Na+/H+ Exchanger NHE3 Activity and Trafficking Are Lipid Raft-dependent*

A previous study showed that ∼25–50% of rabbit ileal brush border (BB) Na+/H+ exchanger NHE3 is in lipid rafts (LR) (Li, X., Galli, T., Leu, S., Wade, J. B., Weinman E. J., Leung, G., Cheong, A., Louvard, D., and Donowitz, M. (2001) J. Physiol. (Lond.) 537, 537–552). Here, we examined the role of LR in NHE3 transport activity using a simpler system: opossum kidney (OK) cells (a renal proximal tubule epithelial cell line) containing NHE3. ∼50% of surface (biotinylated) NHE3 in OK cells distributed in LR by density gradient centrifugation. Disruption of LR with methyl-β-cyclodextrin (MβCD) decreased NHE3 activity and increased K′(H+)i, but Km(Na+) was not affected. The MβCD effect was completely reversed by repletion of cholesterol, but not by an inactive analog of cholesterol (cholestane-3β,5α,6β-triol). The MβCD effect was specific for NHE3 activity because it did not alter Na+-dependent l-Ala uptake. MβCD did not alter OK cell BB topology and did not change the surface amount of NHE3, but greatly reduced the rate of NHE3 endocytosis. The effects of inhibiting phosphatidylinositol 3-kinase and of MβCD on NHE3 activity were not additive, indicating a common inhibitory mechanism. In contrast, 8-bromo-cAMP and MβCD inhibition of NHE3 was additive, indicating different mechanisms for inhibition of NHE3 activity. Approximately 50% of BB NHE3 and only ∼11% of intracellular NHE3 in polarized OK cells were in LR. In summary, the BB pool of NHE3 in LR is functionally active because MβCD treatment decreased NHE3 basal activity. The LR pool is necessary for multiple kinetic aspects of normal NHE3 activity, including Vmax and K′(H+)i, and also for multiple aspects of NHE3 trafficking, including at least basal endocytosis and phosphatidylinositol 3-kinase-dependent basal exocytosis. Because the C-terminal domain of NHE3 is necessary for its regulation and because the changes in NHE3 kinetics with MβCD resemble those with second messenger regulation of NHE3, these results suggest that the NHE3 C terminus may be involved in the MβCD sensitivity of NHE3.

Ϫ , and water (re)absorption (1)(2)(3). Rapid regulation of NHE3 activity occurs as part of normal digestive and renal physiology and in the pathophysiology of diarrhea and some renal diseases of the proximal tubule. The acute regulation of the exchanger seems to be mainly through changes in its V max , but also involves changes in KЈ(H ϩ ) i (4). Regulation of NHE3 involves at least two different mechanisms: regulation by changes in trafficking due to regulated changes in endocytosis and/or exocytosis and changes in turnover number (5)(6)(7)(8)(9)(10)(11). Both these mechanisms often involve changes in NHE3 phosphorylation.
In studies performed to investigate the mechanisms of NHE3 regulation in rabbit ileal Na ϩ absorptive cells, brush border (BB) 4 NHE3 was shown to be partially in lipid rafts (LR) (12). Concerning NHE3 regulation, the LR pool of BB NHE3 is involved in some of its basal endocytosis and exocytosis and in the acute epidermal growth factor increase of the BB amount of NHE3. However, the contribution of LR to NHE3 function has not been examined in detail.
LR are discrete membrane domains that are enriched in glycosphingolipids and cholesterol and that are resistant to solubilization in cold Triton X-100. They are thought to act in the compartmentalization of membrane proteins, separating different biochemical functions and allowing concentration and localization of molecules involved in signal transduction functions (13)(14)(15). Besides the formation of restricted signaling platforms, rafts are implicated in apical protein targeting (13,14,16) and in some aspects of endocytosis in epithelial cells and as a docking site for some pathogens and toxins (17)(18)(19).
The concept that transport proteins distribute in LR and that their activities are LR-dependent is not unique to NHE3, although it has not yet been examined for many transport proteins. Depletion of cholesterol dramatically alters the function of some (Kv2.1 and Kv1.5) but not other (Kv4.2) voltage-gated potassium channels (20,21), decreases SGLT1 (sodium/glucose cotransporter 1) activity (22), and significantly reduces uptake of glutamate by the glial glutamate transporter EAAT2 (23) and the NaCl-dependent serotonin transporter SERT (24). Also, the mouse colonic basolateral membrane Ca 2ϩ -activated potassium channel is activated by cholesterol depletion (25). Other transporters shown to be partially in LR include NHE1 (26), the type IIa Na ϩ /P i cotransporter (27), some connexins (28), and the epithelial Na ϩ channel ENaC (29). In contrast, other transport proteins do not appear to be present or affected by LR. These include the cystic fibrosis transmembrane conductance regulator CFTR in normal tissue, except when exposed to Pseudomonas aeruginosa toxin (30), and intestinal Na ϩ -K ϩ -ATPase (31).
In this study, the opossum kidney (OK) renal proximal tubule epithelial cell line was used to provide a simple model to examine the contribution of LR to NHE3 activity. An advantage of this cell line is that it is a polarized epithelial Na ϩ absorptive cell line that contains NHE3 as the sole plasma membrane Na ϩ /H ϩ exchanger. It also lacks other regulatory elements (nerves, endocrine cells, inflammatory cells) that are present in intact intestine and that might also act by LR-dependent processes. Cell Lines-Studies were carried out in OK/E3V (generously provided by Dr. J. Noël, University of Montreal, Montreal, Canada) and OK/3HA-E3V cell lines. OK/E3V cells are an OK proximal tubule cell line generated by stable transfection of OK-Tina cells, which are OK cells previously selected by acid suicide to lack endogenous NHE3 activity, with a cDNA for rat NHE3 tagged at the C terminus with the vesicular stomatitis virus G (VSV-G) protein epitope (32). The OK/3HA-E3V cell line stably expresses rabbit NHE3 tagged with three copies of the influenza virus HA epitope at the N terminus and with the VSV-G epitope at the C terminus. Cells were cultured in Dulbecco's modified Eagle' medium (DMEM)/nutrient mixture F-12 (Invitrogen) containing 10% (v/v) fetal bovine serum, 100 units/ml penicillin, and 100 g/ml streptomycin in a humidified atmosphere of 95% air and 5% CO 2 at 37°C. Confluent monolayers on plastic dishes, glass coverslips, or filters were serum-depleted for 24 -48 h before study. Cells from every new passage were exposed to an acute acid loading selection to maintain a high level of NHE3 protein expression as described previously (4) with some modification. In brief, cells were exposed to 50 mM NH 4 Cl/saline solution for 1 h, followed by overnight incubation in isotonic 2 mM Na ϩ solution.

Materials
Plasmid Construction and Cell Transfections-For immunological detection (enzyme-linked immunosorbent assay (ELISA)) of NHE3 protein, three copies of the HA epitope (YPYDVPDYA) were inserted into the first extracellular loop of rabbit NHE3 between Glu 37 and Ile 38 by PCR. The pECE plasmid with cDNA encoding rabbit NHE3 with the VSV-G epitope at the C terminus as described (8) and containing unique HindIII and XhoI restriction endonuclease sites inserted into the coding region of NHE3 and XbaI after the VSV-G epitope was used as template in the PCRs. Four primers were designed to introduce the triple HA epitope: sense primer 1 (CCCCAAGCTTATGTCAGGGCGCGGGGGCTGCGGCCC, containing the HindIII endonuclease restriction site (underlined)) and antisense primer 1 (CGCGGATCCagcgtagtcggggacgtcgtaggggtaACCagcgtagtcggg-gacgtcgtaggggtaACCCTCATCGTGATGCTCCTGCTC, containing the first and second HA epitopes (lowercase) and the BamHI restriction endonuclease site (underlined); and sense primer 2 CGCGGATCCtac-ccctacgacgtccccgactacgctGGACGCGTGATCCAGGGCTTCCAGA-TAGTC, containing the third HA epitope (lowercase) and BamHI and MluI restriction endonuclease sites (underlined)) and antisense primer 2 (CTAGTCTAGATTGGTACCTT, containing the XbaI restriction site (underlined)). The primers were extended with Pfu polymerase, resulting in the generation of two mutated cDNAs containing the desired insertion. The two DNA fragments obtained by PCR were ligated to each other and then subcloned into the HindIII-XbaI cloning site of pcDNA3.1(ϩ) (Invitrogen). All constructs studied were sequenced. Mutated NHE3 cDNA was called 3HA-E3V.
The pcDNA3.1/3HA-E3V and empty vector (pcDNA3.1(ϩ)) plasmids were stably transfected into OK-Tina cells using Lipofectamine 2000 (Invitrogen) as recommended by the manufacturer. Several stably transfected clones were picked, expanded, and used for ELISA as well as for transport assays.
The cDNA encoding SGLT1 subcloned into the pEGFP-N1 vector was a gift from Dr. Suketa (University of Shizuoka, Shizuoka, Japan). For transient expression of SGLT1, OK/E3V cells were seeded at 75-80% confluence onto 24-well plates 24 h prior to transfection and then transfected with 1 g of the corresponding plasmid DNAs using Lipofectamine 2000 as described above. After 6 h of incubation with the DNA-lipid complexes, the cells were exposed to serum-containing DMEM/nutrient mixture F-12. 3 days after transfection, cells were used for Na ϩ -dependent glucose uptake assays.
Measurement of Na ϩ /H ϩ Exchange Activity-Na ϩ /H ϩ exchange activity was determined as the initial rate of Na ϩ -induced recovery of cytosolic pH (pH i ) after an acute acid load caused by prepulsing with NH 4 Cl, and pH i was measured fluorometrically using BCECF-AM as described previously (33). Fluorescence measurements (excitation at 490 and 440 nm with emission at 530 nm) were made using SLM-Aminco SPF-500C and Photon Technology International spectrofluorometers. Briefly, OK/E3V or OK/3HA-E3V cells were grown on glass coverslips to 100% confluency. The monolayers were incubated in serum-free DMEM/nutrient mixture F-12 for 24 -48 h prior to use. Cells were loaded with 10 M BCECF-AM in Na ϩ /NH 4 Cl medium (88 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgSO 4 , 1 mM NaH 2 PO 4 , 25 mM glucose, 20 mM HEPES, and 50 NH 4 Cl, pH 7.4) for 30 min at 37°C. During the dye loading and NH 4 Cl prepulse, cells were treated with test agents or vehicle. The cells were initially perfused with TMA ϩ medium (130 mM tetramethylammonium (TMA) chloride, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgSO 4 , 1 mM NaH 2 PO4, 25 mM glucose, and 20 mM HEPES, pH 7.4), resulting in stable acidification of the cells. Then, Na ϩ medium (138 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgSO 4 , 1 mM NaH 2 PO 4 , 25 mM glucose, and 20 mM HEPES, pH 7.4) was added, which induced alkalinization of the cells. To determine the kinetics for external Na ϩ ions, the Na ϩ concentration in Na ϩ medium was varied (5, 10, 15, 75, and 138 mM) while maintaining the osmolarity with TMA chloride. For these experiments, cells were acidified to the same level with 50 mM NH 4 Cl. To calibrate the relationship between the excitation ratio (F 500/450 ) and pH i , the K ϩ /nigericin method was used. As described previously (33), Na ϩ /H ϩ exchange rates (H ϩ efflux) were calculated as the product of Na ϩ -dependent change in pH i and the buffering capacity at each pH i and were analyzed using the nonlinear regression data analysis program Origin, which allows fitting of data to a general allosteric model described by the Hill equation where v is velocity, [S] is the substrate concentration, n app is the apparent Hill coefficient, and K is the affinity constant), with estimates for V max and KЈ(H ϩ ) i and their respective errors (S.E.), as well as fitting to a hyperbolic curve such as would be expected with Michaelis-Menten kinetics. Data from each coverslip were calculated and analyzed as described above. For each independent experiment, results from all coverslips for each condition were analyzed together.
Uptake Studies-Uptake of MGP or L-alanine was assayed in the presence and absence of Na ϩ as described previously (34,35). For uptake experiments, OK/E3V cells were plated onto 24-well plates. Confluent cell monolayers were incubated in serum-free medium for 48 h before the uptake experiments. Transiently transfected OK/E3V cells (for MGP uptake) were used ϳ72 h after transfection and incubated in serum-free DMEM/nutrient mixture F-12 for 6 h before study. Cell monolayers were treated with 10 mM M␤CD or with H 2 O as a vehicle for 30 min at 37°C and then washed twice with TMA ϩ medium (no glucose). MGP or L-alanine uptake was carried out at room temperature and initiated by addition of 0.1 mM MGP/[ 14 C]MGP (0.4 Ci/ml) or 0.2 mM L-alanine/L-[ 3 H]alanine (2 Ci/ml). Uptake of substrates was arrested after an appropriate incubation time by aspirating off the radioactive medium and washing three times with ice-cold TMA ϩ medium without substrate. The radioactivity of isotopes extracted from cell monolayers with 0.5 ml of 1 N NaOH (neutralized with HCl) was assayed by liquid scintillation spectrometry. The amount of accumulated substrate is expressed as cpm/min/well.
Measurement of Surface NHE3-The percentage of total cell NHE3 on the apical surface of OK cells was determined separately by cell-surface biotinylation and modified ELISA (7). For biotinylation, OK/E3V cells were grown to confluent monolayers on plastic dishes, and then the growth was arrested by incubation with serum-free medium for 48 h. Confluent monolayers were treated either with test agent or vehicle at 37°C under the same conditions used for the detection of NHE3 transport activity and surface-labeled with biotin at 4°C as described previously (8). All subsequent manipulations were performed at 4°C. Cells were washed twice with phosphate-buffered saline (PBS; 150 mM NaCl and 20 mM Na 2 HPO 4 , pH 7.4); incubated with arginine-and lysine-reactive succinimidyl 2-(biotinamido)ethyl-1,3-dithiopropionate (0.5 mg/ml; Pierce) in 154 mM NaCl, 10 mM boric acid, 7.2 mM KCl, and 1.8 mM CaCl 2 , pH 9.0; and washed extensively with quenching buffer containing 20 mM Tris and 120 mM NaCl, pH 7.4, to scavenge the unbound biotin. Cells were solubilized with 1 ml of N ϩ buffer (60 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM KCl, 5 mM Na 3 EDTA, 3 mM EGTA, and 1% Triton X-100) and protease inhibitor mixture (catalog number P8340, Sigma), and lysates were centrifuged at 2300 ϫ g for 20 min to remove insoluble cell debris and unbroken cells. Supernatants were diluted with N ϩ buffer to an equal protein concentration, applied to avidin-agarose beads (Pierce) at 4°C, and incubated for 16 h. The remaining supernatant was retained as the intracellular fraction. Finally, the avidin-agarose beads were washed five times with N ϩ buffer, and the biotinylated proteins were recovered from the beads in Laemmli buffer. Several fractions of total, intracellular, and surface pools were separated by SDS-PAGE (9%) and transferred onto nitrocellulose membranes (Schleicher & Schüll). The membranes were first incubated with anti-VSV-G monoclonal antibody (P5D4 hybridoma supernatant) as the primary antibody and horseradish peroxidase-conjugated anti-mouse IgG as the secondary antibody. Immunoreactive bands were detected by enhanced chemiluminescence (ECL kit, PerkinElmer Life Sciences). The films were scanned, and the signals were quantified using ImageQuant Version 4.2a software.
For ELISA, OK/3HA-E3V cells were used. Cells were plated onto 24-well plates and grown to confluency. 24 -48 h before the experiments, cells were incubated in serum-free growth medium. Monolayers were exposed to 10 mM M␤CD or vehicle for 30 min at 37°C. Cells were incubated with anti-HA antibodies (1:1000 dilution) for 1.5 h at 4°C to prevent endocytosis. The cells were washed up to six times with 1:9 (v/v) cold growth medium/PBS to remove unbound antibodies and then fixed in 3% formaldehyde in PBS for 10 min at room temperature. The cell monolayers were washed three times with PBS, incubated in 100 mM glycine in PBS for 15 min at room temperature, and then blocked with PBS containing 5% fetal bovine serum and 1% bovine serum albumin at room temperature for 30 min. Following blocking, the cells were incubated with horseradish peroxidase-conjugated anti-mouse antibodies (1:500 dilution) for 1 h at room temperature and washed six times with growth medium/PBS. For detection of peroxidase activity, 1 ml of oPD was added to each well and incubated for 15 min at room temperature, and the reaction was stopped by 0.75 M HCl. The supernatants were collected, and the absorbance was measured at 492 nm. The absorbance varied linearly with the amount of peroxidase bound. Serial dilution titration analyses were performed to determine the optimal concentration of reagents used in the ELISA as described (36). The optimal concentrations of primary and secondary antibodies as well as developing reagent (oPD) and cells were serially diluted and analyzed by crisscross matrix analysis. The analyses showed optimal dilutions of 1:1000 and 1:500 for anti-HA primary antibodies and horseradish peroxidase-conjugated anti-mouse secondary antibodies, respectively, as well as 1 ml for oPD.
NHE3 Endocytosis-The reduced GSH-resistant endocytosis assay described previously by us (37) was used with slight modifications. OK/E3V cells were labeled with 1.5 mg/ml sulfosuccinimidyl 2-(biotinamido)ethyl-1,3-dithiopropionate for 1 h at 4°C and quenched at 4°C. The cells were then warmed to 37°C and treated with 10 mM M␤CD or vehicle for 1 h at 37°C. Cells were rinsed with 3ϫ ice-cold phosphatebuffered saline at 4°C, and then surface biotin was cleaved by washing with 50 mM Tris-HCl and 150 mM GSH, pH 8.8. The freshly endocytosed, biotin-labeled proteins were protected from cleavage with GSH. Cells were solubilized in N ϩ buffer; biotinylated proteins were retrieved and assayed for endocytosed NHE3 as described above. Fluorescently labeled, IRDye TM 800-conjugated goat anti-mouse secondary antibodies (Rockland Immunochemicals, Inc.) were used for immunoblotting. The fluorescence intensity of NHE3 protein bands was visualized using the Odyssey system (LI-COR Biosciences) and quantitated with Meta-Morph Version 5.0r1 software (Universal Imaging Corp., Downingtown, PA).
Sucrose Gradient Density Flotation-To localize NHE3 in OK/E3V cells to LR, total lysate was fractionated by discontinuous sucrose step gradients as described previously (12,38). OK/E3V cells were grown to 100% confluency, serum-starved for 24 -48 h, and then treated with 10 mM M␤CD or vehicle for 30 min at 37°C. The monolayers were biotinylated (see "Measurement of Surface NHE3") and lysed in N ϩ buffer supplemented with 5 mM dithiothreitol, 1 mM Na 3 VO 4 , 50 mM NaF, and protease inhibitor mixture. Total lysates were loaded on 11 discontinuous sucrose step gradients (30, 27.5, 25, 22.5, 20, 17.5, 15, 12.5, 10, 7.5, and 5%). Each step gradient was prepared with sucrose and N ϩ buffer with 0.1% Triton X-100. Centrifugation was done in a Beckman SW 41Ti rotor at 150,000 ϫ g overnight at 4°C. Surface NHE3 in each fraction was precipitated by avidin-agarose beads. One-quarter of each fraction (total and surface NHE3) was analyzed by SDS-PAGE, Western blotting, and densitometric analysis using ImageQuant Version 4.2a software.
Labeling the Apical Cell Surface with Fluorescent Lectin and Fluorescence Microscopy-To examine the changes in apical membrane structure caused by M␤CD treatment, surface labeling with a fluorescent lectin was used (39). Lectins bind to specific sugar residues of the glycocalyx; and at 4°C, the fluorescent markers remain on the apical surface of the monolayers for several hours. OK/E3V cells were grown and treated under the same conditions that were used for Na ϩ /H ϩ exchange activity assay. The apical surfaces of control and treated cells were labeled with fluorescein isothiocyanate (FITC)-conjugated wheat germ agglutinin (WGA) epitope (Molecular Probes). Briefly, confluent monolayers were cooled for 30 min at 4°C and then washed three times with ice-cold PBS. FITC-conjugated WGA was applied to the monolayers of control and M␤CD-treated cells for 1 h at 4°C to prevent endocytosis of lectin. Unattached lectin was removed by washing the monolayers with ice-cold PBS, and then cells were fixed in 3% formaldehyde in PBS. Fixed cells were washed several times with PBS and mounted on Dako fluorescence mounting medium. The binding sites of lectin were observed by confocal laser scanning microscopy (Zeiss LSM 410) using xy and xz scanning planes. The fluorescence intensity of apically bound, FITCconjugated WGA was calculated using MetaMorph Version 5.0r1 software (Universal Imaging Corporation, PA).
Synthesis of Sterol-M␤CD Complexes-To replete monolayers with cholesterol, cholesterol and its inactive analog cholestane-3␤,5␣,6␤triol were complexed with M␤CD as described (40). Briefly, 6 mg of either cholesterol or cholestane-3␤,5␣,6␤-triol was dissolved in 80 l of isopropyl alcohol/CHCl 3 (2:1). M␤CD (200 mg) was dissolved in 2.2 ml of water and heated to 80°C with stirring. The sterol was added in small aliquots, and the solution was stirred until clear. This yielded a solution that contained 6.8 mM sterol. Complexes were diluted in serum-free DMEM/nutrient mixture F-12 with a final sterol concentration of 0.2 mM.
Statistics-Values are presented as the means Ϯ S.E. Statistical significance was determined using Student's unpaired t test. p Ͻ 0.05 was considered significant.

There Is an NHE3 Pool in Detergent-resistant Membranes in OK/E3V
Cells-To understand the relationship between LR and NHE3, we used OK cells initially processed by acid suicide to greatly decrease endoge-nous NHE3 and then stably transfected with rat NHE3 cDNA (called OK/E3V cells). As reported, these OK/E3V cells represent a non-overexpression model for the study of NHE3 in polarized epithelial cells (38). Important criteria to establish the association of proteins with LR are to demonstrate that the protein is insoluble in cold Triton X-100 and that it shifts from lighter to heavier membrane fractions after cholesterol depletion with M␤CD based on density gradient fractionation (16). OK/E3V cells were treated with 10 mM M␤CD for 30 min at 37°C (which was lowered to 4°C), biotinylated by exposure to succinimidyl 2-(biotinamido)ethyl-1,3-dithiopropionate, and lysed in 1% Triton X-100, and then total fractions were mixed with the highest density sucrose (30%) and applied to the bottom of the sucrose density gradients for ultracentrifugation. The Western blot analyses (Fig. 1A) of fractions of total and biotinylated (surface) NHE3 from untreated cells revealed that NHE3 distributed as a single immunoreactive band with the highest amount in fractions 1, 2 (heaviest), and 17-20 and with lower and almost equal density in fractions 3-16. In M␤CD-treated cells, upon cholesterol depletion, the lighter fractions of total and surface NHE3 shifted toward heavier gradient fractions. These results suggest that NHE3 in OK/E3V cells is partially in LR. Densitometry of surface NHE3 (Fig. 1B) showed that ϳ50% of surface NHE3 was present in LR (Fig. 1A, surface NHE3 Ϯ m␤CD). (Fractions 8 -20 were shifted by M␤CD from light to heavy sucrose gradient fractions.) In contrast, ϳ17% of total NHE3 was in LR (Fig. 1A, total NHE3 Ϯ m␤CD). ϳ15% of total OK cell NHE3 is in the plasma membrane (38), which indicates that ϳ11% of intracellular NHE3 is in LR ((0.5)(0.15) ϩ (x)(0.85) ϭ (0.17)(100)), where x is the fraction of intracellular NHE3 in LR).
Disruption of LR with M␤CD Decreases NHE3 Exchange Activity-To evaluate the effect of disruption of LR by cholesterol removal on the basal activity of NHE3, the monolayers of OK/E3V cells were incubated in the presence or absence of 10 mM M␤CD or vehicle at 37°C 30 min. The Na ϩ /H ϩ exchange activity was significantly inhibited in cells treated with M␤CD compared with control cells (initial rates in ⌬pH/ FIGURE 1. A pool of NHE3-VSV-G in OK cells is LR-associated. OK/E3V cells were treated with or without 10 mM M␤CD for 30 min at 37°C and then surface-biotinylated as described under "Experimental Procedures." Total lysates were fractionated by floatation on an 11-step sucrose gradient. Surface (biotinylated) NHE3 in each fraction was precipitated by avidin-agarose beads. NHE3 was separated by SDS-PAGE and identified by Western blotting (A), and densitometric analysis of surface NHE3 was performed using ImageQuant Version 4.2a software (B). The quantitated data from the two upper panels in A are shown in B as the percentage of total surface NHE3 in each fraction. min ϭ 0.23 Ϯ 0.09 and 0.54 Ϯ 0.08 for M␤CD-treated and control cells, respectively; p Ͻ 0.05) (Fig. 2B). The effect of M␤CD was reversed completely by treatment of OK/E3V monolayers with cholesterol ( Fig.  2A). For these experiments, M␤CD-treated cells were washed free of M␤CD and incubated for an additional 30 min in the presence of excess cholesterol or its inactive analog cholestane-3␤,5␣,6␤-triol. Monolayers pretreated with M␤CD and then recovered in the presence of cholesterol exhibited a rate of pH recovery almost equal to that of the control cells (initial rates in ⌬pH/min ϭ 0.48 Ϯ 0.07 and 0.54 Ϯ 0.07 for M␤CD-treated/reversed and control untreated cells, respectively). By contrast, monolayers treated with the inactive cholesterol analog exhibited only a partial or no recovery (initial rates in ⌬pH/min ϭ 0.25 Ϯ 0.09). Hence, the effect of M␤CD on NHE3 activity was fully reversible and specific for cholesterol. These initial results indicate that ϳ50% of NHE3 transport activity is cholesterol-dependent.
To better characterize the effect of M␤CD on basal NHE3 activity, full kinetic curves were generated. A representative experiment from five similar experiments is shown in Fig. 2A To determine whether disruption of LR in the plasma membrane of OK/E3V cells alters the affinity of protein for Na ϩ , the activity of NHE3 in the presence of varying concentrations of medium Na ϩ was examined. Analysis of the Eadie-Hofstee plot showed that M␤CD affected only the V max , but not K m(Na ϩ ) , of the Na ϩ /H ϩ exchange activity of NHE3 in OK/E3V cells (Fig. 3). The kinetics for external Na ϩ in both control and treated cells followed a classical Michaelis-Menten model with K m(Na ϩ ) values of 12 Ϯ 3 M for control cells and 13 Ϯ 6 M for M␤CD-treated cells (Fig. 3).
Effect of M␤CD on NHE3 Is Not a General Effect on Transport-The inhibitory effect of M␤CD on NHE3 activity might be explained by a general inhibitory effect of cholesterol depletion on membrane dynamics (fluidity/viscosity). To test this hypothesis, we studied the effect of cholesterol depletion on Na ϩ /glucose cotransport activity driven by SGLT1 and Na ϩ -dependent L-alanine transport in OK/E3V cells. It was reported recently that SGLT1 is localized to LR and that LR disruption decreases Na ϩ /glucose transport activity (22). Thus, we used SGLT1 as a positive control. Generally, OK cells express SGLT1 in the apical membrane. However, we found that OK/E3V cells did not exhibit endogenous Na ϩ -dependent glucose uptake (Fig. 4A, inset), probably as a consequence of selection procedures during isolation of OK cells lacking endogenous NHE3. Therefore, the cDNA encoding SGLT1 was transiently transfected into OK/E3V cells, and Na ϩ /glucose cotransport activity was assessed as Na ϩ -dependent uptake of MGP, a non-metabolized analog of D-glucose, in the presence and absence of 10 mM M␤CD. Na ϩ -dependent glucose uptake was significantly reduced in M␤CD-treated cells versus control cells (Fig. 4A). By contrast, cholesterol depletion of OK/E3V monolayers did not affect Na ϩ -dependent L-alanine uptake. The rates of L-alanine uptake (both Na ϩ -dependent and Na ϩ -independent) by M␤CD-treated and control cells were comparable (Fig. 4B). These findings indicate that the inhibitory effect of cholesterol depletion with M␤CD is not the result of nonspecific effects   JUNE 30, 2006 • VOLUME 281 • NUMBER 26

JOURNAL OF BIOLOGICAL CHEMISTRY 17849
on plasma membrane dynamics and that Na ϩ -dependent L-alanine uptake does not occur by a cholesterol-dependent transporter(s).
M␤CD Does Not Affect the Surface Amount of NHE3-The M␤CD decrease in the basal transport rate of NHE3 might have been due to a decrease in the amount of exchanger in the apical membrane of OK cells. To test this hypothesis, the amount of surface NHE3 protein in control cells and cells treated with 10 mM M␤CD for 30 min at 37°C was determined. Surface protein biotinylation and ELISA were performed to examine the surface amount of NHE3 using OK/E3V and OK/3HA-E3V cell lines. Despite the pronounced effect of M␤CD on the transport rate, M␤CD did not change the percentage of surface NHE3 compared with control cells (108 Ϯ 10% of control cells, n ϭ 3 (not significant)). Representative Western blotting of NHE3 is shown in Fig. 5A, and summarized data (densitometric scanning analysis) of three experiments are shown in Fig. 5B.
A second independent method to assess the percentage of surface NHE3 based on a modified ELISA was established to quantify the surface amount of NHE3. For this purpose, an OK/3HA-E3V cell line that contained an external epitope was generated. Two clones were initially tested to show that insertion of the triple HA tag into the first extracellular loop plus the VSV-G tag at the C terminus of NHE3 did not alter the expression, ion exchange activity, and trafficking properties of NHE3 (data not shown). As shown in Fig. 6A, HA-NHE3 in clones II and III of OK/3HA-E3V cells was expressed, and ϳ80 -85 kDa immunoreactive bands were detected in total cell lysates with both anti-HA and anti-VSV-G antibodies. The rat NHE3 protein with only a VSV-G epitope failed to react with anti-HA antibody as a negative control (Fig.  6A), showing that the immunoreactivity was specific to the HA epitope. Insertion of triple HA epitope plus VSV-G tags did not measurably alter the expression of rabbit NHE3 compared with rat NHE3 tagged with only the VSV-G epitope in OK/E3V cells (data not shown).
The functional NHE3 activity in OK/3HA-E3V cells was established using BCECF as described above. Full kinetic curves were generated to determine the basal Na ϩ /H ϩ exchange activity of control and OK3/ 3HA-E3V cells treated with M␤CD (clone II). Representative curves in Fig. 6B show that 3HA-E3V is functional and that M␤CD had a similar percent effect on the antiport activity of 3HA-E3V as it had on NHE3-VSV-G in Fig. 2.
Binding of anti-HA antibody to the external epitope of 3HA-E3V was visualized by confocal microscopy. The accessibility of anti-HA antibody in intact (non-permeabilized) cells was examined first. Cells were incubated with anti-HA antibody at 4°C for 1 h and then fixed and stained with Alexa Fluor 488-labeled anti-mouse secondary antibodies. Serial xy sections of OK/3HA-E3V cells demonstrated that the external HA epitope of NHE3 in OK/3HA-E3V cells was accessible (Fig.  6, C (bottom) and D (top)). For ELISA, antibody to the extracellular HA epitope was added to intact control OK/3HA-E3V cells and treated with M␤CD to detect exchangers that were only at the surface. In five separate ELISA experiments, the number of cell surface-exposed exchangers was similar in control and treated cells (98 Ϯ 15% of the untreated control (not significant)) (Fig. 5C). The results obtained with both methods of surface NHE3 measurement were similar and led to the conclusion that the inhibition of NHE3 activity caused by cholesterol depletion was not due to a change in the surface amount of apical NHE3 in OK cells.

Disruption of LR Does Not Cause Topological Alterations in the Apical Membranes of OK/E3V Cells-
The inhibition of NHE3 activity in OK/E3V cells upon M␤CD treatment might be explained by an alteration in membrane structure because of cholesterol depletion. To consider this possibility, apical membranes of OK/E3V cells treated or not with 10 mM M␤CD for 30 min at 37°C were chilled to 4°C and labeled with FITC-conjugated WGA. WGA binds specifically to N-acetylneuraminic acids and N-acetylglucosamine residues in the glycocalyx at 4°C (41,42) and can be used to delineate the apical membrane of epithelial cells. Fig. 7 (A and B) shows the distribution of WGA (10 g/ml) on the apical side of OK/E3V monolayers in control and treated cells. The appearance of M␤CD-treated cells was not altered. Also, monolayers of control and M␤CD-treated cells bound WGA in irregularly distributed patches on the apical surface as well as in discontinuous rings at the level of the tight junctions. Some cells in both monolayers failed to bind WGA. The morphometric analysis of the fluorescent intensity of FITC-conjugated WGA in xz planes from seven independent fields for control cells (intensity, 15.5 Ϯ 5.2) and 12 independent fields for M␤CD-treated cells (intensity, 15.6 Ϯ 3.2) revealed no differences (Fig.  7, C and D). The similar distribution of WGA in control and treated cells The Na ϩ -dependent uptake of MGP in the presence of Na ϩ or TMA ϩ by SGLT1-transfected OK/E3V cells and control non-transfected cells is shown in A and the inset, respectively. L-Alanine uptake by OK/E3V cells in the presence of Na ϩ or TMA ϩ is shown in B. L-Alanine uptake was at room temperature and initiated by addition of radioactively labeled substrate L-[ 3 H]alanine (2 Ci/ml) and 0.2 mM unlabeled L-alanine. After the incubation times, the uptake of substrate was stopped, and the radioactivity of the extracted isotope was assayed by liquid scintillation spectrometry. The amounts of accumulated substrates are expressed as cpm/min/ml (means Ϯ S.E.) of four experiments for each substrate. *, p Ͻ 0.05 (paired t test; M␤CDtreated versus control cells). and the similar level of fluorescent intensity of WGA suggested that depletion of cholesterol with 10 mM M␤CD did not cause major changes in the apical membrane structure in OK/E3V cells. Therefore, the inhibition of NHE3 activity in cells with disrupted LR was not due to gross alterations in the apical membranes of OK/E3V cells.
cAMP and M␤CD Additively Decrease the Basal Transport of NHE3-Regulation of NHE3 activity is modulated by a variety of hormones and second messengers, including cAMP. Elevation of intracellular cAMP activates protein kinase AII, which phosphorylates NHE3 and inhibits Na ϩ /H ϩ exchange activity (43)(44)(45). The possibility that the inhibitory effect of cAMP and the disruption of LR with M␤CD share a common mechanism of action in the inhibition of NHE3 activity was tested. The Na ϩ /H ϩ exchange activity of NHE3 upon treatment with both M␤CD and a maximal concentration of 8-Br-cAMP (membrane-permeable analog of cAMP) was examined. Control as well as M␤CD-treated cells were preincubated with 0.1 mM 8-Br-cAMP for 30 min prior to determining the rate of Na ϩ /H ϩ exchange. Fig. 8 shows that both 8-Br-cAMP and M␤CD inhibited NHE3 activity to a similar degree. Analysis of the full kinetic curves generated (data not shown) to determine the kinetic parameters for Na ϩ /H ϩ exchange activity revealed that both 8-Br-cAMP and M␤CD reagents affected the V max . The V max values were 486 Ϯ 22 and 503 Ϯ 34 M H ϩ /s (mean Ϯ S.E.) for 8-Br-cAMP-and M␤CD-treated cells, respectively, whereas the V max for control cells was 733 Ϯ 22 M H ϩ /s. Simultaneous addition of 8-Br-cAMP and M␤CD caused further reduction of NHE3 activity; the V max dropped to 277 Ϯ 8 mM H ϩ /s. That the effects of 8-Br-cAMP and M␤CD together were additive suggests that they have different mechanisms of inhibiting NHE3 activity. These results suggest that LR are not involved in 8-Br-cAMP inhibition of the NHE3 V max .
LY-294002 and M␤CD Non-additively Inhibit the Basal Transport of NHE3-Under basal conditions, NHE3 traffics between the cell surface and recycling endosomes in a phosphatidylinositol 3-kinase (PI3K)-dependent manner (9,38,46). Wortmannin, an inhibitor of PI3K, reduces the transport rates and surface level of NHE3 by inhibiting the exocy-tosis of the exchanger back to the plasma membrane. Wortmannin also significantly reduces the Na ϩ /H ϩ exchange activity of NHE3 in OK cells accompanied by a corresponding reduction in surface NHE3 amount (38). The reduction of NHE3 activity by cholesterol depletion and PI3K might have the same molecular mechanism of action. The same approaches as described above under "cAMP and M␤CD Additively Decrease the Basal Transport of NHE3" were used to study the relationship between the PI3K pathway and LR in NHE3 activity. As shown in Fig. 9, LY-294002 (another inhibitor of PI3K that interacts with the ATP-binding site of the enzyme; 50 M, 30 min) and M␤CD both reduced the Na ϩ /H ϩ exchange activity of NHE3, with a more profound effect of M␤CD. Analysis of the kinetic parameters indicated that PI3K inhibitor-mediated inhibition of NHE3 activity as well as M␤CD altered the V max (710 Ϯ 46 M H ϩ /s for LY-294002 and 454 Ϯ 47 M H ϩ /s for M␤CD) compared with the V max in control cells (1025 Ϯ 49 M H ϩ /s). Addition of both LY-294002 and M␤CD simultaneously did not further inhibit the activity of NHE3. Thus, the inhibitory effects of both agents were similar and are not additive. This suggests that LY-294002 and M␤CD may share a common inhibitory mechanism. These results suggest that LR might be involved in regulation of basal activity by a mechanism that involves PI3K.
M␤CD Decreases the Endocytosis of NHE3-M␤CD treatment did not alter the surface amount of NHE3 despite the fact that the basal exocytosis of NHE3 was LR-dependent. Endocytosis studies were undertaken to resolve what appeared to be a contradiction. To quantitate NHE3 endocytosis, apical membrane proteins were labeled with N-hydroxysulfosuccinimidobiotin before treatment with M␤CD or vehicle. After 1 h of biotinylation at 4°C followed by 1 h of incubation at 37°C with M␤CD or vehicle, cells were exposed to reduced GSH at 4°C. This reagent cleaves biotin only from surface proteins. With this protocol, biotinylated NHE3 represented the pool of NHE3 that was initially present on the apical membrane and that was subsequently endocytosed and thus protected from GSH. As shown in Fig. 10 (A and B), treatment of OK/E3V cells with 10 mM M␤CD for 1 h greatly reduced the endocytosis of biotinylated NHE3. The total amount of NHE3 in M␤CD-treated cells was not affected (Fig. 10A). The M␤CD inhibition of basal endocytosis is in agreement with data described previously by us for ileal BB (12). Thus, disruption of LR by M␤CD decreased both the exocytosis and endocytosis of NHE3 such that the net amount of surface NHE3 was not changed.

DISCUSSION
These studies of the role of LR in NHE3 activity characterized in polarized OK cells extended our largely biochemical studies that evaluated NHE3 regulation in ileal absorptive cells (12). The goal was to increase the understanding of the contribution of LR to NHE3 activity and the targeting of a BB transport protein that traffics between the BB and recycling compartment under basal conditions and that undergoes rapid regulated changes in trafficking as part of normal physiology. LR are involved in multiple aspects of NHE3 trafficking. (a) 25-50% of NHE3 in the rabbit ileal Na ϩ absorptive cell BB membrane localizes in LR. (b) LR also accounts for most of the rapid increase in the amount of BB NHE3 caused over minutes by epidermal growth factor. (c) The amount of NHE3 associated with EEA1 is also LR-dependent, indicating a role in the basal endocytosis of NHE3. The results of the present work FIGURE 6. Insertion of a triple HA epitope at the N terminus of NHE3-VSV-G does not affect NHE3 expression and LR-dependent basal NHE3 transport activity in OK/3HA-E3V cells. A, expression of 3HA-E3V in two clones of OK/3HA-E3V cells. Cells were solubilized in 1% Triton X-100, and cell extracts (50 g) were separated on two 10% SDS-polyacrylamide gels run in parallel and probed with either anti-HA antibody (Åb; left) or anti-VSV-G antibody (P5D4 hybridoma; right). Lysates from OK/E3V cells were used as a negative control (left). Immunoreactive bands of ϳ85 kDa were recognized by both antibodies. B, M␤CD decreases NHE3 activity in OK/3HA-E3V cells. OK/3HA-E3V cells (one clone of two studied is shown) were grown on glass coverslips. Cell monolayers were simultaneously loaded with BCECF and treated with M␤CD or vehicle in Na ϩ /NH 4 Cl medium. The cells were rinsed with NH 4 ϩ /Na ϩ -free TMA ϩ medium. The recovery of pH i through NHE3 activity was initiated by addition of Na ϩ medium. Representative (three experiments) full kinetic curves for Na ϩ -dependent pH recovery for control (F) and M␤CD-treated (E) cells are shown. C and D, surface staining of OK/3HA-E3V cells with antibody to the HA epitope. Cells grown as confluent monolayers on glass coverslips were used for immunofluorescence. Cells were chilled, rinsed three times with ice-cold PBS, and incubated with anti-HA antibody (1:1000 dilution) for 2 h at 4°C, and unbound antibodies were washed out with PBS. Monolayers were fixed and incubated with Alexa   show that, in a polarized Na ϩ absorptive epithelial cell, (a) 50% of apical membrane NHE3 and 17% of total NHE3 associated with LR. In OK/E3V cells, ϳ15% of NHE3 is localized to the apical membrane according to our calculations (38), and 85% resides intracellularly. This means that ϳ11% of intracellular NHE3 resides in LR. This difference in the percentage of NHE3 that is LR-associated on the apical surface versus in the intracellular pool is striking. NHE3 functions both on the apical surface (Na ϩ absorption) and intracellularly in the endosomes (albumin uptake (47) and acidification of early endosomes (7, 38)).
Because BB NHE3 regulation is LR-dependent, whereas little intracellular NHE3 associates with LR, regulation of NHE3 on the apical surface and intracellularly is likely to be very different. Although extensive studies of regulation of BB NHE3 have been reported, regulation of intracellular NHE3 has not been evaluated in detail. (b) ϳ50% of basal NHE3 activity was cholesterol-dependent, which was virtually identical to the percentage of apical NHE3 in LR. 50% of surface NHE3 is in LR, and there is a 50% decrease in NHE3 activity with cholesterol depletion, but no changes in total BB NHE3 amount, indicating that LR NHE3 is active. What can be concluded about the 50% of surface NHE3 that is not in LR? That the depletion of cholesterol is expected to disrupt LR and displace NHE3 from them but is not likely to destroy NHE3, at the least, suggests that LR NHE3 contributes more to NHE3 activity than the non-LR component. However, our data do not let us conclude that non-LR NHE3 is inactive or has reduced NHE3 activity. (c) Inhibition of BB NHE3 activity by cAMP, which occurs by both change in turnover number and by increased endocytosis with a decrease in plasma membrane NHE3 (44,48), was not LR-dependent. This is consistent with results in ileal BB, in which cAMP-stimulated endocytosis of NHE3 occurred via an LR-independent mechanism, and indicates that basal endocytosis and cAMP-stimulated endocytosis occur by different mechanisms. (d) Basal NHE3 activity and amount, which were both ϳ50% PI3K-dependent and involved trafficking of NHE3 from an intracellular site to BB, were also ϳ50% LR-dependent.
Thus, the study of NHE3 in two Na ϩ absorptive cell models shows that there is a large apical membrane component of NHE3 that is present in LR and that LR contribute to basal NHE3 activity and do so by effects on the turnover number (change in NHE3 V max and KЈ(H ϩ ) i without changing K m(Na ϩ ) or the surface amount of NHE3). LR are also involved in multiple aspects of basal trafficking of NHE3 (both endoand exocytosis). Moreover, acutely stimulated NHE3 activity appears to be LR-dependent (based on M␤CD inhibition of epidermal growth factor-and ␣ 2 -agonist-induced increases in the amount of NHE3 in BB) (12), 5 whereas other aspects of stimulated trafficking do not involve LR. For instance, 8-Br-cAMP inhibition of NHE3, which involves stimulated endocytosis, does not appear to be LR-dependent. Please note the similarity in the role of LR in regulation of NHE3 in the apical membrane of ileal Na ϩ absorptive cells and the OK renal proximal tubule cell line. This is despite a different percentage of total cell NHE3 present in apical membrane LR (7.5% in OK cells (50% of apical membrane NHE3 is in LR, and 15% of total cell NHE3 is in the apical membrane) and 20% in ileal Na ϩ absorptive cells (ϳ25% of apical membrane NHE3 is in LR, and 80% of total cell NHE3 is in the apical membrane)). In contrast, the percentage of total cellular NHE3 in the detergent-resistant membrane (Triton X-100-insoluble) fraction in apical membranes of OK cells (95% of 15% ϭ 14%) and ileal Na ϩ absorptive cells (50% of 80% ϭ 40%) is also different such that the detergent-resistant membrane/LR NHE3 ratio is similar (ϳ2-fold) in the apical membranes of both cell types. This result suggests that the cytoskeleton/LR association of NHE3 in the apical membranes of both cell types is similar. Given the current understanding that the cytoskeleton is involved in multiple aspects of NHE3 regulation, we speculate that the cytoskeleton has a major role in LR-dependent aspects of NHE3 trafficking and perhaps more generally in NHE3 regulation.
Both the endocytosis and exocytosis of NHE3 were LR-dependent, with both basal and stimulated exocytosis being LR-dependent, but only basal (and not 8-Br-cAMP-stimulated) endocytosis was similarly LRdependent. The LR dependence of exocytosis was expected because 5 M. Donowitz and X. Li, unpublished data.   10. M␤CD inhibits the endocytic rate of NHE3 in OK/E3V cells: GSH-resistant endocytosis assay. Endocytosis was initiated by exposure to 10 mM M␤CD or vehicle for 1 h at 37°C. Cells were then surface-labeled with N-hydroxysulfosuccinimidobiotin at 4°C. The surface biotin was cleaved with reduced GSH at 4°C. Biotinylated proteins protected from GSH by internalization were retrieved by avidin precipitation (lanes 2 and  3). Parallel control experiments were performed with cells always at 4°C to arrest endocytosis (lanes 1 and 4, with lane 1 subtracted from lanes 2 and 3). Aliquots of lysates were used for total NHE3 determination (lanes 4 -6). NHE3 was analyzed by SDS-PAGE and immunoblotting with anti-VSV-G antibody and fluorescently labeled secondary antibodies. Fluorescence intensity was analyzed using MetaMorph Version 5.0r1 software. A representative immunoblot from one experiment is shown in A, and the fluorescence intensity (gray levels) of endocytosed NHE3 expressed as a percentage of total NHE3 is shown as the means Ϯ S.E. from three experiments in B. The higher bands in lanes 5 and 6 represent a non-biotinylatable, nonspecific band. both basal and stimulated endocytic recycling of NHE3 are PI3K-dependent, and in ileal BB, both PI3K and the downstream Akt2 are present in LR in the BB (49,50). Similarly, that NHE3 associates with the early endosomal marker EEA1 by an LR-dependent process (12) further supports the involvement of LR in basal endocytosis. Rather, it is the lack of requirement of LR for cAMP-stimulated endocytosis of NHE3 that is not understood, with there being a possible role for the absence of NHERF1 and NHERF2 in LR in ileal BB, although both take part in cAMP inhibition of NHE3 (12).
Concerning mechanisms by which cholesterol depletion from the membrane affects NHE3 function, 1) this was not due to a change in general membrane structure and function because the surface area was not significantly altered, as was also found previously in similar studies in porcine intestine (31); 2) not all apical membrane transport proteins were inhibited (Na ϩ -dependent L-alanine absorption was normal); and 3) not all kinetic aspects of NHE3 activity were altered (no effect on NHE3 K m(Na ϩ ) ). We suggest that all effects of cholesterol depletion on NHE3 activity, trafficking, and regulation are consistent with the target of cholesterol depletion being the NHE3 C terminus, although we do not have direct experimental evidence that this is true. For instance, the cytoplasmic domain of NHE3 is necessary for all identified NHE3 regulation (osmotic shrinkage of NHE2 may involve an extracellular loop of the N terminus, but no similar role has been identified for NHE3 (51)) and associates with multiple regulatory proteins (2). Moreover, the changes in the kinetics of NHE3 when acutely inhibited by M␤CD treatment are similar to the C terminus-mediated changes that occur when NHE3 is acutely inhibited by elevated cAMP (44,48), cGMP (52), or Ca 2ϩ (53). NHE3 exists in large complexes of up to 1000 kDa, which require the NHE3 C terminus (38,50). These involve multiple associating proteins (2), only some of which have been identified. Because one of the known functions of LR is compartmentalization of signal transduction to localized domains to increase specificity and efficiency of signaling, it would be predicted that regulation of NHE3 is LR-dependent. The putative involvement of the NHE3 C-terminal regulatory domain in determining the LR association of NHE3 is further supported by the recent observations that the cytosolic domains of the Ca 2ϩ -sensitive adenylyl cyclases determine their targeting to plasma membrane LR (54) and that raft targeting of a protein named LAT (linker for activation of T cells) requires both palmitoylation and its intracellular domain via interactions with additional proteins, presumably via protein/protein interactions (55).
It has yet to be experimentally demonstrated 1) whether and how the NHE3 regulatory domain links NHE3 to LR and 2) whether NHE3-associating proteins are involved in targeting NHE3 to LR.
Given the role of NHE3 in digestive physiology, what are the potential implications of a large LR pool of BB NHE3 in Na ϩ absorptive cells in which NHE3 continually traffics between the BB and the recycling system and exists in large multiprotein complexes (38,50)? LR increase the efficiency of signal transduction by providing a restricted spatial environment that contains multiple proteins, which are involved in a signaling pathway. It is possible that delivery of NHE3 with its associating binding proteins and other components of signaling contributes to the rapid postprandial increase in sodium absorption. LR dependence of basal (but not cAMP-stimulated) endocytosis appears to identify two separate NHE3 pools that are involved in different aspects of endocytosis. Perhaps limiting the NHE3 pool involved in stimulated endocytosis protects against dehydration by decreasing the magnitude of NHE3 inhibition in digestion and in pathophysiological conditions.