The NHE1 Na+/H+ Exchanger Recruits Ezrin/Radixin/Moesin Proteins to Regulate Akt-dependent Cell Survival*

Apoptosis results in cell shrinkage and intracellular acidification, processes opposed by the ubiquitously expressed NHE1 Na+/H+ exchanger. In addition to mediating Na+/H+ transport, NHE1 interacts with ezrin/radixin/moesin (ERM), which tethers NHE1 to cortical actin cytoskeleton to regulate cell shape, adhesion, motility, and resistance to apoptosis. We hypothesize that apoptotic stress activates NHE1-dependent Na+/H+ exchange, and NHE1-ERM interaction is required for cell survival signaling. Apoptotic stimuli induced NHE1-regulated Na+/H+ transport, as demonstrated by ethyl-N-isopropyl-amiloride-inhibitable, intracellular alkalinization. Ectopic NHE1, but not NHE3, expression rescued NHE1-null cells from apoptosis induced by staurosporine or N-ethylmaleimide-stimulated KCl efflux. When cells were subjected to apoptotic stress, NHE1 and phosphorylated ERM physically associated within the cytoskeleton-enriched fraction, resulting in activation of the pro-survival kinase, Akt. NHE1-associated Akt activity and cell survival were inhibited in cells expressing ERM binding-deficient NHE1, dominant negative ezrin constructs, or ezrin mutants with defective binding to phosphoinositide 3-kinase, an upstream regulator of Akt. We conclude that NHE1 promotes cell survival by dual mechanisms: by defending cell volume and pHi through Na+/H+ exchange and by functioning as a scaffold for recruitment of a signalplex that includes ERM, phosphoinositide 3-kinase, and Akt.

Apoptosis is necessary for organogenesis, resolution of tissue injury, and pathobiologic cell deletion programs. Typical morphologic features of apoptotic cells include membrane blebbing, which is invariably followed by cytoplasmic shrinkage and chromatin condensation. Pathways regulating apoptotic volume decrease (AVD) 1 have been extensively studied and include activation of ion channels that mediate Na ϩ , K ϩ , and Cl Ϫ efflux (1). However, less is known about counter-regulatory pathways, which promote cell survival by maintaining intracellular volume.
In contrast to lymphocytes and mesenchymal cells, epithelial cells are relatively resistant to shrinkage from hypertonic stimuli or AVD (2,3), because of robust intracellular volume expansion through activation of regulatory volume increase (RVI) pathways (4 -7). Activation of the ubiquitously expressed NHE1 isoform of the Na ϩ /H ϩ exchanger is an important RVI component (5)(6)(7), resulting in Na ϩ influx and cytoplasmic volume expansion. Concomitant NHE1-dependent H ϩ extrusion also leads to intracellular alkalinization, which may defend against apoptosis by inhibiting caspase catalysis (8,9); by preventing conformational changes in Bax, a pro-apoptotic Bcl-2 family member (10); or by extruding cytosolic H ϩ resulting from apoptosis-induced mitochondrial H ϩ release (8).
We previously demonstrated that NHE1 is cleaved by caspase-3 in renal tubular epithelial cell apoptosis, and apoptosis in an NHE-deficient cell line was diminished by NHE1 reconstitution (11), indicating that NHE1 activity is critical for cell survival. Furthermore, mutant NHE1 expression studies demonstrated that domains regulating Na ϩ /H ϩ exchange, as well as interaction with ezrin/radixin/moesin (ERM) proteins, were required for cell survival (11). ERM proteins bind directly to the membrane-proximal NHE1 cytoplasmic domain (12), which tethers NHE1 to the cortical actin cytoskeleton and regulates Na ϩ /H ϩ transport-independent functions, such as maintenance of cell shape, focal adhesion formation, and specific aspects of cell migration (12,13), implying that NHE1 directs separable Na ϩ /H ϩ exchange-and ERM-dependent cell functions.
Linkage of NHE1 and ERM to downstream cell survival signaling pathways has not been described. ERM proteins were originally identified as molecular linkers between cytoskeleton and plasma membrane proteins and have therefore been viewed primarily as structural proteins. However, ezrin has subsequently been shown to directly interact with signaling enzymes, such as phosphoinositide 3-kinase (PI3K) (14 -16), suggesting that ERM proteins may also regulate cell phenotype by functioning as molecular scaffolds for assembly and integration of cytoskeleton-based signalplexes. Because PI3K phosphorylates diverse protein substrates, including precursors in a cascade that results in activation of the survival kinase Akt, we have hypothesized that cell survival is promoted by a signaling complex that includes NHE1, ERM, PI3K, and Akt. We show that apoptotic or hypertonic stress causes NHE1-mediated Na ϩ /H ϩ exchange, as well as activation of a new pathway that involves NHE1 interaction with ERM proteins within a cytoskeletal compartment, which up-regulates Akt activity and cell survival.
NHE1 Stimulation-NHE1 was activated by the addition of NH 4 Cl (50 mM for 25 min at 37°C) to stimulate cytosol acidification and immediate Na ϩ /H ϩ exchange following NH 4 Cl washout (19) or by hypertonic stress following the addition of sucrose to isotonic media (20).
pH i Measurement-Following experimental treatments, BCECF (1 M for 20 min at room temperature) was added to adherent cells, which were maintained in serum-free medium in a dark environment. The cells were lifted with trypsin-EDTA, pelleted, and washed in phosphatebuffered saline. BCECF fluorescence was measured immediately at 37°C with a SLM-Aminco model 8100 spectrofluorometer (Rochester, NY) at 439-nm excitation and 535-nm emission wavelengths. Calibrations were performed by adding pH buffers to cells permeabilized with nigericin (5 mM stock, 10 l in a 2-ml cuvette volume). pH i was calculated from fluorescence ratios using Intracellular Probe Math software (Rochester, NY). Apoptosis Assays-The cells were plated on glass coverslips at 0.25 ϫ 10 6 cells/ml density, grown to 80% confluence, and then maintained in serum-free medium with apoptotic stimuli. In some experiments, the cells were preincubated with PI3K or Akt inhibitors before apoptosis induction. Apoptotic cells were identified by simultaneous fluorescence labeling of chromatin with DAPI and externalized phosphatidylserine with annexin V as previously described (11,21,22). Random fields were viewed at 40ϫ magnification with a Nikon epifluorescence microscope (Tokyo, Japan), and the percentage of apoptotic cells was separately determined by two blinded observers from a total of 200 cells/experimental condition.
Plasmid Transfections-Plasmids were transformed into DH-5␣competent bacterial cells according to the manufacturer's protocol (Invitrogen), extracted using a Maxiprep kit (Qiagen), and amplified by culture in Luria-Bertani-ampicillin broth. cDNAs were transiently transfected into cells according to previously described methods (22). Briefly, the cells were plated into six-well plates (0.25 ϫ 10 6 cells/well) and cultured overnight in Dulbecco's modified Eagle's medium with Ham's F-12 medium plus 10% fetal bovine serum to achieve 80% confluence. The transfection mixture, which contained 2.0 g of plasmid DNA and 6 l of Fugene 6 transfection reagent (Roche Applied Science) in 100 l of serum-free Dulbecco's modified Eagle's medium (Invitrogen) was mixed for 20 min at room temperature and then added to each well with complete medium for 24 h. The cells were evaluated at 24 -48 h post-transfection.
Immunoprecipitation from Cytoskeleton Fractions-The cells were lysed in buffer containing 1% Nonidet P-40, 0.5% deoxycholic acid, 0.05% SDS, 50 mM HEPES, 135 mM NaCl, 3 mM KCl, 1 mM EDTA, and protease inhibitors (10 g/ml aprotinin, 10 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride) (4°C, 1 h). The lysates were centrifuged at 12,000 rpm for 30 min at 4°C. Cytoskeleton-rich pellets were resuspended in 100 l of 1% SDS and then diluted 1:10 in lysis buffer. The protein content was quantified, and lysates containing 700 g of protein were immunoprecipitated with anti-NHE1 antibodies (0.5 g/ml, 4°C, overnight) according to established methods (24). The samples were incubated with ␥-bind protein G-Sepharose beads (4°C for 1 h) and then washed with lysis buffer. Immunoprecipitates were dissolved in 25 l of 2ϫ SDS sample buffer, resolved by 4 -20% SDS-PAGE, and then evaluated as described under "Immunoblot Analysis." RNA Interference-Small interfering RNA (siRNA) inhibition of Akt expression was achieved with the SignalSilence Akt siRNA kit according to the manufacturer's instructions (Cell Signaling Technology, Beverly, MA). Briefly, LLC-PK1 cells at 50% confluence were incubated with Akt siRNA, negative control ␤8 integrin (24), or fluorescein-labeled, nonspecific siRNA (to assess transfection efficiency) (50 nM each). At 48 h post-transfection, transfection efficiency was verified to be Ͼ90%. The cells were then induced to undergo apoptosis, lysed, and probed for Akt expression by immunoblot analysis. The blots were stripped and reprobed for ␣-tubulin as a control for protein loading and nonspecific RNA interference effects. In parallel experiments, siRNAtransfected cells were assayed for apoptosis by annexin V labeling and DAPI staining of chromatin, as described above.
Statistics-The data are representative of three experiments/condition. The graphical results are expressed as the means Ϯ S.E. Comparisons between groups were made by one-way analysis of variance with the Student-Newman-Keuls or Kruskal-Wallis test for multiple comparisons of parametric and nonparametric data, respectively. Comparison between two groups was made by paired t test methods. Statistical significance is defined as p Ͻ 0.05.

NHE1
Defends against Apoptotic Stress-To assess whether NHE1 is activated in cells subjected to apoptotic stress, HRPT cells were stimulated with staurosporine in the presence and absence of NHE1 selective inhibitor EIPA. Intracellular pH was determined by BCECF-acetoxymethyl ester fluorescence. HRPT cells demonstrated initial intracellular alkalinization in response to staurosporine, which was blunted by EIPA ( Table  I), suggesting that NHE1-dependent RVI is a defense against apoptosis. Staurosporine exposure beyond 2 h was associated with cytosol acidification (Table I), in agreement with the kinetics of NHE1 degradation by caspase-3 (11). Similar results were observed following UV light exposure (30 J/m 2 ), a dissimilar stimulus of mitochondrial apoptosis (not shown). These data are consistent with a recent report by Bortner and Cidlowski (25), which shows that apoptotic stress induces a transient increase and then a profound decrease in Na i ϩ . To examine the specificity of NHE1 in cell survival, NHEdeficient PS120 cells were transiently transfected to express NHE1 or NHE3 and then assessed for sensitivity to staurosporine-induced apoptosis. Fig. 1 shows that NHE1 expression resulted in ϳ50% decrease in apoptosis, consistent with our prior report (11). Unlike NHE1, ectopic NHE3 expression did not rescue PS120 cells from apoptosis. These data are in agreement with observations that cell shrinkage stimulates NHE1 activity, whereas NHE3 activity is decreased (26 -28).
NHE1 Rescues Cells from KCl Efflux-stimulated Apoptosis-Many apoptosis models predict irreversible commitment to cell death following AVD, which is characterized by intracellular K ϩ and Cl Ϫ loss, even in the presence of counter-regulatory NHE1 activity (7). To address this issue, NHE-null PS120 cells, transfected with NHE1 cDNA, were incubated with NEM, a thiol compound that activates K ϩ -Cl Ϫ co-transport (29), to induce K ϩ and Cl Ϫ efflux. Fig. 2A demonstrates that NEM caused apoptosis, with a threshold effect at ϳ10 M, a concentration that diminished K i ϩ (not shown). Importantly, Fig. 2B shows that NHE1 expression significantly blunted NEM-mediated apoptosis, consistent with staurosporine data from Fig. 1. These studies demonstrate that NHE1 is important for cell survival, because activation was sufficient to block AVD, even under circumstances associated with robust K ϩ and Cl Ϫ efflux.
NHE1 Stimulation Leads to ERM Activation-To explore the role of ERM proteins in the mechanism of NHE1-regulated cell survival, HRPT cells were pulsed with NH 4 Cl to stimulate cytosol acidification and immediate Na ϩ /H ϩ exchange following NH 4 Cl washout (19, 30 -32). Inactive, cytosolic ERM requires phosphorylation at conserved carboxyl-terminal Thr residues (ezrin, Thr 567 ; radixin, Thr 564 ; and moesin, Thr 558 ) (33) to cause unfolding, translocation to the plasma membrane, and cross-linking between integral membrane proteins and cytoskeleton (34). Lysates from NH 4 Cl-treated cells were assayed for the active ERM conformation by immunoblot analysis with anti-phospho-ERM antibodies. Fig. 3A (upper panel) shows increased ERM protein phosphorylation within 5 min, which is sustained for 2 h. Enhanced phospho-ERM content was not due to increased ERM expression, as evidenced by equivalent ezrin expression at all time points (Fig. 3A, lower panel).
Because NHE3 activity is also (a) associated with ERM binding (35), (b) abundant in proximal tubule in vivo, and (c) activated by NH 4 Cl pulse, HRPT cells were screened for NHE3 expression by immunoblotting. These studies revealed undetectable HRPT cell NHE3 protein expression by immunoblot analysis (not shown), consistent with suppressed NHE3 expression in many proximal tubule cells lines. 2 Specificity of ERM regulation by NHE1 was further verified by hypertonic  1. NHE1 defends against apoptotic stress. NHE-deficient PS120 cells were transiently transfected with empty vector, NHE1, or NHE3 cDNAs (2 g/well). Transfection efficiencies were equivalent between groups (not shown). NHE1-and NHE3-expressing cells were stimulated with STS (5 M for 5 h at 37°C). The percentages of apoptotic cells were determined by annexin V labeling. *, p Ͻ 0.05 compared with other two groups by ANOVA. shrinkage, which activates NHE1 and inhibits NHE3 (26 -28). The addition of 100 mM sucrose to cell culture medium induced peak ERM phosphorylation at 15 min, which remained sustained at 60 min (Fig. 3B).
Specificity of ERM phosphorylation following NHE1 activation was also assessed by comparing phospho-ERM content in NH 4 Cl-stimulated NHE-deficient PS120 fibroblasts versus PS120 cells transiently transfected to ectopically express NHE1. As expected, hypertonicity caused minimal ERM phosphorylation in PS120 cells (Fig. 3C, upper panel), whereas NHE1-transfected cells exhibited significantly greater basal and stimulated ERM phosphorylation, with a kinetic pattern similar to HRPT cells (Fig. 3C, lower panel). Furthermore, phospho-ERM content was significantly diminished in cells pretreated with 5 M EIPA, suggesting that NHE1-dependent Na ϩ /H ϩ exchange is linked to ERM activation. Collectively, the Fig. 3 data provide evidence that NHE1 activation stimulates ERM phosphorylation.
NHE1 and ERM Directly Interact in the Cytoskeleton Fraction-To determine whether NHE1-regulated ERM phosphorylation leads to cytoskeleton-based, physical NHE1-ERM interaction, NHE1 was stimulated with either NH 4 Cl (Fig. 4A) or sucrose (Fig. 4B) and then co-precipitated with phospho-ERM from Nonidet P-40-insoluble cytoskeleton fractions. Fig. 4 shows NHE1 interaction with phosphorylated ERM in the cytoskeleton-enriched compartment. NHE1-phospho-ERM binding was time-dependent, with peak interaction at 15 min, which was sustained for at least 60 min. The data represent the first evidence of direct association between NHE1 and phospho-ERM proteins.

NHE1-Ezrin Interaction Protects against Apoptosis-
We have previously demonstrated that cells expressing a mutant NHE1 construct, with nonconserved Ala substitutions for membrane-proximal Lys and Arg residues (KR/A), which impairs binding to negatively charged proteins, such as ERM (12), were more susceptible to apoptosis (11). To determine whether NHE1-dependent cell survival requires ERM interaction, LLC-PK1 cells stably transfected to express constitutively active and dominant negative ezrin were induced to undergo apoptosis by anisotonic shrinkage (Fig. 5A) or staurosporine exposure (Fig. 5B). Hypertonicity caused modest dose-dependent apoptosis in wild-type ezrin-expressing cells (Fig. 5A), consistent with previous results for hypertonicity-induced apoptosis in other kidney tubule cell lines (11). In cells expressing constitutively active ezrin (T567D), apoptosis was suppressed, whereas cells expressing dominant negative ezrin mutants (nonphosphorylatable point mutant (T567A), amino-terminal ezrin polypeptide, and Y353F point mutant that prevents PI3K binding) exhibited enhanced apoptosis (Fig. 5A). In cells expressing wild-type ezrin, staurosporine caused ϳ30% apoptosis (Fig. 5B), whereas cells expressing constitutively active ezrin (T567D) were relatively resistant to apoptosis, and dominant negative ezrin (T567A) expression accentuated apoptosis in response to staurosporine. Taken together, data from Figs. 4 and 5 demonstrate that activation and physical association of NHE1 and ezrin rescues cells from apoptosis, suggesting that NHE1-ERM interaction is an upstream step in the formation of a cell survival signalplex.
Because LLC-PK1 cells expressing Y353F ezrin mutants (Fig. 5A), which inhibit ERM-PI3K interaction, displayed increased susceptibility to apoptosis, the role of PI3K in LLC-PK1 cell apoptosis was specifically tested in cells pretreated with dissimilar PI3K inhibitors, wortmannin and LY294002. Fig. 5C demonstrates that like HRPT cells (11), untransfected LLC-PK1 cells were susceptible to staurosporine-induced apoptosis. Basal and staurosporine-stimulated apoptosis were enhanced by both PI3K inhibitors, consistent with a mechanism whereby PI3K promotes cell survival through interaction with the NHE1-ERM complex.
Apoptosis Is Regulated by Akt Activity-Although PI3K phosphorylates a variety of substrates, Akt is a well described effector molecule for PI3K and a key signaling enzyme in many cell survival pathways (36). To determine whether Akt regulates cell survival in our system, HRPT cells were pretreated with a chemical Akt inhibitor (Akt I) and then assessed for apoptosis. Fig. 6A reveals that Akt I accentuated staurosporine-induced apoptosis. Similar results were observed in cells induced to undergo apoptosis by hypertonic stimulation (not shown). To test for the role of Akt in cell survival by a different strategy, LLC-PK1 cells were pretreated with Akt siRNA, which resulted in significant suppression of Akt expression (Fig. 6B). Fig. 6C shows enhanced staurosporine-induced apoptosis in the Akt siRNA-treated group, indicating that Akt regulates cell survival. The data are consistent with a role for Akt as a downstream effector in a survival cascade that includes NHE1, ERM, and PI3K.
NHE1 Stimulation Regulates Akt Activity-To determine whether NHE1 activation is linked to Akt activity, NHE1 was stimulated by NH 4 Cl protocol, and cell lysates were probed for Akt activity with anti-phospho-Akt antibodies. These experiments revealed time-dependent Akt phosphorylation, peaking after 30 -60 min of NHE1 stimulation (Fig. 7A). The slower kinetics of Akt phosphorylation compared with ERM phosphorylation (Fig. 3A) suggests that Akt is downstream from ERM in the cell survival cascade. NHE1 activation by hypertonic sucrose stress also induced a 3-fold increase in phospho-Akt content (Fig. 7B). Anisotonic stress-induced Akt phosphorylation was diminished by selective NHE1 inhibition with EIPA (Fig. 7B), suggesting that NHE1-dependent Na ϩ /H ϩ exchange also influences Akt activity. NHE1 Regulates Akt by an ERM-dependent Mechanism-To establish that NHE1 and ERM proteins are linked to Akt activity, NHE1 was activated by NH 4 Cl pulse acidification in LLC-PK1 cell lines stably expressing wild-type and dominant negative ezrin constructs, and Akt activation was assessed by phospho-Akt immunoblot analysis. In wild-type, ezrin-transfected cells, Akt phosphorylation peaked at 30 min and remained increased at 60 min (Fig. 8A), consistent with data in HRPT cells (Fig. 7). Importantly, Akt was not activated by NHE1 stimulation in cells expressing dominant negative ezrin (Fig. 8A).
To confirm by a different approach that Akt is a downstream effector in an NHE1/ERM interaction-initiated cascade, LLC-PK1 cells rendered NHE-deficient were transiently transfected to express wild-type NHE1, the KR/A NHE1 mutant that does not bind ERM proteins (12), or the Na ϩ /H ϩ translocationdefective NHE1 point mutant, E266I (12) (Fig. 8B); then stimulated with NH 4 Cl or sucrose; and probed for Akt phosphorylation by immunoblot analysis. NHE1 stimulation in wild-type FIG. 4. NHE1 and ERM directly interact in the cytoskeleton fraction. To determine whether NHE1 and ERM physically interact, HRPT NHE1 was stimulated by NH 4 Cl and then permitted to recover for indicated time periods (A) or by sucrose for the indicated times (B). The cells were lysed in Nonidet P-40 buffer and centrifuged, and cytoskeleton-rich pellets were lysed in 1% SDS buffer. The lysates were immunoprecipitated with affinity-purified anti-NHE1 IgG, resolved by 8% SDS-PAGE, and immunoblotted with anti-phospho-ERM IgG. NHE1-or E266I-expressing cells resulted in increased phospho-Akt content (Fig. 8C). However, stimulation of KR/A NHE1 mutant did not cause Akt activation, thereby implicating ERM binding to activated NHE1 as a mechanism for Akt-dependent cell survival.

DISCUSSION
NHEs are expressed in all mammalian cells, and the ubiquitously expressed NHE1 isoform is most extensively characterized as a regulator of cell volume and pH i . Because apoptotic cells undergo shrinkage and cytosol acidification, processes consistent with NHE1 inactivation, we have proposed that NHE1 is a cell survival factor. Our previous work has demonstrated that NHE1 expression and Na ϩ /H ϩ exchange are associated with resistance to apoptosis (11). Cell survival in NHEdeficient cells was incompletely restored upon expression of NHE1 mutants with impaired ERM protein binding capacity, suggesting that NHE1-ERM interaction is required for cell survival (11). The current studies extend these findings by demonstrating that NHE1 is activated by apoptotic stress, which leads to direct interaction with ERM in the cytoskeletonenriched cell fraction, and this interaction mediates recruitment of PI3K and Akt to regulate cell survival.
Previous studies demonstrate that NHE1-null cells are more sensitive to apoptosis compared with wild-type cells (11,37), supporting a direct role for NHE1 in survival. The relative importance of NHE1 as a survival factor is established in the current studies by showing that ectopic NHE1 expression partially rescued cells induced to undergo apoptosis following cel- 5. NHE1-ezrin interaction protects against apoptosis. A, to determine whether NHE1 regulates ERM proteins and PI3K, NHE1 was stimulated with sucrose (6 h at 37°C) addition to medium in LLC-PK1 cell lines stably transfected with wild-type ezrin, constitutively active (T567D) ezrin, dominant negative ezrin (T567A), amino-terminal ezrin (N-term) ezrin, or Y353F ezrin that prevents PI3K interaction. B, wild-type, T567D, and T567A cells were incubated with or without STS (5 M for 6 h). *, p Ͻ 0.05 compared with wild-type, STS-treated cells by ANOVA. lular K ϩ and Cl Ϫ efflux. In contrast to previous conceptions that apoptosis-stimulated, intracellular K ϩ and Cl Ϫ depletion irreversibly commit cells to die, we demonstrate that NHE1 activation prevents apoptosis in the context of concurrent robust ion efflux.
At least nine NHE isoforms have been identified, and all of the isoforms regulate electroneutral Na ϩ influx and H ϩ efflux, which could account for cell survival. NHE1 and NHE3 were the focus of our studies because both isoforms are expressed in kidney proximal tubule, from which cell lines for most of the experiments were derived, and both NHEs interact with ERM proteins (12,35). Initial studies demonstrated transient intracellular alkalinization in response to apoptotic stress, consistent with either NHE1 or NHE3 activation. However, pH i changes were blunted by EIPA at concentrations that inhibit NHE1 but not NHE3 activity (26). Furthermore, apoptosis was rescued by ectopic NHE1 but not NHE3 expression, indicating that cell survival is NHE1-specific. Because NHE1 and NHE3 display divergent responses to anisotonic shrinkage, with cell volume reduction stimulating NHE1 and inhibiting NHE3 activity (26 -28), selective rescue of apoptosis by NHE1 is in agreement with an NHE1-specific, cytoplasmic volume preservation mechanism of cell survival.
Cell survival signaling was initiated by interaction between NHE1 and ERM proteins. Although originally identified as cytoskeleton cross-linkers and important components of cell structure, ERM proteins are now recognized to also assemble and integrate actin-based signaling modules, with diverse downstream effects, that include adhesion, migration, secretion, and immunological synapse formation (16,38). We speculate that NHE1 regulates cytoskeleton linkage to plasma membrane to maintain cell volume and shape, which facilitate proper enzyme-substrate targeting and spacing (39), as well as avoidance of molecular crowding, which has been implicated in osmotic shrinkage and spontaneous caspase activation (40,41). In this regard, NHE1 interaction with ERM was specifically localized to cytoskeleton, indicating that compartmentalization is required for activation of appropriate downstream survival effectors. Studies with cells, either lacking NHE1, or with impaired NHE1-ERM interaction did not support cell survival, suggesting that signaling molecules were mislocated and/or excluded from the signaling complex. Cytoskeleton-based, NHE1-directed signaling mechanism is analogous to the focal adhesion complex, which clusters specific proteins, including NHE1 (12,13,42) to regions of cytoskeleton-plasma membrane contact. Because focal adhesion formation also promotes cell survival (43), we propose that NHE1 may be a key integrator for multiple, cooperative signaling cascades, which culminate in cell survival.
In the inactive state, ERM proteins reside within the cytosol in a closed conformation through head-to-tail interactions between the amino-and carboxyl-terminal domains. ERM proteins become activated following phosphorylation at conserved carboxyl-terminal Thr residues and interaction with negatively charged phosphatidylinositol 4,5-bisphosphate (PIP 2 ), which causes protein unfolding and targeting to plasma membrane (33,38). The mechanism of ERM activation in cell survival was not specifically investigated, but Rho kinase and protein kinase C have been shown to phosphorylate ERM (33,38). In addition, the polybasic, membrane-proximal NHE1 domain that binds ERM, also binds PIP 2 (44), suggesting that PIP 2 could coordinately regulate assembly of an NHE1-ERM survival signalplex. Although we cannot exclude that enhanced apoptosis in NHE1 KR/A mutant-expressing cells could be due to impaired interaction with negatively charged molecules, such as PIP 2 , rather than ERM, concordant findings in experiments with NHE1 KR/A and dominant negative ERM mutants support a role for ERM as an intermediary between NHE1 and cell survival signaling.
The hypothesis that NHE1-ERM interaction promotes cell FIG. 9. NHE1-dependent mechanisms of cell survival. Apoptotic stress elicits two major, NHE1-regulated cell survival pathways. Decreased cytoplasmic volume leads to NHE1-dependent Na ϩ /H ϩ exchange, which mediates regulatory volume increase and cytosolic alkalinization. NHE1 activation also stimulates phosphorylation and recruitment of cytoskeleton ERM linkers to the NHE1 cytosolic tail. The NHE1-ERM interaction leads to the formation of a signaling complex that includes phosphoinositide 3-kinase and ultimately, the pro-survival kinase, Akt, which phosphorylates multiple substrates, some of which are depicted in the figure, resulting in apoptosis inhibition. Apaf1, apoptotic protease-activating factor-1; BAD, pro-apoptotic Bcl-2 family member; FKHR, forkhead transcription factor; PH, pleckstrin homology domain; PIP3, phosphatidylinositol 3,4,5-trisphosphate. FIG. 8. NHE1 regulates Akt by an ERM-dependent mechanism. A, NHE1 was stimulated by NH 4 Cl in LLC-PK1 cells stably transfected with wild-type (WT) or dominant negative (T567A) ezrin. Akt activity was determined 30 and 60 min post-NH 4 Cl recovery by immunoblotting with anti-phospho-Akt antibodies. The blots were stripped and reprobed with anti-Akt1 antibodies as a loading control. B, LLC-PK1 cells rendered NHE-deficient by H ϩ suicide protocol were transiently transfected to express HA-tagged, wild-type NHE1 (WT), ERM binding-defective NHE1 (KR/A), or Na ϩ /H ϩ -defective NHE1 (E266I). The cell lysates were immunoprecipitated with anti-NHE1 antibodies and then probed by immunoblot analysis for ezrin (upper panel) and HA (lower panel) expression. C, NHE1 was stimulated in NHE-deficient LLC-PK1 cells transiently transfected to express wild-type (WT), KR/A, or E266I NHE1 mutants as in B. The lysates were probed by immunoblot analysis for phospho-Akt or Akt1 expression.
survival was predicated upon the observation that ERM proteins directly bind to PI3K (14,38). Multiple stimuli can activate PI3K, which then phosphorylates numerous substrates, thereby regulating cell survival, cytoskeletal rearrangement, and transformation (45). PI3K catalyzes phosphorylation of PIP 2 to PIP 3 , which serves as a docking site for pleckstrin homology domain proteins, including Akt. Activated Akt facilitates cell growth, cell cycle entry, and cell migration, but the best described Akt function is mediation of cell survival, through phosphoregulation of multiple apoptosis-related proteins (13,36). Interestingly, Akt has also been shown to regulate cell volume (45)(46)(47), suggesting that NHE1-dependent RVI and cell survival may be integrated through Akt signaling pathways. Our data indicate that NHE1-regulated ERM binding is critical for Akt activity, but the role of NHE1-dependent Na ϩ /H ϩ exchange in Akt activation is less clear, because EIPA preincubation inhibited Akt phosphorylation (Fig. 7), whereas cell expressing the Na ϩ /H ϩ -deficient E266I NHE1 mutant maintained the capacity to activate Akt (Fig. 8). This discrepancy suggests that EIPA may exert effects beyond the inhibition of Na ϩ /H ϩ translocation.
We previously demonstrated that NHE1 is a caspase substrate and that upon apoptotic cleavage, the loss of NHE1 function may have pathophysiologic consequences (11). Based upon our previous and current work, we propose that NHE1 is a critical cell survival factor, by mechanisms that involve Na ϩ /H ϩ exchange to induce RVI and pH i homeostasis, as well as through ERM protein binding that leads to sequential PI3K and Akt kinase activation (Fig. 9). NHE1 is commonly described as a housekeeping protein, which infers that it is merely a pedestrian molecule or a convenient control for biochemical experiments. However, many housekeeping tasks are required for normal cell function, and the current studies demonstrate that a vital NHE1 task is the regulation of cell survival.