Hypertonicity mediated by chloride upregulates the expression of the γ-subunit of Na-K-ATPase in cultured cells derived from the murine inner medullary collecting duct (IMCD3; Capasso JM, Rivard CJ, Enomoto LM, and Berl T. Proc Natl Acad Sci USA 100: 6428–6433, 2003). The purpose of this study was to examine the cellular locations and the time course of γ-subunit expression after long-term adaptation and acute hypertonic challenges induced with different salts. Cells were analyzed by confocal immunofluorescence and immunoelectron microscopy with antibodies against the COOH terminus of the Na-K-ATPase γ-subunit or the γb splice variant. Cells grown in 300 mosmol/kgH2O showed no immunoreactivity for the γ-subunit, whereas cells adapted to 600 or 900 mosmol/kgH2O demonstrated distinct reactivity located at the plasma membrane of all cells. IMCD3 cell cultures acutely challenged to 550 mosmol/kgH2O with sodium chloride or choline chloride showed incorporation of γ into plasma membrane 12 h after osmotic challenge and distinct membrane staining in ∼40% of the cells 48 h after osmotic shock. In contrast, challenging the IMCD3 cells to 550 mosmol/kgH2O by addition of sodium acetate did not result in expression of the γ-subunit in the membranes of surviving cells after 48 h. The present results demonstrate that the Na-K-ATPase γ-subunit becomes incorporated into the basolateral membrane of IMCD3 cells after both acute hyperosmotic challenge and hyperosmotic adaptation. We conclude that the γ-subunit has an important role in the function of Na-K-ATPase to sustain the cellular cation balance over the plasma membrane in a hypertonic environment.
- inner medullary collecting duct cells
- cell culture
- osmotic challenge
- sodium pump subunits
- immunocytochemical localization
the γ-subunit of Na-K-ATPase is a small (7 kDa) transmembrane protein present in the kidney in different nephron segments (1, 18, 24, 25), including the inner medullary collecting duct (IMCD) (17). However, the γ-subunit is absent in immortalized murine cells from the IMCD (IMCD3) cultured under isotonic conditions (6). Acute hyperosmolality caused by sodium chloride (NaCl) stimulates Na-K-ATPase gene expression in IMCD3 cells (16); hence this osmoregulated protein seems to have an important role in the adaptation to the hypertonicity (5).
Recent studies performed with Western blotting demonstrate a significant upregulation of the γ-subunit in IMCD3 cells acutely exposed to NaCl or choline chloride (ChoCl) or adapted to hypertonicity with small increments in NaCl concentration (6, 7). Importantly, this acute upregulation appears to be mediated by Cl and not by Na. If stock solutions of NaCl or ChoCl are added to culture media to reach a final osmolality of 550 mosmol/kgH2O and cell cultures are harvested after 48 h, both γa and γb can be detected on Western blots.
However, while Western blot analysis identifies an increase in γ-subunit protein, location within the cell remains unknown. The aim of the present study was therefore to determine the cellular location and the time course of γ-subunit expression in IMCD3 cells grown in isotonic medium and acutely exposed to 550 mosmol/kgH2O with media containing NaCl, ChoCl, or Na acetate (NaAc). Cell cultures were fixed at different time intervals of upregulation or downregulation, immunostained with the antibodies against the COOH terminus of the γ-subunit (intracellular epitope) or the NH2 terminus of γb (extracellular epitope) of Na-K-ATPase, and analyzed by confocal and electron microscopy. The results show that both acute osmotic shock with Cl-containing media and gradual adaptation to higher osmolality induce γ-subunit expression in the basolateral plasma membrane of IMCD3 cells.
MATERIALS AND METHODS
Rabbit polyclonal antibodies against the COOH terminus of the γ-subunit and the γb splice variant of Na-K-ATPase and corresponding purified peptides were kindly provided by Dr. Steven J. D. Karlish (The Weizmann Institute of Science, Rehovot, Israel). Alexa 488-conjugated goat anti-rabbit IgG and the nuclear stain To-Pro were from Molecular Probes (Leiden, The Netherlands).
A murine IMCD cell line (mIMCD3), originally established by Rauchman et al. (20), was propagated in 300, 600, and 900 mosmol/kgH2O medium. Adaptation of IMCD3 cells to hyperosmotic conditions was made as described by Capasso et al. (5). Briefly, when the cultures were 60–70% confluent, the appropriate volume of 5 M NaCl was added to the growth medium to increase the osmotic pressure of the medium by 50 mosmol/kgH2O. When the cultures were confluent, they were subcultured, and, after at least four passages in the hypertonic medium, some of the cultures were frozen in liquid nitrogen, whereas for others the process was restarted with a further increment of 50 mosmol/kgH2O. For the present experiments, control and adapted cells were grown in tissue culture flasks in 1:1 GIBCO DMEM/NUT mix F-12 containing 10% fetal calf serum and penicillin/streptomycin (100 U/ml and 100 μg/ml, respectively). GIBCO cell culture media, serum, and antibiotics were purchased from Invitrogen. Cell cultures for immunofluorescence microscopy and immunoelectron microscopy were grown on 12-mm-diameter cover glasses or polycarbonate coverslips in 24-well cell culture plates. To stimulate γ-subunit expression, confluent IMCD3 cells grown under isotonic conditions (300 mosmol/kgH2O) were acutely adjusted to 550 mosmol/kgH2O by addition of either 5 M NaCl, 5 M ChoCl, or 5 M NaAc according to Capasso et al. (6). The pH was adjusted to 7.5, and osmolality was measured using an Advanced Wide-Rage Osmometer 3W2 (Needham Heights, MA).
To study the time course of γ-subunit expression and incorporation into the plasma membrane, cell cultures were fixed at 0, 6, 12, 24, and 48 h after osmotic challenge with NaCl or ChoCl. Some of the cultures challenged for 48 h were returned to isotonic medium (300 mosmol/kgH2O) for an additional 48 h and fixed and prepared for immunocytochemistry.
Cultures for confocal immunocytochemistry of Na-K-ATPase subunits were fixed with acetone at −20°C or with 4% paraformaldehyde (PFA) in PBS at room temperature. PFA-fixed cells were permeabilized, and unspecific binding sites were blocked for 30 min with 10% fetal calf serum, 0.5% BSA, and 0.05 M glycine in PBS at room temperature with agitation. Fixed cultures were incubated with primary antibody in a solution containing 0.5% BSA and 0.1% Tween 20 in PBS in a humid chamber for 1 h at room temperature. After being rinsed, the cultures were incubated in a humid chamber for 1 h at room temperature with Alexa 488-conjugated goat anti-rabbit IgG (1:200) in the same solution as the primary antibody. The cells were counterstained for DNA with To-Pro (1:1,000) in PBS, then mounted with DAKO fluorescence mounting medium (DAKO Danmark, Glostrup, Denmark).
Confocal laser-scanning microscopy.
Cell cultures were examined with a Leica DM RXE fluorescence microscope equipped with a Leica TCS-SL confocal laser-scanning microscope using Leica Plan Apochromat ×40/1.25 numerical aperture and ×100/1.30 numerical aperture oil-immersion objectives. The fluorophores were excited using an Ar laser line at 488 nm and HeNe laser lines at 633 nm. Emission wavelengths were monitored between 500 and 535 (FITC) and 650 and 748 nm (CY5), respectively. Scanning resolution was 1,024 × 1,024 pix, and the line frequency was 400 Hz. The images were recorded at a z-level that demonstrated the lateral plasma membrane staining and at the same time usually also the nucleus. In some experiments, up to 30 optical sections in a series were obtained from IMCD3 cell cultures for three-dimensional (3D) reconstruction. The sections were recorded at 0.3-μm intervals in the z-direction.
In control experiments, IMCD3 cells adapted to 600 mosmol/kgH2O were labeled with γ-subunit antibodies preadsorbed with the relevant γ-peptide or with irrelevant antibodies (anti-human growth hormone). All controls were negative.
Cell cultures for immunoelectron microscopy were split, fixed with 4% PFA in PBS, pH 7.4, for 15 min, rinsed in PBS, sedimented at 2,000 g and resuspended in warm 10% gelatin, cooled on ice and cut into small blocks, infiltrated with 2.3 M sucrose overnight, and then frozen in liquid nitrogen (13). Ultrathin cryosections were incubated with polyclonal anti-γ-Na-K-ATPase COOH-terminal IgG for 45 min at room temperature, which was detected with goat anti-rabbit IgG antibody conjugated to 10-nm colloidal gold particles. The sections were stained with uranyl acetate and examined in a FEI Morgagni 268 transmission electron microscope. For general cell ultrastructure, IMCD3 cells grown at 300 mosmol/kgH2O or adapted to 600 mosmol/kgH2O were fixed in 2% glutaraldehyde and embedded in Epon.
The presence of the γ-subunit in membranes from IMCD3 cultures adapted to hypertonic conditions or acutely challenged was verified by Western blotting. The cells were rinsed with PBS and scraped off the culture dish on ice in 10 mM imidazole buffer, pH 7.2, containing protease inhibitors (catalogue no. 1 873 580, Roche). Cell lysates were homogenized, centrifuged at 2,000 g for 5 min at 4°C, and a fraction enriched with cell membranes was obtained from the supernatant by centrifugation at 200,000 g for 10 min. The pellet was dissolved in Laemmli buffer (12) containing 2% SDS and 5% DTT. The samples were run on 10% tricine-Tris polyacrylamide gels. The protein content was estimated using a BCA Protein Assay (Bio-Rad, Hercules, CA) based on the method of Bradford (4). Twenty micrograms of prepared sample were loaded in each well and separated on polyacrylamide gels. After transfer by electroelution to nitrocellulose membranes, the blots were blocked with 5% milk in PBS and incubated with antibody against the γ-subunit (COOH terminus) diluted 1:100. The labeling was visualized with horseradish peroxidase-conjugated secondary antibodies (P217; DAKO) diluted 1:5,000 using the enhanced chemiluminescence system (Amersham, Little Chalfont, UK).
Immunocytochemical localization of Na-K-ATPase γ-subunit in IMCD3 cells adapted to different osmolalities.
IMCD3 cells grown in 300 mosmol/kgH2O media did not show a immunocytochemical reaction to antibodies against the γ-subunit COOH terminus (Fig. 1A) or the γb splice variant (not shown). However, the plasma membrane of all IMCD3 cells adapted to 600 mosmol/kgH2O media revealed strong plasma membrane immunofluorescence with the anti-γ-subunit COOH-terminus antibody (Fig. 1B). A similar pattern was also observed for IMCD3 cultures adapted to 900 mosmol/kgH2O (not shown). The labeling of the plasma membrane in confluent cultures was distinct and continuous and included all cells. IMCD3 cells adapted to 600 and 900 mosmol/kgH2O medium also revealed distinct expression of the γb splice variant in the plasma membrane (Fig. 1, C and D, respectively). However, the intensity of the immunofluorescence with this antibody was lower overall than with the anti-γ-subunit COOH-terminus antibody. All labeled cells exhibited very low background fluorescence in the cytoplasm between the nucleus and the lateral cell membrane.
Transmission electron micrographs demonstrated confluent monolayers of IMCD3 cells in cultures grown in 600 mosmol/kgH2O media (Fig. 2). Interestingly, the basolateral membrane of the cultured cells did not show the complex basal membrane amplification characteristic of the IMCD3 cells in situ but was mostly tightly attached to the growth support (Fig. 2B, arrows). The cells exhibited scattered small microvilli at the apical plasma membrane. The cytoplasm was rich in mitochondria, rough endoplasmic reticulum, and prominent Golgi regions close to the nucleus (Fig. 2B). The ultrastructure of the adapted cells was qualitatively similar to cells growing under control conditions, but cell shape and dimensions varied with the age of the culture. Immunoelectron microscopy of split IMCD3 cells adapted to 600 mosmol/kgH2O substantiated that the γ-subunit of Na-K-ATPase was present in the plasma membrane (Fig. 2C). Here, immunogold particles marking the γ-subunit lined the plasma membrane, whereas the cytoplasm was unlabeled except for occasional small cytoplasmic vesicles.
Localization of the γ-subunit after acute osmotic challenge with different solutes.
IMCD3 cells exposed to an acute osmotic challenge by incubation in 550 mosmol/kgH2O media adjusted with either NaCl or ChoCl showed distinct upregulation of the γ-subunit within 48 h. The immunolocalization of the γ-subunit COOH terminus to the plasma membrane was distinct in large groups of cells (Fig. 3, A and B). Immunostaining for the γ-subunit COOH terminus after the cells were challenged with NaAc, however, revealed no plasma membrane staining and only faint granular fluorescence in the cytoplasm of some cells (Fig. 3C). These cultures also exhibited areas with decreased cell density and cell detachment.
When the osmolality of the medium was first increased from 300 to 550 mosmol/kgH2O for 48 h with ChoCl (Fig. 3D) or NaCl (not shown) and then adjusted back to 300 mosmol/kgH2O for 48 h, the membrane staining for the γ-subunit COOH terminus was completely absent in most cells, whereas in some cells faint fluorescence was still detected.
Plasma membrane labeling for the γb splice variant was equally distinct in large areas of the IMCD3 cell cultures incubated in 550 mosmol/kgH2O for 48 h with either NaCl or ChoCl but was weaker than for the γ-subunit COOH terminus (Fig. 3, E and F). Additionally, there was some labeling in the cytoplasm, possibly reflecting a slower or less complete transfer of the γb splice variant to the cell membrane in these cultures.
Optical sectioning of IMCD3 cells immunolabeled for the γ-subunit COOH terminus was used to analyze the 3D distribution of the γ-subunit in the cells. An example of a series of 30 optical sections, at intervals of 0.3 μm, is presented in Fig. 4, A–F, for IMCD3 cells challenged to 550 mosmol/kgH2O with ChoCl for 48 h and shows 6 optical sections separated by 0.9-μm intervals. The first image is located at the apex of the cells (A) and the last (F) at the base of the cells. The total thickness of the cell layer was 6.6 μm. The optical sections demonstrated that the γ-subunit was not present in the apical plasma membrane (Fig. 4A) but was predominantly along the lateral parts of the basolateral plasma membrane (Fig. 4, B–E) and clearly also, but less distinct, at the basal part (Fig. 4F). The absence of label in the apical membrane and presence in the basolateral membrane, in particular along its midportions, were well demonstrated in the 3D reconstruction (Fig. 5).
Time course of γ-subunit incorporation into cell membrane.
To elucidate when the γ-subunit was incorporated into the plasma membrane, the time course of its immunocytochemical appearance was investigated by challenging IMCD3 cells to hyperosmotic shock for 6, 12, 24, and 48 h compared with cultures kept in 300 mosmol/kgH2O medium at 0 and for 48 h (Fig. 6, A and B). The γ-subunit started to appear in the cytoplasm of challenged cells at 6 h after the onset of incubation, when faint fluorescing spots were visible in a few cells (Fig. 6C). After 12 h, more cells exhibited stronger fluorescence, mostly close to the nucleus, corresponding to the location of the Golgi apparatus, and in a few cells also at the plasma membranes (Fig. 6D). After 24 h, larger groups of cells showed labeled plasma membranes (Fig. 6E), and this labeling became more extensive after 48 h (Fig. 6F).
Immunoblotting with γ-subunit COOH-terminus antibody performed on membrane fractions from IMCD3 cells adapted to 600 mosmol/kgH2O (Fig. 7, lane 1) and IMCD3 cells acutely challenged with ChoCl to 550 mosmol/kgH2O (Fig. 7, lane 2) showed both variants of the γ-subunit (γa and γb). This is consistent with previous Western blot analyses of cell lysates of these cultures (7) and supports the immunocytochemical membrane observations described above.
Hypertonic NaCl and ChoCl cause IMCD3 cells to express the γ-subunit in plasma membranes. The upregulation of the γ-subunit of Na-K-ATPase has been previously demonstrated by immunoblotting in cell lysates of mIMCD3 cells adapted to hypertonic medium or acutely exposed to hyperosmotic challenge (6, 7). While these studies clearly demonstrated the upregulation of the γ-subunit, they neither localized it in the cell nor did they provide the time course for the appearance of the protein. This study demonstrates for the first time that the upregulated γ-subunit is routed to the plasma membrane in IMCD3 cells, both after long-term hyperosmotic adaptation and after acute hyperosmotic challenge for 48 h with either NaCl or ChoCl but not NaAc. These studies also demonstrate that 24 h are needed for the distinct localization of the protein to the plasma membrane. 3D confocal laser-scanning microscopy of the upregulated IMCD3 cells shows that the γ-subunit becomes located in the basolateral, but not apical, plasma membrane. This location thus resembles the location of the γ-subunit (COOH terminus) in the basolateral membrane of IMCD cells in vivo (17), demonstrating polarity also in the induced IMCD3 cells in vitro. Upregulation of γ-subunit expression was detected both with antibodies against the γ-subunit COOH terminus and against the NH2 terminus of splice variant γb. The observed overall stronger immunofluorescence with the former may be due to the simultaneous labeling of both splice variants γa and γb, which have the same COOH terminus (11), or to the higher affinity of the antibody or accessibility of the epitope. Also, our Western blot observations support the presence of both splice variants in the isolated membrane fractions.
Capasso et al. (5) concluded that MAPKs play a role in the response to acute changes in tonicity but that they are not central to the chronic adaptive response. Instead, other osmoprotective proteins, including Na-K-ATPase, appear to be central in the adaptive process. Importantly, Capasso et al. (7) have also shown that Cl, not Na, stimulates expression of the γ-subunit of Na-K-ATPase in IMCD3 cells and that the Jun kinase 2 (JNK2) is activated by hypertonicity. Replacement of NaCl with NaAc or pretreatment of IMCD3 cells with a Cl channel inhibitor completely blocked γ-subunit upregulation, inhibited JNK activation, and caused a significant decrement in cell survival in hypertonic conditions (7). These data confirmed that the replacement of Cl with Ac entirely abolished the appearance of the protein and suggested increased cell death as illustrated by the detachment of cells. Thus the absence of the γ-subunit may increase the adverse effects of hypertonicity, whereas its prescence is consistent with improved cell survival.
Incubation of cells from primary cultures of IMCD3 cells of rats (16), as well as Madin-Darby canine kidney cells (3) in media made hyperosmotic by addition of NaCl, increases both Na-K-ATPase mRNA and enzyme activity. Hypertonicity is also known to regulate many other proteins, for example, aquaporin-2 (23), glucose uniporter (2), growth arrest and DNA damage-inducible proteins (10), heat shock proteins (8, 15), cyclooxygenase 2 (26, 27), and p38 kinase, which is a member of the MAPK family (9). In these studies, hypertonicity has been achieved by using NaCl or NaCl and urea together.
Unlike acute hyperosmolarity, chronic hyperosmolarity failed to activate MAPKs (22). The mechanisms that determine whether cells survive, or go to apoptosis and die because of the hyperosmotic medium, seem to be very complex and depend crucially on factors such as the time course of the osmotic challenge, range of the osmotic steps, and combinations of osmolytes (14, 19, 21). Kültz et al. (10) demonstrated that growth of IMCD3 cells is arrested for ∼18 h after the onset of hyperosmotic shock (600 mosmol/kgH2O) but without an indication of imminent cell death. They also showed that ERK, SAPK1 (JNK), and SAPK2 (p38) were hyperosmotically activated in IMCD3 cells.
In conclusion, because the synthesis and incorporation of the γ-subunit into the plasma membrane of IMCD3 cells are induced by the hypertonic environment and reversibly downregulated when cell cultures are returned to an isotonic environment, the γ-subunit appears to have an important role in the function of membrane-bound Na-K-ATPase in maintaining cellular cation gradients in hypertonic environments. It is also of interest that when cells are exposed to hypertonicity with NaAc, they fail to express the γ-subunit,reflecting the importance of this protein in the adaptive process.
This work was supported by the Water and Salt Research Center established and supported by the Danish National Research Foundation (Grundforskningsfonden), the Danish Medical Research Council, and the University of Aarhus, and National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-19928 (to T. Berl).
We thank Tina Drejer and Albert Meier for excellent technical assistance.
Part of this work was presented at the 2003 Annual Meeting of the American Society of Nephrology in San Diego, CA, and has been published in abstract form (J Am Soc Nephrol 14: 314A, 2003).
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.
- Copyright © 2005 the American Physiological Society