Abstract

The molecular mechanisms associated with reestablishment of renal epithelial polarity after injury remain incompletely delineated. Stress proteins may act as molecular chaperones, potentially modulating injury or enhancing recovery. We tested whether overexpression of heat shock protein 70 (HSP70) would stabilize Na-K-ATPase attachment to the cytoskeleton, under conditions of ATP depletion, and whether a direct association existed between Na-K-ATPase and HSP70 in cultured renal epithelial cells. LLC-PK1 cells were transfected with a tagged HSP70 (70FLAG) or vector alone (VA). Detachment of Na-K-ATPase was detected in Triton soluble lysate after ATP depletion. 70FLAG cells demonstrated a significant (P < 0.01) decrease in detachment of Na-K-ATPase after either 2 or 4 h of ATP depletion. Interactions between HSP70 and Na-K-ATPase were determined by coimmunoprecipitation of 70FLAG and Na-K-ATPase, by direct and competitive binding assays and by immunocytochemical localization. Binding of HSP70 and Na-K-ATPase increased dramatically following injury. Interactions were: 1) reversible; 2) reciprocal to changes in the HSP70 binding protein clathrin; and 3) present only when ATP turnover was inhibited in cell lysate, an established characteristic of HSP binding. These studies indicate that 1) overexpression of HSP70 is associated with decreased detachment of Na-K-ATPase from the cytoskeleton following injury; 2) HSP70 binds to Na-K-ATPase; and 3) binding of HSP70 to Na-K-ATPase is dynamic and specific, increasing in response to injury and decreasing during recovery. Interaction between the molecular chaperone HSP70 and damaged or displaced Na-K-ATPase may represent a fundamental cellular mechanism underlying maintenance and recovery of renal tubule polarity following energy deprivation.

  • cytoprotection
  • renal cell injury
  • immunoprecipitation of HSP70
  • immunoprecipitation of Na-K-ATPase
  • overexpression of HSP70

ischemic renal injury is associated with detachment of Na-K-ATPase from the cytoskeleton, resulting in loss of tubule polarity and impairment of renal function (11, 18, 19, 21, 27, 31). During recovery from reversible renal injury, restitution of cellular polarity appears to be through recycling of displaced Na-K-ATPase into the basolateral membrane (27).

Heat shock proteins have been implicated in the modulation of cellular injury, acting as molecular chaperones for damaged or displaced proteins. Overexpression of 27- or 70-kDa heat shock proteins (HSP70) has been associated with cytoprotection in a variety of renal epithelial cell lines (6, 24, 28).

Previous studies of ischemic injury of renal epithelia have demonstrated: 1) temporal and spatial relationships between HSP70 and Na-K-ATPase following in vivo ischemia (25); 2) HSP72 complexed with aggregated cellular proteins, including Na-K-ATPase, in an ATP-dependent manner (2); and 3) the movement of HSP70 into cytoskeletal pellets is associated with dose-dependent stabilization of Na-K-ATPase (4). In addition, recent work from our laboratory (22) indicates that the differential inhibition of key components of the heat shock response impairs restitution of cellular polarity following ischemic renal injury. However, a direct interaction between members of the heat shock protein family and Na-K-ATPase has not been previously demonstrated. Consequently, we sought to determine whether overexpression of HSP 70 would stabilize Na-K-ATPase attachment to the cytoskeleton and whether there was a direct association between HSP70 and Na-K-ATPase following ATP depletion in cultured renal epithelia cells.

METHODS

To examine the relationship between HSP70 and the attachment of Na-K-ATPase to the cytoskeleton, a tagged human HSP70 fusion protein was constructed and stably transfected into LLC-PK1 cells. Protein interactions were determined by coimmunoprecipitation, by direct and competitive binding assays and by immunocytochemical localization.

Fusion protein construction.

A tagged human HSP70 cDNA was constructed and inserted into a mammalian expression vector. Briefly, PCR primers were constructed based on the sequence of an inducible human HSP70 gene, accession number M11717. The sense primer incorporated a HindIII restriction site, a Kozak translation initiation sequence and an NH2-terminal FLAG tag sequence: 5′-aagcttgccaccatggattacaaggatgacgacgataagatggccaaagccgcggcagtcgg-3′. The antisense primer incorporated a stop codon and an XbaI site: 5′-ttatctagaatctacctcctcaatggtggggcctg-3′. PCR was performed using Ready-to-Go PCR beads (Amersham) and a MiniCycler (MJ Research). Plasmid pAT153 containing genomic DNA coding for human HSP70 served as a template for the reaction (ATCC 57494). The 1.9-kb product of this reaction was ligated into the mammalian expression vector pcDNA 3.1. Following amplification in E. coli (TP10, Invitrogen, Carlsbad, CA), plasmid was purified, sequenced, and transfected into LLC-PK1 cells using Lipofectamine (Invitrogen). Transfected cells were selectively cultured using geneticin (GIBCO), and colonies were picked and expanded to form stably transfected clonal lines. In parallel, LLC-PK1 cells were transfected with empty pcDNA 3.1 vector to act as controls.

Cell preparation.

Both control cells, containing an empty expression vector (VA), and cells transfected with the FLAG-tagged HSP70 construct (70FLAG), were grown in α-MEM (Cellgro, Mediatech, Herndon, VA) with 10% fetal bovine serum and 1 mg/ml of geneticin at 37°C in 5% CO2. Cell culture filters (25 mm; BioCoat, BD Biosciences, Bedford, MA) coated with 40 μg of collagen IV (BD Biosciences) were plated with 2.8 × 105 cells. Studies were conducted 4 days following passage, with cells having achieved confluence.

Cell injury model.

ATP depletion was induced by incubation with prewarmed substrate-free media and 0.1 μM of the mitochondrial inhibitor antimycin A. Substrate-free media lacked the amino acids normally found in standard α-MEM and contained 5 mg/dl l-glucose to maintain osmolarity. Cells were subjected to 2 or 4 h of energy deprivation. An additional group of cells underwent 2 h of energy deprivation followed by a single rinse in PBS and 4-h recovery in standard growth media (α-MEM). Cellular ATP levels were determined in each experimental group using methods previously described (26, 28).

Na-K-ATPase detachment.

The effects of injury on Na-K-ATPase detachment from the cytoskeleton were studied in groups of VA and 70FLAG cells injured in parallel. Injured and uninjured cells were harvested by scraping into chilled extraction buffer (PHEM) containing 0.1% Triton X-100, 60 mM piperazine-N,N′-bis(2-ethanesulfonic acid), pH 6.8, 25 mM HEPES, 10 mM EGTA, 2 mM magnesium chloride, 1 mM benzamidine, 2 mM sodium vanadate, and Complete Protease Inhibitor Cocktail (Roche). Samples were centrifuged at 36,000 g for 30 min to separate the Triton-soluble from the insoluble protein fraction. The supernatant was separated from the pellet and stored at −80°C. The pattern of change within the soluble fraction, which represents detachment from the cytoskeleton, was determined as an index of cellular injury (22).

HSP70 FLAG immunoprecipitation (Na-K-ATPase coprecipitation).

Interactions between FLAG-tagged HSP70 and Na-K-ATPase were studied in groups of VA and 70FLAG cells injured in parallel. Injured and uninjured cells were harvested by scraping into chilled extraction buffer containing 0.1% Triton X-100, 50 mM Tris (pH 7.4), 50 mM NaCl, 1 mM EDTA, 1 mM EGTA, 2 mM CDTA, 60 mM PIPES, 50 nM PMSF, and Complete Protease Inhibitor Cocktail (Roche). Harvested cells underwent three cycles of freeze-thaw using an ethanol dry ice bath. The resulting cell lysate was centrifuged at 12,000 g for 10 min, using an Eppendorf 5415C at 4°C, and the pellet was discarded. Aliquots of supernatant were removed for protein estimation and as loading controls; the remaining supernatant was stored at −80°C.

Supernatants from lysed VA and 70FLAG cells underwent immunoprecipitation using a kit (FLAGIPT-1, Sigma). Briefly, equal volumes of cell supernatant, adjusted to contain equal quantities of protein (500 μg), were added to 40 μl of agarose gel suspension labeled with antibody directed against the FLAG epitope. Following overnight incubation, rotating at 4°C, samples were spun at 10,000 g for 5 s and the supernatant was discarded. The agarose pellet was washed three times with a buffer containing 50 mM Tris·HCl (pH 7.4) and 150 mM NaCl. Bound proteins were eluted using 50 μl of a 150-ng/μl FLAG peptide solution, and the sample was centrifuged for 5 s. The pellet was discarded and the supernatant was mixed with an equal volume of 2× Western loading buffer. Samples were stored at −80°C.

Na-K-ATPase immunoprecipitation (HSP70 coprecipitation).

To verify HSP70 FLAG coimmunoprecipitation of Na-K-ATPase, a parallel series of experiments was performed in which cell extracts from VA and 70FLAG cells underwent immunoprecipitation using antibodies directed against Na-K-ATPase. Cells underwent the same injury, harvest, and protein estimation as for the HSP70 FLAG immunoprecipitation (see above). Equal volumes of cell supernatant, adjusted to contain equal quantities of protein, were precleared three times by incubation with 10 μl of agarose-bound Protein G (Santa Cruz Biotechnology) for 30 min at 4°C, followed by centrifugation for 30 s at 12,000 g using an Eppendorf 5415C at 4°C. Precleared supernatants were divided into two aliquots and incubated for 1 h at 4°C with 100 μg of either mouse monoclonal antibody (IgG1) directed against Na-K-ATPase (15) or normal mouse IgG (sc-2025, Santa Cruz Biotechnology). Subsequently, 20 μl of agarose-bound Protein G were added to each tube and the mixture was incubated overnight rotating at 4°C. Agarose-bound protein G was then separated by centrifuging for 30 s at 12,000 g, and the supernatant was discarded. The remaining pellets were washed three times with 500 μl of wash buffer containing 0.1% wt/vol Triton X-100, 50 mM Tris·HCl, pH 7.4, 300 mM NaCl, and 0.02% wt/vol sodium azide, suspended in an equal volume of 2× Western loading buffer, and stored at −80°C. Samples were boiled for 3 min before Western blotting.

Immunofluorescence/cytochemical localization.

LLC-PK1 cells, either uninjured or ATP depleted, were fixed with 2% paraformaldehyde for 15 min at room temperature and washed with PBS. Fixed cells were permeabilized with 0.1% saponin and blocked with 2% BSA. Cells were incubated with a monoclonal antibody against Na-K-ATPase and a polyclonal antibody against HSP70 (Stressgen). Molecules were visualized with Alexa 488 goat anti-mouse and Alexa 568 goat anti-rabbit antibody, respectively (Molecular Probes). Immunostained samples were observed and imaged using a Olympus AX-70 fluorescent microscope system and processed with Open labs software. Using a piezo-driven objective, the apical membrane was isolated in the focal plane and photographed.

In vitro binding assay/competition assay.

To evaluate interactions between HSP70 and Na-K-ATPase, a binding assay and competition assay were developed using purified proteins. HSP70 FLAG was isolated from 70FLAG cells by immunoprecipitation in extraction buffer lacking the ATPase inhibitor CDTA. Na-K-ATPase was obtained from rabbit renal outer medulla as previously described (14).

Binding assay.

Purified Na-K-ATPase (12 μg) was added to HSP70 FLAG bound to agarose beads. After overnight incubation, the beads were washed three times and the HSP70 FLAG was eluted using FLAG peptide. Samples were centrifuged for 5 s, the pellet was discarded, and the supernatant was mixed with 2× Western loading buffer. Proteins were separated by denaturing SDS-PAGE electrophoresis.

Competition assay.

A fixed amount of purified Na-K-ATPase (12 μg) was added to varing concentrations (250 to 25 μg/ml) of HSP70 FLAG bound to agarose beads and incubated overnight. Immunoprecipitation was carried out with antibody to α-subunit of Na-K-ATPase (2 μg) and protein G as described above.

Protein determination and Western blotting.

Protein concentrations of cell lysates harvested for immunoprecipitation were determined by Bradford assay (Bio-Rad). Protein concentrations of cell lysates harvested in PHEM were determined, following protein precipitation using the Compat-Able Protein Assay Preparation Reagent Set (Pierce, Rockford, IL), by a bicinchoninic acid method using BSA as a protein standard (BCA Protein Assay Kit, Pierce). Equal amounts of protein (10 μg) were mixed with an equal volume of 2× loading buffer containing 100 mM Tris, pH 6.8, 4% SDS, 0.2% bromophenol blue, 20% glycerol, and 200 mM DTT and heated to 50°C for 5 min. Proteins were separated by SDS-PAGE electrophoresis on 4–20% gradient gels (Criterion, Bio-Rad), transferred onto nitrocellulose membranes (Biotrace NT, Pall, FL), and nonspecific binding sites were blocked with 5% skimmed milk in 10 mM Tris, pH 7.5, 37.5 mM sodium chloride, and 0.5% Tween 20. Membranes were incubated for 1 h with monoclonal antibodies directed against inducible HSP72 (SPA-810 Stressgen, BC), Na-K-ATPase, and clathrin (610500, BD Transduction), respectively. After repeated washings, the membranes were incubated with an appropriate species-specific secondary antibody for 1 h. After further washing, immunoreactive antigen was detected with enhanced chemiluminescence. Western blotting analysis reagents and protocols were supplied by the manufacturer (Pierce). Chemiluminescence was detected by exposure of photographic film (Kodak X-OMAT AR, Rochester, NY) and quantified by means of densitometry. Films were scanned using a Linoscan 14000 and the resulting images were analyzed using Scion Image software (Scion, Frederick, MD). The linearity of measurements within the experimental range was confirmed by dilution of samples and subsequent densitometry.

Sample size and statistical evaluation.

Comparisons of Na-K-ATPase detachment in VA and 70FLAG cells were carried out in triplicate. Average values for conditions performed in triplicate were calculated and normalized in terms of the expression of protein in the uninjured untreated cells. Each experimental group was studied on a minimum of five occasions. Values are expressed as means ± SE. Comparison between experimental groups was made using ANOVA and Student's t-test. Values were considered significantly different if P < 0.05. Immunoprecipitation was performed on a minimum of three occasions.

RESULTS

Transfected LLC-PK1 cells were screened and selected for maximal FLAG-tagged HSP70 expression. Cell lines transfected with vector alone demonstrated similar amounts of endogenous HSP70 as seen in nontransfected LLC-PK1 cells. As expected, cells transfected with vector alone (VA) demonstrated a rapid induction of HSP70 with substantial abundance after 2 and 4 h of injury. Protein expression remained slightly elevated during recovery but significantly less than that at 2 and 4 h of injury. HSP70 protein levels were dramatically increased in cells transfected with the tagged human HSP70 fusion protein (70FLAG) with signal from the fusion protein swamping any fluctuation in endogenous expression as a consequence of injury. Both the magnitude and pattern of change in cellular ATP were similar in VA and 70FLAG cells in all experimental conditions.

Densitometry was used to provide semiquantitative analysis of changes in soluble Na-K-ATPase following injury in both VA and 70FLAG cells (Fig. 1). Five samples were studied, in triplicate, for each cell type under each experimental condition. Consequently, the pattern of change in the soluble fraction, i.e., Na-K-ATPase, which was detached from the cytoskeleton, was determined as an index of the severity of epithelial cell injury.

Fig. 1.

Effect of heat shock protein 70 (HSP70)FLAG transfection on detachment of Na-K-ATPase from the cytoskeleton following injury by ATP depletion. The graph shows densitometry of the Triton-soluble fraction on Western blots stained with antibodies directed against Na-K-ATPase. Protein levels have been normalized in terms of change from uninjured control values.

After 2 h of ATP depletion, detached Na-K-ATPase was substantially reduced in 70FLAG cells (139% of uninjured abundance, n = 5) compared with VA cells (244%, n = 5). After 4 h of injury, detachment was greater in VA cells (618%, n = 5) compared with 70FLAG cells (421%, n = 5). Differences in Na-K-ATPase solubility in VA and 70FLAG cells at 2 and 4 h of injury were statistically significant (P < 0.01). In addition, both VA and 70FLAG cells had more soluble Na-K-ATPase after 4-h ATP depletion compared with 2 h of injury (P < 0.01). Following 2 h of injury and 4 h of recovery, the proportion of soluble Na-K-ATPase had returned to a comparable level in both groups.

Immunoprecipitation of VA cell extracts with agarose-bound FLAG antibody yielded no detectable HSP70 on Western blotting. In contrast, incubation of 70FLAG cell extracts with agarose-bound FLAG antibody produced comparable quantities of HSP70 FLAG protein under all experimental conditions (Fig. 2A).

Fig. 2.

Immunoprecipitation of HSP70 FLAG. 70FLAG cell extracts were incubated with agarose-bound FLAG antibody. Shown are Western blots of 70FLAG immunoprecipitate stained with antibody directed against HSP70 (A), Na-K-ATPase (B), and clathrin (C). Experiments were performed on minimum of 3 occasions to ensure consistency.

Samples obtained by FLAG-immunoprecipitation were also analyzed for the presence of candidate coprecipitating proteins; specifically Na-K-ATPase, clathrin, and ezrin. To allow a qualitative comparison of the relative abundance of each protein, membranes were stripped and probed with antibodies in succession. Na-K-ATPase coprecipitated (Fig. 2B) with HSP70 FLAG under control conditions. After 2 and 4 h of injury, the amount of Na-K-ATPase coprecipitating with HSP70 FLAG increases substantially. The binding of Na-K-ATPase and HSP70 was reversible; 4 h after a 2-h injury, the level of Na-K-ATPase that coprecipitated with HSP70 FLAG was comparable to that seen in uninjured control cells. In contrast, the association of clathrin and HSP70 FLAG diminished after either 2 or 4 h of ATP depletion (Fig. 2C). The divergence in relative abundance of coprecipitated proteins under the same experimental conditions, and from the same membrane, suggests that the increase in Na-K-ATPase coprecipitation at 2 and 4 h of injury is not due to a protein loading artifact. Coprecipitation was not observed: 1) for the membrane-associated protein ezrin; 2) in extracts from VA cells obtained and processed in parallel; or 3) when immunoprecipitation was undertaken in extraction buffer lacking the ATPase inhibitor CDTA.

To confirm an association between HSP70 FLAG and Na-K-ATPase, the immunoprecipitation strategy used to perform coimmunoprecipitation was reversed; Na-K-ATPase immunoprecipitate was Western blotted for HSP70. Incubation of 70FLAG cell extracts with anti-Na-K-ATPase and protein G produced a strong signal for Na-K-ATPase (112 kDa) by Western blotting (Fig. 3A).

Fig. 3.

Immunoprecipitation of Na-K-ATPase. 70FLAG cell extracts were incubated with Na-K-ATPase antibody. Shown are Western blots of proteins immunoprecipitated with Na-K-ATPase antibody and stained with antibody directed against Na-K-ATPase (A) and HSP70 (B).

In uninjured control cells, baseline levels of immunoprecipitated Na-K-ATPase were observed. After 2 and 4 h of injury, the abundance of immunoprecipitated Na-K-ATPase was reduced substantially. Following 4 h of recovery, levels were comparable to those seen in uninjured cells. To ensure specificity of immunoprecipitation using anti-Na-K-ATPase antibody, cell extracts were incubated with normal mouse immunoglobulin and protein G. Na-K-ATPase was not immunoprecipitated by normal mouse immunoglobulin under any of the experimental conditions.

Samples obtained by immunoprecipitation using anti-Na-K-ATPase antibody and normal mouse immunoglobulin were analyzed for HSP70. Coprecipitation of HSP70 was observed in extracts from 70FLAG cells incubated with anti-Na-K-ATPase (Fig. 3B). The abundance of HSP70 coprecipitated with Na-K-ATPase following 2 or 4 h of injury was greater than that seen in both uninjured cells and cells allowed to recover from injury. The increased level of HSP70 after injury (Fig. 3B) contrasts with the decreased immunoprecipitation of Na-K-ATPase (Fig. 3A) occurring at the same time points. Reciprocal changes seen on stripped and reprobed membranes cannot be accounted for by differences in protein loading and substantiate the observed changes in protein-protein interaction during and after injury. Coprecipitation was not observed in samples incubated with normal mouse immunoglobulin.

Decreased immunoprecipitation of Na-K-ATPase at 2 or 4 h of injury (Fig. 3A) occurred when maximal induction of HSP70 would be expected, suggesting the possibility of competitive binding between HSP70 and the Na-K-ATPase antibody. To evaluate this possibility, a competitive binding assay was undertaken in which the same quantity of purified Na-K-ATPase was incubated with serial dilutions of HSP70 FLAG and then immunoprecipitated with a fixed amount of Na-K-ATPase antibody. Indeed, the immunoprecipitation of Na-K-ATPase was found to be inversely proportional to the concentration of HSP70 (Fig. 4). The abundance of Na-K-ATPase which could be immunoprecipitated increased as the concentration of HSP70 with which the purified Na-K-ATPase was incubated decreased.

Fig. 4.

Immunoprecipitation of Na-K-ATPase in the presence of HSP70. Equal amounts of purified Na-K-ATPase (12 μg) were incubated with different concentrations of HSP70: lane A 25 μg/ml, lane B 125 μg/ml, and lane C 250 μg/ml. Immunoprecipitation using an antibody to α-subunit Na-K-ATPase (2 μg) was then carried out. Shown is a Western blot probed with anti Na-K-ATPase.

To further explore a potentially direct relationship between HSP70 and displaced Na-K-ATPase, immunocytochemical localization was used. Because detached Na-K-ATPase is largely relocated to the apical membrane after ATP depletion (18, 19, 21), the apical membrane of LLC-PK1 cells was isolated using a piezo-driven objective (Fig. 5).

Fig. 5.

Immunocytochemical localizations of apical HSP70 and Na-K-ATPase in LLC-PK monolayers. LLC-PK cells were labeled by indirect immunofluorescence under uninjured control conditions (A, C, E) or after ATP depletion (B, D, F). Deconvoluted images of the apical domain were obtained using a piezo-driven object. In uninjured cells, minimal apical expression of Na-K-ATPase is present (A); moderate apical expression of HSP70 is observed (C) while the merged image (E) HSP70 (red) and Na-K-ATPase (green) have a negligible overlap. After ATP depletion, Na-K-ATPase is clearly present in the apical region (B). In addition, there is increased expression of apical HSP70 (D). The merged image demonstrates significant colocalization of the two labels (yellow) after injury (F).

In uninjured control cells: 1) minimal Na-K-ATPase is identified on the apical surface as would be expected (Fig. 5A); 2) moderate expression of HSP70 occurs (Fig. 5C); and 3) the overlap between these proteins is negligible (Fig. 5E, Na-K-ATPase red and HSP70 green). After energy deprivation 1) Na-K-ATPase appears at the apical membrane (Fig. 5B); 2) HSP70 becomes more prominent, as well (Fig. 5D); and 3) the merged image (Fig. 5F) demonstrates substantial colocalizations (yellow). These findings are consistent with the immunoprecipitation and binding studies and provide confirmation of interactions between these two proteins in LLC-PK1 cells following renal cell injury.

To confirm a direct binding of HSP70 FLAG to Na-K-ATPase, a protein binding assay was performed in which purified HSP70 FLAG was used to immunoprecipitate purified Na-K-ATPase (Fig. 6).

Fig. 6.

Protein binding assay. Shown are Western blots stained for Na-K-ATPase (top) and HSP70 (bottom). Lanes represent purified HSP70 FLAG (lane A); purified Na-K-ATPase (lane B); purified Na-K-ATPase boiled for 3 min (lane C); immunoprecipitation of purified Na-K-ATPase by purified HSP70 FLAG (lane D); and immunoprecipitation of boiled Na-K-ATPase by purified HSP70 FLAG (lane E).

Purified HSP70 did not contain Na-K-ATPase (lane A). Similarly, purified Na-K-ATPase did not contain HSP70 (lane B). Following incubation of the purified proteins, immunoprecipitation of HSP70 FLAG coprecipitated Na-K-ATPase (lane D) indicating binding of these proteins. Although the anti- Na-K-ATPase antibody was able to recognize boiled Na-K-ATPase (lane C), coprecipitation did not occur after incubation with purified HSP70.

DISCUSSION

Detachment of Na-K-ATPase from the cytoskeleton and migration to the apical membrane is a cardinal feature of early ischemic renal cell injury (18, 19, 21, 27, 31). Rapid restitution of polarity occurs following reperfusion, with domain-specific components, including Na-K-ATPase, integrins, phosphatidylcholine and phosphatidylinositol, relocating to the basolateral membrane (31). Mechanisms underlying the recovery of polarity remain largely undefined but likely include synthesis and recycling of integral membrane proteins, coupled with extensive cytoskeletal repair and reorganization.

Members of the heat shock family of proteins act as inducible molecular chaperones, preventing changes in protein conformation, facilitating repair, and mediating the delivery of proteins to their appropriate intracellular domain. An increasing body of evidence indicates that overexpression of heat shock proteins protects renal epithelia from injury (6, 7, 16, 17, 23, 24, 28, 32, 33). A diversity of potential mechanisms have been postulated including improved focal adhesion complex reassembly (17), mitochondrial membrane stabilization (16), inhibition of apoptosis (23) and altered protein kinase activity (34). Important biophysical properties of injury-induced heat shock proteins suggest their potential to play a role in the preservation of polarity or in the restitution of cellular integrity following renal injury.

Studies in our laboratory and others indicate that rapid induction of heat shock proteins, particularly HSP70, occurs in response to both in vivo and in vitro renal cell injury (4, 10, 23, 26, 25, 2931, 33). Indirect evidence suggests that interaction between heat shock proteins and Na-K-ATPase occurs following injury: 1) marked similarities in temporal and spatial distribution of HSP70 and Na-K-ATPase occur during recovery from ischemia in vivo (25); 2) ischemic preconditioning with induction of HSP70 prevents subsequent dissociation of Na-K-ATPase in rat kidneys exposed to repeat renal ischemia (1); 3) HSP70, HSP25 and HSP90 stabilize Na-K-ATPase in cytoskeletal fractions of ischemic rat renal cortex (4, 5); and 4) attachment of Na-K-ATPase from the cytoskeleton is preserved in a rat which is resistant to ischemic renal injury that also has constitutively higher HSP25 and HSP70 expression (3).

To provide direct evidence of interactions between HSP70 and Na-K-ATPase, a tagged human HSP70 fusion protein was constructed and stably transfected into porcine proximal tubule cells. The effects of expression of the fusion protein on renal response to injury were studied in a well-characterized model of sublethal renal cell injury induced by ATP depletion (9, 1821, 27, 31). HSP70 FLAG overexpression resulted in diminished detachment of Na-K-ATPase from the cytoskeleton after 2 and 4 h of ATP depletion. As previously demonstrated in LLC-PK1 cells overexpressing HSP27 (28), there was no relationship between cellular ATP levels and cytoprotection in the 70 FLAG cells. Moreover, recent work from our laboratory (22) indicates that differential inhibition of HSP70 causes increased detachment of Na-K-ATPase from the cytoskeleton following ischemic renal injury. The observation that overexpression of HSP70 reduces the detachment of Na-K-ATPase, combined with the previous demonstration that diminished expression of HSP70 is associated with increased dissociation of Na-K-ATPase, supports the concept that HSP70 plays a key role in the preservation of polarity, and the restitution of cellular integrity, following renal injury.

To delineate the dynamics of specific interactions between HSP70 and Na-K-ATPase, a series of parallel studies was performed using the 70FLAG cells. Antibody directed against the FLAG epitope was used to precipitate the HSP70 FLAG fusion protein; proteins bound to HSP70 were then coprecipitated. A direct interaction between HSP70 and the α-subunit of the Na-K-ATPase was demonstrated under all experimental conditions. Binding of Na-K-ATPase to HSP70 is dynamic, with increased association demonstrable following injury and levels of adherence falling on recovery, indicating reversibility of interaction.

Clathrin coprecipitation was examined, as clathrin is known to bind HSP70 (13). In contrast to HSP70-Na-K-ATPase interactions, binding of clathrin to HSP70 decreased following injury, suggesting that injury-induced binding of HSP70 to Na-K-ATPase is explicit and selective.

To further demonstrate the specificity of the observed interaction a parallel series of studies was performed in which antibody directed against the α-subunit of Na-K-ATPase was used to precipitate native Na-K-ATPase; proteins bound to Na-K-ATPase were coprecipitated. HSP70 coprecipitated with Na-K-ATPase following 2 and 4 h of injury. The observation that each protein is present as a coprecipitate when an antibody to the other protein is used to perform immunoprecipitation is consistent with a specific interaction between HSP70 and Na-K-ATPase after ATP depletion.

ATP binds to the NH2-terminus of HSP70, causing a conformational change in HSP70. Subsequent hydrolysis of ATP results in release of bound proteins (8, 12). Thus, in the presence of free ATP, HSP70 binds and releases proteins; and coprecipitation would not be anticipated. Indeed, in the present studies Na-K-ATPase does not coprecipitate with HSP70 in buffer lacking the ATPase inhibitor CDTA, which is consistent with the established biophysical properties of HSP70 (34) and supports the specificity of the coprecipitation reaction.

Both coprecipitation strategies identified reciprocal trends in protein interaction occurring as a consequence of 2 and 4 h of injury compared with uninjured or recovery samples. Binding of HSP70 to Na-K-ATPase was increased (Fig. 3B) whereas binding to clathrin was decreased (Fig. 3C), suggesting a relative selectivity of HSP binding during injury. Immunoprecipitation of Na-K-ATPase decreased following 2 or 4 h of injury (Fig. 4A), despite an increase in the proportion of soluble Na-K-ATPase present at these time points (Fig. 1). At 2 and 4 h of ATP depletion, HSP70 would be induced and the expression of this protein would be more abundant than in uninjured cells or during recovery from injury. Consequently, the observed decrease in immunoprecipitated Na-K-ATPase after injury could be explained by decreased immunoprecipitation of Na-K-ATPase due to reduced antigen availability and competitive binding between the Na-K-ATPase antibody and HSP70. In fact, the abundance of Na-K-ATPase which could be immunoprecipitated was found to be inversely proportional to the concentration of HSP70 in the incubation media (Fig. 4). Moreover, coprecipitation of HSP70 with Na-K-ATPase increased during injury despite reduced Na-K-ATPase immunoprecipitation (Fig. 3B). Both of these observations are consistent with increased HSP70 affinity for Na-K-ATPase during injury. Another possible explanation for reduced immunoprecipitation of Na-K-ATPase occurring during injury might be diminished antibody binding due to injury-induced conformational change. Since the same antibody is able to detect soluble Na-K-ATPase 1) after 2 or 4 h of injury, 2) at the apical membrane after ATP depletion, and 3) after boiling purified Na-K-ATPase, impaired antibody binding because of conformational changes seems unlikely.

Detached Na-K-ATPase has been demonstrated to relocate at the apical domain (18, 19, 21). In the present studies, energy deprivation increased the expression of HSP70 and Na-K-ATPase in the apical domain (Fig. 5). Substantial colocalization of these proteins occurred at this site after ATP depletion, which was not observed in uninjured cells (Fig. 4). Whether this interaction between HSP70 and Na-K-ATPase is preferential compared with other proteins which are affected by renal epithelial cell injury remains to be determined (16, 23, 34). Nonetheless, these observations are consistent with 1) the known aberrations in polarity after ATP depletion (18, 19, 21, 27, 31), 2) previous, in vivo, studies of the temporal and spatial localization of these proteins following an ischemic insult (25), and 3) the present immunoprecipitation experiments.

Because the coprecipitation of Na-K-ATPase by HSP70 could be due to either a direct binding between these proteins or indirectly via cytoskeletal or associated proteins, a direct binding assay was performed in which purified proteins were studied in vitro. These studies indicated that Na-K-ATPase bound directly to HSP70 in isolation (Fig. 6). This observation does not preclude a role for cytoskeletal or other chaperone proteins in mediating interactions between Na-K-ATPase and HSP70 in the context of renal cell injury. Taken together however, this sequence of studies provides substantial data to support the concept that HSP70 and Na-K-ATPase have pathobiologically important interactions following renal cell injury.

The present studies indicate that 1) overexpression of HSP70 diminishes the detachment of Na-K-ATPase from the cytoskeleton following injury and 2) a direct interaction between HSP70 and damaged or displaced Na-K-ATPase occurs following renal injury. The characteristics and dynamics of these interactions are consistent with HSP70 functioning as a critical molecular chaperone involved in the stabilization of the cytoskeleton and the restitution of cellular polarity in injured renal epithelial cells.

GRANTS

This work was supported by a National Institutes of Health Grant PO1-HD-32573. Additional funding was provided by an Eden Fellowship from the Royal College of Physicians, London, an American Heart Association Fellowship, a Ruth L. Kirschstein National Research Service Award, and a National Kidney Foundation Research Fellowship.

Acknowledgments

The authors thank M. Campbell for administrative and technical assistance.

Footnotes

  • 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

View Abstract