Epidermal Growth Factor Activates Na^+/H⁺ Exchange in Podocytes through a Mechanism that Involves Janus Kinase and Calmodulin

Abstract

Sodium-proton exchanger type 1 (NHE-1) is ubiquitously expressed, is activated by numerous growth factors, and plays significant roles in regulating intracellular pH and cellular volume, proliferation and cytoskeleton. Despite its importance, little is known about its regulation in renal glomerular podocytes. In the current work, we studied the regulation of NHE-1 activity by the epidermal growth factor receptor (EGFR) in cultured podocytes. RT-PCR demonstrated mRNAs for NHE-1 and NHE-2 in differentiated podocytes, as well as for EGFR subunits EGFR/ErbB1, Erb3, and ErbB4. EGF induced concentration-dependent increases in proton efflux in renal podocytes as assessed using a Cytosensor microphysiometer, were diminished in the presence of 5-(N-methyl-N-isobutyl) amiloride or in a sodium-free solution. Furthermore, pharmacological inhibitors of janus kinase (Jak2) and calmodulin (CaM) attenuated EGF-induced NHE-1 activity. Co-immunoprecipitation studies determined that EGF induced formation of complexes between Jak2 and CaM, as well as between CaM and NHE-1. In addition, EGF increased levels of tyrosine phosphorylation of Jak2 and CaM. The EGFR kinase inhibitor, AG1478, blocked activation of NHE-1, but did not block EGF-induced phosphorylation of Jak2 or CaM. These results suggest that EGF induces NHE-1 activity in podocytes through two pathways: (1) EGF → EGFR → Jak2 activation (independent of EGFR tyrosine kinase activity) → tyrosine phosphorylation of CaM → CaM binding to NHE-1 → conformational change of NHE-1 → activation of NHE-1; and (2) EGF →EGFR → EGFR kinase activation → association of CaM with NHE-1 (independent of Jak2) → conformational change of NHE-1 → activation of NHE-1.

INTRODUCTION

Epidermal growth factor (EGF) plays a number of key roles in the kidney, contributing to cellular proliferative and survival pathways, renal metabolism [1], regenerative hyperplasia [2], tubulointerstitial injury [3], tubulogenesis [4], transport [5], renal cyst formation [6] and renal development [7]. EGF also has been implicated in the genesis and development of polycystic kidney disease [8, 9]. Despite the importance of EGF in many renal functions, particularly in renal tubules and mesangial cells, little is known about its effects in glomerular podocytes. Podocytes are critical for the maintenance of normal glomerular structure [10, 11], and aberrant podocyte function has been implicated in the pathogenesis of chronic renal diseases [12–17]. These highly specialized cells are characterized by the formation of foot processes that are interconnected by the slit diaphragm, which is a critical component of the glomerular filtration barrier [18, 19]. Podocytes are emerging as the focus of intense investigation because they have been implicated as both targets and instigators of renal injury in various progressive renal diseases [18, 20–23]. Therefore, it would be desirable to elucidate signaling pathways in podocytes, which may be important in delineating the mechanisms through which podocytes contribute to the progression of glomerular injury.

Because of the emerging role of the sodium-proton exchanger type 1 (NHE-1), also known as product of SLC9A1, solute carrier family 9A, type 1 in the regulation of the cytoskeleton [24], apoptosis and cellular proliferation [25], cell cycle control [26], and development and maintenance of the transformed cellular phenotype [27], we thought that it might be of interest to develop a better understanding of its regulation in podocytes. Previous studies have demonstrated that EGF stimulates NHE-1 in non-renal cells [28–30], but the signaling pathways involved in the regulation have not been fully elucidated. Moreover, currently there are no data on the regulation of NHE-1 in podocytes. In that regard, recent studies from our laboratory have shown that NHE-1 can be activated by G protein-coupled receptors or hypertonic medium, through Janus kinase 2 (Jak2)-dependent phosphorylation of CaM, and subsequent interaction between CaM and NHE-1[31–33]. Therefore, we wanted to determine whether EGF is important for regulating NHE-1 activity in podocytes, and to identify key components of the signal transduction pathway linking EGF and NHE-1. In the current study, we demonstrate that EGF stimulates NHE-1 activity in podocytes. In addition, we have shown that Jak2 and CaM play critical roles in the activation of NHE-1 by EGF in podocytes.

MATERIALS AND METHODS

Cell Culture

The podocyte cell line was kindly provided by Dr. Peter Mundel of Mt. Sinai School of Medicine. Podocytes were cultured as previously described [34]. Undifferentiated podocytes were maintained in RPMI-1640 medium containing 10 units/ ml of mouse recombinant γ-interferon, 10% FBS, 100 units/ml of penicillin and 100 μg/ml of streptomycin at 33ºC in 95% air and 5% CO₂. To induce differentiation, podocytes were maintained in the same medium as undifferentiated podocytes without γ-interferon at 37ºC in 95% air and 5% CO₂ for 14 days. All experiments were conducted using differentiated podocytes, unless stated otherwise.

Immunofluorescence Microscopy

Immunolabeling was performed as previously described [34]. Cells were seeded in 35-mm collagen-coated glass-bottom culture dishes (MatTek Corporation) and fixed with 2% paraformaldehyde, 4% sucrose in phosphate-buffered saline (PBS) for 10 min at room temperature. Subsequently, cells were permeabilized with 0.3% Triton X-100 (Sigma) in PBS for 5 min, following which nonspecific binding sites were blocked with 2% fetal calf serum, 2% BSA and 0.2% gelatin in PBS for 1h. Incubations with the appropriate dilutions of primary and secondary antibodies (as directed by the manufacturer) were performed in blocking solution. The primary and secondary antibodies used were: anti-WT1 (1:50) (Santa Cruz Biotechnology); anti-synaptopodin (Research Diagnostics, Inc.) and Alexa Fluor 488 goat anti-mouse IgG (Molecular Probes). Confocal microscopy was performed using a Zeiss LSM 510 META laser-scanning microscope (Carl Zeiss, Inc.).

Microphysiometry

NHE-1 activity studies were conducted on a Cytosensor microphysiometer as previously described for other cell types [32, 35, 36]. Cells were plated on transwell membranes (3 μm pore size, 12 mm size) at a density of 300,000 cells per insert, and serum starved overnight on the day prior to experimentation. On the day of the experiment, the cells were washed with serum-free, bicarbonate-free F-12 medium, prior to being placed into microphysiometer chambers. The chambers were perfused at 37ºC with serum-free media or balanced salt solutions. After establishment of a stable baseline for at least five cycles, cells were exposed to the drugs for 4 cycles (360 s). Podocytes had low basal proton efflux levels (−10 to −40 μV/ sec), which roughly corresponds to millipH units/minute according to the Nernst equation). The extracellular acidification rate (ECAR, rate of accumulation of protons during a stop-flow period) was measured at peak stimulation after initiation of drug treatment, as is standard for microphysiometry studies. This typically occurred after two or three cycles (180 or 270 sec) of exposure to EGF. Rate data were expressed as percentage of baseline values.

RT-PCR and PCR

RNA was prepared from differentiated and undifferentiated podocytes using Trizol reagent following the manufacturer’s protocol (Invitrogen). Five micrograms of total RNA were used for first strand cDNA synthesis (SuperScript III RT-PCR kit, Invitrogen). EGFR/ErbB transcripts were identified using SuperArray’s Multigen-12 RT-PCR profiling kit. To analyze NHE-1 message an already published primer pair was used: 5'-TCTGCCGTCTCAACTGTCTCTA-3' sense, 5'-CCCTTCAACTCCTCATTCACCA-3' antisense [37], which generated a 422 base pair (bp) product. For analyzing other NHE messages, new PCR primer pairs were designed: NHE-2 (5'-ACACACAACATCCGGGTCATT-3' sense, 5'-CGCTTGTGTCTGCCGTCA-3' antisense) resulted in a 982 bp product; NHE-3 (5'-CGCCTGGAGTCCTTTAAGTC-3' sense, 5'-GGAGAACACGGGATTATCAAT-3' antisense) generated a 294 bp product, and NHE-4 (5'-GAGGACATAGAAGCGGTGGAC-3' sense, 5'-AGGAGAAAGCCGCTTGATTC-3' antisense) resulted in a 993 bp PCR product. Target specificities of the PCR primers were confirmed by sequencing the PCR products using the MUSC sequencing core facility.

Immunoprecipitation

For immunoprecipitation of Jak2 and NHE-1, quiescent differentiated podocytes grown on 100 mm collagen-coated tissue culture dishes were pretreated with 50 μM of AG490 or 20 μM of AG1478 for 30 minutes prior to treatment with 10 ng/ml of EGF or vehicle for 5 min, and then lysed in 1 ml/dish of RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.25% deoxycholic acid, 1 % NP-40, 1 mM EDTA) supplemented with protease inhibitors (1 mM AEBSF, 0.8 μM aprotinin, 50 μM bestatin, 15 μM E-64, 20 μM leupeptin, 10 μM pepstatin A, 1 mM sodium orthovanadate, 1 mM sodium fluoride and 1 mM PMSF). Equal amounts of proteins (1.5 mg) were precleared by incubation with protein A/G sepharose beads for 30 min at 4°C. After a brief centrifugation, the supernatants were removed and incubated with either agarose conjugated anti-JAK2 antibody (Upstate) or anti-NHE-1 antibody (Chemicon) overnight at 4°C. Immunoprecipitates were captured with 50 μl of protein A/G beads at 4°C for 1 hr. Then, the samples were centrifuged and washed thrice with 1 ml of RIPA buffer, and the proteins were eluted from the beads using 2x Laemmli sample buffer. Samples subsequently were separated by SDS-PAGE and transferred to PVDF membrane. Blots were probed with anti-calmodulin antibody (Upstate), and, to ensure equal NHE-1 and Jak-2 precipitation from the samples, with NHE-1 monoclonal antibody (Chemicon) or Jak-2 antiserum (Upstate).

For phosphotyrosine immunoprecipitation experiments, quiescent podocytes grown onto 100 mm collagen-coated tissue culture dishes were pretreated with AG-490 (50 μM), or with AG-1478 (20 μM) or vehicle for 30 min, then stimulated with 10 ng/ml EGF or vehicle for 5 min and lysed in 0.5 ml/100 mm dish of RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 1% Nonidet P-40, 1 mM NaF, 1 mM Na₃VO₄, 1 mM phenylmethylsulfonyl fluoride, aprotinin, leupeptin, and pepstatin at 1 μg/ml each). Cell lysates were precleared by incubating with protein A-agarose bead slurry for 30 min at 4°C. Precleared lysates (1 μg/μl total cell protein) were incubated with monoclonal anti-phosphotyrosine antibody conjugated to protein A-agarose (Cell Signaling) overnight at 4°C. The agarose beads were collected by centrifugation, washed twice with RIPA buffer and once with PBS, resuspended in 2x Laemmli sample buffer, boiled for 5 min, and subjected to SDS-PAGE and subsequent immunoblot analyses with polyclonal anti-phosphotyrosine, anti-EGFR, anti-Jak2, or with monoclonal anti-CaM antibodies (Upstate).

Statistical Analysis

Data were analyzed by paired, two-tailed Student’s t test and analysis of variance using GraphPad Statistics Software. P values < 0.05 were considered significant.

RESULTS

Immunohistochemical confirmation of podocyte differentiation

Podocytes were stained for WT-1 and synaptopodin. Undifferentiated podocytes did not stain for synaptopodin (Figure 1, Panel A); however, the cells did stain for WT-1 (Figure 1, Panel C). Differentiated podocytes stained for synaptopodin (Figure 1, Panel B) and WT-1 (Figure 1, Panel D). The results of the staining confirm that in our hands, the cultured podocytes showed hallmarks of differentiation.

Figure 1. Immunofluorescence analysis of podocyte markers.

Cells were fixed and then stained for synaptopodin or WT-1 to confirm differentiated from undifferentiated podocytes. Undifferentiated podocytes did not stain for synaptopodin (A). In contrast, differentiated podocytes stained for synaptopodin (B). Undifferentiated and differentiated podocytes stained for WT-1 (C and D).

EGFR mRNAs are expressed in podocytes

Epidermal growth factor (EGF) receptors constitute a family of four prototypical receptor tyrosine kinases (ErbB1-4). EGF receptor (EGFR) subunits dimerize upon ligand binding, resulting in the formation of activated receptors. We determined which EGFR subunit mRNAs were expressed in podocytes using RT-PCR. Undifferentiated podocytes expressed the mRNAs for EGFR/ErbB1, Neu/HER2, ErbB3, and ErbB4 (Figure 2A). Differentiated podocytes expressed the mRNAs for EGFR/ErbB1, Erb3, and ErbB4. Neu/HER2 mRNA was detectable at very minute levels in differentiated podocytes (Figure 2A).

Figure 2. Presence of functional EGFR in podocytes.

A. RT-PCR analyses of EGFR/ErbB receptor subunit mRNA expression. Expression of different EGFR subunits in undifferentiated and differentiated podocytes is shown: EGFR (183 bp), Neu/HER2 (130 bp), ErbB3 (153 bp) and ErbB4 (204 bp). B. Microphysiometry was used to analyze the concentration-dependent effects of EGF on extracellular acidification rate (ECAR) in podocytes. Independent experiments were performed a minimum of three times in triplicate, and the data are presented as mean ± SEM of peak values.

EGF induces concentration-dependent increases in ECAR

Having established that podocytes express EGFR mRNAs, we next determined whether the cells expressed functional EGFR. We measured EGF-induced increases in extracellular acidification rates using microphysiometry under stop flow conditions. Figure 2B shows that EGF increased proton efflux in a concentration-dependent manner, confirming the presence of functional EGFR in differentiated podocytes.

EGF activates Na⁺/H⁺ exchange in podocytes

We next sought to determine the nature of the proton efflux pathway activated by EGF. Because EGF has been shown to stimulate sodium-proton exchangers in fibroblasts, esophageal epithelia and chondrocytes [30, 38–40], we studied the expression of mRNAs encoding plasma membrane localized sodium-proton exchangers NHE-1, NHE-2, NHE-3, and NHE-4. Figure 3A shows that differentiated podocytes express mRNA for NHE-1 and NHE-2, with the levels of NHE-1 mRNA predominating. Undifferentiated podocytes express only the mRNA for NHE-1 (Figure 3A). The mRNAs for NHE-3 and NHE-4 were not detected in undifferentiated or differentiated podocytes. Thus, it is possible that EGF-mediated proton efflux from differentiated podocytes involves NHE-1 or NHE-2.

Figure 3. EGF stimulates NHE-1 activity in podocytes.

A. RT-PCR analysis of NHE isoforms. Expression of different NHE isoform mRNAs in (a) whole mouse kidney lysates; (b) undifferentiated podocytes; and in (c) differentiated podocytes are shown: NHE-1 (422 bp), NHE-2 (982 bp), NHE-3 (294 bp) and NHE-4 (993) bp. B. Microphysiometry was used to characterize EGF-induced increases in ECAR. Podocytes were stimulated with 10 ng/ml of EGF in the presence or absence of sodium-free medium or MIA (a NHE-1 inhibitor). Values depicted are mean ± SEM of peak values from 3 independent experiments performed in triplicate. **P < 0.01 vs drug alone samples.

In order to test the involvement of sodium-proton exchangers in the stimulation of proton efflux by EGF, we isotonically substituted tetramethylammonium (TMA) for sodium in the extracellular perfusate, thereby removing the extracellular substrate for sodium-proton exchangers. Figure 3B shows that EGF stimulated proton efflux in a medium containing sodium, and that this effect was nearly abolished in medium in which sodium was replaced by TMA. In addition, 5 μM of 5-(N-methyl-N-isobutyl) amiloride (MIA), an inhibitor of NHE-1 and NHE-2, attenuated EGF-induced proton efflux by nearly 60% (Figure 3B). These findings suggest that EGF-induced increases in ECAR are due to NHE-1 or NHE-2 in podocytes.

Calmodulin inhibitors, phosphotyrosine inhibitors and Jak2 inhibitors attenuate EGF-induced NHE-1 activity

NHE-1 has two CaM-binding domains that are critical for its activation by many stimuli [41, 42], whereas the role of CaM in the regulation of NHE-2 is much less certain [43]. Although elevations of intracellular calcium increase the activity of NHE-2 [43], CaM has been shown to exert tonic inhibition on NHE-2 [34]. To determine whether CaM is involved in EGF induced increases in ECAR, we analyzed the effects of a panel of CaM inhibitors on EGF-induced proton efflux in podocytes. The results in Figure 4A demonstrate that W-7, fluphenazine, and ophiobolin A, each inhibited EGF-induced increases in ECAR by ≥ 60%. Because none of those agents reduced the basal levels of proton efflux in podocytes, the results are most consistent with EGF activation of NHE-1.

Figure 4.

Effects of CaM inhibitors and tyrosine kinase inhibitors on EGF-induced ECAR in podocytes. Microphysiometry was used to measure ECAR in podocytes. A. Podocytes were pretreated with CaM inhibitors for 30 min prior to stimulation with 10 ng/ ml of EGF. B. Podocytes were pretreated with AG1478 (20μM) or AG490 (50 μM) for 30 min prior to stimulation with 10 ng/ml of EGF. Values are mean ± SEM of peak values from 3 independent experiments performed in triplicate. * P < 0.05; **P < 0.01 vs drug alone samples.

Because previous studies from our laboratory demonstrated that Jak2 is important for NHE-1 activation by hypertonicity and by G_q-coupled receptors [32, 33], we analyzed the effects of a Jak2 inhibitor, AG490, on EGF-induced activation of NHE-1 in podocytes. AG490 (50 μM) inhibited EGF-induced increases in ECAR by >50% (Figure 4B). The EGFR tyrosine kinase inhibitor AG1478 (20 μM) also inhibited ECAR in podocytes that were stimulated with EGF by >95%. These results support the involvement of Jak2 and the EGFR in the EGF-induced increases in ECAR.

EGF increases formation of complexes of Jak2 and NHE-1 with CaM

To further examine a role for Jak2 in EGF-induced signaling, we determined whether EGF stimulates the formation of signaling complexes between Jak2, NHE-1, and CaM. To explore this possibility, we performed co-immunoprecipitation experiments using cell lysates from podocytes pretreated with vehicle or with inhibitors of Jak2 or EGFR tyrosine kinases. Figure 5A shows that CaM was present in Jak2 immunoprecipitates, and that the amount of CaM present in these immunoprecipitates was doubled after EGF stimulation. Pretreatment of cells with a Jak2 inhibitor, AG-490 (50 μM for 30 min) significantly decreased the amount of CaM in Jak2 immunoprecipitates, whereas pretreatment with an EGFR kinase inhibitor, AG1478 (20 μM for 30 min) did not have such effect. This result suggests that EGF-induced Jak2 activity (but not EGFR kinase activity) is necessary for formation of the complex between Jak2 and CaM. In addition, Figure 5B shows that there was a marked increase in the amount of CaM in NHE-1 immunoprecipitates after treatment with EGF. In contrast, there was not an increased formation of complexes between Jak2 and NHE-1 in podocytes after treatment with EGF (not shown). Pretreatment of cells with a Jak2 inhibitor, AG490 (50 μM) or EGFR kinase inhibitor, AG1478 (20 μM) decreased the amount of CaM in NHE-1 immunoprecipitates. The latter result suggests that both EGFR kinase activity and Jak2 activity are required to induce formation of a complex between CaM and NHE-1.

Figure 5. EGF induces formation of signaling complexes between Jak2 and CaM, and NHE-1 and CaM in podocytes.

Co-immunoprecipitation experiments were performed as described under Methods. A. Podocytes were pretreated with AG-490 (50 μM), or with AG-1478 (20 μM), or vehicle for 30 minutes, and then stimulated with 10 ng/ml of EGF for 5 min prior to being immunoprecipitated with agarose conjugated anti-Jak2 antibody and then immunoblotted with anti-CaM or anti-Jak2 antibody. The insert is a representative immunoblot (CaM). The same blot was stripped and re-probed with antibodies for total Jak2 to ensure a equal loading of proteins (Jak2). B. After stimulation of podocytes with 10 ng/ml of EGF for 10 min in the presence or absence of AG490 (50 μM) or AG1478 (20 μM), precleared cell lysates were immunoprecipitated with agarose conjugated anti-NHE-1 antibody and then immunoblotted with anti-calmodulin or anti-NHE-1 antibody. Values are mean ± SEM from 3 independent experiments. * P < 0.05 vs control, ^‡ P < 0.05 vs EGF.

EGF Induces Tyrosine Phosphorylation of Jak and CaM

In order to examine further the signaling mechanisms involved in the activation of NHE-1 by EGF, we next considered that EGF could stimulate tyrosine phosphorylation of CaM. The data presented in Figure 6 (panel A) demonstrate that EGF increased the amount of EGFR in phosphotyrosine immunoprecipitates, and that this effect is unchanged in the presence of Jak2 inhibitor, but is completely abolished after pretreatment with AG1478. This result demonstrates that AG1478 effectively inhibits intrinsic EGFR tyrosine kinase activity in podocytes. Figure 6 (panel B) shows that EGF induces tyrosine phosphorylation of Jak2, which is inhibited by pretreatment with AG-490, but not with AG-1478. These results provide strong evidence that EGF induces tyrosine phosphorylation of EGFR and Jak2 via auto-phosphorylation of these kinases, and also demonstrate that AG-490 and AG-1478 were effective under our experimental conditions. The results also suggest that EGFR kinase activity is not required for Jak2 activation by EGF. Figure 6 (panel C) demonstrates that EGF increases the amount of CaM in phosphotyrosine immunoprecipitates and that this effect can be significantly decreased by pretreatment of cells with AG-490, but not with AG-1478, suggesting that tyrosine phosphorylation of CaM is induced by Jak2, and does not require EGFR kinase activity. In that regard, we demonstrated previously that CaM is a bona fide substrate for Jak2 [31].

Figure 6.

Tyrosine phosphorylation of Jak2 and CaM in response to EGF treatment of podocytes. Co-immunoprecipitation experiments were performed as described under Methods. A. Podocytes were pretreated with AG-490 (50 μM), or with AG-1478 (20 μM), or vehicle for 30 minutes, and stimulated with 10 ng/ml of EGF for 5 min prior to cell lysates being immunoprecipitated with anti-phosphotyrosine/protein A-agarose, and then immunoblotted with polyclonal anti-EGFR antibody (A), or with polyclonal anti-Jak2 antibody (B). C. To determine if CaM was tyrosine phosphorylated, podocytes were stimulated with 10 ng/ml of EGF for 5 min after pretreatment with AG-490 (50 μM), or with AG-1478 (20 μM) or vehicle for 30 minutes, immunoprecipitated with monoclonal anti-phosphotyrosine (PY) antibody conjugated with protein A-agarose and then immunoblotted with a monoclonal anti-CaM antibody. The same blots were stripped and re-probed with polyclonal PY antibody to confirm that CaM is indeed tyrosine-phosphorylated in response to stimulation with EGF (not shown). Values are mean ± SEM from 5 independent experiments. * P < 0.05 vs control, ^‡ P < 0.05 vs EGF.

DISCUSSION

What is new about this work is that we have demonstrated that EGF activates NHE-1 through the intermediary actions of Jak2 and CaM in renal podocytes. The work expands recent studies demonstrating that hypertonicity and G_q-coupled receptors activate NHE-1 in several cell types through a pathway involving sequential phosphorylation and activation of Jak2, tyrosine phosphorylation of CaM, CaM binding to NHE-1, and activation of NHE-1. The current work is significant in that we have (1) demonstrated that a prototypical receptor tyrosine kinase (the EGFR) utilizes this pathway and a second pathway, both of which are required for full activation of NHE-1; (2) refined the previously identified pathway as follows: EGF → EGFR → Jak2 activation (independent of EGFR tyrosine kinase activity) → tyrosine phosphorylation of CaM →CaM binding to NHE-1 → activation of NHE-1; (3) characterized a second activation pathway as follows: EGF → EGFR → EGFR kinase activation → association of CaM to NHE-1 (independent of Jak2) → activation of NHE-1 (Figure 7). (4) We also have identified mRNAs for various isotypes of plasma membrane NHEs, and for EGFR-related subunits, in renal podocytes. Because podocytes have been implicated as playing key roles in the initial stages of various glomerular diseases, this new information may have relevance to the processes that link podocyte dysfunction to progressive renal diseases.

Figure 7.

Schematic representation of two converging pathways of activation of NHE-1 in podocytes by EGF. P-CaM indicates tyrosine-phosphorylated CaM. All other abbreviations are as described in the text.

The evidence implicating Jak2 in the increase in proton efflux is that (1) Jak2 is activated as demonstrated by its tyrosine phosphorylation in response to EGF, (2) AG490 blocks the increased proton efflux induced by EGF, and (3) Jak2 forms a complex with CaM in response to EGF. Although our work does not prove definitively that tyrosine phosphorylation of Jak2 is required for activation of NHE-1 by EGF, this seems likely in that (1) EGF does not increase intracellular calcium levels under our conditions (data not shown), (2) CaM is tyrosine phosphorylated through a pathway that is inhibited by AG490, and (3) CaM is a bona fide substrate for Jak2 [31].

The evidence implicating CaM in the increase in proton efflux is that (1) a panel of CaM inhibitors greatly attenuates the increased proton efflux induced by EGF, (2) CaM is tyrosine phosphorylated in response to EGF, and (3) CaM is induced to form complexes with Jak2 and NHE-1 in response to EGF. The evidence that the proton efflux is mediated by NHE-1 is that it is (1) dependent upon extracellular sodium, (2) inhibited by MIA, (3) dependent upon CaM activity, and (4) associated with increased binding of CaM to NHE-1.

The precise mechanism through which Jak2 activates NHE-1 has not been fully elucidated. We propose that Jak2 tyrosine phosphorylates CaM, thereby increasing its affinity for NHE-1. This would result in increased binding of CaM to NHE-1. A number of kinases have been shown to phosphorylate CaM on serine, threonine and tyrosine residues [44, 45], and to alter the activity of CaM with reference to specific CaM targets [46]. In that regard, our group has recently demonstrated that CaM is directly tyrosine phosphorylated by purified Jak2 [31]. Thus, Jak2 almost certainly phosphorylates CaM on one or both of the tyrosine residues within the CaM sequence, Tyr-99 and Tyr-138. Based on the crystal structure of CaM, Tyr-99 is the more likely target for phosphorylation in that Tyr-99 is located within the third Ca²⁺-binding domain, and is somewhat more exposed than is Tyr-138 [45]. However, Jak2-induced tyrosine phosphorylation of CaM appears to be critical or necessary, but not sufficient to fully activate NHE-1, because EGFR tyrosine kinase activity also is required. Indeed, the effectiveness of AG1478 to block NHE-1 activation suggests that EGFR tyrosine kinase activity also is essential for CaM to bind to NHE-1 and to activate it. It should be noted that we have not formally tested the idea that CaM binding to NHE-1 induces a conformational change that results in activation of NHE-1. However, this idea is intuitively pleasing, and has been supported by experimental evidence in the form of mutation studies by [43], and by solution phase spectroscopy studies of the interaction between CaM and the large regulatory intracellular carboxyl terminus of NHE-1 by Fliegel’s group [41].

It is important to elaborate on our findings that the EGFR kinase inhibitor AG1478 did not decrease the amount of Jak2 and CaM in phosphotyrosine immunoprecipitates (Figure 5B and 5C, respectively), which suggests that there is another factor that allows EGF to regulate tyrosine phosphorylation of CaM independent of EGFR kinase activity. This finding is supported by previous reports that suggest that some EGF-mediated signals such as the JAK/STAT pathway are independent of EGFR kinase activity [47, 48]. Two groups demonstrated that AG1478-independent effects of EGF might be mediated by ErbB2 (Neu/HER2), possibly through oligomerization with ErbB1/EGFR [47, 48]. It is unlikely that this mechanism can account for our findings in that we detected little to no Neu/HER2 mRNA in differentiated podocytes (Figure 2A).

An alternative explanation for the dual Jak2- and EGFR tyrosine kinase-dependent pathways of activation of NHE-1 is that both EGFR and Jak2 could tyrosine phosphorylate CaM. This idea is reasonable because the EGFR has been shown to phosphorylate CaM on Tyr-99 and/or Tyr-138 in other cell systems [44, 49]. Indeed, the EGFR possesses a juxtamembrane CaM-binding motif at residues 624–639, which Martin-Nieto and Villalobo demonstrated could bind to CaM in a calcium-dependent manner, with an affinity of ≈400 nM [50]. However, it seems unlikely that the EGFR directly phosphorylates CaM in podocytes in that the Jak2 inhibitor, AG490, significantly suppresses EGF-induced tyrosine phosphorylation of CaM, whereas AG1478 has no significant effect (Figure 6B).

Because AG1478 attenuates ECAR more than CaM or Jak2 inhibitors, it appears that the receptor tyrosine kinase activity of EGFR might be a bit more necessary than the non-receptor tyrosine kinase pathway involving Jak2-CaM for activating NHE-1. Both pathways clearly converge upon the physical association of NHE-1 and CaM, and are required for effective activation of NHE-1. Additionally, because isotonic substitution of sodium with TMA more effectively attenuates EGF-stimulated ECAR than does MIA, it is possible that there is another sodium-dependent proton-efflux pathway that is insensitive to 5 μM MIA. The possibility is the subject of future work.

What is the significance of our findings to podocyte biology? Although the significance of EGF and/or NHE-1 in podocyte biology is not known, we speculate that NHE-1 could participate in the regulation of the cytoskeleton of podocytes, as NHE-1 is indirectly tethered to, and regulates, the actin cytoskeleton of fibroblasts [24]. NHE-1 is intimately linked to cytoskeletal regulatory proteins such as Rho, and NHE-1 can regulate cytoskeletal architecture through both ion channel regulation and protein-protein interaction [24]. Inasmuch as the structural integrity of the cytoskeleton of podocytes is critical for maintaining the podocyte foot processes and the glomerular slit diaphragm, key cytoskeletal regulatory proteins like NHE-1 clearly could play key roles in maintaining or regulating glomerular architecture and protein permeability. Further work would be necessary to test this possibility. NHE-1 also has been implicated in cellular proliferation and apoptosis [25], so it could also play complex roles in podocyte physiology and pathophysiology. EGF is a mitogen and cell survival factor that also regulates regenerative hyperplasia [2]. Thus, it could regulate important podocyte functions independently of, or in concert with NHE-1.

We conclude that EGF stimulates NHE-1 activity in podocytes through two pathways, each of which is required for significant activation to occur (Figure 7). These pathways converge upon CaM, being essential for its physical engagement with NHE-1. The first can be depicted as follows: EGF → EGFR → Jak2 activation (independent of EGFR tyrosine kinase activity) → tyrosine phosphorylation of CaM → CaM binding to NHE-1 → activation of NHE-1; and the second pathway as follows: EGF → EGFR →EGFR tyrosine kinase activation → association of CaM to NHE-1 (independent of Jak2) → activation of NHE-1 (Figure 7).