Am J Physiol Cell Physiol Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Cell Physiol 292: C1854-C1866, 2007. First published January 17, 2007; doi:10.1152/ajpcell.00474.2006
0363-6143/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
292/5/C1854    most recent
00474.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hoffmann, E. K.
Right arrow Articles by Pedersen, S. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hoffmann, E. K.
Right arrow Articles by Pedersen, S. F.

MEMBRANE TRANSPORTERS, ION CHANNELS, AND PUMPS

Shrinkage insensitivity of NKCC1 in myosin II-depleted cytoplasts from Ehrlich ascites tumor cells

Else K. Hoffmann and Stine F. Pedersen

Department of Molecular Biology, University of Copenhagen, Copenhagen, Denmark

Submitted 1 September 2006 ; accepted in final form 11 January 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Protein phosphorylation/dephosphorylation and cytoskeletal reorganization regulate the Na+-K+-2Cl– cotransporter (NKCC1) during osmotic shrinkage; however, the mechanisms involved are unclear. We show that in cytoplasts, plasma membrane vesicles detached from Ehrlich ascites tumor cells (EATC) by cytochalasin treatment, NKCC1 activity evaluated as bumetanide-sensitive 86Rb influx was increased compared with the basal level in intact cells yet could not be further increased by osmotic shrinkage. Accordingly, cytoplasts exhibited no regulatory volume increase after shrinkage. In cytoplasts, cortical F-actin organization was disrupted, and myosin II, which in shrunken EATC translocates to the cortical region, was absent. Moreover, NKCC1 activity was essentially insensitive to the myosin light chain kinase (MLCK) inhibitor ML-7, a potent blocker of shrinkage-induced NKCC1 activity in intact EATC. Cytoplast NKCC1 activity was potentiated by the Ser/Thr protein phosphatase inhibitor calyculin A, partially inhibited by the protein kinase A inhibitor H89, and blocked by the broad protein kinase inhibitor staurosporine. Cytoplasts exhibited increased protein levels of NKCC1, Ste20-related proline- and alanine-rich kinase (SPAK), and oxidative stress response kinase 1, yet they lacked the shrinkage-induced plasma membrane translocation of SPAK observed in intact cells. The basal phosphorylation of p38 mitogen-activated protein kinase (p38 MAPK) was increased in cytoplasts compared with intact cells, yet in contrast to the substantial activation in shrunken intact cells, p38 MAPK could not be further activated by shrinkage of the cytoplasts. Together these findings indicate that shrinkage activation of NKCC1 in EATC is dependent on the cortical F-actin network, myosin II, and MLCK.

F-actin; Na+-K+-2Cl– cotransporter; myosin light chain kinase; protein kinase A

THE UBIQUITOUS PLASMA MEMBRANE Na+,K+,2Cl– cotransporter, NKCC1 (also known as SLC12a2 or BSC2) is activated by cell shrinkage, low intracellular Cl– concentration ([Cl–]i), and a wide variety of hormones and growth factors and plays important roles in transepithelial transport, maintenance of [Cl–]i, cell volume regulation, and possibly cell cycle progression (18, 23, 30, 51). NKCC1 exhibits consensus sites for multiple protein kinases, including protein kinase C (PKC), casein kinase (CK2), and in most species also protein kinase A (PKA) (51), and Ser/Thr phosphorylation of NKCC1 is increased in vivo in response to activating stimuli (18, 51). Consistent with the notion that phosphorylation is determinant in regulation of NKCC1, Ser/Thr protein phosphatase inhibitors such as calyculin A (CLA) potently stimulate NKCC1 in a variety of cell types, including Ehrlich ascites tumor cells (EATC) (11, 29, 35, 36, 41, 47), and the Ser/Thr protein phosphatases PP1 (6) and PP2A (34) both have been shown to associate with and regulate NKCC1.

Several Ser/Thr kinases have been proposed to play a role in the shrinkage-induced regulation of NKCC1. There is strong evidence for the involvement of the Ste20/SPS1-related proline-alanine-rich kinase (SPAK) in shrinkage-activation of NKCC1 (9, 13), and the SPAK-related oxidative stress response kinase 1 (OSR1) also interacts directly with NKCC1 (49). For other kinases, evidence is less substantial or incompatible with a role in shrinkage-activation. PKA is not required for shrinkage-activation of NKCC1, although it does activate NKCC1 after other stimuli (18). In several cell types, including EATC, p38 mitogen-activated protein kinase (MAPK) is activated by cell shrinkage (1, 12, 15, 46, 50), and a complex consisting of SPAK and NKCC1 has been suggested to regulate p38 MAPK activity (48), yet, p38 MAPK has, conversely, also been proposed to inhibit NKCC1 (17). The c-Jun NH2-terminal kinase (JNK) was assigned a role in phosphorylation and activation of NKCC1 by cell shrinkage (27); however, shrinkage-activation of NKCC1 can occur in the absence of increased JNK activity (33).

The myosin light chain (MLC) is phosphorylated by osmotic shrinkage (3, 28, 54), and the MLC kinase (MLCK) inhibitor ML-7 has been shown to inhibit shrinkage-activation of NKCC1 in a variety of cell types, including EATC (4, 28, 29, 40), albeit with highly cell type-dependent IC50 values. On the other hand, findings in kidney epithelial cells strongly indicate that shrinkage-activation of NKCC1 does not require increased MLC phosphorylation, raising the question of whether the effect of high concentrations of ML-7 on NKCC1 activity in fact reflects inhibition of MLCK (3; see Ref. 5). A certain basal myosin activity did, however, appear to be involved in shrinkage-activation of NKCC1 in kidney epithelial cells (3). Consistent with this, myosin II translocates to the cortical region after osmotic shrinkage in EATC (43) and in Dictyostelium (31).

Substantial evidence also points to a role for F-actin in regulation of NKCC1 (23, 26, 37, 38, 53; see Refs. 22, 44). In EATC and T84 intestinal epithelial cells, depolymerization of F-actin by cytochalasins or cell swelling elicits a modest increase in NKCC1 activity (23, 25, 26, 37, 38). Conversely, cell shrinkage, which in EATC and other suspended cells is associated with cortical F-actin polymerization (see Ref. 44), elicits robust NKCC1 activation inhibitable by cytochalasins and unaffected by phalloidin (26, 38). Moreover, regulatory volume increase (RVI), which in EATC is strongly NKCC1 dependent, is inhibited by the F-actin-disrupting toxin cytochalasin B (45). On the basis of these observations, we suggested a three-state model in which 1) F-actin integrity is necessary to maintain NKCC1 in a silent state, 2) if F-actin is disrupted, NKCC1 enters a partially activated, not further activatable, state, and 3) if the actin cytoskeleton is intact, cell shrinkage causes NKCC1 to enter a fully activated state (43–45).

The relationship between protein phosphorylation events and the actin cytoskeleton in shrinkage-induced NKCC1 activation is not clear. In airway epithelial cells, the role of F-actin in NKCC1 regulation was linked to PKC{delta} (53); however, in most cells, including EATC, PKC does not play a major role in shrinkage-activation of NKCC1 (29, 47; see Ref. 18). By treatment of EATC with cytochalasin B, sealed plasma membrane vesicles (cytoplasts) can be isolated (19). In these EAT cytoplasts, NKCC1 appears to be partially activated (20) and F-actin and myosin II levels appear to be reduced (39), suggesting that analysis of NKCC1 regulation in cytoplasts might contribute to the understanding of shrinkage-induced NKCC1 regulation. Thus the aim of this study was to examine whether, in cytoplasts, NKCC1 is shrinkage-activated in an MLCK-dependent manner and whether its regulation by other stimuli is similarly affected.

The findings indicate that in EAT cytoplasts, NKCC1 is partially activated under steady-state conditions, cannot be further activated by shrinkage, and is insensitive to the MLCK inhibitor ML-7, whereas regulation by other stimuli is intact. It is suggested that in EATC, an interaction among cortical F-actin, myosin II, and MLCK is required for shrinkage-induced NKCC1 activation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Materials, Ringer solutions, and stock solutions. Unless otherwise stated, all reagents were of analytical grade and obtained from Sigma-Aldrich (St. Louis, MO) or Mallinckrodt Baker (Deventer, The Netherlands). Heparin was obtained from Leo (Ballerup, Denmark). ML-7 and H89 were obtained from Calbiochem (Bad Soden, Germany) and was dissolved at 5 mM in 96% ethanol and at 1 mM in DMSO, respectively. Staurosporine and cytochalasin B were from Sigma-Aldrich and were dissolved at 500 µM in DMSO and 20.8 mM in 96% ethanol, respectively. CLA was obtained from Alomone Laboratories (Jerusalem, Israel) and was dissolved at 20 µM in 96% ethanol. Poly-L-lysine (250 mg/ml) was dissolved in double-distilled H2O (ddH2O). All these reagents were stored at –20°C until use. Stock solutions of paraformaldehyde (20% wt/vol in ddH2O) were prepared fresh regularly and kept at 4°C. Dowex 50 Wx8 was obtained from Fluka (Geneva, Switzerland). 86Rb was obtained from Risø (Roskilde, Denmark).

Monoclonal antibodies against NKCC1 and nonmuscle myosin II (both Developmental Studies Hybridoma Bank, University of Iowa, IA) were used in both Western blotting and immunocytochemistry at 1:250 dilution, and CMII23 at 1:100 dilution. SPAK antibodies (kind gifts from E. Delpire, Vanderbilt University Medical Center, Nashville, TN; H. Ushiro, Mie University School of Medicine, Japan; and G. Naselli, The Walter and Eliza Hall Institute of Medical Research, Victoria, Australia) were used at 1:50 or 1:100 dilution. Rabbit polyclonal antibodies against p38 MAPK, JNK, and phosphorylated p38 MAPK (Thr180/Tyr182) (all from Cell Signaling Technology, Beverly, MA) were used at 1:100 (p38 MAPK), 1:300 (JNK), and 1:100 dilution (p-p38 MAPK), respectively. Mouse monoclonal antibody against chicken gizzard MLCK (Sigma-Aldrich) was used at 1:200 dilution. Alkaline phosphatase-coupled secondary antibodies were obtained from Jackson ImmunoResearch Europe and used at 1:600 dilution.

The standard isotonic Ringer's solution contained (in mM) 143 NaCl, 5 KCl, 1 MgSO4, 1 Na2HPO4, 1 CaCl2, 3.3 MOPS, 3.3. TES, and 5 HEPES, pH 7.4. The hypertonic (600 mosM) medium was prepared by doubling the concentrations of all ions except Ca2+ compared with the standard medium (310 mosM) while maintaining the same concentrations of MOPS, TES, and HEPES.

Cell suspensions. EATC were grown in the ascites fluid of female NMRI mice by weekly intraperitoneal transplant. The transplantations were well tolerated by the mice, which showed no signs of discomfort or pain. Eight days after transplant the mice were killed by cervical dislocation according to the guidelines of the local ethical committee, and the cells in the ascites fluid were harvested in standard isotonic Ringer with added heparin (2.5 IU/ml), as described previously (21). All animal experiments were approved by the Danish National Committee for animal research (no. 231101-118). The cells were washed two to three times in heparin-free standard Ringer by centrifugation (700 g, 45 s, 37°C) and resuspended at a cytocrit of 4%. Before experiments, the cells were incubated for 30 min in the standard Ringer in a water bath (37°C) with gentle shaking.

Cytoplasts. Cytoplasts were prepared from the EATC essentially as described in Ref. 20. Cells were centrifuged and resuspended in a nominally Ca2+-free standard Ringer at a cytocrit of 9%. After 10 min, cytochalasin B (42 µM) was added, and the suspension was shaken for 2 min. This caused the formation of plasma membrane blebs, which were sheared off the cells by gentle homogenization (4 min) in a Dounce homogenizer using a loose-fitting pistil. These isolated blebs form the cytoplasts, intact vesicles devoid of organelles (which are contained in the remainder of the cells, the karyoplasts). This mixture was diluted threefold with Ca2+-free standard Ringer at 25°C, and the karyoplasts were removed by centrifugation at 250 g for 2 min. The supernatant was again centrifuged at 3,500 g for 2 min, and the pellet was resuspended and washed by centrifugation twice in standard Ringer (25°C) containing 1% BSA and finally suspended in standard Ringer without BSA at 20 mg wet wt/ml. To remove any remaining karyoplasts, we centrifuged the suspension at 250 g for 2 min and saved the supernatant. The pellet was suspended once more and centrifuged again (250 g, 2 min). The two last supernatants were pooled and centrifuged (3.500 g, 2 min) to collect the cytoplasts, which were then suspended in ice-cold standard Ringer at a concentration of 20–30 mg wet wt/ml and kept on ice. Cytoplasts were used within 90 min of preparation. Figure 1 shows the partially purified preparation before the two final centrifugation steps, containing both cells and cytoplasts (A), and the final cytoplast preparation (B), visualized by rhodamine-phalloidin labeling of F-actin (see below).


Figure 1
View larger version (42K):
[in this window]
[in a new window]

  		Fig. 1. Preparation of cytoplasts from Ehrlich ascites tumor cells (EATC). F-actin is shown in cytoplasts detaching from cytochalasin B (CB)-treated EATC (A) and in the final cytoplast preparation (B). The cytoplasts were prepared by treating EATC with 42 µM CB, followed by a series of purification procedures as described in MATERIALS AND METHODS. Cytoplasts were paraformaldehyde-fixed, permeabilized, and stained for F-actin using rhodamine-conjugated phalloidin as described in MATERIALS AND METHODS. Confocal images were acquired using a x100/1.4-NA objective and the 488- and 567-nm argon-krypton laser lines, and images were frame-averaged and are shown in RGB pseudocolor. Experiments shown are representative of 7 independent experiments.

 
Unidirectional 86Rb fluxes in cytoplasts. Unidirectional K+ influxes were measured using 86Rb as a tracer for K+, essentially as described previously (20 and 25). Cytoplasts were pelleted and suspended in the desired Ringer solution at 37°C for 1 min. The flux was initiated by diluting the cytoplast suspension 2:1 with Ringer of the same composition but containing 86Rb (350 kBq/ml) and ouabain (final concentration 1 mM) with or without addition of bumetanide (final concentration 30 µM), and samples were removed at the indicated times for separation of cytoplasts from the Ringer by ion exchange chromatography as previously described (10, 25). In brief, a 160-µl cytoplast suspension was applied to chilled cation exchange columns in the Tris form (Dowex 50, mesh 50–100, prepared in Pasteur pipettes) at the indicated times and washed through with 2x 750-µl ice-cold sucrose-BSA solution (250 mM sucrose, 10 mM MOPS, 1% BSA, pH 7.4 with Tris), thereby removing extracellular 86Rb from the cell suspension. Cytoplasts and Ringer free of 86Rb emerged in seconds and were transferred to counting vials and counted by liquid scintillation counting (Ultima Gold; Packard). Binding of cytoplasts to the columns was reduced by the presence of BSA and by preloading the columns by running the column with a sample of the cell suspension just before the experiment. Unidirectional K+ influx is presented as micromoles per gram of protein per minute, calculated from the initial linear increase in isotope counts in cytoplast lysates, the specific activities of the Ringer, and the protein content of the cytoplast suspension.

Coulter counter measurements of cytoplast volume. Cytoplast volume was measured as previously described for cell volume (21) by electronic cell sizing using a Coulter Multisizer II (Coulter, Luton, UK) after diluting 3 ml of cytoplast suspension into a final volume of 50 ml of standard Ringer solution. At the time indicated, osmolarity was increased to 650 mosM by addition of 3.6 ml of a 2.5 M NaCl stock solution. The tube orifice was 100 µm. The median cytoplast volume was calculated as the median of the cytoplast volume distribution curves after calibration with latex beads (diameter 39.2 µm).

Measurements of Cl– concentrations. Cell suspension samples (1 ml) and cytoplast suspension samples (1 ml) were transferred to preweighed, predried Eppendorf tubes for determination of ion content and cell water. The vials were centrifuged (20,000 g, 60 s), the supernatant was removed by suction, and the samples were weighed. Two 100-µl aliquots of the supernatant were saved and processed in parallel with the cell and cytoplast pellets for determination of extracellular Cl– concentration. Excess supernatant was carefully removed by suction, and the wet weight of the pellet was determined. The packed cells and cytoplasts were lysed in 800 and 200 µl of distilled water, respectively, deproteinized by addition of 100 and 25 µl of perchloric acid (70%), respectively, and centrifuged (20,000 g, 10 min). Supernatants were saved for determination of cellular ion content, and the dry weight of the PCA precipitate was determined after samples were dried for 48 h at 90°C and converted to cell dry weight by multiplication by 0.77, which was previously found to be the relation between cell dry weight and PCA precipitate dry weight (32). In one of the cytoplast preparations, the dry weight of the isotonic sample was lost, and that for the otherwise identical, parallel hypertonic sample was employed. The Cl– content in both extracellular and cellular samples was measured by coulometric titration using a CMY10 chloride titrator (Radiometer, Copenhagen, Denmark) as previously described (23). [3H]inulin (Amersham Biosciences) was used as a marker of extracellular space. The Cl– concentration is given as micromoles per gram of wet weight after correction for extracellular Cl– in the trapped volume ([3H]inulin space) in the cell pellet.

SDS-PAGE and Western blot analysis. Cells or cytoplasts were pelleted by centrifugation, washed quickly in ice-cold PBS, and added to 500 µl of boiling lysis buffer (10 mM Tris, pH 7.5, 1% SDS, 1 mM Na3VO4). Cells or cytoplasts were homogenized by repeated transfer through a 27-gauge needle. The lysates were cleared by brief centrifugation (16,000 g) to precipitate nonsoluble material. Protein content of cell and cytoplast pellets were determined by a modified Lowry method as described previously (20). Aliquots of 14–22 µg of protein (equal amounts in each well for a given experiment, calculated by protein content determination and verified by Ponceau S staining) were resolved on 10% NuPAGE Bis-Tris gels using NuPAGE MOPS SDS running buffer (NP0002), Fermentas protein standards, and a Novex Xcell (E19001) system (Novex, San Diego, CA). Separated proteins were electrotransferred to nitrocellulose membranes using the XCell II blot module (Novex), followed by staining in 1% Ponceau S red solution. Membranes were blocked [blocking buffer: 5% nonfat dry milk in 1x TBST (0.01 M Tris·HCl, pH 7.4, 0.15 M NaCl, and 0.1% Tween 20)] for 2 h at room temperature or overnight at 4°C before incubation with primary antibody in blocking buffer for 2 h at room temperature. The membranes were extensively washed, incubated with alkaline phosphatase-coupled secondary antibody for 1 h, washed, developed using a 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazolium solution (Kirkegaard and Perry Labs, Gaithersburg, MD), and quantified by densitometric scanning (UN-SCAN-IT software).

Immunocytochemistry. Cytoplasts were allowed to adhere on poly-L-lysine-coated no. 1 coverslips and fixed in 2% paraformaldehyde for 10 min at room temperature, followed by 30 min on ice. After three washes in TBS (150 NaCl, 10 Tris·HCl, 1 MgCl2, 1 EGTA, pH 7.3), the cytoplasts were permeabilized (0.2% Triton X-100 in TBS for 10 min), washed three times in TBS, blocked for 30 min in normal goat serum (1:10 in TBS plus 1% BSA), incubated with primary antibody [monoclonal anti-nonmuscle myosin (CMII23, 1:100)], SPAK (1:100; the 2 antibodies used were from Mie University School of Medicine and The Walter and Eliza Hall Institute of Medical Research), or monoclonal anti-NKCC1 (T4, 1:250) for 2 h, washed three times in TBS plus 1% BSA, incubated with FITC-coupled secondary antibody (1:400) plus rhodamine-conjugated phalloidin (2 U/ml) in TBS + 1% BSA for 1 h, and finally washed three times in TBS plus 1% BSA. The coverslips were mounted using antifade mounting medium (N-propyl-galleate 2% wt/vol in glycerol plus 10% 10x PBS) on glass slides with spacers to prevent cell compression.

Statistics and data analysis. Data are shown as individual experiments representative of at least three independent experiments or as means ± SE of at least three independent experiments. Significance was evaluated using two-sided Student's t-test, with P < 0.05 taken to indicate a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
To examine whether NKCC1 was regulated by osmotic shrinkage in EAT cytoplasts, we exposed suspensions of cytoplasts to a hypertonic challenge (600 mosM, compared with 300 mosM under isotonic conditions), and bumetanide-sensitive 86Rb influx was measured as a correlate of cotransporter activity (see Ref. 26). As shown, the bumetanide-sensitive 86Rb influx in the cytoplasts was unaltered by hypertonic exposure and was ~11 µmol·g protein–1·min–1 under both conditions (Fig. 2). This is more than an order of magnitude higher than the isotonic flux in intact EATC, yet several times lower than the shrinkage-induced flux in the intact cells (see DISCUSSION). To exclude the possibility that the refractoriness to activation by hypertonic exposure simply reflected that the cytoplasts did not exhibit osmotic shrinkage, we monitored cytoplast volume over time after hypertonic exposure using the Coulter Counter technique. The volume distribution curves for the cytoplasts in isotonic Ringer's solution and at various times after hypertonic exposure are shown in Fig. 3A, and the median volumes are shown in Fig. 3B. It should be noted that given the very small volume of the cytoplasts, the lower parts of the volume distribution curves are lost as background even when the smallest orifice possible is used; hence, the median volume, which is normally used as a measure of cell size at a given time, is not obtained from a full normal distribution curve. However, as shown by both the volume distribution curves and the median values, the cytoplasts shrank when exposed to hypertonic Ringer's solution and did not recover their volume within the time monitored (10 min).


Figure 2
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 2. Na+-K+-2Cl– cotransporter (NKCC1) is partially activated in EAT cytoplasts and cannot be further activated by hypertonic challenge. NKCC1 activity was estimated as bumetanide-sensitive, unidirectional 86Rb influx. Briefly, cytoplasts, prepared as described in MATERIALS AND METHODS, were suspended in the desired Ringer solution at 37°C for 1 min, after which the flux was initiated by mixing the cytoplast suspension 1:3 with Ringer containing 86Rb (350 kBq/ml) and ouabain (final concentration 1 mM) in the absence or presence of bumetanide (30 µM). Aliquots were removed at the indicated times for separation of cytoplasts from the Ringer on chilled cation exchange columns as described in MATERIALS AND METHODS. Cytoplasts free of extracellular 86Rb emerged in seconds and were transferred to counting vials and counted by liquid scintillation counting. The K+ influx is presented as µmol·g protein–1·min–1, calculated from the initial linear increase in counts in cell lysates, corrected for the specific activity of the Ringer and the protein content of the cytoplast suspension. Data are shown as means ± SE of 5 independent sets of paired experiments. In comparison, note that the bumetanide-sensitive 86Rb influx in intact EATC is ~0–1 µmol·g protein–1·min–1 under isotonic conditions and ~35 µmol·g protein–1·min–1 within 3 min after salt addition to double extracellular osmolarity (23, 25).

 

Figure 3
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 3. Effect of hypertonic exposure on cytoplast volume. Cytoplasts were prepared as described in MATERIALS AND METHODS, and cytoplast volume was measured as a function of time by electronic cell sizing using a Coulter Multisizer II. Size distribution curves were measured in isotonic standard Ringer and as a function of time after addition of NaCl to obtain a final osmolarity of 650 mosM. A: size distribution curves under isotonic conditions and at the indicated time points after hypertonic exposure. B: median cytoplast volume as a function of time after addition of NaCl. Data in A are representative of 5 independent sets of experiments, and the data in B are means ± SE of these 5 experiments.

 
Since cytosolic Cl– concentration is known to play a pivotal role in modulating the volume sensitivity of NKCC1, a low Cl– concentration being permissive for NKCC1 activity (see Ref. 51), we considered the possibility that a change in the Cl– concentration in cytoplast compared with that in intact cells might explain the observed differences in NKCC1 activity. However, this was not the case, because the Cl– concentrations were 65 ± 24.6 (n = 3) and 62 ± 14.7 (n = 4) µmol/g wet wt in cells and cytoplasts, respectively, under isotonic conditions (the extracellular Cl– concentration under these conditions was measured at 153 ± 1.8 mM) and 94 ± 24 (n = 3) and 77 ± 10 (n = 4) µmol/g wet wt in cells and cytoplasts, respectively, under hypertonic conditions (under which the extracellular Cl– concentration was measured at 313 ± 8.9 mM), with no significant differences between the intracellular/"intracytoplastic" Cl– concentrations in cells and cytoplasts.

Another possible caveat relates to the fact that, obviously, the relative levels of individual proteins are expected to differ between intact cells and cytoplasts. In particular, it seems probable that the relative plasma membrane area, and hence the levels of plasma membrane proteins such as NKCC1, is increased in cytoplasts compared with intact cells, and hence, the increased bumetanide-sensitive 86Rb influx might simply reflect this fact. It was therefore pertinent to compare the relative levels of NKCC1 protein in intact EATC and cytoplasts. Western blotting for NKCC1 in crude membrane lysates of EATC and cytoplasts indicated that the amount of NKCC1 in the cytoplasts, when normalized to micrograms of total protein, was increased ~2.5-fold compared with that in intact EATC (Fig. 4A), i.e., substantially less than the increase in bumetanide-sensitive 86Rb influx in cytoplasts compared with cells (see above). Immunocytochemical labeling of EAT cytoplasts with a monoclonal antibody against NKCC1, followed by visualization by confocal laser scanning microscopy (CLSM), confirmed that NKCC1 was present in the cytoplasts, where it localized to the cytoplast membrane along with F-actin (Fig. 4B). Thus these findings indicate that in EAT cytoplasts, NKCC1 appears to exhibit elevated basal activity yet to have lost the ability to be further activated by hypertonic stress.


Figure 4
View larger version (32K):
[in this window]
[in a new window]

  		Fig. 4. Expression and localization of NKCC1 in EAT cytoplasts. A: comparison of the relative levels of NKCC1 protein in intact cells and cytoplasts by Western blotting. EATC and cytoplasts were incubated in isotonic Ringer and lysed, and equal amounts of protein per lane were subjected to SDS-PAGE and Western blotting analysis using a mouse monoclonal antibody (T4) against NKCC1. Relative band intensity was estimated by densitometric analysis. At top is shown a representative Western blot of NKCC1 staining in cells and cytoplasts as indicated, and at bottom, the relative band intensity is shown as means ± SE (n = 3). *P < 0.05, significantly different from the isotonic control in intact EATC. B: localization of F-actin (red) and NKCC1 (green) in cytoplasts. Cytoplasts were paraformaldehyde-fixed, permeabilized, blocked with normal goat serum, and washed in TBST as described in MATERIALS AND METHODS. F-actin was labeled with rhodamine-conjugated phalloidin, and NKCC1 was labeled with T4 antibody followed by FITC-conjugated secondary antibody. Images were obtained as described in the legend to Fig. 1. In the absence of primary antibody, FITC fluorescence was essentially undetectable (not shown). The image shown is representative of 7 independent experiments.

 
The next step was to asses which elements of NKCC1 regulation might differ between intact EAT cells and cytoplasts and hence potentially underlie the observed differences in NKCC1 activity in the two preparations. In intact EATC, we have previously found that iso- and hypertonic NKCC1 activity was potentiated by the PP1 and PP2A inhibitor CLA, shrinkage-activation of NKCC1 was inhibited by the broad-spectrum Ser/Thr kinase inhibitor staurosporine and the MLCK inhibitor ML-7, and both shrinkage- and CLA-mediated NKCC1 activation were partially inhibited by the PKA inhibitor H89 (29). We therefore evaluated the effects of these compounds on NKCC1 activity in EAT cytoplasts, measured as bumetanide-sensitive 86Rb influx. As shown in Fig. 5, the basal NKCC1 activity in the EAT cytoplasts was strongly potentiated by 100 nM CLA and blocked by 1 µM staurosporine. Moreover, the basal NKCC1 activity in the cytoplasts was attenuated by ~50% after preincubation with H89 (2 µM) to inhibit PKA. Thus NKCC1 activity in the cytoplasts was potentiated by inhibition of PP1 and PP2A and was dependent on Ser/Thr protein kinase(s) and, at least in part, on PKA for its activity. In marked contrast, compared with intact cells (29), the basal NKCC1 activity in the cytoplasts exhibited a greatly reduced sensitivity to ML-7, with an IC50 of ~60 µM (Fig. 6, A and B). Thus NKCC1 activity in the cytoplasts was essentially unaffected by inhibition of MLCK, arguing against a role for this kinase in the high basal activity of NKCC1 in this preparation. To examine whether this in fact reflected a refractoriness of NKCC1 to regulation by MLCK or simply an absence of MLCK from the cytoplasts, we compared the relative levels of MLCK protein between intact EATC and EAT cytoplasts by Western blotting using a monoclonal antibody against MLCK (Fig. 6C). As shown, MLCK was present in the cytoplasts, although the amount of MLCK protein relative to total protein tended to be lower in cytoplasts than in intact EATC, although the difference was not quite statistically significant (Fig. 6C, P = 0.07).


Figure 5
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 5. Effects of Ser/Thr protein phosphatase and protein kinase inhibitors on NKCC1 activity in EAT cytoplasts. Effect of the Ser/Thr protein phosphatase inhibitor calyculin A (CLA; 100 nM, 1-min preincubation), the broad protein kinase inhibitor staurosporine (STS; 1 µM, 3-min preincubation), and the PKA inhibitor H89 (2 µM, 1-min preincubation) on NKCC1 activity in EAT cytoplasts. NKCC1 activity was estimated as bumetanide-sensitive, unidirectional 86Rb influx as described in the legend to Fig. 2, except for the presence of inhibitors as indicated. Data are shown as means ± SE of 29 (Ctrl), 4 (CLA), 3 (STS), and 4 (H89) independent experiments. *P < 0.05, significantly different from the control (Ctrl) condition.

 

Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 6. NKCC1 in EAT cytoplasts are insensitive to the myosin light chain kinase (MLCK) inhibitor ML-7. A and B: effect of ML-7 on NKCC1 activity. NKCC1 activity was estimated as bumetanide-sensitive, unidirectional 86Rb influx as described in the legend to Fig. 2, except that where indicated, ML-7 was present at the concentration indicated (3-min preincubation and present throughout the experiment). A representative experiment is shown in A. In B, fluxes are calculated relative to that under isotonic conditions in the same experiment, and data shown are means ± SE of 15 (Ctrl), 8 (10, 40, and 80 µM ML-7), 4 (20 µM ML-7), or 5 (60 µM ML-7) independent experiments. C: EATC and cytoplasts were incubated in isotonic Ringer and lysed, and equal amounts of protein per lane were subjected to SDS-PAGE and Western blotting using a mouse monoclonal antibody raised against chicken gizzard MLCK. The relative band intensity was estimated by densitometric analysis. In each experiment, the band intensity in both samples (cells and cytoplasts) was divided by the band intensity in the sample from the cells, which thus obtained the value of 1. It was subsequently tested using Student's t-test whether the relative value in the cytoplasts was significantly different from unity. A representative Western blot of MLCK staining in cells vs. cytoplasts is shown at top, and at bottom, the relative band intensity is shown as means ± SE (n = 3). CPs, cytoplasts. Note that the relative values of MLCK in cells and cytoplasts are not quite significantly different (P = 0.07).

 
As noted above, osmotic shrinkage of intact EATC is associated with cortical F-actin polymerization and translocation of myosin II to the cortical region (43, 45). It was therefore of interest to determine whether the distribution of F-actin and myosin II differed between cytoplasts and intact cells. This was investigated by immunocytochemistry and CLSM of a partially purified cytoplast preparation containing both cytochalasin-treated cells and cytoplasts. F-actin was labeled using rhodamine-conjugated phalloidin, and myosin II was labeled using a monoclonal antibody against nonmuscle myosin II. As shown, F-actin was present in both the cytoplasts and the whole cells (Fig. 7). In the cytochalasin-treated cells, the cortical ring of F-actin was irregular and appeared disrupted compared with control cells due to the cytochalasin treatment, in accordance with our previous findings in EATC (39). Similar to the cytochalasin-treated cells, the cytoplasts appeared to exhibit a disrupted, irregular cortical ring of F-actin. Myosin II was present in the cells predominantly in a perinuclear/Golgi region, in good agreement with our previous observations (39, 43). In marked contrast, myosin II was consistently absent from the cytoplasts (Fig. 7). The organization of F-actin and myosin II in an intact, non-cytochalasin-treated, osmotically shrunken EATC is shown for comparison (Fig. 7, inset).


Figure 7
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 7. Absence of nonmuscle myosin II in EAT cytoplasts. Cytoplasts were paraformaldehyde-fixed, permeabilized, blocked with normal goat serum, and washed in TBST as described in MATERIALS AND METHODS. F-actin (red) was labeled with rhodamine-conjugated phalloidin, and nonmuscle myosin II (green) was labeled with mouse monoclonal CMII23 antibody, followed by FITC-conjugated secondary antibody. In the absence of primary antibody, FITC fluorescence was essentially undetectable (not shown). Inset shows the organization of myosin II and F-actin in intact, osmotically shrunken EATC (similar data were previously published in Ref. 43). Fluorescence was analyzed by confocal laser scanning microscopy (CLSM) using a x100/1.4-NA objective and the 488- and 567-nm argon-krypton laser lines, and images were frame-averaged and are shown in RGB pseudocolor. Images shown are representative of 5 independent experiments.

 
Recent evidence has implicated the Ste20-related kinases SPAK and OSR1 in the activation of NKCC1 by cell shrinkage (9, 13, 14). It was therefore of interest to address the possible differences in the interactions between SPAK and NKCC1 in cytoplasts compared with intact cells. The localization of SPAK in intact EATC and cytoplasts under iso- and hypertonic conditions is shown in Fig. 8. As shown, SPAK was present in both the intact cells and the cytoplasts, and under isotonic conditions, SPAK exhibited a broad distribution in both preparations. Interestingly, upon hypertonic shrinkage, SPAK appeared to translocate to the cortical region in the intact EATC, whereas in cytoplasts, no shrinkage-induced translocation was detectable.


Figure 8
View larger version (34K):
[in this window]
[in a new window]

  		Fig. 8. Localization of Ste20-related proline- and alanine-rich kinase (SPAK) in EATC and cytoplasts under iso- and hypertonic conditions. Cells and cytoplasts prepared as described in MATERIALS AND METHODS and exposed to either iso- or hypertonic Ringer solution for 5 min, as indicated, were prepared for immunocytochemistry as described in the legend to Fig. 7. F-actin (red) was labeled with rhodamine-conjugated phalloidin, and SPAK (green) was labeled with polyclonal SPAK antibody, followed by FITC-conjugated secondary antibody. In the absence of primary antibody, FITC fluorescence was undetectable (not shown). Fluorescence was analyzed by CLSM using a x100/1.4-NA objective and the 488- and 567-nm argon-krypton laser lines, and images were frame-averaged and are shown in RGB pseudocolor. Images shown are representative of 4 (cytoplasts) and 3 (intact EATC) independent sets of experiments.

 
To further evaluate the involvement of SPAK and OSR-1 in NKCC1 regulation in EATC and cytoplasts, we compared the relative protein levels of SPAK and OSR1 in cytoplasts with the levels in intact cells. At a dilution of 1:100, the antibody employed (from Vanderbilt University) recognizes both SPAK and OSR1 (48, 49). In congruence with this, two bands of estimated molecular weights of ~60 and 67 kDa were detected in lysates of intact EATC as well as EAT cytoplasts (Fig. 9A). These bands are likely to correspond to OSR1 and SPAK, respectively (see Ref. 48), although the presence of splice variants of these proteins in EATC cannot be excluded. Compared with the amount relative to total protein in intact cells, the levels of both proteins were increased in cytoplasts about fourfold for the 60-kDa presumptive OSR1 band and about twofold for the 66-kDa presumptive SPAK band. Although protein levels are obviously not directly comparable between the two preparations because many other proteins are also expected to vary in their distribution, it may be noted that in contrast to the increase in SPAK and OSR1 levels, the amount of MLCK protein relative to total protein was decreased in the cytoplasts (see above), and the amount of p38 protein relative to total protein was unchanged (see below). This strongly indicates that different cytoplasmic proteins from the EATC sort differently to the cytoplasts (see also DISCUSSION).


Figure 9
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 9. Relative levels of oxidative stress response kinase 1 (OSR1), SPAK, p38 MAPK, and phosphorylated p38 MAPK in EAT cytoplasts compared with intact EATC. Intact EATC or cytoplasts were incubated in isotonic Ringer (300 mosM) for 20 min, and at time 0, NaCl (to a final osmolarity of 650 mosM) or isotonic Ringer was added, the cells/cytoplasts were incubated for 5 min and lysed, and equal amounts of protein per lane were subjected to SDS-PAGE and Western blot analysis using the relevant antibodies as described in MATERIALS AND METHODS. A: protein levels of OSR1, SPAK, and p38 MAPK in cytoplasts under isotonic conditions, relative to that in intact EATC under the same conditions (indicated by dashed line). B: specific activity of p38 MAPK [phosphorylated (p-)p38 MAPK/total p38 MAPK] in cytoplasts and EATC under iso- and hypertonic conditions. Inset: relative specific activity of p38 MAPK after osmotic shrinkage, i.e., the specific activity in hypertonic medium compared with that in isotonic medium (indicated by dashed line) in cells and cytoplasts. Results are shown as means ± SE of 3 independent experiments. *P < 0.05, significant difference compared with the value in cells under isotonic conditions. #P < 0.05, significant difference compared with the value in the same preparation (cells or cytoplasts, respectively) under isotonic conditions. P values for OSR1 and p38 MAPK relative to intact cells (A) are 0.08 and 0.06, respectively.

 
SPAK has been proposed to form a complex with NKCC1 and p38 MAPK regulating p38 MAPK activity, and an inhibitory role for p38 MAPK in regulation of NKCC1 has also been proposed (see Introduction). Hence, the relative level of total as well as active p38 MAPK protein was compared between intact EATC and EAT cytoplasts by Western blotting of crude membrane lysates from the two preparations with antibodies against total p38 MAPK and against the activated, phosphorylated form of this protein (phospho-Thr180/Tyr182 p38 MAPK), indicative of activation. As shown in Fig. 9A, the relative level of total p38 MAPK protein was unaltered in the cytoplasts compared with that in intact EATC. A marked difference in the pattern of p38 MAPK phosphorylation was, however, noted between the two preparations (Fig. 9B). First, the relative phosphorylation of p38 MAPK under isotonic conditions was increased ~1.5-fold in the cytoplasts compared with that in intact cells. Second, whereas in accordance with our previous findings (46) a 5-min hypertonic exposure elicited an approximately twofold increase in p38 MAPK phosphorylation in intact EAT, p38 MAPK phosphorylation in the cytoplasts was not further increased by hypertonic exposure (Fig. 9B, inset). Thus it appears that in cytoplasts, p38 MAPK activity is elevated under basal conditions yet cannot be further stimulated by hypertonic exposure.

Finally, it may be noted that the relative level of another MAPK implicated in shrinkage-induced NKCC1 regulation, JNK (27), was significantly reduced in the cytoplasts. Thus JNK1 (p54) and JNK2 (p46) protein levels in the cytoplasts were reduced to 0.53 ± 0.050 (n = 3, P < 0.05) and 0.30 ± 0.095 (n = 3, P < 0.05), respectively, relative to those in intact cells. The possible role of JNK in the altered NKCC1 regulation in the cytoplasts was, however, not further evaluated in the present study.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Although studies in several cell types point to roles for F-actin and myosin and/or MLCK in NKCC1 regulation, the mechanisms involved are incompletely elucidated. We previously noted that F-actin and myosin II content appeared dramatically altered in cytochalasin-induced membrane blebs on EAT cells (39) and that basal NKCC1 activity appeared to be increased in the sealed membrane vesicles (cytoplasts) sheared off from these cytochalasin-treated cells (20). These findings indicated that studies of EAT cytoplasts might provide important information regarding the relationship between cytoskeletal components and protein phosphorylation/dephosphorylation events in NKCC1 regulation.

The present study confirmed the usefulness of the cytoplast preparation in studying the involvement of F-actin, myosin II, and SPAK in the regulation of NKCC1. The cytoplasts are devoid of myosin II and exhibit a disrupted F-actin cytoskeleton while retaining many of the essential features of the intact cells (e.g., Cl– concentration, ability to shrink upon hypertonic exposure, and NKCC1 regulation by a number of protein kinase and phosphatase pathways). This notwithstanding, the cytoplast preparation is doubtlessly limited by the possible confounding roles of other differences not evaluated in the present study, including, for example, the presumably larger surface-to-volume ratio in the cytoplasts, and possible differences in the relative amounts of membrane transport proteins other than NKCC1 between the two preparations. Rather than providing direct causal evidence, the observations made by comparison of cytoplasts and intact cells are thus by definition correlational. However, similar limitations hold for other means of manipulating F-actin and myosin II, and the cytoplast preparation provides a valuable alternative approach.

The first and central observation made in the present study was that in the cytoplasts, NKCC1 appeared to be partially active but could not be further activated by osmotic shrinkage. The cytoplasts shrunk osmotically when exposed to a hypertonic challenge, yet the bumetanide-sensitive 86Rb flux was unaltered by shrinkage, amounting to ~11 µmol·g protein–1·min–1 under both conditions. In comparison, previously reported values in intact EATC were ~0–1 µmol·g protein–1·min–1 under isotonic conditions and ~35 µmol·g protein–1·min–1 within 3 min after salt addition to double extracellular osmolarity (23, 25). Justifying this comparison, it may be noted that with one single exception (29) in which the isotonic bumetanide-sensitive 86Rb fluxes were measured at ~7 µmol·g protein–1·min–1, bumetanide-sensitive 86Rb flux data from EATC are extremely reproducible and robust. Given the presumably larger surface-to-volume ratio in the cytoplasts, the relative amount of membrane proteins may be expected to increased. Consistent with this notion, the amount of NKCC1 per microgram of total protein in the cytoplasts was ~2.5 times that in intact EATC. If the bumetanide-sensitive 86Rb is corrected for the greater amount of NKCC1 in the cytoplasts, it can be estimated that in the cytoplasts, the bumetanide-sensitive 86Rb flux per cotransporter unit was increased at least about five times {11/([0.1–1]·2.5)} compared with that in isotonic EATC, and still about eight times [(35·2.5)/11] lower compared with that in hypertonically shrunken EATC.

The lack of shrinkage-induced NKCC1 regulation in cytoplasts does not reflect a complete refractoriness to regulation by increased Ser/Thr phosphorylation, since exposure to the Ser/Thr protein phosphatase inhibitor CLA activated NKCC1 by ~2.5 fold, whereas the broad-spectrum Ser/Thr kinase inhibitor staurosporine completely blocked NKCC1 activity, comparable to findings in intact EATC, with the exception that CLA potentiated the flux by ~8-fold in the intact cells (29). This indicates that in the cytoplasts, similarly to intact cells, NKCC1 is activated by inhibition of PP1 and/or PP2A and is dependent on Ser/Thr protein kinase(s) for its activity. The complete block of the high steady-state NKCC1 activity in the cytoplasts by staurosporine substantiates the notion that the high basal flux in the cytoplasts involves a deregulation of NKCC1. Together, the effects of staurosporine and CLA indicate that a Ser/Thr phosphorylation-dependent activation pathway, which is unable to stimulate NKCC1 under isotonic conditions in intact cells, leads to its constitutive activation in the cytoplasts. The lesser potentiation by CLA in cytoplasts compared with intact EATC (29) is consistent with the interpretation that the protein phosphatases involved in NKCC1 inactivation are less active in the cytoplasts compared with the intact cells. However, to unequivocally distinguish between increased kinase activity and decreased phosphatase activity in the cytoplasts, a more thorough analysis after activation with CLA, such as relaxation kinetic analysis, is required (see e.g., Ref. 24).

Which Ser/Thr kinase(s) is responsible for the increased basal NKCC1 activity in the cytoplasts? The high basal activity of NKCC1 in cytoplasts appears to in part (~50%) reflect a deregulation of the PKA-mediated activation of NKCC1, as judged from the inhibitory effect of the PKA inhibitor H89. Although this compound is generally considered a relatively specific inhibitor of PKA, it may be noted that at high concentrations, this compound has also been found to inhibit PKC (2) and, more surprisingly, to have some nonspecific inhibitory effects on various K+ channels not related to phosphorylation by PKA (42). Hence, as is true for all inhibitor studies, this finding should be interpreted cautiously. The pathway of PKA-mediated NKCC1 activation is not fully elucidated. The role of PKA in NKCC1 regulation has been proposed to be involve inhibition of PP1 (18), yet this seems incompatible with the finding that, at least in EATC, H89 inhibits CLA-mediated NKCC1 activation (29).

In contrast, NKCC1 activity in the cytoplasts was essentially unaffected by the MLCK inhibitor ML-7, which blocks shrinkage-induced activation of NKCC1 in intact EATC with an IC50 value of ~0.4 µM (29). The corresponding IC50 value in cytoplasts was ~150 time higher (~60 µM), a concentration at which ML-7 inhibits both PKA and PKC (IC50 ~21 µM for PKA and 42 µM for PKC, respectively, as reported in the Calbiochem catalog). It is notable that the loss of regulation by MLCK coincides with the loss of shrinkage activation of NKCC1, whereas regulation by other pathways (PKA, PP1/PP2A) is maintained. Although the ratio of MLCK to NKCC1 protein level was reduced in the cytoplasts, this cannot account for the complete refractoriness to regulation by MLCK in this preparation. A more plausible scenario was suggested by the finding that myosin II was absent from the cytoplasts and cortical F-actin organization disrupted. Hypertonic cell shrinkage has been shown to elicit MLC phosphorylation in a variety of cell types (3, 4, 28, 52, 54), and the effects of ML-7 on NKCC1 have therefore been proposed to reflect a dependence on MLC phosphorylation. However, in many cell types, non-MLCK-specific concentrations of ML-7 are required to inhibit NKCC1 activity (for a discussion, see Ref. 3), and at least in some cells, MLC phosphorylation is not necessary for shrinkage-induced activation of NKCC1. This was clearly demonstrated in kidney epithelial cells, where shrinkage-induced MLC phosphorylation was Rho kinase-dependent, whereas NKCC1 activation was Rho kinase-independent (3). Interestingly, in these cells, F-actin did not seem to play an important role in NKCC1 activation (3). Myosin per se does, however, appear to be important for shrinkage-induced NKCC1 activation also in kidney epithelial cells, because the myosin ATPase inhibitor blebbistatin reduced shrinkage-induced NKCC1 activation (3). In contrast to the findings in kidney epithelial cells, however, the IC50 for the inhibitory effect of ML-7 on the NKCC1 activity in intact EATC was so low (~0.4 µM) that it is highly probable that it reflects a dependence on MLCK.

In EATC, osmotic shrinkage is associated with a rapid increase in cortical F-actin and translocation of myosin II to the cortical region (43). In the present study, we showed that when cortical F-actin was disrupted and myosin II was absent, NKCC1 could be activated by shrinkage and was no longer regulated by MLCK, whereas regulation by other stimuli was maintained. Although we cannot exclude the possibility that other differences between the intact EATC and cytoplasts contribute to the observed effects, the present findings support the hypothesis that in EATC, the cortical F-actin-myosin II network plays an essential role in shrinkage-activation of NKCC1 and that MLCK is involved in this process. It seems likely that the controversy in the literature with respect to the roles of F-actin and MLC/MLCK in NKCC1 regulation may reflect the substantial differences in cytoskeletal arrangement between spherical suspended cells such as EATC, fibroblasts, and epithelial cells. In this regard it may also be noted that the relative roles of MLCK and ROK in regulation of MLC phosphorylation differ with subcellular localization such that MLCK is more important for peripheral MLC phosphorylation and ROK for stress-fiber associated MLC phosphorylation (55). It thus seems likely that in EATC, MLC-phosphorylation is MLCK-dependent; however, this remains to be experimentally verified. Another possible mechanism of regulation of NKCC1 expected to be affected by cytoskeletal reorganization is the translocation of NKCC1 to the plasma membrane, which has been shown to be modulated by other stimuli activating NKCC1 (7).

Shrinkage-induced NKCC1 regulation was recently proposed to be mediated by a sequence of events involving WNK4-mediated activation of the Ste20-related kinase SPAK, followed by SPAK-mediated phosphorylation and activation of NKCC1 (13, 14). Similar to SPAK, the closely related OSR1 has been shown to associate directly with NKCC1 (9, 13). The refractoriness of NKCC1 to shrinkage-activation in cytoplasts is not due to a lack of SPAK or OSR1, since the relative protein level of these kinases were increased in cytoplasts compared with intact cells. In contrast, we observed a marked difference in the effect of osmotic shrinkage on SPAK localization in the two preparations, in that osmotic shrinkage elicited SPAK translocation to the cortical region in intact cells yet not in cytoplasts. The shrinkage-induced SPAK translocation in intact cells is in congruence with the previously reported translocation of SPAK to F-actin upon osmotic stress (56). The present finding that SPAK translocates to the cortical region in the myosin II-containing intact cells, yet not in the myosin II-deficient cytoplasts, in conjunction with the lack of shrinkage-induced NKCC1 activation in the latter, strongly supports the notion that the cortical F-actin/myosin II network is important for the SPAK-dependent regulation of NKCC1 (see Ref. 8).

Several MAPKs, and in particular p38 MAPK and JNK, are activated by osmotic stress in various cell types, including EATC (12, 15, 46). Interestingly, p38 MAPK phosphorylation in cytoplasts follows the same pattern as seen for NKCC1, being increased under basal conditions yet unaffected by cell shrinkage. Because p38 MAPK has been assigned an inhibitory, rather than a stimulatory, role in NKCC1 regulation (16, 17), it seems unlikely that the altered p38 MAPK activity underlies the altered NKCC1 regulation; however, this remains to be directly determined. Since an NKCC1-SPAK complex has been proposed to associate with p38 MAPK and act as a scaffold for p38 MAPK signaling (48), it is also possible that, conversely, the altered regulation of p38 MAPK activity in cytoplasts may in fact be downstream from the corresponding change in NKCC1 regulation.

Figure 10 illustrates a tentative working model for NKCC1 regulation. It is suggested (state I) that the integrity of the actin cytoskeleton (including myosin II) is required to keep NKCC1 in a silent state under isotonic conditions, presumably due to a scaffolding role in controlling the activity and/or proximity of the kinases and phosphatases modulating NKCC1 phosphorylation. Under these conditions, the rates of phosphorylation and dephosphorylation are similar. F-actin depolymerization and loss of myosin II from the cortical region alter this control in favor of phosphorylation such that NKCC1 is rendered partially activated (state II). This is seen in EAT cytoplasts as well as in hypotonically swollen EATC (20, 25, 26), in which F-actin is also depolymerized and myosin II translocates away from the cortex to the Golgi/perinuclear region (43, 45). Osmotic shrinkage elicits cortical remodeling and the formation of a cortical F-actin-myosin II network (43) to which the kinase(s) mediating shrinkage-induced NKCC1 activation (e.g., SPAK) may translocate, resulting in NKCC1 phosphorylation and activation (state III). In the cytoplasts, this is prevented by the disruption of the cortical actin cytoskeleton, the lack of myosin II, and possibly also the reduction in MLCK, rendering NKCC1 refractory to shrinkage-induced activation. Finally, the basal level of NKCC1 activity is increased in the cytoplasts by a mechanism that appears to be entirely dependent on Ser/Thr kinase activity, in part mediated by PKA.


Figure 10
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 10. Summary model for regulation of NKCC1 in intact cells under isotonic conditions, after osmotic shrinkage and swelling, and in cytoplasts. The model summarizes findings in our present and previous studies regarding the relationship among NKCC1, Ser/Thr kinases and phosphatases, and F-actin/myosin II in EATC and EAT cytoplasts. See text for details. Data are from the present study and from Refs. 25, 26, 42, and 44.

 

   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This study was supported by Danish Natural Sciences Research Foundation Grants 21-01-0507 and 21-04-0535 (to E. K. Hoffmann and S. F. Pedersen), Carlsberg Foundation Grant 0894-10 (to E. K. Hoffmann), Danish Cancer Society Grant DP05072 (to E. K. Hoffmann and S. F. Pedersen), and Novo Nordic (to E. K. Hoffmann).


   ACKNOWLEDGMENTS
 
We are indebted to Birthe J. Hansen for excellent technical assistance.


   FOOTNOTES
 

Address for reprint requests and other correspondence: E. K. Hoffmann, Dept. of Molecular Biology, Univ. of Copenhagen, 13 Universitetsparken, Dk-2100 Copenhagen, Denmark (e-mail: ekhoffmann{at}aki.ku.dk)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
1. Bildin VN, Wang Z, Iserovich P, Reinach PS. Hypertonicity-induced p38MAPK activation elicits recovery of corneal epithelial cell volume and layer integrity. J Membr Biol 193: 1–13, 2003.[CrossRef][ISI][Medline]

2. Chijiwa T, Mishima A, Hagiwara M, Sano M, Hayashi K, Inoue T, Naito K, Toshioka T, Hidaka H. Inhibition of forskolin-induced neurite outgrowth and protein phosphorylation by a newly synthesized selective inhibitor of cyclic AMP-dependent protein kinase, N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide (H-89), of PC12D pheochromocytoma cells. J Biol Chem 265: 5267–5272, 1990.[Abstract/Free Full Text]

3. Ciano-Oliveira C, Lodyga M, Fan L, Szaszi K, Hosoya H, Rotstein OD, Kapus A. Is myosin light-chain phosphorylation a regulatory signal for the osmotic activation of the Na+-K+-2Cl– cotransporter? Am J Physiol Cell Physiol 289: C68–C81, 2005.[Abstract/Free Full Text]

4. Ciano-Oliveira C, Sirokmany G, Szaszi K, Arthur WT, Masszi A, Peterson M, Rotstein OD, Kapus A. Hyperosmotic stress activates Rho: differential involvement in Rho kinase-dependent MLC phosphorylation and NKCC activation. Am J Physiol Cell Physiol 285: C555–C566, 2003.[Abstract/Free Full Text]

5. Ciano-Oliveira C, Thirone AC, Szaszi K, Kapus A. Osmotic stress and the cytoskeleton: the R(h)ole of Rho GTPases. Acta Physiol (Oxf) 187: 257–272, 2006.[CrossRef][Medline]

6. Darman RB, Flemmer A, Forbush B. Modulation of ion transport by direct targeting of protein phosphatase type 1 to the Na-K-Cl cotransporter. J Biol Chem 276: 34359–34362, 2001.[Abstract/Free Full Text]

7. Del Castillo IC, Fedor-Chaiken M, Song JC, Starlinger V, Yoo J, Matlin KS, Matthews JB. Dynamic regulation of Na+-K+-2Cl– cotransporter surface expression by PKC-{epsilon} in Cl–-secretory epithelia. Am J Physiol Cell Physiol 289: C1332–C1343, 2005.[Abstract/Free Full Text]

8. Delpire E, Gagnon KB. SPAK and OSR1, key kinases involved in the regulation of chloride transport. Acta Physiol (Oxf) 187: 103–113, 2006.[CrossRef][Medline]

9. Dowd BF, Forbush B. PASK (proline-alanine-rich STE20-related kinase), a regulatory kinase of the Na-K-Cl cotransporter (NKCC1). J Biol Chem 278: 27347–27353, 2003.[Abstract/Free Full Text]

10. Dunham PB, Jessen F, Hoffmann EK. Inhibition of Na-K-C1 cotransport in Ehrlich ascites cells by antiserum against purified proteins of the cotransporter. Proc Natl Acad Sci USA 87: 6828–6832, 1990.[Abstract/Free Full Text]

11. Flatman PW. Regulation of Na-K-2Cl cotransport by phosphorylation and protein-protein interactions. Biochim Biophys Acta 1566: 140–151, 2002.[Medline]

12. Friis MB, Friborg CR, Schneider L, Nielsen MB, Lambert IH, Christensen ST, Hoffmann EK. Cell shrinkage as a signal to apoptosis in NIH 3T3 fibroblasts. J Physiol 567: 427–443, 2005.[Abstract/Free Full Text]

13. Gagnon KB, England R, Delpire E. Volume sensitivity of cation-Cl– cotransporters is modulated by the interaction of two kinases: Ste20-related proline-alanine-rich kinase and WNK4. Am J Physiol Cell Physiol 290: C134–C142, 2006.[Abstract/Free Full Text]

14. Gagnon KB, England R, Delpire E. Characterization of SPAK and OSR1, regulatory kinases of the Na-K-2Cl cotransporter. Mol Cell Biol 26: 689–698, 2006.[Abstract/Free Full Text]

15. Gillis D, Shrode LD, Krump E, Howard CM, Rubie EA, Tibbles LA, Woodgett J, Grinstein S. Osmotic stimulation of the Na+/H+ exchanger NHE1: relationship to the activation of three MAPK pathways. J Membr Biol 181: 205–214, 2001.[ISI][Medline]

16. Gosmanov AR, Lindinger MI, Thomason DB. Riding the tides: K+ concentration and volume regulation by muscle Na+-K+-2Cl– cotransport activity. News Physiol Sci 18: 196–200, 2003.[Abstract/Free Full Text]

17. Gosmanov AR, Thomason DB. Insulin and isoproterenol differentially regulate mitogen-activated protein kinase-dependent Na+-K+-2Cl– cotransporter activity in skeletal muscle. Diabetes 51: 615–623, 2002.[Abstract/Free Full Text]

18. Haas M, Forbush B, III. The Na-K-Cl cotransporter of secretory epithelia. Annu Rev Physiol 62: 515–534, 2000.[CrossRef][ISI][Medline]

19. Henius GV, Laris PC, Woodburn JD. The preparation and properties of cytoplasts from Ehrlich ascites tumor cells. Exp Cell Res 121: 337–345, 1979.[CrossRef][ISI][Medline]

20. Hoffmann EK, Jessen F, Dunham PB. The Na-K-2Cl cotransporter is in a permanently activated state in cytoplasts from Ehrlich ascites tumor cells. J Membr Biol 138: 229–239, 1994.[ISI][Medline]

21. Hoffmann EK, Lambert IH, Simonsen LO. Separate, Ca2+-activated K+ and Cl– transport pathways in Ehrlich ascites tumor cells. J Membr Biol 91: 227–244, 1986.[CrossRef][ISI][Medline]

22. Hoffmann EK, Mills JW. Membrane events involved in volume regulation. Curr Topics Membr 48: 123–196, 1999.

23. Hoffmann EK, Sjoholm C, Simonsen LO. Na+,Cl– cotransport in Ehrlich ascites tumor cells activated during volume regulation (regulatory volume increase). J Membr Biol 76: 269–280, 1983.[CrossRef][ISI][Medline]

24. Jennings ML, al Rohil N. Kinetics of activation and inactivation of swelling-stimulated K+/Cl– transport. The volume-sensitive parameter is the rate constant for inactivation. J Gen Physiol 95: 1021–1040, 1990.[Abstract/Free Full Text]

25. Jensen BS, Jessen F, Hoffmann EK. Na+-K+-2Cl– cotransport and its regulation in Ehrlich ascites tumor cells. Ca2+/calmodulin and protein kinase C dependent pathways. J Membr Biol 131: 161–178, 1993.[CrossRef][ISI][Medline]

26. Jessen F, Hoffmann EK. Activation of the Na+-K+-2Cl– cotransport system by reorganization of the actin filaments in Ehrlich ascites tumor cells. Biochim Biophys Acta 1110: 199–201, 1992.[Medline]

27. Klein JD, Lamitina ST, O'Neill WC. JNK is a volume-sensitive kinase that phosphorylates the Na-K-2Cl cotransporter in vitro. Am J Physiol Cell Physiol 277: C425–C431, 1999.[Abstract/Free Full Text]

28. Klein JD, O'Neill WC. Volume-sensitive myosin phosphorylation in vascular endothelial cells: correlation with Na-K-2Cl cotransport. Am J Physiol Cell Physiol 269: C1524–C1531, 1995.[Abstract/Free Full Text]

29. Krarup T, Jakobsen LD, Jensen BS, Hoffmann EK. Na+-K+-2Cl– cotransport in Ehrlich cells: regulation by protein phosphatases and kinases. Am J Physiol Cell Physiol 275: C239–C250, 1998.[Abstract/Free Full Text]

30. Kregenow FM. The response of duck erythrocytes to hypertonic media. Further evidence for a volume-controlling mechanism. J Gen Physiol 58: 396–412, 1971.[Abstract/Free Full Text]

31. Kuwayama H, Ecke M, Gerisch G, Van Haastert PJ. Protection against osmotic stress by cGMP-mediated myosin phosphorylation. Science 271: 207–209, 1996.[Abstract]

32. Lambert IH, Hoffmann EK, Jorgensen F. Membrane potential, anion and cation conductances in Ehrlich ascites tumor cells. J Membr Biol 111: 113–131, 1989.[CrossRef][ISI][Medline]

33. Liedtke CM, Cole TS. Activation of NKCC1 by hyperosmotic stress in human tracheal epithelial cells involves PKC-delta and ERK. Biochim Biophys Acta 1589: 77–88, 2002.[Medline]

34. Liedtke CM, Wang X, Smallwood ND. Role for protein phosphatase 2A in the regulation of Calu-3 epithelial Na+-K+-2Cl–, type 1 co-transport function. J Biol Chem 280: 25491–25498, 2005.[Abstract/Free Full Text]

35. Lionetto MG, Pedersen SF, Hoffmann EK, Giordano ME, Schettino T. Roles of the cytoskeleton and of protein phosphorylation events in the osmotic stress response in eel intestinal epithelium. Cell Physiol Biochem 12: 163–178, 2002.[CrossRef][Medline]

36. Lytle C. Activation of the avian erythrocyte Na-K-Cl cotransport protein by cell shrinkage, cAMP, fluoride, and calyculin-A involves phosphorylation at common sites. J Biol Chem 272: 15069–15077, 1997.[Abstract/Free Full Text]

37. Matthews JB, Smith JA, Hrnjez BJ. Effects of F-actin stabilization or disassembly on epithelial Cl– secretion and Na-K-2Cl cotransport. Am J Physiol Cell Physiol 272: C254–C262, 1997.[Abstract/Free Full Text]

38. Matthews JB, Smith JA, Mun EC, Sicklick JK. Osmotic regulation of intestinal epithelial Na+-K+-2Cl– cotransport: role of Cl– and F-actin. Am J Physiol Cell Physiol 274: C697–C706, 1998.[Abstract/Free Full Text]

39. Mills JW, Falsig PS, Walmod PS, Hoffmann EK. Effect of cytochalasins on F-actin and morphology of Ehrlich ascites tumor cells. Exp Cell Res 261: 209–219, 2000.[CrossRef][ISI][Medline]

40. O'Donnell ME, Martinez A, Sun D. Endothelial Na-K-Cl cotransport regulation by tonicity and hormones: phosphorylation of cotransport protein. Am J Physiol Cell Physiol 269: C1513–C1523, 1995.[Abstract/Free Full Text]

41. Palfrey HC, Pewitt EB. The ATP and Mg2+ dependence of Na+-K+-2Cl– cotransport reflects a requirement for protein phosphorylation: studies using calyculin A. Pflügers Arch 425: 321–328, 1993.[CrossRef][ISI][Medline]

42. Pearman C, Kent W, Bracken N, Hussain M. H-89 inhibits transient outward and inward rectifier potassium currents in isolated rat ventricular myocytes. Br J Pharmacol 148: 1091–1098, 2006.