Elevated levels of intracellular Ca2+ ([Ca2+]i) inhibit Na+/H+ exchanger 3 (NHE3) activity in the intact intestine. We previously demonstrated that PLC-γ directly binds NHE3, an interaction that is necessary for [Ca2+]i inhibition of NHE3 activity, and that PLC-γ Src homology 2 (SH2) domains may scaffold Ca2+ signaling proteins necessary for regulation of NHE3 activity. [Ca2+]i regulation of NHE3 activity is also c-Src dependent; however, the mechanism by which c-Src is involved is undetermined. We hypothesized that the SH2 domains of PLC-γ might link c-Src to NHE3-containing complexes to mediate [Ca2+]i inhibition of NHE3 activity. In Caco-2/BBe cells, carbachol (CCh) decreased NHE3 activity by ∼40%, an effect abolished with the c-Src inhibitor PP2. CCh treatment increased the amount of active c-Src as early as 1 min through increased Y416 phosphorylation. Coimmunoprecipitation demonstrated that c-Src associated with PLC-γ, but not NHE3, under basal conditions, an interaction that increased rapidly after CCh treatment and occurred before the dissociation of PLC-γ and NHE3 that occurred 10 min after CCh treatment. Finally, direct binding to c-Src only occurred through the PLC-γ SH2 domains, an interaction that was prevented by blocking the PLC-γ SH2 domain. This study demonstrated that c-Src 1) activity is necessary for [Ca2+]i inhibition of NHE3 activity, 2) activation occurs rapidly (∼1 min) after CCh treatment, 3) directly binds PLC-γ SH2 domains and associates dynamically with PLC-γ under elevated [Ca2+]i conditions, and 4) does not directly bind NHE3. Under elevated [Ca2+]i conditions, PLC-γ scaffolds c-Src into NHE3-containing multiprotein complexes before dissociation of PLC-γ from NHE3 and subsequent endocytosis of NHE3.
elevated levels of intracellular calcium ([Ca2+]i) inhibit electroneutral sodium absorption in the small intestine (11, 14, 17, 33). In the rabbit ileum, carbachol (CCh) elevates apical domain [Ca2+]i via basolateral membrane (BLM) muscarinic (M3) receptors and inhibits Na+/H+ exchanger 3 (NHE3) activity by 40% (11, 14, 23, 46) by decreasing NHE3 surface expression through increased endocytosis (4, 8, 12, 15, 20, 22, 30, 32, 39, 56, 58). This effect is accompanied by increased size of multiprotein NHE3-containing complexes (30). The multi-PDZ domain-containing proteins NHERF2 and NHERF3 are necessary to mediate Ca2+ regulation of NHE3 activity by directly binding other signaling proteins and also dynamically associate with the NHE3 C terminus (5, 10, 13, 27, 30, 34, 42, 57, 60). Thus these data suggest that signaling complexes bind NHE3 and are necessary for the regulation of NHE3 activity (16). As an example of elevated [Ca2+]i signaling, CCh increased brush border (BB) levels of diacylglycerol (DAG) and translocation of activated PLC-γ and PKCα to the BB (24). We have previously suggested a role for apical PLC-γ in [Ca2+]i inhibition of NHE3 activity by demonstrating that PLC-γ directly binds NHE3 (59). This interaction is necessary for [Ca2+]i inhibition of NHE3 activity, and we suggested that the Src homology 2 (SH2) domains of PLC-γ might scaffold Ca2+ signaling proteins necessary for regulation of NHE3 activity (59). Moreover, in intact rabbit small intestine and in Caco-2/BBe cells, [Ca2+]i-mediated inhibition of NHE3 activity is c-Src dependent (30). In these studies, activation of c-Src through increased tyrosine phosphorylation is necessary for inhibition of NHE3 activity (30). Another previous study suggested that activated c-Src directly binds the SH2-SH3 domains of PLC-γ (25). In addition, conflicting studies have demonstrated that activated c-Src operates both downstream (48) and upstream (3) of PLC-γ. Although an interaction between PLC-γ and c-Src has been proposed, a role for this interaction in the regulation of NHE3 activity has not been demonstrated.
c-Src is a member of a family of nine membrane-associated nonreceptor tyrosine kinases (nRTKs; Refs. 6, 45, 47). Src tyrosine kinases have multiple biological functions, including roles in cell adhesion assembly and turnover and in motility, proliferation, and survival, all of which are important in disease contexts (47). All Src family proteins share a similar structure, with molecular masses in the range of 52–62 kDa (6). They are comprised of six functional domains: a myristylation domain, which enables interaction with the plasma membrane; a unique domain; SH3 and SH2 domains for binding to other proteins; a kinase domain, containing an autophosphorylation site at Y416; and a C-terminal regulatory region (6, 26, 36, 47). Phosphorylation of Y527 in the c-Src C-terminal domain takes part in the regulation of c-Src tyrosine kinase activity; c-Src undergoes a conformational change that “opens” or “closes” the protein depending on the state of phosphorylation of Y527 and the molecular interactions involving this residue. When Y527 is phosphorylated (i.e., negative regulation), it interacts with the c-Src-SH2 domain (36), the SH3 domain becomes engaged with the SH2-kinase linker region, and Y416 in the activation loop of the kinase becomes unphosphorylated, rendering Src closed and catalytically inactive (26).
In the previous study by Li et al. (30), the c-Src inhibitor PP2 prevented CCh-mediated inhibition of NHE3 activity, suggesting that this regulatory process is c-Src dependent. While CCh induced elevation of [Ca2+]i increased the amount of tyrosine phosphorylated c-Src in the BB, this study did not address whether c-Src is a part of NHE3-containing multiprotein complexes. Moreover, although the study demonstrated an increase in the size of NHE3-containing complexes in response to elevated [Ca2+]i, the presence of c-Src in these complexes was not demonstrated. Our previous studies using a phosphorylated peptide specific for binding the SH2 domain of PLC-γ demonstrated that blocking the PLC-γ SH2 domains abolished [Ca2+]i-mediated inhibition of NHE3 activity (59). Based on these data, we proposed that activated c-Src may exist in close proximity to NHE3 to mediate calcium inhibition. Therefore, we tested the hypothesis that c-Src-dependent calcium inhibition of NHE3 activity occurs through the scaffolding of c-Src to the SH2 domain(s) of PLC-γ, which directly binds to NHE3.
MATERIALS AND METHODS
CCh was from Sigma. BODIPY 577/618 maleimide was from Invitrogen.
Affinity-purified mouse monoclonal antibody to human PLC-γ and rabbit polyclonal antibody to phospho-Src (Y527) were from Millipore. Rabbit polyclonal antibody to phosphorylated (Y416) c-Src was from AbCam. Rabbit polyclonal antibody to phosphorylated human PLC-γ (Y783) was from Cell Signaling. Monoclonal anti-HA antibody was from Covance.
Caco-2/BBe cells express all four members of the NHERF gene family and small amounts of NHE3 (13). Triple HA-tagged rabbit NHE3 was expressed by adenovirus into Caco-2/BBe cells (termed Caco-2/BBe/NHE3) for biochemical analysis and more accurate transport measurements since endogenous NHE3 activity is ∼10% of Caco-2/BBe cells infected with adenovirus NHE3 (41). Caco-2/BBe cells were grown on Anapore filters (Nunc) until postconfluent for 12 days in DMEM supplemented with 25 mM NaHCO3, 10 mM HEPES, 0.1 mM nonessential amino acids, 50 U/ml penicillin, 50 μg/ml streptomycin, and 10% fetal bovine serum (complete media) in a 5% CO2-air incubator at 37°C. Cells were serum starved overnight and then treated with 6 mM EGTA in serum free complete media for 2 h at 37°C. Caco-2/BBe cells were then exposed to 3HA-NHE3 adenovirus for 6 h at 37°C. Cells were allowed to recover in normal media over the next 40 h before study (42, 57).
Fusion proteins of full-length PLC-γ- and PLC-γ-specific array domains (i.e., 2 SH2 domains, SH3 domain) were generated as GST-tagged fusion proteins, as described previously (54). Briefly, PLC-γ cDNA inserts were subcloned into pGex vector (GE Life Sciences). GST-tagged fusion proteins were generated after transformation into Escherichia coli and subsequently purified with glutathione agarose beads (Sigma). Full-length His6-tagged human recombinant PLC-γ purified protein was from Calbiochem. Full-length active human c-Src was from Millipore.
Peptide synthesis and purification were performed by the Synthesis and Sequencing Facility at Johns Hopkins University School of Medicine. Coupling of the peptide to BODIPY 577/618 maleimide was performed according to the manufacturer's protocol (Invitrogen). PLC-γ SH2 domain binding (i.e., “active”) and negative control (i.e., “inactive”) peptides have been previously described (7, 59).
Measurement of Na+/H+ exchange.
Cellular Na+/H+ exchange activity in Caco-2/BBe/NHE3 cells grown to 14 days postconfluency on Transwell filters was determined fluorometrically using the intracellular pH-sensitive dye 2,7-bis(carboxyethyl)5–6-carboxyfluoresceinacetoxy-methyl ester (BCECF-AM; 5 μM; Molecular Probes, Eugene, OR), as described previously (28). Caco-2/BBe/NHE3 cells were exposed to 50 mM NH4Cl during a 45-min dye loading, as described previously (42, 57, 59). Cells were perfused initially with TMA+ solution alone or with 10 μM CCh for 1–10 min (130 mM tetramethylammonium chloride, 5 mM KCl, 2 mM CaCl2, 1 mM MgSO4, 1 mM NaH2PO4, 25 mM glucose, and 20 mM HEPES, pH 7.4) before being switched to Na+ solution (130 mM NaCl instead of tetramethyl-ammoniumchloride) for the Na+-dependent pHi recovery. At the end of each experiment, the fluorescence ratio was calibrated to pHi using the high potassium/nigericin method. Na+/H+ exchange activity data were calculated as the ratio of Na+-dependent changes in pHi over initial time (ΔpH/min) of Na+-dependent pH recovery using at least three coverslips per condition in a single experiment. Initial rates were analyzed by using Origin (Microcal Software) to determine statistical significance among individual experiments. Means ± SE were determined from at least three separate experiments.
Protein overlay (Far Western) assays were used to examine the interaction of PLC-γ purified proteins (2 μg of each full-length and γ-specific array domains) on blots with recombinant c-Src (overlay) by subsequent incubation of blots with monoclonal anti-Src antibody, as described previously (55). For peptide competition studies, PLC-γ purified proteins were separated on SDS-PAGE, and proteins were transferred to nitrocellulose membranes and exposed to 40 μg of either the “active” or “inactive” peptides conjugated to BODIPY 577/618 for 1 h at room temperature. Binding of peptides was visualized using a fluorescent Typhoon Imager (Johns Hopkins University School of Medicine Proteomics Core). Membranes were then exposed to 4 μg purified full-length Src recombinant protein overnight at 4°C. c-Src binding was determined by Western blotting as described above. Results were obtained from three individual experiments.
PLC-γ or NHE3 were immunoprecipitated (IP) from the total lysate of Caco-2/BBe/NHE3 cells (in the presence of 1% Triton X-100). All IPs were done at 4°C with constant mixing on a rotary shaker. Briefly, each sample (1 mg of total cell lysate per IP) was first precleared with either protein A-Sepharose beads (Sigma) or protein A beads conjugated to rabbit anti-mouse secondary antibody for 1 h. The precleared lysate was then incubated with 4 μg of antibodies to PLC-γ or c-Src for 1 h. Protein A-Sepharose beads were then added to each IP mixture, and incubation was continued for another 1 h. The beads were washed four times with phosphate-buffered saline containing 0.1% Tween 20 (Sigma). The IP pellets were analyzed by SDS-PAGE and Western blotted with corresponding antibodies. Caco-2/BBe cells were grown on 10-cm2 Transwell Petri dishes until postconfluent for 12 days and infected with adenovirus 3HA-NHE3 construct as described above. After infection, cells were allowed to recover in normal media for 40 h before CCh treatment. On day 14 postconfluency, cells were serum-starved again for 4 h and treated either with vehicle or 10 μM CCh (Sigma) for 1–10 min at 37°C. Adenovirus infected Caco-2/BBe cells were washed three times in ice-cold phosphate-buffered saline containing 50 mM Tris. Cells were collected and lysed in 500 μl of ice-cold lysis buffer (10 mM HEPES, 50 mM NaCl, 5 mM EDTA, 1 mM benzamidine, and 0.5% Triton X-100). Cell lysate was solubilized for 30 min at 4°C with end-over-end rotation and subsequently homogenized 10 times using a 23-gauge needle. Cellular debris was cleared by centrifugation at 14,000 rpm for 15 min. Supernatant was incubated with either anti-PLC-γ or anti-Src antibodies conjugated to protein G agarose beads (Pierce) for 2 h with end-over-end rotation at 4°C. Samples were washed five times with lysis buffer, and immunoprecipitated proteins were eluted from beads with 2× sample buffer. Samples were resolved by 10% SDS-PAGE, and proteins were detected with anti-HA (NHE3), anti-Src, and anti-PLC-γ antibodies and visualized on an Odyssey Infrared Imaging System (Li-Cor, Lincoln, NE).
Sucrose density gradient centrifugation.
Caco-2/BBe/NHE3 cells were seeded on Transwell filters in 10-cm2 Petri dishes and studied on day 14. Cells were solubilized in 1 ml of N+ buffer (50 mM HEPES, Tris, pH 7.4, 150 mM NaCl, 3 mM KCl, 5 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, 1 mM aprotinin, 1 mM pepstatin, 1 mM iodoacetamide, and 1% Triton X-100), sonicated, and spun briefly to remove unbroken cells, nuclei, and cell debris. Solubilized cell extracts (4°C) were applied to the top of discontinuous 2.5–30% sucrose gradients (increasing at increments of 2.5% sucrose containing 0.1% Triton X-100). After centrifugation for 16.5 h at 4°C at 150,000 g in an SW40i rotor, the gradients were fractionated (20 fractions; 0.6 ml each) from the bottom with a perfusion pump, and NHE3, c-Src, and PLC-γ expression was visualized by Western blot with anti-HA monoclonal antibody (for NHE3 in Caco-2/BBe cells), anti-PLC-γ monoclonal antibody, and anti-Src monoclonal antibody as described above. NHE3, Src, and PLC-γ protein expression was detected with either IRdye700 or IRdye800 anti-mouse secondary antibodies (Rockland) using the Odyssey Infrared Imaging System (Li-Cor).
Caco-2/BBe/NHE3 cells were exposed to vehicle or CCh for various time points from 1–60 min. Cells were solubilized in N+ buffer as described above. Total cell lysates were separated on SDS-PAGE and proteins transferred to nitrocellulose membranes. Phosphorylation of c-Src at Y416 and Y527 at each time point was determined using phospho-Src-specific antibodies described above and compared with the amount of total c-Src as determined using monoclonal anti-Src antibody. The amount of phosphorylation at each site was normalized to the amount of total c-Src and to GAPDH, which served as a loading control. Proteins were visualized on Odyssey Infrared Imaging System as described above.
AlphaScreen-AlphaScreen (Perkin Elmer) was used to detect the interaction between PLC-γ and c-Src and determine whether this interaction is prevented in the presence of the active peptide that binds to PLC-γ SH2 domains. To develop the binding assay, Glutathione Donor Beads (Perkin Elmer) were conjugated with GST-tagged PLC-γ SH2 or SH3 (20 nM each) domains for 1 h at 4°C. Similarly, anti-His Acceptor Beads (Perkin Elmer) were conjugated to His6-tagged full-length c-Src (20 nM) for 1 h at 4°C. Conjugations were performed in assay buffer containing 25 mM HEPES, 100 mM NaCl, 0.1% (wt/vol) BSA, and 0.05% (vol/vol) Tween 20, pH 7.4. Conjugated acceptor beads (50 μl for each well and 20 μg/ml final concentration) were transferred to a 96-well white opaque OptiPlate (Perkin Elmer) in triplicate were added and incubated for 1 h. PLC-γ conjugated donor beads (25 μl for each well and 20 μg/ml final concentration) were then added and incubated for another 1 h at room temperature. The plate was read on an EnVison plate reader with AlphaScreen capability (Perkin Elmer). The specificity of the observed AlphaScreen binding signals between PLC-γ and c-Src was verified by AlphaScreen competition assays using the active and inactive PLC-γ SH2 peptides (See above). The assay procedure is similar to the association assay described above except that the active or inactive peptides were serially diluted (0.02–20 μM) in the assay buffer containing PLC-γ-conjugated donor beads. The solutions were then incubated, transferred, and assayed as described above. Additional binding controls were performed in which only the donor or the acceptor beads were conjugated to PLC-γ or c-Src, respectively, and paired with unconjugated donor or acceptor beads. The binding curve and statistics were generated using Origin software. All assays were performed in triplicate and a minimum of three independent experiments were performed for each assay.
Results were expressed as means ± SE. Statistical evaluation was by ANOVA or Student's t-test.
c-Src is necessary for CCh inhibition of NHE3 activity in Caco-2/BBe/NHE3 cells.
Previous studies in the rabbit ileum demonstrated that activated c-Src is necessary for CCh inhibition of NHE3 activity. To determine the role for c-Src in regulation of NHE3 activity, we first tested whether c-Src is also necessary for CCh inhibition of NHE3 activity in a polarized intestinal epithelial cell model. To accurately measure NHE3 activity, Caco-2/BBe cells were grown on filters until postconfluent for 12 days and were then infected with an adenovirus 3HA-NHE3 construct (see materials and methods), as previously described (41). NHE3 activity was determined 48 h after the infection. Caco-2/BBe/NHE3 cells were pretreated with either vehicle or 10 μM PP2 for 30 min and then exposed to vehicle or CCh for 5 min before Na+-dependent recovery of pHi. As shown in Fig. 1, in vehicle-pretreated controls, NHE3 activity was significantly inhibited after 10 μM CCh treatment (Fig. 1, n = 4). CCh inhibition of NHE3 activity was abolished in cells pretreated with the Src inhibitor PP2, consistent with the finding that c-Src is necessary for CCh-mediated inhibition of NHE3 activity (30). Since NHE3 has been previously demonstrated to exist in multiprotein complexes in the intact intestine, we tested whether c-Src also resides in similar complexes.
c-Src resides in similar sized multiprotein complexes as NHE3 and PLC-γ.
We have previously demonstrated that NHE3 and PLC-γ reside in similar multiprotein complexes in Caco-2/BBe/NHE3 cells (59). In the intact intestine and Caco-2/BBe cells, CCh increases the size of NHE3-containing complexes, which includes NHERF2, α-actinin-4, and activated PKCα. Since CCh inhibition of NHE3 activity requires activated c-Src, we asked whether c-Src associates with NHE3 in similar multiprotein complexes. As determined by sucrose density gradient centrifugation, c-Src exists in similar sized multiprotein complexes (range: ∼200 to 1,200 kDa) as both NHE3 and PLC-γ in Caco-2/BBe/NHE3 cells under basal conditions (Fig. 2). Since the size of NHE3-containing complexes changes with acute regulation, we determined whether c-Src complexes were regulated to a similar degree. In Caco-2/BBe/NHE3 cells treated with CCh, c-Src-containing complexes were observed to shift to larger sized fractions. A similar pattern was also observed for PLC-γ and NHE3 (Fig. 2). The results of these experiments were consistent with c-Src interacting with NHE3; therefore, we further investigated the nature of this interaction by coimmunoprecipitation and in vitro studies.
c-Src coimmunoprecipitates and directly binds to PLC-γ, but not NHE3, in Caco-2/BBe/NHE3 cells.
To test whether c-Src associates with NHE3 in vivo, coimmunoprecipitation studies were performed. Total cell lysates were prepared from 14-day postconfluent Caco-2/BBe/NHE3 cells. c-Src was immunoprecipitated and subjected to SDS-PAGE for NHE3 and PLC-γ interactions. Under basal conditions, immunoprecipitated c-Src associated with PLC-γ but not NHE3 (Fig. 3A). Additionally, this association was dynamic in that the association of c-Src and PLC-γ increased after 10 μM CCh treatment (2 min). A similar result was observed when we coprecipitated PLC-γ with c-Src 5 min after CCh treatment (Fig. 3B). We have previously demonstrated that NHE3 and PLC-γ directly bind under basal conditions, and this interaction was lost 10 min after CCh treatment. Furthermore, we have demonstrated that CCh-induced endocytosis of NHE3 occurs as early as 1 min after CCh in Caco-2/BBe/NHE3 cells, suggesting that CCh-mediated signal transduction occurs earlier than 10 min (56). Therefore, we tested whether the association of NHE3 and PLC-γ differs between 1 and 5 min from that previously described at 10 min. As shown in Fig. 3B, immunoprecipitated PLC-γ associates with NHE3 and c-Src under basal conditions at 2 and 10 min. The association of PLC-γ and NHE3 increased at 2 min after CCh but was lost by 10 min, supporting our previous observation (59). In addition, this increased interaction was still detected at 5 min post-CCh although to a lesser degree (data not shown). At 10 min post-CCh treatment, the association of PLC-γ and c-Src was increased compared with untreated controls at the same time point (Fig. 3B). These results suggest that CCh signaling rapidly (2–5 min) increases the association of c-Src with PLC-γ and also increases the PLC-γ association with NHE3 during the same time course. These events precede the dissociation of PLC-γ from NHE3 that occurs after 10 min, which we demonstrated here and have previously described (59). Therefore, we next determined whether PLC-γ directly binds c-Src.
PLC-γ is comprised of protein-protein interacting domains that are specific for this PLC isoform. This region is composed of two split pleckstrin homology (PH) domains that are separated by two Src homology 2 (SH2) domains and an SH3 domain. To determine direct binding, we generated GST-tagged fusion proteins of each individual PLC-γ-specific protein-protein interacting domain as well as full-length PLC-γ. Purified PLC-γ proteins were separated on SDS-PAGE and transferred to nitrocellulose membranes, which were incubated with purified full-length active recombinant c-Src protein (His6-tagged; see materials and methods). As shown in Fig. 4, far Western blot analysis demonstrated that c-Src directly bound to full-length PLC-γ, as demonstrated below (see Fig. 6). This interaction occurred through direct binding to the SH2, not the SH3, domains of PLC-γ. These data suggest that activated c-Src (through phosphorylation of Y416) directly interacts with SH2 domains, which are protein-interacting domains that bind phosphorylated tyrosine residues. Since c-Src is activated by phosphorylation of Y416 and it dynamically associates with PLC-γ in response to CCh treatment, we asked whether CCh signaling increases phosphorylation of c-Src.
CCh signaling induces rapid activation of c-Src in Caco-2/BBe/NHE3 cells.
Dormant c-Src is autoinhibited by phosphorylation of Y527, which binds to its own SH2 domain(s) and maintains a closed conformation. Dephosphorylation of Y527 opens c-Src and phosphorylation of Y416 activates c-Src (26, 36). To demonstrate that CCh activates c-Src, we used site-specific antibodies to determine the degree of c-Src phosphorylation at Y416 and Y527. Caco-2/BBe/NHE3 cells were treated with vehicle or CCh for 1–60 min. At each time point, cells were washed with ice-cold PBS and scraped, and cell lysates obtained for Western blot analysis. CCh treatment increased the amount of c-Src phosphorylated at Y416 as early as 1 min (Fig. 5A). Increased Y416 phosphorylation was observed for up to 10 min before decreasing at 60 min to levels that matched untreated controls (Fig. 5B, bar graph). In contrast to Y416, phosphorylation of Y527 decreased over the same time course (1–10 min) and was increased to control values by 60 min (Fig. 5, bar graph). These data suggest that CCh activates c-Src over the same time course in which NHE3 activity is decreased and the interaction of c-Src and PLC-γ is increased. In addition, we measured phosphorylation of PLC-γ at Y783 after CCh treatment at similar time points and determined that PLC-γ phosphorylation was not increased after CCh treatment (data not shown). This result supports our previous finding in the intact ileum that CCh signaling induced translocation of PLC-γ to the BB without increasing PLC-γ Tyr phosphorylation. Since CCh signaling involves elevated [Ca2+]i, we asked whether the dynamic association of c-Src to PLC-γ is Ca2+ dependent. We performed far Western blotting of full-length active c-Src direct binding to the full-length and SH2 domains of PLC-γ in the presence or absence of 2 μM free Ca2+. As shown in Fig. 6, direct binding of c-Src to either the full-length or SH2 domain of PLC-γ is not Ca2+ dependent. Therefore, these data suggest that the interaction of c-Src to PLC-γ SH2 domains requires c-Src phosphorylation induced by CCh signaling in Caco-2/BBe/NHE3 cells that is separate from the elevation in [Ca2+]i.
Using a specific PLC-γ SH2 domain binding peptide, we previously demonstrated that protein-protein interactions at this site are necessary for CCh-mediated inhibition of NHE3 activity in Caco-2/BBe/NHE3 cells (7, 59). Since our data suggest that phosphorylated c-Src directly binds to the PLC-γ SH2 domain, we tested the hypothesis that blocking PLC-γ SH2 domain using the specific phosphorylated peptide would prevent c-Src binding to PLC-γ. To test whether we could compete with the binding of c-Src to PLC-γ using the active peptide, far Western blot analysis was performed. As demonstrated in Fig. 7, top, BODIPY-conjugated active and inactive peptides (40 μg each) were exposed to purified PLC-γ proteins (2 μg each) separated on nitrocellulose membranes. A fluorescent band (from conjugated BODIPY; excitation = 577 nm; emission = 618 nm) indicating active peptide, but not inactive peptide, bound to PLC-γ SH2 domain was detected using a Typhoon Imager (see materials and methods). After peptide binding, the membranes were exposed to purified, full-length active c-Src (4 μg) for 2 h. Membranes were then washed, and c-Src was detected by specific c-Src monoclonal antibody (Fig. 7, bottom). c-Src only bound to PLC-γ SH2 domain protein that was exposed to inactive peptide, which does not bind PLC-γ SH2 domains. The active peptide completely prevented binding of c-Src to PLC-γ SH2 domain.
We further demonstrated the direct interaction of PLC-γ SH2 domain and c-Src using AlphaScreen (Amplified Luminescent Proximity Homogeneous Assay) technology (Perkin Elmer). Briefly, with the use of this fluorescence resonance energy transfer-like technique, PLC-γ and c-Src were conjugated to donor and acceptor beads, respectively. Protein-protein interactions were detected by luminescence emission at 540 nm after donor excitation at 680 nm (40, 50). As shown in Fig. 8A, the PLC-γ SH2, but not SH3, domain directly binds full-length, active c-Src. This interaction was prevented when c-Src was pretreated with calf intestinal phosphatase (control = 352,790 ± 9,923 counts/s vs. calf intestinal phosphatase treated = 14,112 ± 2,976 counts/s; P < 0.01; n = 3). AlphaScreen counts were nearly undetectable in assays in which only either the donor or acceptor beads were conjugated to PLC-γ or c-Src, respectively (Fig. 8A). To demonstrate that the interaction between PLC-γ and c-Src required PLC-γ SH2 domains, we performed peptide competition assays using the active and inactive peptides. Pretreatment of PLC-γ conjugated donor beads with 5 μg of the active peptide, but not the inactive peptide, significantly reduced binding to c-Src (Fig. 8A). We also determined the dose-dependent response curve for active peptide blocking PLC-γ/c-Src direct binding. In Fig. 8B, the active peptide decreased the interaction of PLC-γ and c-Src with an IC50 of 517 nM, while the inactive peptide did not prevent binding up to 20 μM. The results of these assays demonstrate that c-Src directly binds to the PLC-γ SH2 domain and that phosphorylation of c-Src at Y416 is required for this interaction. Therefore, these data, in combination with our previous findings, suggest that CCh inhibition of NHE3 activity involves direct interaction between Src and PLC-γ, which scaffolds c-Src through its SH2 domains.
This study provides additional insights into how signaling complexes are involved in regulation of intestinal Na+ absorption by controlling NHE3 activity. Regulation of NHE3 activity occurs through direct binding of regulatory proteins to its C-terminal domain (16, 35, 58). These protein-protein interactions often form multiprotein signaling complexes that take part in both stimulation and inhibition of NHE3 activity, which occurs under both physiologic and pathophysiologic conditions (35, 58). One aspect of NHE3 regulation in which the role of signaling complexes has been partially defined is inhibition of NHE3 by CCh, which mimics the regulation that occurs after meals. c-Src has been shown to be an important component of this regulation, with demonstration that 1) c-Src forms a signaling complex with PLC-γ through the PLC-γ SH2 domains, and 2) this signaling complex is dynamic, with the interaction between PLC-γ and c-Src increasing as early as 2 min post-CCh, which is before release of the PLC-γ and c-Src from NHE3, which occurs 10 min after CCh. Since both PLC-γ SH2 domains and c-Src are necessary for CCh inhibition of NHE3 activity, we propose a model that these proteins interact to rapidly inhibit NHE3 activity in polarized intestinal epithelial cells.
We previously characterized the role of c-Src in acute inhibition of NHE3 following CCh exposure in the intact rabbit ileum (30). The CCh-induced decrease in NHE3 activity was associated with an increase in the size of NHE3-containing multiprotein complexes, and this increase in complex size was prevented by pretreatment with the c-Src inhibitor PP2 (30). Moreover, these CCh effects were associated with rapid activation and translocation of c-Src to the BB of intestinal epithelial cells. However, the mechanism by which c-Src took part in regulation of NHE3 activity was not further understood. The current results provide more insights into these mechanisms.
In the CCh-treated ileum, the increased size of NHE3 complexes was c-Src-dependent; however, whether c-Src directly bound to NHE3 or was a part of NHE3-containing complexes was not determined (30). We previously showed that the increase in size was due to proteins being indirectly linked to NHE3 via binding to NHERF2; after CCh this was true for the increased association of PKC-α and α-actinin-4 with NHE3 (27, 30). We demonstrate here that the circumstances are similar for c-Src; although it exists in similar size multiprotein complexes as NHE3 (Fig. 2), it does not associate directly with NHE3 [by coimmunoprecipitation (Fig. 3) or in vitro binding assays (data not shown)]. Rather, c-Src is brought into the NHE3 complex by binding to PLC-γ via its previously demonstrated scaffolding function (59). This is an example in which multiple scaffolds bind NHE3 to take part in forming its dynamic signaling complexes. In describing the interactions of PLC-γ with NHE3, we suggested that this was by lipase-independent functions of PLC-γ and involved it acting as a scaffold protein (59). In these studies, blocking the PLC-γ SH2 domains, using a phosphorylated peptide that specifically binds to these domains, prevented CCh-mediated inhibition of NHE3 activity in Caco-2/BBe/NHE3 cells (59).
Our current study was designed to understand how c-Src takes part in NHE3 regulation and the relationship between this regulation and the binding of c-Src to PLC-γ. The results of our study demonstrated that c-Src and PLC-γ directly bind each other and that their association is dynamic, rapidly increasing in response to CCh (2–10 min), which occurs before the CCh-related dissociation of PLC-γ from binding to NHE3 (10 min). Moreover, we demonstrate (Fig. 8) that direct binding of c-Src to PLC-γ requires activated c-Src (i.e., Y416 phosphorylated) and that the amount of pY416 c-Src was increased after CCh treatment (Fig. 5). Whether c-Src phosphorylation at Y416 is necessary for the association to PLC-γ observed by coimmunoprecipitation studies remains to be determined. We also showed that the direct binding of active c-Src to PLC-γ does not require Ca2+ in addition to pY416. These data suggest that CCh-initiated signaling effectors separate from elevated [Ca2+]i are required for the association of c-Src and PLC-γ. Since CCh regulation of NHE3 is Ca2+ dependent, the data from this study suggest that elevated [Ca2+]i due to CCh signaling allows for increased phosphorylation of c-Src at Y416 that results in increased binding to PLC-γ and increased association with NHE3-containing complexes. Whether one or the other or both PLC-γ SH2 domains are required for this effect remains to be studied. Furthermore, future studies are required to determine the downstream effects of activated c-Src in CCh-induced NHE3 inhibition.
Dynamic changes in protein-protein interactions involving scaffolds that take part in forming NHE3-containing complexes have also been observed with NHERF2 (27, 30). With acute stimulation of NHE3 activity by lysophosphatidic acid (9), NHERF2 transiently dissociates from NHE3. In this study, NHERF2 initially bound NHE3, but completely lost association after 30 min lysophosphatidic acid treatment before binding NHE3 again after 1 h. We and others have demonstrated that several proteins, including NHERF2, bind to the C terminus of NHE3 in the same area (aa 585–605), which has been shown to be necessary for PLC-γ binding. Since it is unlikely that several proteins simultaneously bind the same 21-aa region of NHE3, their individual association must change as a part of signal transduction. Furthermore, since CCh inhibition of NHE3 occurs by increased endocytosis, the change in the association of NHE3 with PLC-γ may be correlated with the change in subcellular localization of NHE3 due to trafficking. Another possibility may involve a change in the conformation of the NHE3 C terminus, possibly due to phosphorylation or dephosphorylation of sites that may alter the affinity of NHE3 for PLC-γ. It is possible that the loss of PLC-γ binding after 10 min may allow another protein complex to bind NHE3; however, the identity of a second complex that increases its association with NHE3 after 10 min CCh remains to be determined.
Whether additional proteins exist in the c-Src/PLC-γ signaling complex remains to be identified. c-Src contains an SH2 and an SH3 domain that have been demonstrated to directly bind c-Src substrates for phosphorylation (2, 21, 29, 38, 43, 44, 52). The presence of these protein-protein interacting domains gives c-Src the potential to link other proteins, in addition to PLC-γ, to the complex and provide a possible explanation to the increased size of c-Src complexes observed after CCh treatment (Fig. 2). Although c-Src is necessary for Ca2+ regulation of NHE3 activity and formation of larger NHE3 complexes, NHE3 is not tyrosine phosphorylated under basal or regulated conditions (35, 58). Similarly, NHE3 is not phosphorylated by PKC-α under elevated [Ca2+]i conditions as well (51, 53). Therefore, despite the presence of two active kinases being brought into NHE3-containing complexes after CCh, neither c-Src and PKC-α phosphorylate NHE3 directly and rather are thought to phosphorylate other components of the complex or other downstream effectors that result in inhibition of NHE3 activity through increased endocytosis (2, 29, 52). Our current results combined with our previous studies demonstrate that c-Src is necessary for the formation of a signaling complex required to inhibit NHE3 after CCh treatment in intestinal epithelial cells.
Based on our results, we propose a model in which a c-Src/PLC-γ signaling complex regulates NHE3 activity under elevated [Ca2+]i conditions. NHERF2, NHERF3, and PLC-γ directly bind to NHE3, and their direct interaction with NHE3 is necessary for elevated [Ca2+]i inhibition of NHE3 activity. Since NHE3 is highly regulated, it is possible that different signaling complexes may specifically form with NHERF2 and NHERF3 under elevated [Ca2+]i conditions. Please note that PLC-γ/c-Src is one of several signaling complexes demonstrated to be involved in acute regulation of NHE3 activity (16, 35). Other multiprotein complex models have been described for cAMP and cGMP regulation of NHE3 activity (16, 35).
We previously demonstrated that CCh initiates signal transduction and inhibition of NHE3 activity by increased endocytosis as early as 1 min post-CCh treatment (24, 56). CCh induces rapid increase in elevated [Ca2+]i, increased inositol 1,4,5-trisphosphate and DAG production, increased PKC activity, and increased translocation of PLC-γ and c-Src to detergent-resistant membrane fractions of the BB, which occur simultaneously to increased endocytosis of NHE3 (24, 30, 56). In the current study, we demonstrate that activation of c-Src also occurs as early as 1 min after CCh treatment through increased phosphorylation of Y416 (active site) and dephosphorylation of Y527 (inactive site). Although CCh signaling through the M3 muscarinic receptor has been well described, downstream signaling involving the activation of c-Src by CCh in intestinal epithelial cells is not well understood. Several studies have linked downstream effects of activated c-Src after muscarinic receptor stimulation supporting the concept of c-Src as an intermediate effector. A recent report by Hassan et al. (18) demonstrated that CCh inhibits oxalate transport through the anion exchanger SLC26A6 partially through activation of PKCδ by c-Src in T84 cells. Currently, the kinases/phosphatases involved in CCh-mediated stimulation of c-Src in intestinal Na+ absorptive epithelial cells have not been identified.
Activated c-Src has been demonstrated to play an integral role in several biological processes including establishing epithelial polarity and protein trafficking (6). NHE3 inhibition occurs through changes in trafficking either by increased endocytosis or, less frequently, decreased exocytosis (58). A role for c-Src in regulation of endocytosis has been suggested by studies in which c-Src activates members of the Rho GTPase family, which alter the actin cytoskeleton as well as stimulating phospholipase D, to produce conical lipids, phosphatidic acid, and DAG, which are involved in creating negative curvature required for formation of endocytic vesicles (1, 19, 31, 37, 49). Whether Rho GTPases and/or phosphoplipase D contribute to the mechanism responsible for CCh inhibition of NHE3 activity also remains to be determined.
The results of our study showed that PLC-γ directly binds c-Src through its SH2 domains, and this interaction is necessary for CCh mediated inhibition of NHE3 activity in Caco-2/BBe/NHE3 cells. Moreover, c-Src is linked to NHE3 via binding to PLC-γ; nonetheless, it is likely that there are additional components of this signaling complex likely formed on PLC-γ. Whether c-Src contributes directly to CCh mediated endocytosis of NHE3 remains to be studied. In conclusion, this study demonstrates that CCh inhibition of NHE3 activity involves the dynamic formation of a signaling complex that includes PLC-γ and c-Src, which take part in regulating NHE3 trafficking.
This work was supported, in whole or in part, by National Institute of Diabetes and Digestive and Kidney Diseases Grants K01-DK-080930, R03-DK-091482, R01-DK-26523, R01-DK-61765, P01-DK-072084, and P30-DK-089502 (The Hopkins Conte Digestive Disease Basic and Translational Research Core Center).
M. Donowitz is a part owner of Tranzmembrane, Incorporated.
N.C.Z. conception and design of the research; N.C.Z. and L.J.L. performed the experiments; N.C.Z. and L.J.L. analyzed the data; N.C.Z., L.J.L., O.K., X.L., and M.D. interpreted the results of the experiments; N.C.Z. prepared the figures; N.C.Z. drafted the manuscript; N.C.Z., O.K., X.L., and M.D. edited and revised the manuscript; N.C.Z. approved the final version of the manuscript.
- Copyright © 2013 the American Physiological Society