In cells of the murine renal inner medullary collecting duct (mIMCD3) cell line, acute hypotonic shock (50% dilution of medium with sterile water but not with sterile 150 mM NaCl) increasedEgr-1 mRNA abundance 2.5-fold at 6 h, as determined by Northern analysis. This increase was accompanied by increased Egr-1 transcription, as quantitated by luciferase reporter gene assay. Increased transcription was dose dependent, additive with otherEgr-1 transcriptional activators, and occurred in the absence of overt cytotoxicy, as quantitated via a fluorometric viability assay. In addition, hypotonic stress increasedEgr-1 protein abundance, which was accompanied by augmentedEgr-1-specific DNA binding ability, as measured via electrophoretic mobility shift assay. Increased DNA binding was further associated with increased transactivation byEgr-1, demonstrated through transient transfection of mIMCD3 cells with a luciferase reporter gene driven by tandem repeats of the Egr-1 DNA consensus sequence. Taken together, these data indicate that hypotonic stress activates Egr-1 transcription, translation, DNA binding, and transactivation in renal medullary cells. This phenomenon might play a role in the acquisition of the adaptive phenotype in response to hypotonic stress in cells of the renal medulla in vivo.
hypotonic stress is encountered systemically in prokaryotes and simple eukaryotes in response to environmental changes in ambient tonicity. In higher eukaryotes, exposure to hypotonicity is generally limited to relatively few epithelia in the absence of profound systemic disturbances in water balance. In response to water loading or diuresis, cells of the renal medulla can generate and therefore encounter a markedly hypotonic urine with an osmolality approaching 50 mosmol/kgH2O. In addition, in the face of avid water conservation, cells of this tissue may be subjected to total solute burdens in excess of 2,000 mosmol/kgH2O (4, 15); following diuresis, such concentrations may fall by >50% in <1 h (3). Cell volume under anisotonic conditions is maintained through rapid influx or efflux of inorganic ions (e.g., K+ and Cl−), followed by accumulation or dumping of osmotically active organic solutes called osmolytes (15). In the case of adaptation to hypotonic stress, the relatively rapid efflux of inorganic and organic solutes mediates the regulatory volume decrease (RVD) that occurs in response to an acute and potentially detrimental influx of water.
Although the molecular mechanisms underlying the RVD that follows cell swelling have received considerable attention, the earliest genetic events engendered by hypotonic stress are obscure. The zinc finger-containing transcription factor,Egr-1, as an immediate-early gene, is rapidly inducible at the transcriptional level by diverse cell stressors (40). Egr-1 has been implicated in the cellular response to anisotonicity in other contexts, including the renal epithelial cell (12) and myocardial (46) responses to hypertonic stress and the renal medullary response to elevated concentrations of the kidney-specific solute, urea (7-11).
The Egr-1 protein, a zinc finger-containing transcription factor (40), binds to its GC-rich consensus sequence and thereby activates transcription of a series of downstream targets, including genes encoding growth factors and cytokines and their receptors (25), as well as adhesion molecules (31), transcription factors (43), and enzymes and structural proteins (14,26, 28, 30, 36, 39, 42). In this fashion, regulatedEgr-1 expression can govern a coordinated program of gene expression and permit a stressed cell to acquire an adaptive phenotype. Although likely not required for cell growth in all contexts (20, 27, 29),Egr-1 expression is essential for acquisition of resistance to several forms of cell stress (17, 21). Therefore, the ability of hypotonic stress to activateEgr-1 at multiple levels of regulation was investigated in cultured renal medullary cells.
It is shown here that, in cells of the murine renal inner medullary collecting duct (mIMCD3) cell line (37), hypotonic stress increased mRNA abundance and transcription ofEgr-1, which was accompanied by increased Egr-1 protein abundance and DNA binding, as well as by increased Egr-1-associated transactivation. These data suggest a role forEgr-1 in mediating transcriptional events in response to hypotonic stress in vivo.
Cell culture and hypotonic or solute treatment. mIMCD3 cells were maintained in Dulbecco’s modified Eagle’s medium-Ham’s F-12 (DMEM/F12) medium (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum (JRH, Lenexa, KS), as described previously (37). Cells were growth suppressed in DMEM/F12 without serum for 24 h prior to treatment with medium supplemented with sterile water or urea. Unless indicated, hypotonic treatment represented a 50% final dilution achieved by the removal of 5 ml of medium from a 10-ml dish and replacing it with 5 ml of sterile tissue culture water. Of note, for all experiments, control treatment with 50% medium dilution with sterile 150 mM NaCl resulted in no appreciable difference from control treatment. When used, urea was added to a final concentration of 200 mosmol/kgH2O.
Transient transfection and reporter gene analysis. mIMCD3 cells were transiently transfected via electroporation (10, 11); luciferase and β-galactosidase activities were monitored as previously described (10, 11), following treatment commencing 48 h after transfection. Data are expressed as means + SE, except where noted. The construction ofEgr-1-Luc (632-Luc) has been previously described (11). TheEgr-1-responsive luciferase reporter construct, EBS3Luc, was a modification of plasmid no. 802 (kindly provided by V. P. Sukhatme, Beth Israel Hospital, Boston, MA) and was prepared by subcloning the fragment of no. 802, which contained three tandem repeats of theEgr-1 DNA binding site (EBS) upstream of the c-fos minimal promoter, into the luciferase reporter vector pXP2 (33).
Northern analysis. RNA was prepared from cell monolayers using Trizol reagent (Life Technologies) in accordance with the manufacturer’s directions. Total RNA (10 μg) was subjected to agarose-formaldehyde gel electrophoresis (12) and transferred to Nytran membrane according to standard technique (38). Blots were hybridized with unpurified randomly primed [α-32P]CTP-labeled murine Egr-1 cDNA (41) in hybridization buffer (19) in the presence of BioKey Blocking Reagent (BioKey, Beaverton, OR; 150 μl/hybridization) overnight at 41°C. After two low-stringency washes, a 30-min high-stringency wash with 0.2× standard sodium citrate-0.1× sodium dodecyl sulfate at 55°C was performed. Densitometry was performed as described (12); Egr-1 mRNA expression was normalized to 18S ethidium bromide staining following densitometric scanning of the negative of a gel photograph.
Cell viability assay. The propidium iodide assay was performed in accordance with an unpublished protocol kindly provided by C. M. Yuan (Walter Reed Army Medical Center). Confluent mIMCD3 cells were passaged at 1:10 into 96-well dishes. Approximately 72 h later, cells received the indicated treatment for 5 h, after which they were incubated with propidium iodide for an additional 1 h. Fluorescence at 645 nm was measured in a Cytofluor II (PerSeptive Biosystems) with excitation at 530 nm. After overnight lysis with saponin (0.1%) in the dark, fluorescence was again quantitated. After background (for wells containing medium alone) was subtracted from each determination, cell death was expressed as the ratio of prelysis fluorescence to postlysis fluorescence.
Western analysis. For Western analysis, lysates were prepared, subjected to electrophoresis, and transferred to polyvinylidene difluoride as previously described (10,16). All incubations were performed at 25°C with gentle rocking. Membranes were blocked with 5% nonfat dry milk (NFDM) in 1× phosphate-buffered saline (PBS)-0.1% Tween for 1.5 h and washed three times with standard incubation buffer (SIB; 1× PBS-0.1% Tween-1% NFDM), incubated with a 1:500 dilution of anti-Egr-1 antibody (Santa Cruz Biotechnology) in 1× PBS-1% NFDM for 1 h, washed three times with SIB, incubated with 1:4,000 dilution of the horseradish peroxidase-coupled appropriate secondary antibody (Pierce, Rockford, IL) in SIB, and washed three times with 1× PBS-0.3% Tween-1% NFDM. Detection was via ECL enhanced chemiluminescence according to the manufacturer’s directions (Amersham, Arlington Heights, IL).
Hypotonic stress activates Egr-1 mRNA expression. Northern analysis was performed using the full-length murine Egr-1 cDNA and total RNA prepared from control and hypotonically stressed mIMCD3 cells. Hypotonic treatment (in the form of 50% medium dilution with sterile water) increased Egr-1 mRNA abundance 2.5-fold at 6 h of treatment (Fig.1). In contrast to treatment with urea, wherein Egr-1 mRNA abundance was increased at 30 and 60 min and then promptly declined (8), hypotonic stress resulted in an initial decrease inEgr-1 mRNA abundance, followed by prolonged elevation at 4 and 6 h (inset, Fig. 1).
Hypotonic stress activates Egr-1 transcription. To determine whether the enhancedEgr-1 mRNA abundance was associated with increased Egr-1 transcription, as was the case with urea-inducible Egr-1expression, mIMCD3 cells were transiently transfected with a luciferase reporter gene driven by 1.2 kb of the murineEgr-1 5′-flanking sequence. At 6 h of treatment, reporter gene activity was increased ∼10-fold by hypotonic stress (50% dilution, Fig. 2). This was comparable to the degree of upregulation noted with elevated urea concentrations (200 mM for 6 h; Fig. 2 and Ref. 11).
Dose response of hypotonic stress-inducible immediate-early gene transcription. The effect of hypotonic stress on Egr-1transcription was evident after as little as a 25% dilution (Fig.3). The effect peaked at 50% dilution, and greater degrees of hypotonic stress potently suppressedEgr-1 transcription. At 100% dilution (replacement of medium with sterile water), total loss of cell adhesion was evident following 6 h of treatment.
Effect of water in combination with other activators of Egr-1 transcription. As shown in Fig.4, the peptide mitogen, epidermal growth factor (EGF, 10−7 M), and urea (200 mM) activated Egr-1transcription in the absence of water treatment (−water). The effect of superimposed 50% medium dilution with water (+water) was additive with that of EGF. Importantly, the urea effect was also enhanced in the presence of water stress; however, the effect was slightly less than additive. Of note, doses of urea, EGF, and water associated with maximal Egr-1transcriptional responses were used.
Effect of hypotonic stress on cell viability. Transcription of immediate-early genes such as Egr-1 may be a consequence of nonspecific cytotoxicity. Because of the near complete loss of cell attachment with 75 and 100% dilution and the profound inhibition in basal Egr-1 transcription, the effect of hypotonic stress on mIMCD3 viability was assessed with a propidium iodide exclusion cell viability assay (seemethods). Even at medium dilutions of up to 75%, there was negligible cell death at 6 h of treatment (Fig. 5). At dilutions >75%, however, cell death increased precipitously.
Hypotonic stress increases Egr-1 protein expression. To demonstrate that enhancedEgr-1 transcription was accompanied by increased Egr-1 protein abundance, Western analysis was performed with detergent lysates prepared from mIMCD3 cells treated for various intervals with 50% dilution (Fig.6). IncreasedEgr-1 protein abundance was not reproducibly evident at 30 min but was detectable at 2 and 4 h. In marked contrast to treatment with urea and classical activators ofEgr-1 protein expression (8), peak effect was not evident until fully 6 h of treatment, after which it decreased. Maximal induction, measured densitometrically, was approximately fourfold.
Hypotonic stress increases Egr-1 binding to its DNA consensus sequence. To confirm that the upregulatedEgr-1 protein was functional, its ability to interact with the Egr-1 DNA consensus sequence, EBS, was evaluated via electrophoretic mobility shift assay (EMSA), as previously described (8). In the absence of added lysate, no retardation of free radiolabeled EBS probe was observed [control (C), Fig. 7]. In the presence of detergent lysates prepared from control-treated mIMCD3 cells (C, Fig. 7), a specific retarded complex was seen (open arrowhead), consistent with previous observations (8). Not shown, this complex could be competed with excess cold (unlabeled) EBS oligonucleotide (8). In the presence of detergent lysates prepared from urea-treated cells (U, 200 mM for 1 h; Fig. 7), a marked increase in probe retardation was observed, again consistent with previous observations (8). Similarly, lysates prepared from hypotonically stressed cells (50% dilution for 6 h, Fig. 7) induced an increase in the retarded band comparable to that of urea treatment. To confirm that this shifted band indeed consisted largely ofEgr-1, supershift experiments were performed. EMSA reactions, using the same lysates from control-, urea-, and water-treated cells, were performed in the presence of specific anti-Egr-1 antiserum. A marked diminution in the abundance of the retarded band was observed under all conditions, and supershifted bands appeared. Densitometric analysis revealed that >80% of the retarded band was supershifted, despite the use of an exceedingly small amount of antibody (100 ng), as a consequence of the extremely dilute nature of the commercially available reagent.
Hypotonic stress increases transactivation by the Egr-1 protein. To confirm that the upregulated expression of Egr-1 protein was associated with increased transactivation byEgr-1, mIMCD3 cells were transfected with a luciferase reporter gene containing the c-fos promoter driven by three tandem repeats of the Egr-1 DNA binding site (EBS3Luc). Hypotonic stress increased transactivation by Egr-1 by >20-fold (Fig. 8). Hypotonic stress increased CMV-Gal expression less than twofold (not shown).
The present data indicate that, in cultured renal medullary cells, hypotonic stress increases Egr-1 mRNA expression, transcription, protein expression, DNA binding, and transactivation. In addition, the effect of hypotonic stress onEgr-1 transcription occurred in a dose-dependent fashion, was additive with otherEgr-1 transcriptional activators and occurred in the absence of overt cytotoxicity.
Cellular stressors activating Egr-1 at the mRNA or transcriptional level include ionizing radiation (reviewed in Ref. 45), reactive oxygen intermediates (13, 34), osmotic stress (12, 46), and urea stress (7-11). Stressors examined specifically in the context of renal physiology and pathophysiology include stress induced by acid (47), mechanical stretch (2), hypertonicity (12), hyperosmotic urea (7-11), heavy metals (32), calcium oxalate crystal exposure (18), and ischemia (35).
Hypotonicity-inducible Egr-1expression appears to be, in part, a transcriptionally mediated event, which may provide insight into the upstream signaling pathways operative in hypotonic stress. TheEgr-1 promoter contains consensus sequences for multiple transcription factors, including the serum response factor, the Ets domain-containing transcription factor,Elk-1,Sp1, activator protein 1 (comprised offos-jun heterodimers), and adenosine 3′,5′-cyclic monophosphate (cAMP) response element-binding protein (5, 6, 24, 44). Therefore, hypotonic stress will likely activate one or more of the well-described signaling intermediates that regulate phosphorylation and activation of these transcription factors, including the mitogen-activated protein kinases, jun kinase and extracellular signal-regulated kinase, and cAMP-dependent protein kinase.
The relation between Egr-1 mRNA and protein expression in response to hypotonicity warrants comment. An initial decrement in Egr-1 mRNA abundance was followed by a sustained increase in mRNA expression (Fig.1), whereas increased Egr-1 protein expression was fully evident by 2 h of treatment (Fig. 6). Given the short half-life of the Egr-1 protein, these data are most consistent with a model wherein a component of the early increase in Egr-1 protein abundance is a consequence of diminished degradation of theEgr-1 protein and not necessarily of enhanced translation. At later time points (e.g., 6 h of hypotonicity), increased Egr-1 transcription, mRNA abundance, and protein expression are all observed.
In addition to the multiple levels ofEgr-1 regulation described above, other levels of complexity are likely operative in the medullary response to hypotonicity. Specifically, posttranslational modification of Egr-1 will also potentially play a role. Recently, regulated phosphorylation ofEgr-1 (in different contexts) has been reported to enhance (22) or diminish (23)Egr-1 interaction with its DNA binding site, affording another level of transcriptional control.
The precise role that Egr-1 plays in the renal medullary response to hypotonicity remains undefined. Although generally described as protective of cell stressors (17, 21),Egr-1 expression has also been observed to be permissive for the adverse manifestations of stress in at least one model (1). In either case,Egr-1 likely mediates its effects through transcriptional regulation of downstream genes, including those encoding growth factors and cytokines and their receptors (25), as well as transcription factors (43), adhesion molecules (31), and other enzymes and structural proteins (14, 26, 28, 30, 36, 39, 42). Which, if any, of these known Egr-1 effectors are operative in the renal medullary cell response to hypotonic stress remains to be determined.
We thank V. P. Sukhatme for theEgr-1 cDNA, promoter, and plasmid no. 802.
Address for reprint requests: D. M. Cohen, PP262, 3314 S.W. U.S. Veterans Hospital Rd., Portland, OR 97201.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-02188 and by the Medical Research Foundation of Oregon.
- Copyright © 1997 the American Physiological Society