Carbachol regulation of rabbit ileal brush border Na+-H+ exchanger 3 (NHE3) occurs through changes in NHE3 trafficking and complex formation and is Src dependent

Abstract

The epithelial brush border membrane (BBM) Na(+)-H(+) exchanger 3 (NHE3) is the major transport protein responsible for ileal electroneutral Na(+) absorption. We have previously shown that ileal BBM NHE3 activity is rapidly inhibited by carbachol, an agonist that mimics cholinergic activation in digestion. In this study, we investigated the mechanisms involved in this NHE3 inhibition. Carbachol decreased the amount of ileal Na(+) absorptive cell BBM NHE3 within 10 min of exposure. Based on OptiPrep gradient centrifugation, carbachol increased the amount of NHE3 in early endosomes and decreased the amount of NHE3 in BBM, consistent with effects on NHE3 trafficking. The decrease in BBM NHE3 occurred in the detergent-soluble BBM fraction with no change in the amount of NHE3 in the BBM detergent-resistant membranes. The size of BBM NHE3 complexes increased in carbachol-exposed ileum, as studied with sucrose gradient centrifugation. The NHE3 complex size increased in the total BBM, but did not change in the detergent-soluble fraction. This suggests that carbachol treatment enhanced the association of proteins with NHE3 complexes specifically in the detergent-resistant fraction of ileal BBM. NHERF2, alpha-actinin-4 and protein kinase C were among those NHE3-associated proteins because they were more efficiently coimmunoprecipitated from total BBM after carbachol treatment. Moreover, Src was involved in the carbachol-mediated inhibition since: (1) c-Src was rapidly activated in the detergent-resistant membranes by carbachol; and (2) carbachol inhibition of ileal Na(+) absorption was completely abolished by the Src family inhibitor 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine (PP2). Moreover, the carbachol-induced increase in the size of NHE3-containing complexes was reversed by PP2. These data demonstrate that regulation of NHE3 activity by carbachol can be achieved at several interrelated levels: (1) the subcellular level, at which NHE3 is rapidly endocytosed from BBM to endocytic vesicles upon treatment with carbachol; (2) multiple BBM pools, in which carbachol selectively decreases the amount of NHE3 in the BBM detergent-soluble fraction but not the detergent-resistant membrane; and (3) the molecular level, at which NHE3 complex-associated proteins can be changed upon carbachol treatment, with carbachol leading to larger BBM NHE3 complexes and increased co-IP of NHERF2 with alpha-actinin-4 and activated PKC. The study further describes NHE3 presence simultaneously in multiple dynamic BBM pools in which NHE3 distribution and associated proteins are altered as part of carbachol-induced and Src-mediated rapid signal transduction, which decreases the amount of BBM NHE3 and thus inhibits NHE3 activity.

Figure 1. Carbachol treatment decreases total BBM NHE3 by decreasing the amount of NHE3 in the DS fraction of ileal BBM, but that in the DRM is not affected

Ileal segments were incubated with (+) or without (−) 10 μm carbachol for 10 min. Ileal villus cells were then isolated by scraping and BBM prepared. BBM were lysed in 0.5% Triton X-100 and soluble (DS) and insoluble (DRM) fractions were obtained as described (Li et al. 2001). A, carbachol decreased the amount of NHE3 only in the DS and not in the DRM. Upper panel, Western blot; lower panel, densitometry of the Western blot shown in the upper panel, in arbitrary densitometry units. One representative of three independent experiments is shown. B, carbachol decreased total BBM NHE3 by 24 ± 3%. Shown is the percentage of total BBM NHE3 after carbachol (▪), in comparison with the untreated control (□), which was set as 100%. Results were calculated from three independent experiments similar to that shown in A (means ±s.d.). P values were in comparison to untreated controls (Student's paired t test).

Figure 2. Carbachol decreases the amount of NHE3 in BBM by shifting NHE3 from BBM to endosomal pools

A, intact ileal segments were incubated with or without carbachol (10 μm) for 10 min. Ileal villus cells were scraped off and total membranes were prepared (1–1.5 mg protein) and fractionated by OptiPrep density gradients. Equal volumes (¹/₆ of the total) of each fraction were analysed by SDS-PAGE and Western blotting. Aa–Ad shows resident marker proteins for different populations of membranes vesicles, which were well separated as detected by antibodies specific to each protein: EEA1 (a) and β-adaptin (b) for early endosomes; sucrase (c) for BBM (apical) and Na⁺–K⁺ ATPase (d) for basolateral membranes (basolateral). Ae and Af show NHE3 localization in the absence (e) or presence of carbachol (f). Carbachol treatment increased the amount of NHE3 in the endosomal pool from 22 to 38% of total NHE3 and decreased BBM NHE3. A representative of three similar independent experiments which gave similar results is shown. Marker labelling was similar in control/carbachol conditions with only control values shown in Aa–d. B, densitometric/ImageQuant analysis of percentages of total NHE3 in the endosomal pool from three independent experiments based on comparison of NHE3 from lanes 1 and 2 compared to total NHE3 from all lanes (means ±s.d.). P values are in comparison to untreated controls (Student's paired t test).

Figure 3. Carbachol treatment increases the size of protein complexes containing NHE3 and α-actinin-4 in ileal BBM

Isolated BBM (1 mg protein) were lysed with 0.5% Triton X-100 for the preparation of total lysates. The total lysates (detergent-resistant membrane, DRM, plus detergent-soluble membrane, (DS) or the DS pool alone were then fractionated by sucrose gradients. Fraction no. 1 has the heaviest density of sucrose. Equal volumes (per cent) of each fraction were analysed by SDS-PAGE and Western blotting. A, Western analysis on each fraction of total lysates for NHE3 (a) and α-actinin-4 (b). Carbachol treatment (10 μm, 10 min) increased the size of NHE3 complexes that contain NHE3 and α-actinin-4, from ∼400 up to 900 kDa. Three molecular weight markers (arrows), which were fractionated by a sucrose gradient identical to those above, are shown at the bottom of A to indicate the sizes of protein complexes along the gradient. B, in the DS fraction, carbachol treatment did not significantly change the size of the NHE3 complexes. A representative of three similar independent experiments is shown.

Figure 4. NHERF2, α-actinin-4 and p-PKC are co-precipitated from the ileal BB in a Ca²⁺-dependent manner

A, BBM isolated from ilea, which were treated with (+Carb; 10 μm, 10 min) or without carbachol (−Carb), were lysed by 1% Triton X-100 and subjected to immunoprecipitation with preimmune serum (control), anti-NHERF2 and anti-α-actinin-4 antibodies, respectively. The IP pellets were analysed by SDS-PAGE and Western blotting. NHERF2, α-actinin-4 and phosphorylated conventional PKC (p-PKC) appear to be in the same complex since they could be coimmunoprecipitated. Carbachol increased the interactions between NHERF2 and α-actinin-4 (upper panels), as well as the interaction between NHERF2 and p-PKC (middle panel). The results shown in the upper and middle panels are from the same blot. NHERF2 from total BBM used for each IP is shown (bottom panel) to indicate that an equal amount of the BBM was used in each IP. A representative of two similar independent experiments which had similar results is shown. B, densitometric scanning/quantification of the experiment in A. The arrows indicate that quantification of α-actinin-4 is shown above and of p-PKC below.

Figure 5. Effect of Src family kinase inhibitor PP2 on carbachol inhibition of rabbit ileal water absorption

Net water transport was determined in 10 cm ileal loops exposed in vitro to carbachol (10 μm, 37°C) on the serosal surface for 30 min. Studies were performed in the presence or absence of luminal PP2 (20 μm). Effects are suggested as being on Na⁺ absorption since Cl⁻ secretion was inhibited by the presence of serosal bumetanide (50 μm) in all conditions. Water fluxes were determined as microlitres of water transported per 30 min per milligram dry ileal weight. The conditions shown were studied in each of 6 animals. Results are means ±s.e.m.P values are in comparison to untreated controls (left panel) except * is a comparison of carbachol versus PP2 plus carbachol.

Figure 6. Carbachol stimulates ileal c-Src activity

A, ileal Na⁺ absorptive cells contain three Src family kinases, c-Yes, c-Fyn and c-Src. Forty-five micrograms of total cellular membranes of ileal villus cells were used for each blot and Western analysis performed. B, only c-Src kinase activity was increased after carbachol treatment. c-Yes, c-Fyn and c-Src were immunoprecipitated from total membrane lysates [500 μg protein in 15 mm HEPES buffer (RIPA) containing: 150 mm NaCl, 1% Triton-X100, 1% deoxycholate, 0.1% SDS, 10 mm FDTA, 1 mm dithiothreitol, 1 mm Na₃VO₄] of ileal villus cells isolated from untreated (−Carb, upper panel) or carbachol-treated lysates (+Carb, 10 μm, lower panel) using antibodies specifically against each kinase. Immunoprecipitated kinases were then probed with antiphosphorylated Src antibody for the assessment of the activity of each kinase. Results were similar in two experiments. Size of c-Src is marked on the right.

Figure 7. Carbachol rapidly activates the BBM DRM pool of c-Src but does not alter the BBM DS pool

Ileal segments were incubated with (+) or without (−) carbachol (10 μm) for 1 min. BBM were then isolated and DS and DRM pools were prepared. The DRM pellets were resuspended in a volume of RIPA buffer to make a final volume equal to that of the DS fraction. c-Src kinase was immunoprecipitated. The IP pellets were then analysed by SDS-PAGE and Western blotted with either anti-Src (Src, upper panel) or antiphosphorylated Src (p-Src, lower panel). The lower band recognized by the anti-p-Src antibody has not been identified. It is possible that the antibody recognizes multiple states of Src phosphorylation. A representative of two similar independent experiments is shown.

Figure 8. Inhibition of Src family kinases reversed the carbachol-stimulated increase of the sizes of NHE3- and α-actinin-4-containing ileal complexes

Ileal segments were treated under three conditions for 10 min, 37°C: without carbachol (control), with carbachol (+Carb, 10 μm), and with both carbachol and PP2 (+Carb+PP2, 10 and 20 μm, respectively). A, ileal villus cells were then isolated and total cellular membranes (45 μg protein) were prepared and analysed for c-Src activity (p-Src) and amount (a and b), as well as actin as a loading control (c). Carbachol increased ileal membrane Src family kinase activity. The Src family kinase inhibitor, PP2, decreased carbachol-induced p-Src activity to a level below that of controls. Results of all three panels (a, b and c) were obtained from the same blot via stripping and reprobing starting with a. B, protein complexes from total ileal membranes of NHE3 (a) and α-actinin-4 (b) in total membranes were separated and determined by sucrose density gradient fractionation as described in Fig. 3. PP2 diminished the carbachol-induced increase in the size of both NHE3- and α-actinin-4-containing complexes.

Figure 9. Carbachol increases ileal villus epithelial cell p-Src

Ileal mucosa was exposed in vitro to carbachol for 10 min at 37°C and then fixed and permeabilized and processed with antip-Src pab for confocal microscopy using a 100× lens. Serial 1.5 μm sections (15 sections) were obtained and merged to create the single plane 3D reconstructed image shown. In addition, fifteen 44 × 44 pixel boxes were created along the apical surface, excluding any overlap with the lamina propria signal, and quantified with Metamorph software. Carbachol treatment increased activated Src in epithelial cells, most clearly near the BBM, shown by arrows [control 60.7 ± 9.2 grey scale (gs) versus carbachol 77.6 ± 4.3 gs, P < 0.0003, Student's unpaired t test].