Lysophosphatidic acid 5 receptor induces activation of Na⁺/H⁺ exchanger 3 via apical epidermal growth factor receptor in intestinal epithelial cells

Abstract

Na⁺ absorption is a vital process present in all living organisms. We have reported previously that lysophosphatidic acid (LPA) acutely stimulates Na⁺ and fluid absorption in human intestinal epithelial cells and mouse intestine by stimulation of Na⁺/H⁺ exchanger 3 (NHE3) via LPA₅ receptor. In the current study, we investigated the mechanism of NHE3 activation by LPA₅ in Caco-2bbe cells. LPA₅-dependent activation of NHE3 was blocked by mitogen-activated protein kinase kinase (MEK) inhibitor PD98059 and U0126, but not by phosphatidylinositol 3-kinase inhibitor LY294002 or phospholipase C-β inhibitor U73122. We found that LPA₅ transactivated the epidermal growth factor receptor (EGFR) and that inhibition of EGFR blocked LPA₅-dependent activation of NHE3, suggesting an obligatory role of EGFR in the NHE3 regulation. Confocal immunofluorescence and surface biotinylation analyses showed that LPA₅ was located mostly in the apical membrane. EGFR, on the other hand, showed higher expression in the basolateral membrane. However, inhibition of apical EGFR, but not basolateral EGFR, abrogated LPA-induced regulation of MEK and NHE3, indicating that LPA₅ selectively activates apical EGFR. Furthermore, transactivation of EGFR independently activated the MEK-ERK pathway and proline-rich tyrosine kinase 2 (Pyk2). Similarly to MEK inhibition, knockdown of Pyk2 blocked activation of NHE3 by LPA. Furthermore, we showed that RhoA and Rho-associated kinase (ROCK) are involved in activation of Pyk2. Interestingly, LPA₅ did not directly activate RhoA but was required for transactivation of EGFR. Together, these results unveil a pivotal role of apical EGFR in NHE3 regulation by LPA and show that the RhoA-ROCK-Pyk2 and MEK-ERK pathways converge onto NHE3.

na⁺ absorption is a vital process present in all organisms from single cell bacteria to multicellular organisms. The Na⁺/H⁺ exchanger 3, NHE3 (Slc9a3), is highly expressed in the apical membrane of small intestine, colon, and proximal tubules of kidney, where it constitutes the major Na⁺-absorbing process (13, 44). NHE3 is regulated by hormones and growth factors through multiple mechanisms that include trafficking, phosphorylation, transcriptional regulation, and interaction with scaffolding and cytoskeletal proteins (13, 44). Lysophosphatidic acid (LPA) is a small, bioactive glycerophospholipid that acts as an extracellular signaling molecule in most eukaryotic tissues. Five cell surface LPA receptors and three putative LPA receptors that belong to the G protein-coupled receptor (GPCR) superfamily have been identified: LPA₁–LPA₈ (5). The relevance of LPA to various pathologic conditions, such as inflammation, cancer, obesity, and hypertension, is widely investigated. More recently, LPA has been implicated in electrolyte transport in epithelial cells. LPA stimulates NHE3 activity in opossum kidney (OK) cells (23). Inhibition of adenylate cyclase by LPA₂ signaling is necessary for the inhibition of the cystic fibrosis transmembrane conductance regulator (CFTR)-dependent Cl⁻ secretion in the mouse intestine (26). In addition, LPA₂ signaling stimulates Cl⁻/OH⁻ exchange in Caco-2 cells (36). However, unlike the regulation of CFTR by LPA, the absence of LPA₂ did not appreciably affect NHE3-dependent Na⁺ absorption (29). Previously, it was shown that LPA₅ is highly expressed in the intestine (24), and quantitative real-time PCR (qRT-PCR) of isolated intestinal epithelial cells shows that LPA receptor mRNA is expressed in the order of LPA₁>LPA₅>>LPA₂, LPA_3, LPA₄ (28, 29). However, in Caco-2 and the colonal Caco-2bbe cells, LPA₅ expression is low (29) and unless LPA₅ is exogenously expressed, LPA is not able to stimulate NHE3 activity in Caco-2 cells (29). The stimulation of NHE3 activity by LPA₅ acutely occurs within 3–5 min of treatment with low concentrations of LPA, and it requires the presence of Na⁺/H⁺ exchanger regulatory factor 2 (NHERF2), which interacts with the carboxyl terminal motif of LPA₅ through its second PSD95/Dlg/ZO-1 (PDZ) domain. NHERF2 also interacts with LPA₂ via the similar mechanism, and the absence of NHERF2 obliterates the effects of LPA on CFTR and NHE3 (26, 29, 35). In this study, we investigate the mechanism whereby LPA₅ regulates NHE3. We report that LPA₅ stimulates NHE3 activity through transactivation of the epidermal growth factor receptor (EGFR) in the apical membrane of Caco-2bbe cells and that this regulation involves proline-rich tyrosine kinase 2 (Pyk2) and mitogen-activated protein kinase kinase (MEK).¹

EXPERIMENTAL PROCEDURES

Cell cultures.

Caco-2bbe (C2b) cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate, 50 U/ml penicillin, 50 μg/ml streptomycin, and 1× nonessential amino acids. C2b cells stably transfected with human NHE3 with a COOH-terminal vesicular stomatitis virus glycoprotein (VSVG) epitope tag, C2b/E3V, were previously described (29). LPA₅ was expressed by infecting with lentivirus harboring NH₂-terminal hemagglutinin (HA)-tagged human LPA₅, and cells were selected with 10 μg/ml puromycin. However, the cells maintain LPA₅ expression only for nine passages and all the experiments were carried out with cells of less than five passages following the initial infection with virus. The expression of LPA₅ was confirmed by Western immunoblot analysis using an anti-HA antibody on the cells of the same passage that were used for each experiment.

Chemicals and materials.

LPA (18:1, 1-oleoyl-2-hydroxy-sn-glycero-3-phosphate) was purchased from Avanti Polar Lipids (Alabaster, AL) and prepared in PBS, pH 7.2, containing 0.1% BSA. For all experiments, LPA was used at the final concentration of 1 μM for 2–3 min unless otherwise specified and the equal volume of PBS containing 0.1% BSA was added as a control. When needed, cells were pretreated with LY294002 (50 μM), U73122 (5 μM), PD98059 (20 μM), U0126 (10 μM), AG1478 (1 μM), PD153035 (1 μM), Y27632 (50 μM), or C3 (5 μg/ml), and the equal volume of DMSO was added as a vehicle control. All the antibodies were obtained from Cell Signaling (Danvers, MA), Calbiochem (La Jolla, CA), Covance (Emeryville, CA), or Santa Cruz (Santa Cruz, CA). Recombinant human epidermal growth factor (EGF) was obtained from Invitrogen (Carlsbad, CA). All other chemicals were obtained from Sigma (St. Louis, MO) or Calbiochem.

Gene silencing.

Lentiviral vectors (pLKO.1) containing short hairpin RNA (shRNA) targeting Pyk2 or EGFR were obtained from Sigma or Open Biosystems (Huntsville, AL), respectively. Silencing of Pyk2 or EGFR expression was performed using shRNA lentiviral particles. pLKO.1-puro was used to generate control lentivirus.

Na⁺-dependent intracellular pH recovery.

The Na⁺-dependent changes in intracellular pH (pH_i) by NHE3 was determined using the ratio-fluorometric, pH-sensitive dye 2′,7′-bis-(2-carboxyethyl)-5-carboxyfluorescein acetoxymethyl ester (BCECF-AM) as previously described (39). Briefly, cells grown on Transwells (BD Falcon, Franklin Lakes, NJ) for 5–7 days postconfluence were serum starved overnight and the next morning were washed in Na⁺ buffer [130 mM NaCl, 20 mM HEPES, 5 mM KCl, 1 mM tetramethylammonium-PO₄ (TMA-PO₄), 2 mM CaCl₂, 1 mM MgSO₄, and 25 mM glucose] and then dye-loaded by incubating for 20 min with 6.5 μM BCECF-AM in the same solution. When inhibitors were used, the equal volume of an inhibitor or DMSO was added and incubated at room temperature during the dye-loading period. The cells were mounted on a perfusion chamber, placed on an inverted microscope, and superfused with NH₄⁺ buffer (50 mM NH₄Cl, 80 mM TMA-Cl, 20 mM HEPES, 5 mM KCl, 1 mM TMA-PO₄, 2 mM CaCl₂, 1 mM MgSO₄, and 25 mM glucose) and subsequently with TMA⁺ buffer (130 mM TMA-Cl, 20 mM HEPES, 5 mM KCl, 1 mM TMA-PO₄, 2 mM CaCl₂, 1 mM MgSO₄, and 25 mM glucose) containing 1 μM LPA or 0.1% BSA as a control. Hence, cells were preexposed to LPA or control BSA for 2–3 min before perfusion with a Na⁺ buffer that drives Na⁺-dependent pH recovery. Na⁺ buffer was supplemented with 50 μM Hoe 694 to inhibit NHE1 and NHE2 activities (29). Calibration of the fluorescence signal was performed using the K⁺/H⁺ ionophore nigericin as described previously (39). The microfluorometry was performed on a Nikon TE200 inverted microscope with a Nikon CFI Super Fluor ×40 objective, coupled to a Lambda 10–2 filter wheel controller equipped with a multiwavelength filter set designed for BCECF. Photometric data were acquired using Metafluor software (Molecular Devices, Sunnyvale, CA). Na⁺/H⁺ exchange rate was described by the rate of pH_i recovery, which was calculated by determining slopes along the pH_i recovery by linear least-squares analysis over a minimum of 9 s.

Immunoprecipitation and Western blot analysis.

Immunoprecipitation was performed as previously described (14). Briefly, C2b/E3V/LPA₅ cells were washed twice in cold PBS, scraped, and lysed in lysis buffer (Cell Signaling) containing 20 mM Tris·HCl (pH 7.5), 150 mM NaCl, 1 mM β-glycerophosphate, 2.5 mM sodium pyrophosphate, 1 mM Na₂EDTA, 1 mM EGTA, 1 mM Na₃VO₄, 1 μg/ml leupeptin, 1% Triton X-100, and protease inhibitor cocktail tablets (Roche, Indianapolis, IN). The crude lysate was sonicated for 2 × 15 s and spun at 14,000 g for 15 min. Protein concentration was determined by the bicinchoninic acid assay (Sigma). Lysate (500 μg) was precleared by incubation with 30 μl of protein A-Sepharose beads for 1 h, and the supernatant was then incubated overnight with anti-EGFR antibody. Immunocomplex was purified by incubating with 50 μl of protein A-Sepharose beads for 1.5 h, followed by three washes in lysis buffer and two washes in PBS. All the above steps were performed at 4°C or on ice. The bound immunocomplex was eluted by incubating the protein A beads in Laemmli sample buffer for 10 min at 95°C. The proteins were then separated by SDS-PAGE and transferred to a nitrocellulose membrane for Western immunoblotting, as previously described (39).

Confocal immunofluorescence microscopy.

C2b/E3V/LPA₅ cells grown 7 days postconfluence on Transwells were washed twice with cold PBS, fixed in 4% paraformaldehyde in PBS for 10 min at room temperature, permeabilized in 0.2% Triton X-100 in PBS for 5 min, and blocked in PBS containing 5% normal goat serum for 30 min at room temperature. Cells were then stained with anti-EGFR, anti-HA, or anti-villin antibodies for 1 h at room temperature. Following three washes, 10 min each, with PBS, the cells were incubated with Alexa 488-conjugated donkey anti-mouse IgG or Alexa 555-conjugated goat anti-rabbit IgG (Invitrogen) for 1 h at room temperature. After 3 × 10-min washes with PBS, the excised Transwells were mounted with ProLong Gold Antifade Reagent (Invitrogen) and observed under a Zeiss LSM510 laser confocal microscope (Zeiss Microimaging, Thornwood, NY) coupled to a Zeiss Axioplan2e with ×63 Pan-Apochromat oil lenses.

Surface biotinylation.

Surface biotinylation of LPA₅, NHE3, EGFR, and Na⁺-K⁺-ATPase was performed as previously described (14). Briefly, cells grown on permeable filters were treated with 1 μM LPA or 0.1% BSA vehicle for 5 min, then were rinsed twice in PBS and incubated for 10 min in borate buffer composed of 154 mM NaCl, 7.2 mM KCl, 1.8 mM CaCl₂, and 10 mM H₃BO₃, pH 9.0. Cells were then incubated for 40 min with 0.5 mg/ml NHS-SS-biotin (Pierce, Rockford, IL) in borate buffer. Unbound NHS-SS-biotin was quenched with Tris buffer (20 mM Tris, 120 mM NaCl, pH 7.4). Cells were then rinsed with PBS, scraped, lysed in the lysis buffer described above, and sonicated for 2 × 15 s. The lysate was agitated for 30 min and spun at 14,000 g for 15 min to remove the insoluble cell debris. Protein concentration was determined, and 1 mg of lysate was then incubated with streptavidin-agarose beads (Pierce) for 2 h. The streptavidin-agarose beads were washed three times in lysis buffer and twice in PBS. All the above procedures were performed at 4°C or on ice. Biotinylated surface proteins were then eluted by boiling the beads at 95°C for 10 min. Dilutions of the total and surface LPA₅, NHE3, EGFR, and Na⁺-K⁺-ATPase were resolved by SDS-PAGE, and immunoblotted with anti-HA antibody, anti-VSVG antibody, anti-EGFR antibody, and anti-α₁-subunit of Na⁺-K⁺-ATPase antibody, respectively. Densitometric analysis was performed using Scion Image software (National Institutes of Health, Bethesda, MD).

RhoA activation assay.

Activation of RhoA was determined by a modified method described by Zhang et al. (45). Cells were seeded on 60-mm culture dishes. After serum starvation for 24 h, cells were activated with LPA and lysed in ice-cold lysis/binding buffer (25 mM Tris·HCl, pH 7.5, 150 mM NaCl, 5 mM MgCl₂, 1% NP-40, 1 mM DTT, 10% glycerol, and protease inhibitors). Five hundred micrograms of lysate was incubated for 1 h with glutathione-sepharose beads (GE Healthcare, Uppsala, Sweden) coupled with the Rho-binding domain of Rhotekin (GST-RBD). The construct to express GST-RBD was kindly provided by Dr. Martin Schwartz (University of Virginia, Charlottesville). Beads were washed three times with lysis/binding buffer, and the bound RhoA proteins were eluted with Laemmli sample buffer and subjected to Western blotting using an anti-RhoA antibody.

Statistical analysis.

Results are presented as means ± SE. Statistical analyses were performed by Student's t-test for paired comparison. P < 0.05 was considered significant.

RESULTS

LPA-induced NHE3 activation in C2b/E3V/LPA₅ cells is MEK dependent.

We have shown previously that LPA does not regulate NHE3 in Caco-2 and C2b cells due to low LPA₅ expression in these cells (29). To circumvent this problem, we have used C2b cells expressing hNHE3V (29). Consistent with the previous study, transfection of HA-LPA₅ in C2b/E3V cells resulted in a significant increase in NHE3 activity in response to 1 μM LPA, but not in control transfected cells (Fig. 1A). Major LPA signaling pathways include the phosphatidylinositol 3-kinase (PI3K)-Akt, phospholipase C (PLC)-β-Ca²⁺, and MEK-ERK pathways (32). Among them, PI3K is involved in LPA-induced stimulation of NHE3 in OK cells (23) and thus we first assessed the effect of PI3K inhibitor LY294002 on NHE3 activation by LPA. However, LPA-induced NHE3 activation in C2b/E3V/LPA₅ cells was not affected by the presence of the PI3K inhibitor LY294002 (Fig. 1B), suggesting that PI3K does not play a key role in LPA₅-induced activation of NHE3. Similarly, the PLC-β inhibitor U73122 did not block activation of NHE3 by LPA (Fig. 1B). On the other hand, the MEK inhibitor PD98059 and U0126 did not affect the basal NHE3 activity, but completely abrogated LPA-mediated activation of NHE3 (Fig. 1C). As shown previously (29), LPA increased NHE3 protein expression in the apical membrane, which was blocked by MEK inhibitors (Fig. 1D).

Fig. 1.

Lysophosphatidic acid 5 receptor (LPA₅)-dependent Na⁺/H⁺ exchanger 3 (NHE3) activation in Caco-2bbe/human (h)NHE3/LPA₅ cells is dependent on ERK. A: Caco-2bbe/hNHE3 (C2b/E3V) cells were transfected with lentiviral hemagglutinin (HA)-LPA₅ or control virus derived from pCDH-puro vector. Serum-starved C2b/E3V/LPA₅ or C2b/E3V/pCDH cells were loaded with BCECF, and NHE3 activities were determined as the rate of Na⁺-dependent pH recovery, ΔpH/min, in the presence or absence of 1 μM LPA for 3 min. For all experiments, cells were treated with LPA for 2–3 min. n ≥ 12; *P < 0.05. NS, not significant. B: NHE3 activity in C2b/E3V/LPA₅ cells treated with or without LPA was determined in the presence of LY294002, U73122, or DMSO for 30 min. C: NHE3 activity in C2b/E3V/LPA₅ cells treated with or without LPA was determined in the presence of PD98059, U0126, or DMSO. D: surface NHE3 protein abundance was assessed by surface biotinylation to determine the effects of MEK inhibitors (preincubation for 30 min) on surface NHE3 expression in C2b/E3V/LPA₅ cells as described in experimental procedures. Total NHE3 and β-actin were used as loading controls. n = 3.

NHE3 regulation by LPA₅ is mediated via EGFR.

We have demonstrated previously that LPA phosphorylates ERK in Caco-2 cells via LPA₂ receptor (43). To ensure that LPA₅ signaling activates MEK in C2b/E3V cells, we determined phosphorylation of ERK as readout. Figure 2A shows that LPA induced a modest increase in the phosphorylation level of ERK in C2b/E3V cells, but the effect was significantly greater when LPA₅ was expressed, suggesting that LPA-induced phosphorylation of ERK is primarily mediated via LPA₅ in C2b/E3V/LPA₅ cells. It is widely known that the receptor tyrosine kinases are transactivated by GPCRs, including LPA receptors (40). To determine whether EGFR transactivation is involved in LPA-induced ERK activation, C2b/E3V and C2b/E3V/LPA₅ cells were pretreated with the EGFR inhibitors AG1478 or PD153035. As shown in Fig. 2B, both EGFR inhibitors blocked LPA-induced ERK phosphorylation. Consistent with the effects on ERK phosphorylation, AG1478 completely blocked LPA-mediated activation of NHE3, suggesting that EGFR is necessary for regulation of NHE3 by LPA₅ (Fig. 2C). We confirmed the role of EGFR in LPA₅-induced NHE3 regulation by knockdown of EGFR (Fig. 2D). EGFR knockdown in C2b/E3V/LPA₅ cells decreased LPA-induced NHE3 activation (Fig. 2E), confirming the involvement of EGFR in this regulation. In addition, knockdown of EGFR blocked the increase in surface NHE3 abundance by LPA (Fig. 2F). Together, our data show that the presence of EGFR is obligatory for the activation of NHE3 by LPA₅.

Fig. 2.

LPA₅-induced NHE3 activation is mediated by EGF receptor (EGFR) transactivation. A: phosphorylation of ERK was determined in C2b/E3V/pCDH and C2b/E3V/LPA₅ cells treated with LPA (2 min). p-ERK, phosphorylated ERK. β-Actin was used as a loading control. B: phosphorylation of ERK by LPA in C2b/E3V/pCDH and C2b/E3V/LPA₅ cells was determined in the presence or absence of of preincubation with EGFR inhibitors AG1478 or PD153035 for 30 min. C: NHE3 activity in C2b/E3V/LPA₅ cells treated with or without LPA (3 min) in the presence or absence of AG1478 determined as described in experimental procedures. n ≥ 12; *P < 0.05. D: knockdown of EGFR expression in C2b/E3V/LPA₅ cells by short hairpin (sh)EGFR lentivirus is shown. EGFR expression level was decreased by 70% in shEGFR-infected cells compared with control cells. E: stimulation of NHE3 activity by LPA in C2b/E3V/LPA₅ cells with EGFR knockdown was determined. n ≥ 12; *P < 0.05. F: the effect of EGFR knockdown on NHE3 protein expression in the apical membrane was determined by surface biotinylation. Total NHE3 and β-actin were used as the loading control, respectively. n = 3.

LPA transactivates EGFR in the apical membrane.

To determine the cellular locale of LPA₅ and EGFR, confocal immunofluorescence microscopic analysis was performed in Triton X-100-permeabilized C2b/E3V/LPA₅ cells. Figure 3 Aa shows the presence of LPA₅ (green) in the apical membrane and the subapical region. HA-LPA₅ staining at the basolateral membrane was low. Villin (Fig. 3Ab, red) was labeled to outline the apical side of the cell. In contrast to LPA₅, EGFR labeling was evident in both apical and basolateral membrane, although the basolateral staining was more prominent (Fig. 3, Ba and Bb, green). To ensure the apical expression of LPA₅ and EGFR, immunostaining was performed on nonpermeabilized cells, and anti-HA antibody or m528 anti-EGFR antibody that binds to the HA epitope of HA-LPA₅ or an extracellular domain of EGFR, respectively, was added only on the apical side of the cells. The data shown in Fig. 3, Ac and Bc, confirm the apical expression of LPA₅ and EGFR. The polarized expression of LPA₅ and EGFR was confirmed by surface biotinylation of C2b/E3V/LPA₅ cells (Fig. 3C). Western blot analysis of the apical and basolateral fractions showed that LPA₅ expression was more prominent in the apical membrane in which NHE3 was present. EGFR, on the other hand, was highly expressed in the basolateral membrane. However, a weaker band was detected in the apical fraction, confirming the apical expression of EGFR seen in Fig. 3B.

Fig. 3.

LPA transactivates EGFR in the apical (AP) membrane. A: C2b/E3V/LPA₅ cells grown on Transwells were fixed with (a and b) or without (c) permeabilization and stained with anti-HA antibodies (green) or anti-villin antibodies (red). Anti-villin antibodies (red) were used to mark the apical side of the cells. Vertical (x-z) sectional views perpendicular to the confocal focal plane are shown. BL, basolateral. B: C2b/E3V/LPA₅ cells on Transwells were fixed with (a and b) or without (c) permeabilization. Cells were labeled with anti-EGFR monoclonal antibodies with an epitope in the cytoplasmic domain (a and b, green) or m528 anti-EGFR antibody that binds to an extracellular domain of EGFR (c, green). Vertical (x-z) sectional views are shown. C: surface biotinylation was performed to determine the polarity of LPA₅ and EGFR expression in C2b/E3V/LPA₅ cells as described in experimental procedures. Western blot shows protein expression in the apical and basolateral fractions. NHE3 and Na⁺-K⁺-ATPase were used as the apical and basolateral marker, respectively. n = 3. D: C2b/E3V/LPA₅ cells on Transwells were treated with LPA (2 min) from either the apical or the basolateral side of the cells. To inhibit EGFR, m528 anti-EGFR antibody was applied from the apical or basolateral side and cells were gently shaken for 16 h before LPA treatment. The expression levels of phospho-ERK and total ERK were determined. Results are representative of three independent experiments. E: apical EGFR was blocked by apical application of m528 or control antibodies. NHE3 activity was determined in the presence or absence of LPA (3 min). n ≥ 8; *P < 0.05. F: basolateral EGFR was blocked by m528 or control antibodies. NHE3 activities in the presence or absence of LPA are shown. n ≥ 8; *P < 0.05. G: NHE3 activity was determined in C2b/E3V/LPA₅ cells treated with 1 μM LPA, 10 ng/ml EGF, or LPA and EGF together, for 3 min. n ≥ 12; *P < 0.05; **P < 0.01.

To determine the polarity of EGFR transactivation by LPA, cells were treated with LPA from either apical or basolateral side of the cells that were grown in Transwells (Fig. 3D). Apical LPA increased the phosphorylation level of ERK, but basolaterally applied LPA showed no effect, consistent with the apical expression of LPA₅ shown earlier (Fig. 3, A and C). To determine whether apical or basolateral EGFR is transactivated by LPA, EGFR was blocked using neutralizing anti-EGFR antibody m528 from either the apical or the basolateral compartment (31). Inhibition of apical EGFR blocked ERK phosphorylation induced by apical LPA. However, inhibition of basolateral EGFR failed to block ERK phosphorylation. Consistent with the effects on ERK phosphorylation, inhibition of apical EGFR by m528 blocked stimulation of NHE3 activity by LPA in C2b/E3V/LPA₅ cells (Fig. 3E). On the contrary, inhibition of basolateral EGFR did not affect NHE3 activation (Fig. 3F).

It was shown previously that activation of EGFR in the basolateral membrane of enterocytes activates NHE3 in Caco-2 cells and rabbit ileum (16, 19). To examine whether the apical EGFR also regulates NHE3, NHE3 activity was determined following apical addition of EGF. Figure 3G shows that apical EGF activated NHE3 activity, although its effect was small compared with the effect by LPA. Interestingly, when LPA and EGF were added concomitantly, we observed enhanced stimulation of NHE3 in an additive manner. These studies indicate that LPA₅, which is predominantly expressed in the apical membrane of C2b cells, transactivates apical EGFR, but not basolateral EGFR. However, transactivation of EGFR by LPA₅ does not saturate EGFR in the apical membrane.

LPA-induced NHE3 activation in C2b/E3V/LPA₅ cells is Pyk2 dependent.

It was shown that Pyk2 is activated by LPA (1) and that Pyk2 is necessary for regulation of NHE3 by acidification (27, 38). We examined whether LPA activates Pyk2 by determining phosphorylation of Pyk2 at Tyr402 (25). LPA phosphorylated Pyk2 in C2b/E3V/LPA₅ cells, but not in control transfected C2b/E3V cells (Fig. 4A). We next investigated whether Pyk2 plays a role in LPA-induced NHE3 regulation by knockdown of Pyk2 (Fig. 4B). Pyk2 knockdown in C2b/E3V/LPA₅ cells attenuated LPA-induced NHE3 activation (Fig. 4C), demonstrating the importance of Pyk2 in this regulation. Consistent with MEK and EGFR inhibition, Pyk2 knockdown blocked the increase in NHE3 apical abundance by LPA (Fig. 4D).

Fig. 4.

Proline-rich tyrosine kinase 2 (Pyk2) is involved in LPA₅-dependent NHE3 activation. A: phosphorylation of Pyk2 (p-Pyk2) was determined in C2b/E3V/pCDH and C2b/E3V/LPA₅ cells. The expression levels of Pyk2 and β-actin are shown as loading controls. B: C2b/E3V/LPA₅ cells were treated with shControl or shPyk2. Western blot shows that Pyk2 expression level was decreased by 60% by shPyk2. C: NHE3 activity in C2b/E3V/LPA₅ treated with shControl or shPyk2 was determined in the presence or absence of LPA (3 min). n ≥ 12; *P < 0.05. D: the effect of Pyk2 knockdown on surface NHE3 protein abundance was determined by surface biotinylation. Total NHE3 and β-actin were used as the loading control, respectively. n = 3.

LPA₅-dependent transactivation of EGFR independently activates ERK and Pyk2.

A previous study using Pyk2-deficient fibroblasts showed that Pyk2 is necessary for activation of EGFR by LPA (1). However, in our study, knockdown of Pyk2 did not affect phosphorylation of EGFR, indicating that Pyk2 is not involved in EGFR activation by LPA₅ in C2b/E3V/LPA₅ cells (Fig. 5A). On the other hand, inhibition of EGFR by AG1478 completely blocked phosphorylation of Pyk2, indicating that Pyk2 is regulated by EGFR (Fig. 5B, lane 3). We next examined whether Pyk2 is regulated by MEK or vice versa. Inhibition of MEK by PD98059 did not diminish Pyk2 phosphorylation by LPA (Fig. 5B, lane 4). To our surprise, ERK phosphorylation was not altered by Pyk2 knockdown (Fig. 5C), suggesting that Pyk2 and MEK do not influence each other.

Fig. 5.

Pyk2 and ERK are independently activated by LPA₅. A: phosphorylation of EGFR was assessed by determining Tyr phosphorylation levels of immunoprecipitated EGFR in C2b/E3V/LPA₅ cells treated with shControl or shPyk2. B: the effects of AG1478 or PD98059 on Pyk2 or ERK phosphorylation were determined by Western blotting. Results are representative of three independent experiments. C: phosphorylation of ERK in cells treated with shControl or shPyk2 was determined by Western blotting. Results are representative of three independent experiments. D: activation of RhoA by LPA in C2b/E3V/LPA₅ cells in the presence or absence of AG1478 was determined by Rhotekin pulldown. Activated RhoA bound to glutathione S-transferase (GST)-Rhotekin at different time intervals was resolved by SDS-PAGE, transferred, and blotted using an anti-RhoA antibody to determine the extent of RhoA activation. Total RhoA was determined as a loading control. n = 3. E: activation of RhoA by EGF was determined as described above. n = 3. F and G: C2b/E3V/LPA₅ cells were treated with LPA in the presence or absence of C3 toxin (pretreated for 16 h, F) or Y27632 (pretreated for 2 h, G). Phosphorylated and total Pyk2 and ERK were determined. n = 3.

LPA activates Pyk2 via Rho GTPase in IEC-18 intestinal epithelial cells (41). To determine whether LPA activates RhoA in C2b cells, we determined RhoA activation by using GST-RBD. LPA rapidly activated RhoA as shown in Fig. 5D, left. Surprisingly, the presence of AG1478 ablated RhoA activation (Fig. 5D), suggesting that RhoA activation is dependent on EGFR. To ensure that EGFR is capable of activating RhoA, Rhotekin pulldown assay was performed in cells treated with EGF. Figure 5E shows that activation of EGFR by EGF resulted in activation of RhoA, demonstrating that EGFR can directly activate RhoA in C2b cells. Next, we determined whether LPA₅ signaling activates Pyk2 via a Rho GTPase-dependent pathway. Inhibition of RhoA with the C3 exoenzyme blocked phosphorylation of Pyk2 but not ERK (Fig. 5F). Consistently, Y27632, a Rho-associated kinase (ROCK) inhibitor, inhibited phosphorylation of Pyk2, but not phosphorylation of ERK (Fig. 5G). These results suggest that LPA activates Pyk2 through the EGFR-RhoA-ROCK pathway and that Pyk2 and MEK are independently regulated by LPA₅ via EGFR.

DISCUSSION

Recent studies have shown that LPA has stimulatory effects on fluid absorption in the intestine (29). Using cultured C2b cells, we showed previously that LPA stimulates Na⁺ absorption by NHE3 when LPA₅ is transfected in these cells (29). In addition, our in vivo data using LPA₂-null or NHERF2-null mice have been consistent with the notion that LPA₅ is the main LPA receptor in the intestine mediating Na⁺ and fluid absorption (29). In the present study, we extended our study by investigating the mechanism of LPA₅-elicited stimulation of NHE3. We found that EGFR plays a central role in LPA₅-dependent regulation of NHE3 in C2b cells. Several receptor tyrosine kinases, including members of the EGFR family, are rapidly transactivated following GPCR stimulation. The mechanism of EGFR transactivation has been a subject of many studies, and multiple mechanisms have been proposed (40). EGFR is generally thought to be located in the basolateral membrane of polarized epithelial cells (3, 34). Previous studies have shown the role of the basolateral EGFR in regulation of electrolyte transport in intestinal epithelial cells (17, 19). However, it is becoming more evident that EGFR is found at the apical membrane; EGFR is found at the luminal surface of enterocytes in the sucking rat ileum and pig intestine as well as gastric mucosal oxyntic cells (9, 18). Although the functions of apical EGFR are poorly characterized, activation of apical EGFR decreases paracellular permeability of oxyntic mucosal cells and increases tolerance to apical acid (4). Interestingly, overexpression of EGFR in LLC-PK₁ cell sorts a fraction of EGFR to the apical membrane where it transduces cellular signaling to enhance cell proliferation (21). Despite the detection of EGFR in both apical and basolateral membranes of C2b cells, the basolateral expression was much higher. On the other hand, LPA₅ showed an opposite expression profile with a prevalence for the apical expression and the application of LPA to the apical but not the basolateral side of C2b cells activated ERK. Importantly, LPA₅ transactivated EGFR in the apical membrane but not the basolateral EGFR. The pairing up of LPA₅ with apically located EGFR probably makes it spatially more efficient for rapid signal transduction, given their close proximity. We showed previously that LPA₅ interacts with NHERF2 through its carboxyl terminal motif that specially binds to the second PDZ domain of NHERF2 (29). Interestingly, the first PDZ domain of NHERF1 interacts with EGFR to stabilize the receptor expression at the cell surface (22). Our unpublished data indicate that NHERF2 also interacts with EGFR, which is not unexpected given the high degree of similarity between NHERF1 and NHERF2. In intestinal epithelial cells, NHERF2 is predominantly expressed in the apical or near the apical membrane, and hence it remains an intriguing possibility that NHERF2 tethers LPA₅ and EGFR together.

Transactivation of EGFR often involves Src and ectodomain shedding of a transmembrane precursor of EGFR ligand (10, 40). Consistently, we found that inhibition of Src and metalloproteinases by PP2 and TAPI-1, respectively, blocked auto-phosphorylation of EGFR induced by LPA (data not shown). In addition to Src, Pyk2 has been implicated in linking GPCRs to EGFR, although other studies showed that Pyk2 is recruited by transactivated EGFR (1, 33). In addition, Pyk2 is involved in acid-induced activation of NHE3 and Src (27). In the current study, knockdown of Pyk2 blocked LPA₅-dependent activation of NHE3. However, knockdown of Pyk2 did not affect LPA-induced phosphorylation of EGFR. Instead, inhibition of EGFR ablated phosphorylation of Pyk2, indicating that Pyk2 lies downstream of EGFR. It was shown that LPA₅ couples with G_12/13 to activate RhoA and ROCK in B103 neuroblastoma cells (24). Consistent with this earlier study, LPA rapidly activated RhoA in C2b/E3V cells, and C3 transferase or Y27632 attenuated LPA-induced Pyk2 phosphorylation, indicating that LPA₅ activates Pyk2 via the Rho-ROCK pathway. It is generally assumed that GPCRs directly activate RhoA via G proteins. However, we found that the presence of EGFR was necessary for the activation of RhoA in C2b cells, suggesting that LPA activates RhoA via transactivation of EGFR. This unexpected finding was affirmed by demonstrating that activation of RhoA by EGF implying that RhoA is a downstream effect of EGFR. In agreement with our current findings, LPA activates Rho in Swiss 3T3 cells via EGFR (8).

LPA has been described to induce exocytic trafficking of NHE3 to the apical membrane in a NHERF2 dependent manner (29). This regulation is acute occuring within 2–3 min of LPA treatment. We found that increased NHE3 expression in the apical membrane of C2b cells is maintained for at least 15 min after removal of LPA (data not shown). However, whether LPA specifically alters the rate of NHE3 removal from the apcial membrane is not known. In the current study, we found that LPA₅-induced activation of Pyk2 and ERK is associated with increased NHE3 abundance at the apical surface. In addition, we found that phosphorylation of Pyk2 was not blocked by MEK inhibition, indicating that Pyk2 and ERK are activated via two distinct pathways. Based on these findings, we propose a model that LPA₅ signaling transactivates EGFR, which then activates the RhoA-ROCK-Pyk2 and Ras-MEK-ERK pathways (Fig. 6). However, it remains unclear how ERK or Pyk2 regulates NHE3 activity and surface abundance in response to LPA. Regulation of NHE3 activity is dependent on the actin cytoskeleton, and RhoA/ROCK stimulates NHE3 activity by formation of stress fiber that increases NHE3 apical membrane abundance (20, 42). Pyk2 is a member of the focal adhesion kinase (FAK) family, and it regulates actin cytoskeletal organization in combination with FAK (6). Thus, it seems feasible that Pyk2 inhibition may contribute to disruption of the actin cytoskeleton, resulting in decreased NHE3 expression in the apical membrane and hence decreased NHE3 activity. The Ras-MEK-ERK pathway has been the subject of intense study because mutations of the signaling components of this pathway are a hallmark of several human diseases (11). ERK has also been implicated to play a role in regulation of membrane protein trafficking by phosphorylation (7, 15). The p90 kDa ribosomal S6 kinases (RSKs) are a family of Ser/Thr kinases that lie downstream of ERK (2). Sustained acidification activates NHE1 in AP-1 cells via an ERK-dependent pathway, and RSK phosphorylates NHE1 in response to serum (30, 37). Recently, starfish NHE, which is a homolog of mammalian NHE3, was shown to be phosphorylated by RSK at Ser-590, -606, and -673 in the COOH terminus of starfish NHE (12). Given the evolutionary link between starfish NHE and mammalian NHE3, it is possible that NHE3, which contains several R-X-X-pS/T motifs, is also phosphorylated by RSKs. Probable points of cellular communication by Pyk2 and ERK in stress fiber formation and NHE3 phosphorylation, respectively, are in conceptual agreement with our present data that either Pyk2 knockdown or MEK inhibition similarly blocks activation of NHE3 by LPA without a sign of redundancy.

Fig. 6.

Putative model depicting the mechanism of NHE3 regulation by LPA₅-mediated signaling. LPA₅ transactivates EGFR in the apical membrane, which activates the RhoA-RhoA-associated kinase (ROCK)-Pyk2 and the Ras-MEK-ERK pathways that converge onto NHE3. HB, heparin-binding; MMP, matrix metalloproteinases; NHERF, Na⁺/H⁺ exchanger regulatory factor 2.

In summary, we report here that LPA₅ transactivates apical EGFR, followed by two independent activations of ERK and Pyk2. This work highlights the importance of apical EGFR in regulation of NHE3 that EGFR in the apical membrane plays a pivotal role in regulation of NHE3 by activation of the RhoA-ROCK-Pyk2 and Ras-MAPK pathways.

GRANTS

This work was supported by National Institutes of Health Grants (NIH) DK-061418 and DK-061418S1. P. He was supported by a postdoctoral fellowship from the American Heart Association.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).