点击显示 收起
【摘要】 The exclusive basolateral localization of the Na-K-ATPase in kidney epithelium is a critical aspect of nephron function. It has been suggested that mislocalized delivery of the Na-K-ATPase to the apical surface in autosomal dominant polycystic kidney disease (ADPKD) is due to the inappropriate expression of an alternative isoform of the -subunit, the 2 -isoform. It has been reported that heterologous expression of this 2 -isoform in Madin-Darby canine kidney (MDCK) cells results in apical delivery of the Na-K-ATPase. We created a MDCK cell line containing a tetracycline-inducible promoter and expressed either myc -tagged 2 - or flag-tagged 1 -subunits to study the surface localization of these -subunit isoforms in polarized monolayers. We find that the 2 -isoform is targeted to the basolateral surface of the plasma membrane in a polarization pattern indistinguishable from the 1 -isoform. However, inclusion of butyrate in the growth medium leads to upregulation of overexpressed 1 - or 2 -subunits and to their appearance at the apical surface. The 2 -isoform expressed in MDCK cells does not assemble into 1 2 heterodimers with the endogenous 1. Our findings demonstrate that expression of the 2 -isoform does not lead to apical localization of the Na-K-ATPase in MDCK cells and provides evidence for an unexpected effect of butyrate on the trafficking of Na pump subunits.
【关键词】 MadinDarby canine kidney cells apical localization sodium pump subunits
THE NA - K - ATPASE, OR THE SODIUM pump, plays a central role in ion regulation. At the cellular level, the functioning of the pump to transport Na + out and K + into the cell is essential for cell volume homeostasis and the maintenance of the inward Na + gradient that is used to fuel the accumulation of a range of substrates, such as amino acids and sugars. At the tissue level, the Na-K-ATPase has a key role in kidney epithelia to drive the uptake of Na + from the filtrate, thereby allowing the homeostatic control of fluid and electrolyte levels in the body. The directional uptake of Na + across epithelia relies on the polarized localization of the Na-K-ATPase at the basolateral surface in the epithelial layer of cells lining the nephron.
The minimal functional unit of the Na-K-ATPase is comprised of a heterodimer of an -subunit in complex with a -subunit ( 9, 16 ). The -subunit, which contains all the motifs essential for ATP hydrolysis and ion transport, is not trafficked out of the endoplasmic reticulum (ER) until it assembles with the -subunit ( 8, 19 ). Hence, one of the roles attributed to the -subunit is that of a molecular chaperone, involved in the folding, stabilization, and targeting of the -subunit (for a review, see Ref. 9 ). Localization signals ( 3, 13 ) and cytoskeletal binding motifs ( 7 ) have been identified in the -subunit, suggesting that the -subunit has a role in cellular sorting of the functional Na-K-ATPase heterodimer; however, other evidence establishes a role for the -subunit in polarized Na-K-ATPase targeting ( 34 - 36 ). Despite significant efforts to understand the mechanism of pump polarization, the signals and interactions that direct the localization are still not well established.
The most commonly expressed isoform of the -subunit, 1, is present in most tissue types, including the kidney and is generally associated with basolateral targeting of the Na-K-ATPase in polarized epithelial cells. The 2 -isoform, which was first identified as the adhesion molecule on glia (AMOG) ( 11 ), has restricted expression, limited primarily to particular regions in the brain ( 12, 20 ), epithelial cells of the eye ( 38 ), and inner ear ( 12 ), and is only normally found in the kidney during development before a strict basolateral polarization of the Na-K-ATPase is established ( 4 ). Abnormal apical localization of the Na-K-ATPase in kidney epithelia has been established in autosomal dominant polycystic kidney disease (ADPKD) ( 39, 40 ) and has been correlated with the inappropriate expression of the 2 -subunit in ADPKD cyst cells.
In an earlier report, the 2 -isoform, which is not normally expressed in MDCK cells, was observed at the apical surface ( 39 ) in stably transfected MDCK cells. This led to the suggestion that the apical delivery of the Na-K-ATPase in ADPKD is driven by the presence of the apically targeted 2 -isoform. However, a direct interaction of the 2 -isoform with the endogenous 1 -subunit was not established and the molecular factors leading to apical targeting of the 2 -isoform were not identified. In a more recent report, the 2 -isoform fused with YFP was found localized in the basolateral membrane when expressed in MDCK cells ( 34 ). The different 2 -subunit localization in MDCK cells observed in these two reports has not been accounted for.
We investigated the localization of the 1 - and 2 -subunit isoforms stably expressed in MDCK cells under tetracycline (tet) regulation. We find that the 2 -isoform, like the 1 -isoform, is exclusively at the basolateral surface of the cell membrane under normal growth conditions. This strict polarization is disrupted by the presence of butyrate in the growth medium. The presence of butyrate causes the overexpressed 1 - or 2 -subunits to appear at the apical as well as the basolateral surface in an isoform-independent fashion. We find that only the inappropriate localization of the 1 -isoform is associated with mistargeting of the -subunit and that the 2 -isoform does not assemble with the endogenous at a detectable level. Since butyrate was present in earlier work in which the 2 -isoform was observed apically in MDCK cells ( 39 ), our results provide an experimental basis for the observed apical delivery of 2 and document an unexpected effect of butyrate on Na pump localization.
METHODS
Cell maintenance. MDCK cell lines were maintained at 37°C in a humidified incubator with 5% CO 2 in DMEM supplemented with 25 mM HEPES buffer, 10% fetal bovine serum (tet screened), 10 U/ml penicillin, 10 µg/ml streptomycin, 2.5 µg/ml fungizone, and 5 µg/ml plasmocin. Media were supplemented with 500 µg/ml zeocin, 400 µg/ml hygromycin B, and/or 6 µg/ml blasticidin as appropriate. For induction of the gene of interest, 1 µg/ml tet was added to growth media. Media were supplemented with 10 mM sodium butyrate as indicated in specific experiments. Cell lines were split every 2-3 days after trypsin-EDTA treatment to detach cells from tissue culture plates.
Tetracycline-inducible MDCK, MDCK 2 myc, and MDCK 1 flag cells. The background cell line was a type I, high-resistance MDCK cell line ( 6, 7 ), which was converted into a MDCK/FlpIn cell line by using the Flp-In system (Invitrogen). This system creates an isogenic cell line with single Flp recombination target (FRT) sites ( 8 ). This MDCK/FlpIn line was a kind gift to us from the late Dr. R. B. Gunn (Emory University School of Medicine at Atlanta). To generate a tet-inducible line (MDCK/FlpIn/T-Rex), MDCK/FlpIn cells were stably transfected with pcDNA6/TR (GIBCO/Invitrogen) that constitutively expresses the tet repressor under the control of the human CMV promoter. Colonies of cells were selected, screened, expanded, and harvested according to manufacturer's protocol. The selected MDCK/FlpIn/T-Rex host cell line was cotransfected with pOG44 and pcDNA5/FRT/TO expression vector harboring either the rat 2 cDNA fused to a COOH-terminal myc tag ( 2 myc) or the sheep 1 cDNA with a COOH-terminal flag tag ( 1 flag) within the FRT site for homologous recombination. Integration of the gene of interest was selected with 800 µg/ml hygromycin B resistance to create the MDCK/FlpIn/T-Rex/ 2 myc and the MDCK/FlpIn/T-Rex/ 1 flag cell lines. Expression of the gene of interest in this system is repressed by the tet repressor and repression is relieved by the addition of 1 µg/ml tet to the growth media. For ease of communication, in this work the MDCK/FlpIn/T-Rex cells are simply referred to as MDCK, the MDCK/FlpIn/T-Rex/ 2 myc as MDCK 2 myc, and MDCK/FlpIn/T-Rex/ 1 flag as MDCK 1 flag.
MDCK cell membrane preparations. Cells were grown to confluence on 10-cm plates, washed twice with PBS, scraped, pelleted (1,000 g for 10 min), and stored at -20°C. Pellets were thawed and resuspended in ice-cold homogenizing buffer (HB: 10 mM Tris·HCl, 2 mM EDTA, 250 mM sucrose, pH 7.4) and broken open by dounce homogenization and/or passing through a low gauge needle. Unbroken cells were removed by centrifugation at 1,000 g for 10 min. To obtain total membrane preparations, the supernatant was spun in a TLA-55 rotor for 30 min at 55,000 rpm. Alternatively, the supernatant was loaded into a 5-step sucrose gradient and membrane-enriched fractions were separated by ultracentrifugation. Pellets were suspended in HB supplemented with 1 x protease inhibitor cocktail (Roche 1836153), and protein concentrations were determined by the Lowry method as previously described ( 15 ).
Na-K-ATPase activity. The difference in phosphate (P i ) liberated by 50 µg of total membrane protein in the presence or absence of 150 µM ouabain was determined as previously described ( 15 ) and is reported here in nanomoles of liberated P i per milligram of protein per hour.
Confocal imaging. Cells were grown in 6-mm transwells until an elevated transepithelial resistance was observed. Monolayers were washed twice with ice-cold PBS buffer supplemented with 0.1 mM CaCl 2 and 1.0 mM MgCl 2 and then fixed by submerging the transwell insert in 20°C acetone for 30 s. Cells were washed twice with PBS and then the transwell membranes were excised from the inserts and blocked overnight at 4°C in immunofluorescence blocking buffer (IFBB; 1% gelatin, 1% bovine albumin, 6 mg/ml donkey serum, 6 mg/ml goat serum, and 0.05% Na-azide in PBS). Cell monolayers attached to filter membranes were immunoprobed sequentially with 1:50 anti-KETYY (gift from Dr. J. Kyte, University of California, San Diego), 1:50 anti- myc (Cell Signaling no. 2276), 1:1,000 Cy 3-anti-mouse (Jackson Laboratories), and 1:1,000 Alexa 488-anti rabbit (Molecular Probes) diluted into IFBB. Samples were washed four times for 5 min in PBS after 1-h incubation with each antibody. Samples were mounted using vectashield hard-set mounting medium with DAPI (Vector Laboratories), and confocal images were collected on a Zeiss LSM 5 Pascal with a multitracking configuration optimized for detection of Cy 3, Alexa 488, and DAPI.
Growth in transwells. MDCK cells were plated at 50% confluence in Costar transwell-permeable supports with 0.4-µm-pore size. Cells in transwells were maintained as described above with the exception that they were fed daily with fresh supplemented DMEM but never split.
Measurement of transepithelial resistance. Monolayer transepithelial resistance ( R TE ) of MDCK cells grown in transwells was measured daily under sterile conditions with a World Precision Instruments EVOM Epithelial Voltohmmeter according to manufacturer's instructions. Resistance in Ohms was calculated per manufacturer's instruction according to
where d is the diameter of the transwell insert membrane growth area.
Cell surface biotinylation. The biotinylation protocol was based on the previously described method ( 14 ). Cells were grown in 24-mm transwells until polarized monolayers had formed. Plates were placed on ice and washed once with ice-cold DMEM (supplemented only with 25 mM HEPES) and twice with ice-cold PBS buffer supplemented with 0.1 mM CaCl 2 and 1.0 mM MgCl 2. Freshly prepared EZ-LINK Sulfo-NHS-SS Biotin label (Pierce) was added (0.5 mg/ml) to biotinylation buffer (10 mM triethanolamine, pH 7.5, 2 mM CaCl 2, 150 mM NaCl). Biotinylation reagent or buffer was added to the apical (0.5 ml) or basolateral compartment (1.5 ml) and incubated on ice with slow rocking in a 4°C cold room for 25 min. Biotinylation buffer was replaced with 1 ml (apical) and 2 ml (basolateral) quench buffer (PBS supplemented with 0.1 mM CaCl 2, 1.0 mM MgCl 2, and 100 mM glycine) and incubated on ice with slow rocking motion at 4°C for 20 min. This quench-wash step was repeated once, quench buffer was removed, and the permeable membrane was excised from the transwell with a scalpel and transferred to a 1.5-ml tube containing 0.75 to 1 ml lysis buffer (1% Triton X-100, 150 mM NaCl, 5 mM EDTA, 50 mM Tris, pH 7.5) for 1 h at 4°C with end-over-end rotation. The permeable membrane was removed, and samples were spun at 10,000 g for 10 min at 4°C. An aliquot (5-15% of sample) was removed to determine protein concentration and to compare protein expression. The remaining sample was incubated with 75 to 100 µl of streptavidin-coupled sepharose beads in a mini-spin column at 4°C overnight with end-over-end rotation. Buffer was removed from the column by centrifugation (500 g for 1 min at 4°C), and samples were washed three times with lysis buffer, twice with salt wash buffer (0.1% Triton X-100, 500 mM NaCl, 5 mM EDTA, 50 mM Tris, pH 7.5), and once with no-salt wash buffer (10 mM Tris, pH 7.5). After a final spin at 10,000 g for 5 min, 50 to 100 µl of 2 x sample buffer (SB) containing 80 to 100 mM DTT were added to beads. The DTT breaks the sulfhydryl (SS) bond between the biotin and the protein-reactive group in the label to elute labeled proteins from the beads. Samples were separated by SDS-PAGE and subjected to Western blot analysis.
Coimmunoprecipitation. Thirty micrograms of plasma membrane-enriched fractions were solubilized in 1% NP-40, 150 mM NaCl, 10 mM Tris, 2 mM EDTA, pH 7.5, for 1 h at 4°C with rotation. Samples were spun at 55,000 rpm in TLA-55 for 30 min, and supernatant was incubated with 20 µl of anti-Loop or 5 µl of anti- myc antibody for 1 h at 4°C. One hundred microliters of protein G-Sepharose beads were added, and samples were incubated overnight at 4°C with rotation. Samples were spun at 8,000 g for 30 s, supernatant was aspirated, and washed three times with solubilization buffer, once with 0.5 M NaCl, 10 mM Tris, 2 mM EDTA, pH 7.5, and once with water. Precipitate was eluted from beads with SB, separated by SDS-PAGE, and subjected to Western blot analysis.
Western blot. Samples in 1 x or 2 x SB supplemented with 80 to 100 mM DTT were heated to 80°C for 6 min and then subjected to SDS-PAGE and transferred to PVDF membranes. Blots were blocked with 5% milk in PBS for at least 1 h before immunoprobing in PBS with 0.5% milk, 0.1% Tween for 1 h to overnight followed by 3 x 10-min PBS-Tween washes. Antibodies were used for Western blot at the following dilutions: 1:400,000 of anti- 1 (ABR MA3-929), 1:400,000 of anti- 1 (ABR MA3-930), 1:500 of anti- 2 (BD Transduction Laboratories no. 610915), 1:1,000 of anti-KETYY (gift from Dr. J. Kyte, University of California, San Diego), 1:1,000 of anti- myc (Cell Signaling no. 2276), 1:1,000 of anti-flag M2 (Sigma F-3165), 1:500 of anti-E-cadherin (AbCam ab8996), 1:5,000 to 25,000 of horseradish peroxidase (HRP)-conjugated goat anti-mouse or anti-rabbit IgG. HRP-conjugated secondary antibodies and chemiluminescent reagents were used for signal detection.
RESULTS
Tetracycline-regulated overexpression of -subunits in MDCK cells. To study the polarized localization of -isoforms of the Na-K-ATPase, we created a system in which a protein can be selectively expressed in MDCK cells under the control of a tet-inducible promoter. The cDNA for either the rat 2 -isoform with a COOH-terminal myc tag or the sheep 1 -isoform with a COOH-terminal flag tag was introduced into the tet-controlled site and stable lines were isolated. Western blot analysis shown in Fig. 1 A demonstrates tet-induced expression of 2 myc, which is detectable with either the anti- myc ( panel 1 ) or anti- 2 ( panel 2 ) antibodies, in MDCK 2 myc. The expression of the 2 myc subunit does not greatly alter the levels of the endogenous 1 - or -subunits ( Fig. 1 A, panels 3 and 4 ). Western blot signal for the sarcoplasmic reticulum Ca pump (SERCA), a homolog of the Na-K-ATPase -subunit which functions independent of any -subunit, served as a loading control ( Fig. 1, A and B, bottom ).
Fig. 1. Tetracycline (tet)-dependent expression of 2 myc and 1 flag in Madin-Darby canine kidney (MDCK) cells. MDCK cells were transfected with pcDNA5/FRT/TR 2 myc or pcDNA5/FRT/TR 1 flag and stable cell lines were selected by growth in hygromycin B. MDCK 2 myc ( A ) or MDCK 1 flag cells ( B ) were grown in the absence (-) or presence (+) of tet at 1 µg/ml for 3 days and total membrane preparations were obtained. Equal concentrations of total membrane protein were separated by SDS-PAGE and Western blot analysis was performed detecting with anti- myc, anti- 2, anti- 1, anti-, anti-SERCA, or anti-flag antibodies as indicated. C : ouabain-sensitive Na-K-ATPase activity of total membrane preparation isolated from cells grown in the presence of tet was determined.
The 1 flag is only expressed in the presence of tet in the MDCK 1 flag cells, as revealed by Western blot with the anti-flag antibody ( Fig. 1 B, panel 1 ). Tet-induction of 1 flag in MDCK 1 flag cells markedly increases the level of 1 subunits ( Fig. 1 B, panel 2 ). Although an associated increase in translation of has been observed on expression of 1 in MSV-MDCK cells ( 25 ), we find that overexpression of 1 flag in the tet-inducible MDCK cells does not significantly increase the cellular level of the endogenous -subunit ( Fig. 1 B, panel 3 ). The Na-K-ATPase activity measurements presented in Fig. 1 C confirm that overexpression of 1 flag or 2 myc subunits by tet induction does not increase the cellular level of the ouabain-sensitive Na-K-ATPase activity, consistent with a stable cellular level of -subunits observed by Western blot analysis in Fig. 1, A and B.
Cellular localization of 2 myc in tet-induced cells. Confocal microcopy was performed to compare the cellular localization of the 2 myc with the endogenous Na-K-ATPase in tet-induced MDCK 2 myc cells. Figure 2 A, top left, shows the basolateral localization of the endogenous -subunit of the Na-K-ATPase in a focal plane located in the center of the cell monolayer, as detected with the -subunit COOH-terminal antibody, anti-KETYY. This basolateral localization of the endogenous Na-K-ATPase has been well established in MDCK cells. Unlike the endogenous pump, the 2 myc subunit was detected primarily in intracellular compartments by the anti- myc antibody ( Fig. 2 A, top right ). DAPI staining of the nuclei is presented in the bottom left and a merged image in the bottom right of Fig. 2 A. These images are consistent with previous observations ( 17 ) that the majority of the overexpressed 2 myc subunits are located in intracellular compartments.
Fig. 2. Cellular localization of the 2 myc isoform. A : tet-induced MDCK 2 myc cells were grown in polarized monolayers and subject to confocal imaging in which anti-KETYY, an antibody recognizing -subunit of the Na-K-ATPase, was detected with a Cy 3-conjugated secondary ( top left ) and anti- myc was detected with an Alexa 488 secondary ( top right ). DAPI stain indicates nulei ( bottom left ) and a merged image is presented in bottom right. B : alternatively, total membranes were obtained and separated into organelle-enriched fractions by ultracentrifugation in a step sucrose gradient. Equal amounts of protein from each organelle-enriched fraction were separated by SDS-PAGE and Western blot was performed with anti- 2, anti- 1, or anti- antibodies. Subunit carrying core ( C ) or higher-order modified ( H ) N-linked glycosylation chains are indicated.
To further characterize the localization of the 2 myc and to determine whether any of the expressed 2 myc was targeted to the cell surface, we performed fractionation of cell membranes isolated from tet-induced MDCK 2 myc cells. Results presented in Fig. 2 B establish that the 2 myc is present at significant levels in all three fractions ( Fig. 2 B, top ). Whereas the endogenous - and 1 -subunits are primarily located in the plasma membrane-enriched fraction ( Fig. 2 B, middle and bottom ), 2 myc in the plasma membrane-enriched fraction (P) migrates at a higher molecular weight than the predominant form in the endoplasmic reticulum-enriched fraction (E). Deglycosylation experiments using Endo H and PNGase F confirmed that the 2 myc in the ER is a core N-linked glycosylated form ( 2 myc C ) and the shift in molecular weight in the plasma membrane-enriched fraction is due to modification of the core glycosylation chains to a higher-order form ( 2 myc H ) (data not shown).
Polarized basolateral localization of 1 and 2 myc detected by surface labeling of polarized monolayers. To characterize the polarized distribution of surface -isoforms in MDCK cells, surface proteins of an MDCK 2 myc cell monolayer grown on a permeable support were biotinylated from either the apical (A) or basolateral (B) compartment. The Western blot in Fig. 3 A detects endogenous 1 -subunit only at the basolateral surface and not the apical surface. This result is consistent with the well-established selective targeting of the endogenous Na-K-ATPase to the basolateral surface of MDCK cells ( 5, 14 ). Figure 3 B demonstrates that the 2 myc subunit at the cell surface is also localized predominantly at the basolateral and not the apical membrane domain. This result establishes that the heterologous 2 -isoform at the cell surface exhibits the same basolateral polarization as the endogenous 1 -isoform and is in contrast to the 2 myc apical localization report by Wilson et al. ( 39 ).
Fig. 3. Polarity determination of -isoforms by cell surface biotinylation. Monolayers of MDCK 2 myc cells were grown on permeable supports until tight junctions had formed and cells were polarized. 2 myc Expression was induced by addition of 1 µg/ml tet for 2 days in the (+) conditions. An amine-directed biotin reagent was added to either the apical (A) or basolateral (B) compartment to label surface proteins. Cells were lysed in detergent buffer, and the labeled surface proteins were precipitated with streptavidin beads, eluted, and separated by SDS-PAGE. Western blot analysis was performed with anti- 1 ( A ) or anti- 2 ( B ) as detecting antibodies.
Butyrate alters surface polarity of 2 myc. When the 2 -subunit was detected at the apical surface of MDCK cells in previous work, butyrate was present in the growth media ( 39 ). However, butyrate was not present in Fig. 3 or in previous work in which the 2 -subunit fused to GFP was found exclusively in the basolateral surface of MDCK cells ( 34 ). We therefore examined the role of butyrate in the expression and targeting of Na-K-ATPase subunits. We found that butyrate treatment of tet-induced MDCK 2 myc cells increases the tet-induced expression of the 2 myc subunit in these cells ( Fig. 4 A, bottom ). We also observed a decrease in the cellular level of endogenous 1 -subunits in butyrate-treated MDCK 2 myc cells ( Fig. 4 A, middle ), although the cellular level of -subunits remained fairly consistent after butyrate treatment ( Fig. 4 A, top ).
Fig. 4. Alteration of 2 myc cell surface localization in the presence of butyrate. Polarized monolayers of MDCK 2 myc cells were grown on permeable supports in the presence ( A, B ) or absence ( B ) of tet. Growth media were supplemented with 10 mM butyrate (BT) for 48 h in (+) conditions. Surface proteins were labeled with biotin reagent from either the apical (A) or basolateral (B) compartment. Cells were lysed and separated into 2 pools. A : with 15% of tet-induced sample, total protein concentration was determined and equal concentrations of protein were separated by SDS-PAGE. Western blot analysis performed with the anti-, anti- 1, and anti- 2 antibodies. B : with 85% of the sample, labeled proteins were precipitated with streptavidin beads, and precipitant was separated by SDS-PAGE. Western blot analysis was performed with the anti- 2 antibody and then with the anti-KETYY antibody.
To determine whether butyrate influences the polarization of the 2 -subunit in MDCK cells, we biotinylated surface proteins from tet-induced or uninduced MDCK 2 myc cells grown in the presence or absence of butyrate. In the absence of butyrate, the 2 myc is only present on the basolateral surface ( Figs. 3 B and 4 B, top ). However, in the presence of butyrate, the 2 myc subunit can be labeled from both the apical and basolateral compartments ( Fig. 4 B, top ). Western blot analysis of the same samples detecting the Na-K-ATPase -subunit with the anti-KETYY antibody reveals that the endogenous -subunit remains polarized to the basolateral surface during butyrate treatment, regardless of 2 myc induction ( Fig. 4 B, bottom ). This demonstrates that the presence of butyrate does not allow the biotin label to cross the monolayer.
2 myc subunit does not associate with endogenous Na-K-ATPase -subunits in MDCK cells. Our observation that endogenous pump -subunit basolateral localization is not affected by altered distribution of the surface-targeted 2 myc subunit in the presence of butyrate ( Fig. 4 B, bottom ) suggests that the 2 myc does not assemble with the Na-K-ATPase -subunit in MDCK cells to form an 1 2 heterodimer. We therefore performed coimmunoprecipitations from the plasma membrane-enriched fraction of tet-induced MDCK 2 myc cells with antibodies directed at either the myc epitope or the endogenous -subunit to establish whether 1 2 assembly occurs. Figure 5 A, top, shows that the myc antibody is capable of immunoprecipitating the 2 myc subunit from a plasma membrane-enriched sample. However, the myc antibody does not coimmunoprecipitate the -subunit ( Fig. 5 A, bottom ). Likewise, an antibody directed at the major cytoplasmic loop of the -subunit which precipitates the -subunit ( Fig. 5 B, middle ) does not coimmunoprecipitate the expressed 2 myc ( Fig. 5 B, top ), although the endogenous 1 -isoform is coimmunoprecipitated under the same conditions ( Fig. 5 B, bottom ). Butyrate treatment for 2 days at 10 mM enhances the expression of the tet-induced 2 myc (see Fig. 4 A ) but does not alter coimmunoprecipitation results ( Fig. 5 C ). Control precipitations lacking protein input or antibody provided no evidence to suggest these immunoprecipitation results were artifact (data not shown). Thus the findings shown in Fig. 5 support the implications of the biotinylation data shown in Fig. 4 that the - and 2 -subunits do not assemble in the MDCK cells, although 1 does associate with endogenous 1 -subunits. It has previously been shown that 1 2 heterodimers can form in the absence of 1 ( 2, 6 ).
Fig. 5. Endogenous and 2 myc do not coimmunoprecipitate. Plasma membrane-enriched fractions were isolated from tet-induced MDCK 2 myc cells grown in the absence of butyrate ( A, B ) or the presence of butyrate ( C ) for 2 days. After solubilization in NP-40, 10% of the sample was removed to serve as an input control for Western blot. The remaining soluble sample was subject to immunoprecipitation with anti-myc ( A ) or anti-loop ( B and C; recognizes endogenous -subunit) antibody and protein G-Sepharose beads. Western blot analysis was performed with anti-, anti-myc, or anti- 1 as detecting antibodies.
Cellular effects of butyrate on MDCK cells. To further characterize the impact of butyrate on MDCK cells, we studied its effect in the background MDCK cell line. A dramatic decrease of the monolayer transepithelial resistance ( R TE ) is associated with butyrate treatment ( Fig. 6 A ). Normally, a stable high R TE develops as MDCK cells grow together into polarized monolayers and tight junctions are formed. The addition of butyrate causes these polarized cell layers to lose their high R TE. This loss of R TE is reversible on removal of butyrate (data not shown), consistent with previous work which demonstrated that MDCK monolayers remain polarized in the presence of low levels of butyrate ( 22, 24 ).
Fig. 6. Butyrate alters MDCK cell characteristics. MDCK cells were seeded at 50% confluence on permeable supports and allowed to grow into a confluent monolayer before 10 mM butyrate was added to the media for 2 days. A : transepithelial resistance was monitored. Butyrate was added to the monolayers plotted in open symbols and dashed black lines. Control monolayers were kept in normal media (closed symbols and gray solid lines). Cell surface biotinylation was performed on either the apical (A) or basolateral (B) surface and with 15% of sample ( B ), protein concentration was determined and equal amounts of total protein were separated by SDS-PAGE for Western blot analysis with the anti- or anti- 1 antibodies. C : biotinylated proteins were precipitated from 85% of the sample with streptavidin beads, separated by SDS-PAGE, and Western blots were probed with anti- or anti- 1 antibodies.
As we previously observed in the tet-induced MDCK 2 myc cell line ( Fig. 4 A ), a reduction of the total cellular Na-K-ATPase 1 occurs after butyrate treatment of the background MDCK cell line ( Fig. 6 B, bottom ). A smaller but consistent reduction of the -subunit in the total lysates was also observed ( Fig. 6 B, top ). It should be noted that equal protein concentrations were loaded in Fig. 6 B, so the proportionately larger reduction in the endogenous 1 -subunit cannot be explained by inhibited cell growth or cell death in the butyrate condition. Whether this observation reflects different degradation rates for the two subunits is the subject of ongoing work.
During butyrate treatment, both the endogenous - and 1 -subunits of the Na-K-ATPase in MDCK cells remain localized to the basolateral surface ( Fig. 6 C ). A reduced amount of surface-labeled endogenous Na-K-ATPase is seen in the presence of butyrate ( Fig. 6 C ), which is expected given the loss of Na-K-ATPase subunits, especially 1 -subunits during butyrate treatment. The conservation of the basolateral localization of the endogenous subunits during butyrate treatment is in sharp contrast to the loss of polarization of the overexpressed 2 myc isoform in Fig. 4 B.
Butyrate causes mislocalization of Na-K-ATPase overexpressed 1 -subunits. To determine whether the loss of basolateral polarization of the 2 myc subunit by butyrate ( Fig. 4 B ) is isoform specific, butyrate treatment and surface biotinylation were performed with MDCK 1 flag cells. These cells overexpress 1 flag from the same chromosomal site previously used for 2 myc expression. Figure 7 A, bottom, demonstrates that butyrate treatment enhances the tet-induced overexpression of the 1 flag in MDCK 1 flag cells, analogous to the increase of 2 myc observed in Fig. 4 A. Under the tet-induced butyrate-stimulated condition, the core glycosylated form of the 1 flag subunit ( 1 flag C ) is the major form of 1 detected ( Fig. 7 A, middle and bottom ), which is not the case in the absence of butyrate-stimulated overexpression (see Fig. 1 B ). The glycosylation status of this form has been confirmed by deglycosylation experiments with Endo H and PNGase F (data not shown). A decreased mobility of the 1 flag carrying higher-order glycosylation ( 1 flag H ) is also observed ( Fig. 7 A, middle and bottom ).
Fig. 7. Butyrate causes mispolarizaton of overexpressed 1 flag in MDCK 1 flag cells. MDCK 1 flag cells were grown on permeable supports in the presence of 1 µg/ml tet. Butyrate (10 mM) was present for 48 h before apical (A) or basolateral (B) surface labeling where indicated (+). After being labeled, cells were lysed and labeled proteins ( A ) were precipitated from 85% of sample with streptavidin beads and the precipitates were resolved by SDS-PAGE. Western blots were probed for, 1, flag, or E-cadherin. B : with the remaining sample, protein concentration was determined, and equal total protein amounts were separated by SDS-PAGE and subject to Western blot analysis with the anti- or anti-flag antibodies.
Surface labeling of tet-induced MDCK 1 flag cells reveals that the endogenous Na-K-ATPase subunits and the expressed 1 flag are selectively located in the basolateral surface in the absence of butyrate, as expected ( Fig. 7 B ). However, just as with the 2 myc, overexpressed 1 flag subunits are detected at the apical surface as well as the basolateral surface in the presence of butyrate ( Fig. 7 B, middle ). Unlike the mislocalized 2 myc, the endogenous -subunit is clearly mistargeted to the apical surface along with the mislocalized 1 flag ( Fig. 7 B, top ). To confirm that this apical detection is not due to an overall loss of cell polarity, Western blot detection was also performed for E-cadherin, which remains at the basolateral surface ( Fig. 7 B, bottom ) in the butyrate-treated tet-induced MDCK 1 flag cells.
DISCUSSION
In the present work, we overexpressed epitope-tagged forms of the 1 - and 2 -isoforms of the Na-K-ATPase -subunit in a tet-inducible MDCK expression system to compare the processing and polarized targeting of the Na-K-ATPase subunits. We found that both the 1 - and 2 -isoforms are localized exclusively to the basolateral membrane at the cell surface under normal growth conditions. When either the 1 - or 2 -isoform is expressed in the presence of butyrate, the overexpressed -subunit is detected at both the apical and basolateral surfaces.
Effect of -subunit overexpression on endogenous Na-K-ATPase levels. It has been previously reported that overexpression of the 1 -subunit leads to a corresponding increase in the expression of endogenous in some cell systems in which the level is a limiting factor in Na-K-ATPase production ( 10, 25, 27, 28 ). However, we have seen that the tet-induced expression of 2 myc or 1 flag in our MDCK cell system does not lead to a significant change in expression of either endogenous or 1 Na-K-ATPase subunits. Our findings suggest that expression of the -subunit is not a limiting factor for the formation and expression of Na-K-ATPase heterodimers in MDCK cells. This, in turn, suggests that the -subunit is the limiting factor for pump expression and that unassociated -subunits might be expressed in excess to the 1:1 ratio of necessary for the formation of the Na-K-ATPase heterodimer ( 23 ). Since an additional role for the -subunit as an adhesion molecule has been suggested ( 18, 31 ), it is important to determine whether there are unassociated -subunits present in cells and whether the -subunits perform a function beyond their role in the sodium pump.
Surface localization of the 2 -isoform expressed in MDCK cells. Under normal cell growth conditions, the 2 -isoform is strictly localized to the basolateral compartment when expressed in MDCK monolayers, just like the 1 isoform. We showed that the presence of butyrate in the growth media is the likely cause of 2 -isoform apical localization reported in previous work ( 39 ). This effect is not isoform specific, as butyrate disrupts the polarization of expressed -subunits when either the 1 - or 2 -isoforms of the -subunit are overexpressed. Our finding and those of others ( 34 ) demonstrate that there is no difference in the localization of the 2 -isoform when expressed in MDCK cells, contrary to previous claims. Although our work does not directly address the role that the 2 -isoform may play in PKD, it is clear that the expression of the 2 -isoform is not sufficient to account for mistargeting of the Na-K-ATPase in an otherwise normal system.
Cellular effects of butyrate in MDCK cells. Butyrate is a small fatty acid normally located in the digestive tract as a metabolic product of intestinal flora in vivo where it is the primary energy source for colonocytes. In vitro, butyrate has been observed to induce a number of cellular alterations, which vary between cell lines. The primary effect attributed to butyrate is inhibition of histone deacetylase activity, which leads to alterations in protein expression and can lead to cell cycle arrest, and eventually induce apoptosis (for a review, see Refs. 17, 21 ).
We observed a decrease in the total cellular level of the endogenous -subunit in the presence of butyrate. Interestingly, it has been demonstrated that the presence of the Na-K-ATPase -subunit is necessary for normal cell-cell contacts in MDCK cells ( 27 ) and suggested that interacting -subunits between cells may serve an adhesive role ( 31 ). It has been previously noted that the introduction of 1 into MSV-transformed MDCK cells allows those cells to regain elevated R TE values ( 26 ). The loss of R TE we observe when MDCK cells are treated with butyrate along with the drop in the -subunit level fit well with this finding and together suggest that the presence of the 1 -subunit may contribute to the high cellular R TE in MDCK cells.
A reduction of basolateral short-circuit current ( I SC ) attributed to decreased Na-K-ATPase activity has also been found in butyrate-treated MDCK cells ( 24 ). We showed that prolonged treatment with butyrate decreases the total level of endogenous pump subunits ( Figs. 4 A, 6 B, and 7 A ) and can reduce the Na-K-ATPase at the cell surface ( Figs. 4 A and 6 C ). This suggests that butyrate may stimulate the internalization and degradation of the Na-K-ATPase as well as potentially influencing transcriptional regulation of expression.
Although butyrate decreases the levels of endogenous -subunits, it increases expression off the CMV promoter ( 24, 37 ) and therefore leads to stimulated expression of the 2 myc or 1 flag ( Figs. 4 A and 7 A ) in the tet-induced MDCK cell system. A number of observations suggest that it is unlikely that the inappropriate apical localization of these subunits ( Figs. 4 B and 7 B ) is simply a result of the butyrate-enhanced expression overloading the cellular sorting and targeting mechanism. First, in the absence of butyrate, we do not observe inappropriate delivery of overexpressed -subunits even after prolonged tet induction which increases expression (data not shown). Second, in surface labeling experiments, we do not detect the core glycosylated 1 flag C at the plasma membrane ( Fig. 7 B ), although the cellular level of 1 flag C is greatly increased in the presence of butyrate ( Fig. 7 A ). Finally, we do not observe apical localization or decreased surface accessibility of E-cadherin in butyrate-treated cells, which indicates that the sorting and delivery mechanisms are functioning correctly for this basolateral membrane protein. Our finding of maintained E-cadherin polarization in the presence of butyrate is consistent with previous observations in human colon cancer cells ( 1 ). Based on the results presented here, it appears that butyrate specifically alters the delivery of newly expressed Na-K-ATPase -subunits and 1 1 heterodimers when expressed in MDCK cells but does not cause inappropriate localization of other basolaterally directed proteins. From our current results, it is not clear whether the changes in glycosylation that we observe in the presence of butyrate ( Figs. 4 and 7 ) are related to the mistargeting. However, a precedent for glycosylation of -subunits influencing polarization has been previously established ( 33 ).
Isoform-specific association of Na-K-ATPase subunits. Through the use of butyrate, we observed the mistargeting of endogenous MDCK Na-K-ATPase 1 -subunit to the apical surface with exogenously overexpressed 1 -subunit ( Fig. 7 B ). However, the endogenous 1 is not similarly mistargeted with the 2 -subunit expressed under the same conditions ( Fig. 4 B ). Furthermore, co-IP experiments under conditions in which 1 1 heterodimers are observed fail to show 2 association with ( Fig. 5 ). These findings suggest that the 1 -subunit in MDCK cells has a higher affinity for the 1 -isoform than for the 2 -isoform and are consistent with reports of lower affinity between subunits in an 1 2 than an 1 1 or 2 2 heterodimeric configuration of the Na-K-ATPase when expressed in amphibian oocytes or in insect cell lines ( 2, 29, 30 ).
In summary, we established that the 2 -isoform of the Na-K-ATPase is targeted to the basolateral surface of MDCK cells. We also show that the small fatty acid butyrate causes the inappropriate delivery of overexpressed 1 - and 2 -subunits to the apical surface of MDCK cells. This finding identifies a novel effect of butyrate in the polarized delivery of these important membrane proteins.
GRANTS
This work was supported by National Institutes of Health Grants GM-39500 and HL-30315.
ACKNOWLEDGMENTS
We thank the late Dr. R. B. Gunn (Emory University School of Medicine at Atlanta) for generosity in providing the MDCK/FlpIn cell line, Dr. J. Kyte (University of California, San Diego) for the gift of the anti-KETYY antibody, and Dr. R. Mercer (Washington University, St. Louis) for the gift of cDNA. We also thank Dr. S. Lutesenko (Oregon Health Sciences University) and Dr. J. Bystriansky (University of Illinois at Chicago) for critical comments on the manuscript.
【参考文献】
Barshishat M, Polak-Charcon S, Schwartz B. Butyrate regulates E-cadherin transcription, isoform expression and intracellular position in colon cancer cells. Br J Cancer 82: 195-203, 2000.
Blanco G, Mercer RW. Isozymes of the Na-K-ATPase: heterogeneity in structure, diversity in function. Am J Physiol Renal Physiol 275: F633-F650, 1998.
Blostein R, Zhang R, Gottardi CJ, Caplan MJ. Functional properties of an H-K-ATPase/Na-K-ATPase chimera. J Biol Chem 268: 10654-10658, 1993.
Burrow CR, Devuyst O, Li X, Gatti L, Wilson PD. Expression of the beta2-subunit and apical localization of Na + -K + -ATPase in metanephric kidney. Am J Physiol Renal Physiol 277: F391-F403, 1999.
Cereijido M, Shoshani L, Contreras RG. The polarized distribution of Na +, K + -ATPase and active transport across epithelia. J Membr Biol 184: 299-304, 2001.
Crambert G, Hasler U, Beggah AT, Yu C, Modyanov NN, Horisberger JD, Lelievre L, Geering K. Transport and pharmacological properties of nine different human Na, K-ATPase isozymes. J Biol Chem 275: 1976-1986, 2000.
Devarajan P, Scaramuzzino DA, Morrow JS. Ankyrin binds to two distinct cytoplasmic domains of Na-K-ATPase alpha subunit. Proc Natl Acad Sci USA 91: 2965-2969, 1994.
Gatto C, McLoud SM, Kaplan JH. Heterologous expression of Na + -K + -ATPase in insect cells: intracellular distribution of pump subunits. Am J Physiol Cell Physiol 281: C982-C992, 2001.
Geering K. The functional role of beta subunits in oligomeric P-type ATPases. J Bioenerg Biomembr 33: 425-438, 2001.
Geering K, Theulaz I, Verrey F, Hauptle MT, Rossier BC. A role for the beta-subunit in the expression of functional Na + -K + -ATPase in Xenopus oocytes. Am J Physiol Cell Physiol 257: C851-C858, 1989.
Gloor S, Antonicek H, Sweadner KJ, Pagliusi S, Frank R, Moos M, Schachner M. The adhesion molecule on glia (AMOG) is a homologue of the beta subunit of the Na-K-ATPase. J Cell Biol 110: 165-174, 1990.
Gonzalez-Martinez LM, Avila J, Marti E, Lecuona E, Martin-Vasallo P. Expression of the beta-subunit isoforms of the Na-K-ATPase in rat embryo tissues, inner ear and choroid plexus. Biol Cell 81: 215-222, 1994.
Gottardi CJ, Caplan MJ. Molecular requirements for the cell-surface expression of multisubunit ion-transporting ATPases. Identification of protein domains that participate in Na-K-ATPase and H,K-ATPase subunit assembly. J Biol Chem 268: 14342-14347, 1993.
Gottardi CJ, Dunbar LA, Caplan MJ. Biotinylation and assessment of membrane polarity: caveats and methodological concerns. Am J Physiol Renal Fluid Electrolyte Physiol 268: F285-F295, 1995.
Hu YK, Kaplan JH. Site-directed chemical labeling of extracellular loops in a membrane protein. The topology of the Na-K-ATPase alpha-subunit. J Biol Chem 275: 19185-19191, 2000.
Kaplan JH. Biochemistry of Na-K-ATPase. Annu Rev Biochem 71: 511-535, 2002.
Kruh J. Effects of sodium butyrate, a new pharmacological agent, on cells in culture. Mol Cell Biochem 42: 65-82, 1982.
Larre I, Ponce A, Fiorentino R, Shoshani L, Contreras RG, Cereijido M. Contacts and cooperation between cells depend on the hormone ouabain. Proc Natl Acad Sci USA 103: 10911-10916, 2006.
Laughery MD, Todd ML, Kaplan JH. Mutational analysis of alpha-beta subunit interactions in the delivery of Na-K-ATPase heterodimers to the plasma membrane. J Biol Chem 278: 34794-34803, 2003.
Lecuona E, Luquin S, Avila J, Garcia-Segura LM, Martin-Vasallo P. Expression of the beta 1 and beta 2(AMOG) subunits of the Na-K-ATPase in neural tissues: cellular and developmental distribution patterns. Brain Res Bull 40: 167-174, 1996.
Leschelle X, Delpal S, Goubern M, Blottiere HM, Blachier F. Butyrate metabolism upstream and downstream acetyl-CoA synthesis and growth control of human colon carcinoma cells. Eur J Biochem 267: 6435-6442, 2000.
Lever JE. Inducers of mammalian cell differentiation stimulate dome formation in a differentiated kidney epithelial cell line (MDCK). Proc Natl Acad Sci USA 76: 1323-1327, 1979.
Mircheff AK, Bowen JW, Yiu SC, McDonough AA. Synthesis and translocation of Na + -K + -ATPase - and -subunits to plasma membrane in MDCK cells. Am J Physiol Cell Physiol 262: C470-C483, 1992.
Moyer BD, Loffing-Cueni D, Loffing J, Reynolds D, Stanton BA. Butyrate increases apical membrane CFTR but reduces chloride secretion in MDCK cells. Am J Physiol Renal Physiol 277: F271-F276, 1999.
Rajasekaran SA, Gopal J, Willis D, Espineda C, Twiss JL, Rajasekaran AK. Na-K-ATPase beta1-subunit increases the translation efficiency of the alpha1-subunit in MSV-MDCK cells. Mol Biol Cell 15: 3224-3232, 2004.
Rajasekaran SA, Palmer LG, Moon SY, Peralta Soler A, Apodaca GL, Harper JF, Zheng Y, Rajasekaran AK. Na-K-ATPase activity is required for formation of tight junctions, desmosomes, and induction of polarity in epithelial cells. Mol Biol Cell 12: 3717-3732, 2001.
Rajasekaran SA, Palmer LG, Quan K, Harper JF, Ball WJ Jr, Bander NH, Peralta Soler A, Rajasekaran AK. Na-K-ATPase beta-subunit is required for epithelial polarization, suppression of invasion, and cell motility. Mol Biol Cell 12: 279-295, 2001.
Schmalzing G, Gloor S, Omay H, Kroner S, Appelhans H, Schwarz W. Upregulation of sodium pump activity in Xenopus laevis oocytes by expression of heterologous beta 1 subunits of the sodium pump. Biochem J 279: 329-336, 1991.
Schmalzing G, Kroner S, Schachner M, Gloor S. The adhesion molecule on glia (AMOG/beta 2) and alpha 1 subunits assemble to functional sodium pumps in Xenopus oocytes. J Biol Chem 267: 20212-20216, 1992.
Schmalzing G, Ruhl K, Gloor SM. Isoform-specific interactions of Na-K-ATPase subunits are mediated via extracellular domains and carbohydrates. Proc Natl Acad Sci USA 94: 1136-1141, 1997.
Shoshani L, Contreras RG, Roldan ML, Moreno J, Lazaro A, Balda MS, Matter K, Cereijido M. The polarized expression of Na +,K + -ATPase in epithelia depends on the association between beta-subunits located in neighboring cells. Mol Biol Cell 16: 1071-1081, 2005.
Tang MJ, Wang YK, Lin HH. Butyrate and TGF-beta downregulate Na-K-ATPase expression in cultured proximal tubule cells. Biochem Biophys Res Commun 215: 57-66, 1995.
Vagin O, Denevich S, Sachs G. Plasma membrane delivery of the gastric H,K-ATPase: the role of beta-subunit glycosylation. Am J Physiol Cell Physiol 285: C968-C976, 2003.
Vagin O, Turdikulova S, Sachs G. Recombinant addition of N-glycosylation sites to the basolateral Na-K-ATPase beta1 subunit results in its clustering in caveolae and apical sorting in HGT-1 cells. J Biol Chem 280: 43159-43167, 2005.
Vagin O, Turdikulova S, Sachs G. The H,K-ATPase beta subunit as a model to study the role of N-glycosylation in membrane trafficking and apical sorting. J Biol Chem 279: 39026-39034, 2004.
Vagin O, Turdikulova S, Yakubov I, Sachs G. Use of the H,K-ATPase beta subunit to identify multiple sorting pathways for plasma membrane delivery in polarized cells. J Biol Chem 280: 14741-14754, 2005.
Wacker I, Kaether C, Kromer A, Migala A, Almers W, Gerdes HH. Microtubule-dependent transport of secretory vesicles visualized in real time with a GFP-tagged secretory protein. J Cell Sci 110: 1453-1463, 1997.
Wetzel RK, Sweadner KJ. Immunocytochemical localization of NaK-ATPase isoforms in the rat and mouse ocular ciliary epithelium. Invest Ophthalmol Vis Sci 42: 763-769, 2001.
Wilson PD, Devuyst O, Li X, Gatti L, Falkenstein D, Robinson S, Fambrough D, Burrow CR. Apical plasma membrane mispolarization of Na-K-ATPase in polycystic kidney disease epithelia is associated with aberrant expression of the beta2 isoform. Am J Pathol 156: 253-268, 2000.
Wilson PD, Sherwood AC, Palla K, Du J, Watson R, Norman JT. Reversed polarity of Na + -K + -ATPase: mislocation to apical plasma membranes in polycystic kidney disease epithelia. Am J Physiol Renal Fluid Electrolyte Physiol 260: F420-F430, 1991.
作者单位：Department of Biochemistry and Molecular Genetics, University of Illinois at Chicago, Chicago, Illinois