lysates were prepared on day 14, immunoprecipitated for nephrin, and blotted for phosphotyrosine (Py20) and nephrin. Immunoblot of glomerular lysates for nephrin always showed two bands (~180 and ~170 kd) as reported by others (12,21). These double bands are attributed to differential glycosylation. Tyrosine phosphorylated nephrin always migrated at ~180 kd. (Top) Immunoblot. (Bottom) Densitometry. *P versus normal rats; n 6. (B) On day 14 of PHN, glomerular lysates were analyzed by immunoblotting for Src, phospho- (activated)-src, and Fyn. (C) Glomerular lysates were prepared in buffer without phosphatase inhibitors and incubated with 20 U of calf intestinal alkaline phosphatase for 30 min at 37 C before immunoprecipitation (lane 1). Thus, at least some Src-family kinases are in an active form even in normal rat glomeruli. Taken together, these results indicate that the cytoplasmic domain of nephrin could be tyrosine phosphorylated in vitro and in vivo and that Src-family kinases are the likely kinases responsible for this phosphorylation. We next studied how tyrosine phosphorylation of nephrin is Discussion Tyrosine phosphorylation of nephrin has been reported by four other groups (9 12). Verma et al. (11) demonstrated that the Src-family kinase Fyn directly binds to and phosphorylates nephrin. Simons et al. (10) reported that rat nephrin was tyrosine phosphorylated when rats received an injection of an antibody that causes morphologic changes of podocytes. Similarly, Lahdenpera et al. (9) recently reported that clustering of nephrin by antibodies caused strong tyrosine phosphorylation of nephrin in the cytoplasmic domain. However, the direct consequence of nephrin phosphorylation was not addressed in these studies. Nephrin and podocin both are critical for glomerular permselectivity (2). Direct interaction of nephrin and podocin was demonstrated and is believed to be important for their functions (6,7). Thus, the current results demonstrating that tyrosine phosphorylation of nephrin by Fyn augments nephrin podocin interaction suggests that tyrosine phosphorylation of nephrin has a direct implication in normal GEC function. Podocin does not contain any domain known to bind phosphotyrosine (e.g., SH2 domain, PTB domain). Thus, most likely, tyrosine phosphorylation causes conformational changes of nephrin, which facilitate interaction with podocin. Another possibility is that phosphorylated nephrin recruits a phosphotyrosine-binding adapter molecule, which in turn recruits podocin. Of interest, it was recently demonstrated that the cytoplasmic domain of nephrin binds to the p85 regulatory subunit of PI3K and together with CD2AP and podocin activates the AKT signaling pathway (12) It is possible that Fynmediated tyrosine phosphorylation of nephrin not only augments its interaction with podocin but also facilitates recruitment of additional signaling molecules such as PI3K. In addition, it was recently reported that podocin facilitates recruitment of nephrin to the lipid raft (8). It will be of interest to study whether tyrosine phosphorylation of nephrin by Srcfamily kinases would affect this process. However, because Src-family kinases are primarily localized in the lipid raft (13),

9 3014 Journal of the American Society of Nephrology J Am Soc Nephrol 15: , 2004 it is more likely that tyrosine phosphorylation would modulate nephrin podocin interaction after nephrin is recruited to the lipid raft. Precise location where Fyn is modulating nephrin podocin interaction will require further studies. Src-family kinases are known to regulate signaling of adhesion molecules. c-src has a pivotal role in integrin-mediated signal transduction via focal adhesion kinase and also in cadherin-mediated signal transduction via catenins (15,28). Because nephrin acts as an adhesion molecule between adjacent foot processes of GEC, by analogy to integrins and cadherins, it is conceivable that the cytoplasmic domain of nephrin could serve as a scaffold for protein complexes that contain Srcfamily kinases and transmit signals into the cells. A recent review by Benzing (29) highlighted the role of the slit diaphragm proteins in intracellular signaling pathways and the potential role of tyrosine phosphorylation of nephrin and Neph1 in these pathways. Our data indicating that Fyn augments AP-1 promoter activation by nephrin and podocin is in line with these newly emerging concepts. Fyn-knockout mice demonstrate morphologic changes of podocytes and/or proteinuria (11,16). Because nephrin can be phosphorylated by Fyn, it is tempting to hypothesize that Fyn-mediated tyrosine phosphorylation of nephrin is important for the maintenance of normal morphology and permselectivity of GEC. Increased nephrin podocin interaction by Fyn in vitro and tyrosine phosphorylation of nephrin in normal rat glomeruli would be in agreement with this hypothesis. Curiously, in the PHN rat model of membranous nephropathy, nephrin tyrosine phosphorylation was increased by almost fivefold, as compared with normal rats. This seems to contradict the hypothesis that tyrosine phosphorylation by Fyn is important for normal GEC function. There are several possible interpretations for these results. It is possible that the primary target of injury in PHN is not nephrin; thus, hyperphosphorylation of nephrin is either an epiphenomenon or a compensatory mechanism after GEC injury. Alternatively, a delicate balance of nephrin phosphorylation and dephosphorylation might be important for maintenance of normal podocyte function. Although facilitation of nephrin podocin interaction by the action of Fyn may be necessary for normal podocyte function, transient interruption of this association may also be required in vivo. Thus, hyperphosphorylation of nephrin as observed in PHN could lead to inappropriate regulation of nephrin podocin interaction. In addition, we have noted that protein expression of nephrin was consistently decreased when it was expressed with Fyn (Figures 2E, 4D, and 5B). Lahdenpera et al. (9) also noted that the nephrin mutant that lacks most of the cytoplasmic domain (thus cannot be phosphorylated) had a higher protein expression, as compared with the wild-type nephrin. Thus, it is also possible that hyperphosphorylation of nephrin could lead to a decreased protein stability of nephrin. In summary, the current study demonstrates that the cytoplasmic domain of nephrin is tyrosine phosphorylated in vivo and in vitro. This tyrosine phosphorylation influences nephrin podocin interaction and nephrin-mediated signaling. How tyrosine phosphorylation of nephrin is modulated and contributes to proteinuria in kidney diseases requires further studies. Acknowledgment This study was supported by grants from the Canadian Institute of Health Research (T.T., S.L.); the Kidney Foundation of Canada (T.T., S.L.); and the Ministry of Education, Science, Culture and Sports of Japan (H.K.). T.T. and S.L. are recipients of a scholarship from the Canadian Institute of Health Research. We are grateful for the assistance of Dr. Benzing in setting up AP-1 luciferase assay. References 1. Kestila M, Lenkkeri U, Mannikko M, Lamerdin J, McCready P, Putaala H, Ruotsalainen V, Morita T, Nissinen M, Herva R, Kashtan CE, Peltonen L, Holmberg C, Olsen A, Tryggvason K: Positionally cloned gene for a novel glomerular protein nephrin is mutated in congenital nephrotic syndrome. Mol Cell 1: , Kerjaschki D: Caught flat-footed: Podocyte damage and the molecular bases of focal glomerulosclerosis. J Clin Invest 108: , Tryggvason K: Unraveling the mechanisms of glomerular ultrafiltration: Nephrin, a key component of the slit diaphragm. JAm Soc Nephrol 10: , Khoshnoodi J, Ruotsalainen V, Tryggvason K: The role of nephrin in cytoskeleton organization and podocyte morphology [Abstract]. J Am Soc Nephrol 11: A0280, Boute N, Gribouval O, Roselli S, Benessy F, Lee H, Fuchshuber A, Dahan K, Gubler MC, Niaudet P, Antignac C: NPHS2, encoding the glomerular protein podocin, is mutated in autosomal recessive steroid-resistant nephrotic syndrome. Nat Genet 24: , Huber TB, Kottgen M, Schilling B, Walz G, Benzing T: Interaction with podocin facilitates nephrin signaling. J Biol Chem 276: , Schwarz K, Simons M, Reiser J, Saleem MA, Faul C, Kriz W, Shaw AS, Holzman LB, Mundel P: Podocin, a raft-associated component of the glomerular slit diaphragm, interacts with CD2AP and nephrin. J Clin Invest 108: , Huber TB, Simons M, Hartleben B, Sernetz L, Schmidts M, Gundlach E, Saleem MA, Walz G, Benzing T: Molecular basis of the functional podocin-nephrin complex: Mutations in the NPHS2 gene disrupt nephrin targeting to lipid raft microdomains. Hum Mol Genet 21: 21, Lahdenpera J, Kilpelainen P, Liu XL, Pikkarainen T, Reponen P, Ruotsalainen V, Tryggvason K: Clustering-induced tyrosine phosphorylation of nephrin by Src family kinases. Kidney Int 64: , Simons M, Schwarz K, Kriz W, Miettinen A, Reiser J, Mundel P, Holthofer H: Involvement of lipid rafts in nephrin phosphorylation and organization of the glomerular slit diaphragm. Am J Pathol 159: , Verma R, Wharram B, Kovari I, Kunkel R, Nihalani D, Wary KK, Wiggins RC, Killen P, Holzman LB: Fyn binds to and phosphorylates the kidney slit diaphragm component nephrin. J Biol Chem 278: , Huber TB, Hartleben B, Kim J, Schmidts M, Schermer B, Keil A, Egger L, Lecha RL, Borner C, Pavenstadt H, Shaw AS, Walz G, Benzing T: Nephrin and CD2AP associate with phosphoinositide 3-OH kinase and stimulate AKT-dependent signaling. Mol Cell Biol 23: , Dykstra M, Cherukuri A, Sohn HW, Tzeng S-J, Pierce SK: Location is everything: Lipid rafts and immune cell signaling. Ann Rev Immunol 21: , 2003

10 J Am Soc Nephrol 15: , 2004 Fyn Phosphorylates Nephrin Thomas SM, Brugge JS: Cellular functions regulated by Src family kinases. Annu Rev Cell Dev Biol 13: , Cantley LG: Growth factors and the kidney: Regulation of epithelial cell movement and morphogenesis. Am J Physiol 271: F1103 F1113, Yu CC, Yen TS, Lowell CA, DeFranco AL: Lupus-like kidney disease in mice deficient in the Src family tyrosine kinases Lyn and Fyn. Curr Biol 11: 34 38, Abe J, Berk BC: Fyn and JAK2 mediate Ras activation by reactive oxygen species. J Biol Chem 274: , Takeuchi M, Kuramochi S, Fusaki N, Nada S, Kawamura-Tsuzuku J, Matsuda S, Semba K, Toyoshima K, Okada M, Yamamoto T: Functional and physical interaction of proteintyrosine kinases Fyn and Csk in the T-cell signaling system. J Biol Chem 268: , Tezuka T, Umemori H, Akiyama T, Nakanishi S, Yamamoto T: PSD-95 promotes Fyn-mediated tyrosine phosphorylation of the N-methyl-D-aspartate receptor subunit NR2A. Proc Natl Acad SciUSA96: , Letourneur F, Klausner RD: T-cell and basophil activation through the cytoplasmic tail of T-cell-receptor zeta family proteins. Proc Natl Acad Sci U S A 88: , Kawachi H, Koike H, Kurihara H, Yaoita E, Orikasa M, Shia MA, Sakai T, Yamamoto T, Salant DJ, Shimizu F: Cloning of rat nephrin: expression in developing glomeruli and in proteinuric states. Kidney Int 57: , Takano T, Cybulsky AV: Complement C5b-9-mediated arachidonic acid metabolism in glomerular epithelial cells: Role of cyclooxygenase-1 and -2. Am J Pathol 156: , Dustin ML, Olszowy MW, Holdorf AD, Li J, Bromley S, Desai N, Widder P, Rosenberger F, van der Merwe PA, Allen PM, Shaw AS: A novel adaptor protein orchestrates receptor patterning and cytoskeletal polarity in T-cell contacts. Cell 94: , Chan PY, Kanner SB, Whitney G, Aruffo A: A transmembraneanchored chimeric focal adhesion kinase is constitutively activated and phosphorylated at tyrosine residues identical to pp125fak. J Biol Chem 269: , Blom N, Gammeltoft S, Brunak S: Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol 294: , Topham PS, Kawachi H, Haydar SA, Chugh S, Addona TA, Charron KB, Holzman LB, Shia M, Shimizu F, Salant DJ: Nephritogenic mab is directed at the extracellular domain of rat nephrin. J Clin Invest 104: , Yan K, Khoshnoodi J, Ruotsalainen V, Tryggvason K: N-linked glycosylation is critical for the plasma membrane localization of nephrin. J Am Soc Nephrol 13: , Schlaepfer DD, Jones KC, Hunter T: Multiple Grb2-mediated integrin-stimulated signaling pathways to ERK2/mitogen-activated protein kinase: Summation of both c-src- and focal adhesion kinase-initiated tyrosine phosphorylation events. Mol Cell Biol 18: , Benzing T: Signaling at the slit diaphragm. J Am Soc Nephrol 15: , 2004