Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 212 (Pt 15), 2365-77

CFTR Cl- Channel Functional Regulation by Phosphorylation of Focal Adhesion Kinase at Tyrosine 407 in Osmosensitive Ion Transporting Mitochondria Rich Cells of Euryhaline Killifish

Affiliations

CFTR Cl- Channel Functional Regulation by Phosphorylation of Focal Adhesion Kinase at Tyrosine 407 in Osmosensitive Ion Transporting Mitochondria Rich Cells of Euryhaline Killifish

William S Marshall et al. J Exp Biol.

Abstract

Cystic fibrosis transmembrane conductance regulator (CFTR) anion channels are the regulated exit pathway in Cl(-) secretion by teleost mitochondria rich salt secreting (MR) cells of the gill and opercular epithelia of euryhaline teleosts. By confocal light immunocytochemistry, immunogold transmission electron microscopy (TEM), and co-immunoprecipitation, using regular and phospho-antibodies directed against conserved sites, we found that killifish CFTR (kfCFTR) and the tyrosine kinase focal adhesion kinase (FAK) phosphorylated at Y407 (FAK pY407) are colocalized in the apical membrane and in subjacent membrane vesicles of MR cells. We showed previously that basolateral FAK pY407, unlike other FAK phosphorylation sites, is osmosensitive and dephosphorylates during hypotonic shock of epithelial cells (Marshall et al., 2008). In the present study, we found that hypotonic shock and the alpha(2)-adrenergic agonist clonidine (neither of which affects cAMP levels) rapidly and reversibly inhibit Cl(-) secretion by isolated opercular membranes, simultaneous with dephosphorylation of FAK pY407, located in the apical membrane. FAK pY407 is rephosphorylated and Cl(-) secretion rapidly restored by hypertonic shock as well as by forskolin and isoproterenol, which operate via cAMP and protein kinase A. We conclude that hormone mediated, cAMP dependent and osmotically mediated, cAMP independent pathways converge on a mechanism to activate CFTR and Cl(-) secretion, possibly through tyrosine phosphorylation of CFTR by FAK.

Figures

Fig. 1.
Fig. 1.
Immunocytochemistry for cystic fibrosis transmembrane conductance regulator (CFTR) and focal adhesion kinase (FAK) pY407 [primary antibodies: mouse monoclonal anti-human CFTR and rabbit anti-phosphorylated human FAK pY407; secondary antibodies: goat anti-mouse IgG Alexa Fluor 488 (green) and goat anti-rabbit IgG Alexa Fluor 594 (red)]. (A) CFTR immunofluorescence was present in the apical crypt of mitochondria rich (MR) cells (green). The inset shows an XZ view of CFTR immunofluorescence in the apical crypt (N=12). (B) FAK pY407 immunofluorescence was present in the pavement cells and in the apical crypt of MR cells (red). The inset shows an XZ view of FAK pY407 immunofluorescence in the apical crypt (N=12). (C) There was a high degree of colocalization of CFTR and FAK pY407 in the apical crypt (yellow). The inset shows an XZ view of CFTR and FAK pY407 immunofluorescence in the apical crypt (N=12). (D) Same frame as C, but 10 μm below. At this level only FAK pY407 was present. (E) A line scan of fluorescence intensity (arbitrary units) versus distance in μm across the apical crypt of a double labelled MR cell showing good correspondence in the colocalization of CFTR (green line) and FAK pY407 (red line) in the crypt periphery. ac, apical crypt; n, nucleus; pvc, pavement cells; MR cells, mitochondria rich cells. Scale bar, 10 μm.
Fig. 2.
Fig. 2.
Transmission electron micrographs of immunogold double labelling using CFTR (primary: mouse monoclonal anti-human CFTR; secondary: goat anti-mouse IgG–15 nm gold) and FAK pY407 (primary: rabbit anti-phosphorylated human FAK pY407; secondary: goat anti-rabbit IgG–10 nm gold) (N=7). (A) Low magnification micrograph of an MR cell. Scale bar, 1 μm. (B) Magnified view of region 1 of A showing the presence of FAK pY407 only (red arrows) in the pavement cells. (C) Both CFTR and FAK pY407 were detected in the apical region of MR cells. A magnified view of the apical crypt (region 2 of A) showed CFTR (yellow arrowheads) and FAK pY407 colocalization. (D) FAK pY407 was also present in apical membrane and subapical vesicles of the MR cells (region 3 of A). There was a close distribution of the two proteins along the apical crypt edge of the MR cells. CFTR was also found in the subapical vesicles (not shown). (E) Double labelling in the cytoplasmic area of the cells showed only FAK pY407 in the basolateral membrane. ac, apical crypt; m, mitochondria; pvc, pavement cells; v, vesicle. Scale bar for B–E, 0.5 μm.
Fig. 3.
Fig. 3.
Immunocytochemistry for CFTR (green) and FAK pY407 (red) following hypotonic shock pretreatment. (A) In isotonic conditions, FAK pY407 immunofluorescence, shown in bright field (BF), was present in the basolateral membrane (N=12). (B) Basolateral FAK pY407 immunofluorescence was eliminated following a 60 min hypotonic shock pretreatment (N=12). FAK in granular lymphocytes was insensitive to hypotonic shock. (C) CFTR immunofluorescence was still present in the apical crypt following hypotonic shock (N=12). (D) Same frame as C, showing the absence of FAK pY407 in the pavement cells and in the apical crypt. (E) Same frame as C, but an optical section 10 μm below. Neither CFTR nor FAK pY407 was apparent at this level, which we interpret as the absence of CFTR and dephosphorylation of FAK. (F) A line scan of fluorescence intensity (arbitrary units) versus distance in μm across the apical crypt of a double labelled MR cell (indicated by the white line in D) confirmed the presence of CFTR fluorescence (green line) in the apical crypt and the absence of FAK pY407 fluorescence (red line) in that same area and in the pavement cells. gc, granular lymphocytes; ac, apical crypt; pvc, pavement cells. Scale bar for A and B, and C–E, 10 μm.
Fig. 4.
Fig. 4.
Immunocytochemistry for CFTR (green) and FAK pY407 (red) following 60 min hypotonic shock pretreatment and 10 min post-treatment with cAMP activators forskolin (10 μmol l–1, N=9), isoproterenol (1 μmol l–1, N=6) and 3-isobutyl-1-methylxanthine (IBMX, 0.1 mmol l–1, N=6). Shown here in XZ views, FAK pY407 immunofluorescence was restored in the apical crypt and partly in the subapical region following the addition of forskolin (A), isoproterenol (D) and IBMX (G). CFTR distribution was unchanged following the addition of all three cAMP activators (B, E, H). Merged immunofluorescence showed a high degree of colocalization between FAK pY407 and CFTR in the apical crypts (C, F, I). Scale bar, 10 μm.
Fig. 5.
Fig. 5.
Immunofluorescence for CFTR (green) and FAK pY407 (red) following 60 min clonidine (10 μmol l–1) pretreatment and 10 min hypertonic post-treatment (N=11). Clonidine pretreatment had no effect on CFTR immunofluorescence in an XY scan at the level of the apical crypts (A). An XZ view of one MR cell (white arrow in A) showed positive staining for CFTR in the apical crypt (B). However, FAK pY407 immunofluorescence was completely eliminated following clonidine incubation in an XZ scan of the same cell (C). CFTR distribution remained the same following clonidine and hypertonic incubation (XZ scan of MR cell, D), while FAK pY407 immunofluorescence was restored in the apical crypt of the same cell (E). Merged immunofluorescence showed that the high degree of colocalization between FAK pY407 and CFTR in the apical crypts was also maintained under hypertonic conditions (F). Scale bars for A–C and D–F, 10 μm.
Fig. 6.
Fig. 6.
Immunocytochemistry for CFTR and FAK pY576 [primary antibodies: mouse monoclonal anti-human CFTR and rabbit anti-phosphorylated human FAK pY576; secondary antibodies: goat anti-mouse IgG Alexa Fluor 488 (green) and goat anti-rabbit IgG Alexa Fluor 594 (red)]. In isotonic conditions, CFTR (A) and FAK pY576 immunofluorescence (B) were present in the apical crypt of MR cells. The insets show an XZ view of CFTR and FAK pY576 immunofluorescence in the apical crypt (N=12). (C) Merging panels A and B determined the colocalization of CFTR and FAK pY576 (yellow). The inset shows an XZ view of CFTR and FAK pY576 colocalization in the apical crypt (N=12). (D) A line scan of fluorescence intensity (arbitrary units) versus distance in μm across the apical crypt of a double labelled MR cell (indicated by the white line in C) showing a high degree of correspondence in the localization of CFTR (green line) and FAK pY576 (red line) in the crypt periphery following isotonic conditions. (E,F) CFTR and FAK pY576 distribution remained the same following 60 min hypotonic pretreatment (Hypo, N=6). (G,H) Hypotonic pretreatment for 60 min followed by forskolin addition, 10 min (Hypo + forskolin, N=6) and (I,J) 60 min hypotonic pretreatment followed by hypertonic shock, 10 min (Hypo + hyper, N=6). ac, apical crypt. Scale bar, 10 μm.
Fig. 7.
Fig. 7.
Immunoblots of CFTR and FAK pY407 opercular epithelia under isotonic conditions (Iso, N=3), 60 min hypotonic pretreatment (Hypo, N=3), 60 min hypotonic pretreatment followed by forskolin addition, 10 min (Hypo + forskolin, N=3) and 60 min hypotonic pretreatment followed by hypertonic shock, 10 min (Hypo + hyper, N=3). (A) CFTR bands were detected at ∼140 kDa in all treatments. (B) Strong bands were detected for FAK pY407 at ∼135 kDa in isotonic conditions and in pretreated membranes with hypotonic + forskolin and hypotonic + hypertonic shock. A weaker band was detected following hypotonic pretreatment alone (top panel). Reprobing the immunoblot with FAK pY576 (middle panel) revealed visible bands at ∼135 kDa for all treatments. A stronger band was detected (lane 2) following the hypotonic treatment compared with the previous immunoblot probed with FAK pY407. (C) CFTR, FAK pY407 and FAK pY576 were present in gill (positive control, N=3). Visible bands were detected at ∼140 kDa (CFTR) and ∼135 kDa (FAK pY407 and FAK pY576). Both CFTR and FAK proteins were absent from the heart tissue (negative control, N=3). (D) Quantitative western analysis of CFTR and FAK pY576 expression levels did not reveal any significant difference between the control and treated groups. However, FAK pY407 expression was significantly decreased after 60 min hypotonic shock compared with the control group, the hypotonic + forskolin treatment and the hypotonic + hypertonic treatment (**P<0.01). Statistical significance was determined by one way ANOVA followed by Bonferroni's post-test. The immunoreactive bands were quantified by densitometric scanning and normalized with respect to actin.
Fig. 8.
Fig. 8.
Co-immunoprecipitation of CFTR with FAK pY407 in killifish opercular epithelium (N=3). Proteins immunoprecipitated (IP) with CFTR and blotted using anti-FAK pY407 revealed a band at ∼140 kDa (lane 2). Similarly, co-immunoprecipitation of FAK pY407 with CFTR (IP with FAK pY407 and blotted using anti-hCFTR) revealed a band of the same molecular mass (lane 3). Positive controls for CFTR (IP with CFTR and blotted using anti-hCFTR) and FAK pY407 (IP with FAK pY407 and blotted using anti-FAK pY407) showed bands at ∼140 kDa and ∼135 kDa, respectively (lanes 1 and 4). There was an absence of bands in the negative controls (IP without primary antibodies and blotted with anti-hCFTR or anti-FAK pY407, lanes 5 and 6).
Fig. 9.
Fig. 9.
Transport activation in opercular epithelia of unstimulated (isotonic) conditions and following 60 min hypotonic shock pretreatment and 10 min post-treatment with cAMP activators. (A) In isotonic epithelia, transepithelial current (Im, μAcm–2) increased with the addition of forskolin (10 μmol l–1, N=6), isoproterenol (0.1 mmol l–1, N=6) and IBMX (0.1 mmol l–1, N=7). This increase was significant only with the addition of forskolin (*P<0.05). (B) Im decreased significantly following hypotonic shock (**P<0.01, ***P<0.001) with no recovery in the untreated control membranes, but there was a significant increase in Im after the addition of each cAMP activator (P<0.01, same concentrations as in A), returning to levels not significantly different from the initial values for each group. Stimulated levels were all significantly elevated when compared with hypotonic controls run in parallel (P<0.05). Statistical significance was determined by paired t-tests between test and control membranes and by one way ANOVA followed by Bonferroni's post-test for differences between all groups.
Fig. 10.
Fig. 10.
Effects of hypotonicity, clonidine and hypertonicity on opercular epithelia electrophysiology. Hypotonic treatment significantly decreased Im in resting epithelia (*P<0.05). Hypertonic shock after hypotonic inhibition significantly increased Im ( P<0.001) to levels significantly elevated compared with the initial state (P<0.05, N=8). Clonidine addition (10 μmol l–1, 60 min) had a more pronounced inhibitory effect than did hypotonic shock on Im (***P<0.001) and a control isotonic rinse following clonidine incubation produced no recovery of Im (N=18). Hypertonic shock after clonidine produced a moderate increase in Im but this was significantly lower than without clonidine, indicating hypertonicity affecting only part of the overall clonidine inhibition. Isotonicity levels were not restored and remained significantly different from the initial state and the hypertonic treatment without drug addition (***P<0.001 and §P<0.001, N=11). Statistical significance was determined by paired t-tests between test and control membranes and by one way ANOVA followed by Bonferroni's post-test for differences between all groups.

Similar articles

See all similar articles

Cited by 11 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback