Abstract

Materials and Methods

Real-Time PCR.

PCR was performed on 8- and 80-ng cDNA samples per 20-μl reaction in triplicate (n = 1) by using Taqman universal PCR master mix (Applied Biosystems) according to manufacturer instructions with an ABI Prism 7900HT sequence detection system (Applied Biosystems). Primers directed against the human sequence of AR were 5′-ATCGCAGCCAAGCACAATAA-3′ and 5′-AGCAATGCGTTCTGGTGTCA-3′; those for the human BGT1 were 5′-TGTTCAGCTCCTTCACCTCTGA-3′ and 5′-GCAATGCTCTGTGTTCCAAAAG-3′. Fold difference in cDNA abundance (F) was calculated according to the formula F = E ^{(Ct₁−Ct₂₎}. E (= 1.98 ± 0.056, n = 18, mean ± SEM) is the efficiency of the reaction determined by using AR and BGT1 gene-specific primers in reactions containing a 10-fold difference (8 and 80 ng) in cDNA template. Ct₁ and Ct₂ = the number of cycles required to reach the threshold of amplicon abundance (Fig. ​(Fig.77C) for any two experimental conditions.

Open in a separate window

Figure 7

ORE/TonE-dependent AR (A) and BGT1 (B) gene expression is depressed at high extracellular NaCl concentration by herbimycin, a tyrosine kinase inhibitor, and DRB, a serine/threonine CK2 inhibitor. HepG2 cells were grown at 300 mosmol/kg. Fresh medium was substituted that had the same osmolality, was hypotonic (NaCl-free medium, adjusted to 200 mosmol/kg by adding NaCl), or was hypertonic (adjusted to 500 mosmol/kg by adding NaCl). Cells were exposed to DMSO, herbimycin, or DRB in isotonic medium beginning 1 h before altering NaCl concentration. Total RNA was isolated 16 h after altering NaCl concentration and reverse-transcribed. Real-time PCR was used to determine cDNA abundance. Mean ± SEM; n ≥ 3. *, P ≤ 0.05. (C) Representative amplification curve showing position of the amplicon threshold (broken line).

Results

To confirm that the C terminus of TonEBP/OREBP contains a TAD (12), we cloned amino acids 548–1,531 into a GAL4 expression vector to generate the chimera GAL4dbd-TonEBP/OREBP. In this reporter system, generation of luciferase by a cotransfected reporter gene construct depends on the presence of a functional TAD fused to the GAL4dbd. The TAD may be regulated by factors that are endogenous or cotransfected. We examined the effect of different levels of NaCl in the medium on activity of TADs by measuring luciferase activity in HEPG2 cells. Our results confirm that amino acids 548–1,531 contain a TAD that is functional in isotonic medium (300 mosmol/kg; Fig. ​Fig.1).1).

Open in a separate window

Figure 1

The TAD of TonEBP/OREBP is regulated osmotically in contrast to control TAD chimeras that are not. HepG2 cells were cotransfected with pFR-Luc, a GAL4 upstream activating sequence reporter plasmid, and a GAL4dbd expression vector to generate either GAL4dbd alone or one of the following GAL4dbd-TAD chimeras: GAL4dbd-c-Jun, GAL4dbd-polyQ, or GAL4dbd-TonEBP/OREBP. Twenty-four hours after transfection at 300 mosmol/kg, fresh medium was substituted that had the same osmolality, was hypotonic (NaCl-free medium, adjusted to 200 mosmol/kg by adding NaCl), or was hypertonic (500 mosmol/kg by adding NaCl). Sixteen hours after altering the NaCl concentration cells were harvested, and Photinus luciferase activity was determined. The absolute activities of the GAL4dbd fusion proteins (RLUs per μg of total cell protein) depend on which TAD is appended: None, 4 ± 1; c-Jun, 629 ± 251; polyQ, 11 ± 3; TonEBP/OREBP, 3,666 ± 782. Mean ± SEM; n ≥ 3.

The TAD of TonEBP/OREBP is regulated osmotically. The activity of the TAD of TonEBP/OREBP increased at least 8-fold in hypertonic medium (500 mosmol/kg) and decreased greatly when the medium was made hypotonic (200 mosmol/kg; Fig. ​Fig.1).1). As controls in this experiment, the c-Jun and polyQ TADs (17 polyglutamine stretch in the C terminus of TonEBP/OREBP) each showed transactivation activity in isotonic medium, but neither chimera demonstrated osmotic dependence. That c-Jun TAD activity could be regulated was confirmed by marked increase (>50×) when recombinant MEKK, its upstream activator, was cotransfected (data not shown). We assume that the response of the TonEBP/OREBP TAD to NaCl concentration is caused by a change in tonicity and not to NaCl per se. Previous studies have demonstrated that ORE/TonE-driven expression is similar in response to various agents capable of generating a tonic response such as NaCl or raffinose but not to permeating solutes such as urea (1).

With an increase in tonicity, TonEBP/OREBP translocates to the nucleus (3, 4). Tonicity-induced nuclear localization is blocked by 1 μM MG-132, a 26S proteasome inhibitor (14, 15). To determine whether proteasomal degradation of a cofactor or inhibitor is involved in transactivation, we examined the effect of MG-132 on TAD function. We found that 1 μM MG-132 had no significant effect on TAD activity regardless of tonicity (Fig. ​(Fig.2).2). In real-time PCR experiments, HepG2 cells show 40–60% reduction of tonicity-induced AR and BGT1 mRNA expression with 1 μM MG-132 (data not shown). Thus, as reported previously for other cells (14), MG-132 reduces expression of genes whose transcription is regulated by ORE/TonEs also in HepG2 cells. Therefore, although proteasomal degradation apparently is involved in tonicity-induced nuclear translocation of TonEBP/OREBP, we find no evidence that it has a role in tonicity-dependent activation of the TonEBP/OREBP TAD.

Open in a separate window

Figure 2

TonEBP/OREBP TAD activity is not affected significantly by the proteasome inhibitor MG-132. HepG2 cells were cotransfected with pFR-Luc, a GAL4 upstream activating sequence reporter plasmid, and an expression vector to generate a GAL4dbd-TonEBP/OREBP TAD chimera. Twenty-four hours after transfection at 300 mosmol/kg, fresh medium was substituted that had the same osmolality, was hypotonic (NaCl-free medium, adjusted to 200 mosmol/kg by adding NaCl), or was hypertonic (adjusted to 500 mosmol/kg by adding NaCl). Cells were exposed to DMSO or MG-132 in isotonic medium beginning 1 h before altering NaCl. Sixteen hours after altering NaCl concentration cells were harvested, and Photinus luciferase activity was determined (RLUs per μg of total cell protein). Values are normalized to cells at 300 mosmol/kg treated with DMSO. Mean ± SEM; n ≥ 3.

Although the TAD chimera lacks the TonEBP/OREBP N-terminal putative nuclear localization sequence, it does contain the GAL4 nuclear localization sequence, which should lead to constitutive nuclear localization. To confirm this, HepG2 cells were transfected with the TAD construct, and TonEBP/OREBP was visualized by immunofluorescence. We initially attempted to use antibodies specific for the GAL4dbd, but none of the available antibodies worked in our hands for detecting immunofluorescence of the TAD chimera in HepG2 cells. Therefore, we tested an antibody targeted to the C terminus of TonEBP/OREBP itself. A minority of cells displayed bright nuclear immunofluorescence (Fig. ​(Fig.33A). These, presumably, were successfully transfected cells that were overexpressing the TAD chimera. In all these cells, the bright immunofluorescence was localized to the nucleus regardless of the osmolality. The majority of cells, which presumably were not transfected successfully, are invisible in Fig. ​Fig.33A. However, they did appear when the brightness of the images was increased (Fig. ​(Fig.33B). This dim immunofluorescence presumably represents the much lower level of native TonEBP/OREBP. Its location depended on the osmolality as reported (refs. 3 and 4; Fig. ​Fig.33B). At 200 and 300 mosmol/kg the native TonEBP/OREBP is spread diffusely throughout the cell, but at 500 mosmol/kg it is localized in the nucleus.

Open in a separate window

Figure 3

The GAL4dbd-TonEBP/OREBP TAD is constitutively located in the nucleus. HepG2 cells were transfected with the GAL4dbd-TonEBP/OREBP TAD construct. Twenty-four hours after transfection at 300 mosmol/kg, fresh medium was substituted that had the same osmolality, was hypotonic (NaCl-free medium, adjusted to 200 mosmol/kg by adding NaCl), or was hypertonic (adjusted to 500 mosmol/kg by adding NaCl). TonEBP/OREBP was visualized by immunofluorescence using an antibody targeted to its C-terminal amino acids and a secondary antibody linked to Alexa 488. (A) A minority of cells displayed bright nuclear immunofluorescence. These, presumably, were successfully transfected cells that were overexpressing the TAD chimera. In all these cells the bright immunofluorescence was localized to the nucleus regardless of the osmolality. (B) The majority of cells, which presumably were not successfully transfected and are invisible in A, appear when the brightness of the images is increased. Brightness of cells at 300 and 500 mosmol/kg was enhanced equally, whereas cells at 200 mosmol/kg were enhanced further to be able to visualize immunofluorescence. This dim immunofluorescence presumably represents the much lower level of native TonEBP/OREBP. Its location depends on the osmolality. At 200 and 300 mosmol/kg the native TonEBP/OREBP is spread diffusely throughout the cell, but at 500 mosmol/kg it is localized in the nucleus.

Posttranslational modification, particularly phosphorylation, commonly modulates the activity of transcription factors (9). To determine whether phosphorylation of the TAD occurs at high NaCl concentration, we examined Western blots of nuclear protein extracts from HEK293 cells transfected with the GAL4dbd-TonEBP/OREBP TAD construct. HEK293 cells can be transfected with greater efficiency than HepG2 cells but behave as do HepG2 cells in response to tonicity. HEK293 cells demonstrate tonicity dependence of TAD activity and ORE/TonE-driven gene expression (data not shown). The chimera containing the TAD of TonEBP/OREBP from hypertonically treated HEK293 cells exhibited retarded electrophoretic mobility compared with that from cells treated with isotonic medium (Fig. ​(Fig.4).4). This electrophoretic mobility difference was eliminated after phosphatase treatment, which is consistent with tonicity-induced phosphorylation of the ToneBP/OREBP TAD (Fig. ​(Fig.4).4).

Open in a separate window

Figure 4

The chimera containing the GAL4dbd-TonEBP/OREBP TAD from hypertonically treated cells exhibits retarded electrophoretic mobility compared with that from cells treated with isotonic medium. The mobility difference is eliminated after phosphatase treatment. HEK293 cells were transfected with an expression vector to generate the GAL4dbd-TonEBP/OREBP TAD chimera. Twenty-four hours after transfection at 300 mosmol/kg, fresh medium was substituted that either had the same osmolality or was hypertonic (adjusted to 500 mosmol/kg by adding NaCl). Six hours after altering NaCl concentration cells were harvested, and nuclear proteins were extracted. Nuclear proteins from cells at each osmolality were treated with a dual-specificity serine/threonine/tyrosine protein phosphatase (λ-PPase). Proteins were separated by SDS/PAGE. Immunodetection was carried out by using an antibody targeted to the C-terminal amino acids of TonEBP/OREBP.

Tyrosine and serine phosphorylation of TonEBP/OREBP are known to increase with tonicity (11), and the TAD contains both tyrosines and serines. Eleven of the serine residues are contained within putative CK2 sites. Because the TonEBP/OREBP TAD undergoes tonicity-dependent phosphorylation, we examined the effect of herbimycin, a tyrosine kinase inhibitor, and DRB, a serine/threonine protein kinase CK2 inhibitor, on TAD function (Fig. ​(Fig.5).5). Both herbimycin and DRB significantly reduced activity of the TonEBP/OREBP TAD at 500 mosmol/kg (Fig. ​(Fig.5).5). A more modest and statistically insignificant reduction occurred at 300 and 200 mosmol/kg. We reasoned that such inhibition of transactivating activity might reduce transcriptional activity of native TonEBP/OREBP. Therefore, we examined the effect of herbimycin or DRB on a reporter driven by ORE/TonEs, the cognate DNA element of TonEBP/OREBP. When HepG2 cells, transfected with ORELuc, were treated with herbimycin, luciferase activity was reduced significantly at all osmolalities (Fig. ​(Fig.66A). We attribute the effects at 200 and 300 mosmol/kg to partial activation of TonEBP/OREBP at those osmolalities. Partial activation at 300 mosmol/kg was demonstrated previously (15). When treated with DRB, luciferase activity was reduced significantly at 500 mosmol/kg (Fig. ​(Fig.66B). DRB treatment resulted in statistically insignificant reduction at 200 and 300 mosmol/kg (Fig. ​(Fig.66B). We also tested whether herbimycin or DRB treatment affects the mRNA abundance of genes whose transcription is controlled by ORE/TonEs. Both herbimycin and DRB reduced hypertonic induction of both AR and BGT1 genes as measured by quantitative Real-time PCR (Fig. ​(Fig.77 A and B). A representative amplification plot is shown in Fig. ​Fig.77C. Although DRB is known also to inhibit RNA polymerase II (16), the effects seen in the present study are not consistent with a general decrease in mRNA expression. BGT1 mRNA levels at 300 mosmol tend to increase rather than decrease with 10 μM DRB treatment (Fig. ​(Fig.77B). Additionally, 10 μM DRB does not affect β-actin mRNA abundance at any tonicity as measured by real-time PCR (data not shown).

Open in a separate window

Figure 5

TonEBP/OREBP TAD activity at 500 mosmol/kg is decreased by herbimycin, a tyrosine kinase inhibitor (A), and DRB, a serine/threonine CK2 inhibitor (B). HepG2 cells were cotransfected with pFR-Luc, a GAL4 upstream activating sequence reporter plasmid, and an expression vector to generate a GAL4dbd-TonEBP/OREBP TAD chimera. Twenty-four hours after transfection at 300 mosmol/kg, fresh medium was substituted that had the same osmolality, was hypotonic (NaCl-free medium, adjusted to 200 mosmol/kg by adding NaCl), or was hypertonic (adjusted to 500 mosmol/kg by adding NaCl). Cells were exposed to DMSO, herbimycin, or DRB in isotonic medium beginning 1 h before altering NaCl. Sixteen hours after altering NaCl concentration cells were harvested, and Photinus luciferase activity was determined (RLUs per μg of total cell protein). Values are normalized to cells at 300 mosmol/kg treated with DMSO. Mean ± SEM; n ≥ 3. *, P ≤ 0.05.

Open in a separate window

Figure 6

Native TonEBP/OREBP-driven reporter gene expression is decreased by herbimycin, a tyrosine kinase inhibitor (A), and DRB, a serine/threonine CK2 inhibitor (B). HepG2 cells were cotransfected with ORE-Luc containing bp −3,429 to +27 of the rabbit AR gene upstream of the P. pyralis luciferase gene. The sequence −3,429 to +27 includes the AR promoter and three ORE/TonEs in native gene context. Twenty-four hours after transfection at 300 mosmol/kg, fresh medium was substituted that had the same osmolality, was hypotonic (NaCl-free medium, adjusted to 200 mosmol/kg by adding NaCl), or was hypertonic (adjusted to 500 mosmol/kg by adding NaCl). Cells were exposed to DMSO, herbimycin, or DRB in isotonic medium beginning 1 h before altering NaCl. Sixteen hours after altering NaCl concentration cells were harvested, and Photinus luciferase activity was determined (RLUs per μg of total cell protein). Values are normalized to cells at 300 mosmol/kg treated with DMSO. Mean ± SEM; n ≥ 3. *, P ≤ 0.05.

Discussion

Hypertonicity is known to increase the binding of TonEBP/OREBP to its cognate DNA elements, ORE/TonEs, through increased nuclear abundance. This increase occurs through rapid translocation of existing TonEBP/OREBP to the nucleus and also, more slowly, through increased mRNA transcription and protein synthesis (3, 4). Additionally, hypertonicity results in an increase in phosphorylation of TonEBP/OREBP (11) but the role of this posttranslational modification is unknown. In the present study, we show that transactivation activity of TonEBP/OREBP varies directly with NaCl concentration and that this is correlated with apparent tonicity-dependent phosphorylation of the TAD of TonEBP/OREBP. The increase in TonEBP/OREBP TAD activity at elevated NaCl concentrations is independent of tonicity-modulated nuclear translocation. Nuclear localization is necessary for transcription factor-driven gene expression, with translocation typically controlled through change in association with nuclear importins or exportins (YAP1 and NFAT) and/or anchor proteins (NF-κB and MAPK; ref. 5). Native TonEBP/OREBP, under isotonic conditions, is both cytosolic and nuclear. With an increase in tonicity, TonEBP/OREBP translocates to the nucleus (3, 4). Tonicity-modulated nuclear localization is blocked by MG-132, a 26S proteasome inhibitor (14). MG-132 also inhibits the tonicity-induced increase in mRNA abundance of three genes whose transcription is regulated by ORE/TonEs, AR, BGT1, and the sodium-myo-inositol cotransporter (ref. 14 and present study). As a result, a tonicity-dependent increase in nuclear TonEBP/OREBP is proposed as the key event in driving ORE/TonE-dependent genes (15). Although nuclear localization of TonEBP/OREBP is necessary, osmotic regulation of TAD function is independent of tonicity-driven nuclear localization. The TAD chimera lacks the TonEBP/OREBP N-terminal putative nuclear localization sequence and, unlike native TonEBP/OREBP, is constitutively located in the nucleus. Nuclear localization of the TonEBP/OREBP TAD chimera is due to the presence of the GAL4dbd.

TonEBP/OREBP TAD activity apparently is independent of the proteasome. Proteasomal degradation may regulate transcription factor activity independently of nuclear translocation, e.g., precursor degradation may be required to generate a subunit or cofactor [NF-κB (6)]. To examine the involvement of the proteasome in TonEBP/OREBP transactivation, we tested the same concentration of MG-132 known to inhibit ORE/TonE-driven gene expression (14), but we found no significant effect regardless of tonicity. The effect of MG-132 on TonEBP/OREBP regulation, thus, may be limited to inhibition of tonicity-dependent nuclear translocation of the protein. The mechanism could be similar to that of NF-κB, the translocation of which to the nucleus depends on proteasome degradation of the inhibitor IκB (5). Although nuclear localization of TonEBP/OREBP may be blocked by an unknown cofactor analogous to IκB, tonicity-dependent transactivation evidently is independent of the proteasome.

The requirement for dimer formation in TonEBP/OREBP transactivation (12) may be met by the GAL4dbd of the TAD chimera. The TonEBP/OREBP TAD chimera used in the present study lacks the N-terminal dimerization domain, but because the GAL4 DNA binding domain acts as a dimer, any requirement for dimerization could be met by this system. Dimerization of native TonEBP/OREBP is independent of tonicity (12). Although dimerization is required for transactivation, apparently it is not responsible for increased transactivation at high extracellular NaCl concentrations.

The C terminus of TonEBP/OREBP contains a TAD that is regulated osmotically. Similar to CREB and CREM, the C terminus of TonEBP/OREBP contains two homopolymeric glutamine stretches (10). Glutamine-rich domains are thought to act as surfaces for interaction with components of the basal transcription machinery such as cofactors for RNA polymerase II and thus are characteristic of TADs (10, 17–19). We have shown that C-terminal amino acids (548–1,531) of TonEBP/OREBP contain a TAD that is functional in isotonic medium, confirming previous results with a similar construct (12). Additionally, we demonstrate that the TAD of TonEBP/OREBP is regulated osmotically in contrast to control TAD chimeras that are not. These results evidence a new level of tonicity dependence in the regulation of TonEBP/OREBP.

The TonEBP/OREBP TAD apparently becomes phosphorylated, at high NaCl concentrations, coincident with increased transactivation activity. Phosphorylation can potentiate transactivation in several ways. Phosphorylation can regulate transactivation (CREB) by recruiting coactivator proteins that contact the general transcriptional machinery (10). Phosphorylation also may be required for maximal transcriptional activity (STAT1 and NF-κB) or for the formation of stable homodimer-DNA complexes (STAT3α; refs. 6 and 8). Tyrosine and serine phosphorylation of native TonEBP/OREBP increase with tonicity (11); the specific residues have not been identified, and the role of tonicity-dependent phosphorylation is unknown other than that it is unnecessary for DNA binding (11). Here we provide evidence that the TonEBP/OREBP TAD chimera is phosphorylated at high extracellular NaCl concentrations. Additionally, we show that herbimycin, a tyrosine kinase inhibitor, and DRB, a serine/threonine CK2 inhibitor, reduce TonEBP/OREBP activity in several independent ways. Both herbimycin and DRB reduce the activity of the ToneBP/OREBP TAD chimera at 500 mosmol/kg, which may indicate that a phosphorylation event is necessary for maximal transcription activity. In support of this theory, herbimycin and DRB each decrease tonicity-dependent activity of native TonEBP/OREBP. This decreased activity was shown in two ways. Expression of an ORE/TonE-driven reporter was reduced at 500 mosmol/kg by DRB and at all tonicities by herbimycin. ORE/TonE-dependent AR and BGT1 gene expression was depressed at high extracellular NaCl concentrations by both inhibitors. The C-terminal amino acids (548–1,531) of TonEBP/OREBP contain tyrosine residues as well as serine residues within consensus sites for CK2 phosphorylation. Thus, maximal activation of the TonEBP/OREBP TAD may involve tonicity-dependent phosphorylation.

Currently, we can summarize the regulation of TonEBP/OREBP as follows. Hypertonicity leads to an increase in DNA binding of TonEBP/OREBP to its cognate DNA elements through increased abundance of the protein in the nucleus. TonEBP/OREBP abundance increases in the nucleus through tonicity-dependent translocation from the cytosol of existing protein and, in the longer term, through increased protein synthesis and transcription of its mRNA (3, 4). An N-terminal DNA binding domain has been identified as having a putative nuclear localization sequence (4). Nuclear localization may be inhibited by a cytosolic cofactor that is degraded in the proteasome in a fashion analogous to IκB inhibition of NF-κB (14). An N-terminal dimerization domain also has been identified (12). Dimerization is required for DNA binding, but TonEBP/OREBP forms a dimer constitutively, i.e., dimer formation is independent of tonicity (12). Dimerization is required also for transactivation (12). The C terminus of TonEBP/OREBP contains a TAD that is regulated osmotically (this study). Hypertonicity results in increased phosphorylation of TonEBP/OREBP on tyrosine and serine residues (11). Whereas DNA binding is independent of tonicity-induced phosphorylation (11), maximal transactivation may involve tonicity-dependent phosphorylation of the TAD (this study).

Abbreviations

References