HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 41, 32244-32249, October 13, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/32244    most recent M005082200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Park, O. K. Articles by Schaefer, T. S. Search for Related Content PubMed PubMed Citation Articles by Park, O. K. Articles by Schaefer, T. S. Social Bookmarking                      What's this?

Dimer Stability as a Determinant of Differential DNA Binding Activity of Stat3 Isoforms^*

Ohkmae K. Park^§, Laura K. Schaefer^¶, Wenlan Wang, and Timothy S. Schaefer^¶

From the  Kumho Life and Environmental Science Laboratory (KLESL), Kwangju 500-480, Korea, the ^¶ Department of Neurosurgery, University of Texas M.D. Anderson Cancer Center, Houston, Texas 77030, and the  Department of Clinical and Medical Science, A. I. duPont Hospital for Children, Wilmington, Delaware 19803

Received for publication, June 13, 2000, and in revised form, July 18, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Stat3 and Stat3 are two Stat3 isoforms with marked quantitative differences in their DNA binding activities. To examine the molecular basis of the differential DNA binding activities, we measured DNA binding strength and dimer stability, two possible mechanisms responsible for these differences. Stat3 and Stat3 showed no difference in DNA binding strength, i.e. they had similar association and dissociation rates for DNA binding. However, competition analyses performed with dissociating reagents including an anti-phosphotyrosine antibody, SH2 domain protein, and a phosphopeptide demonstrated that Stat3 dimers are more stable than Stat3 dimers. We report here that dimer stability of activated forms plays a critical role in determining DNA binding activity of Stat3 isoforms. We found that C-terminal deletions of Stat3 increased both DNA binding activity and dimer stability of Stat3. Our findings suggest that the acidic C-terminal region of Stat3 does not interfere with the DNA binding of activated Stat3 dimers, but destabilizes the dimeric forms of Stat3. We propose that dimer stability described in vitro may be the underlying mechanism of in vivo stability of activated Stat3 proteins, regulating dephosphorylation of tyrosine 705.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Extracellular signaling molecules such as cytokines and growth factors modulate the properties of cells through their binding to specific cell-surface receptors. For transducing the signals to the nucleus, the ligand-receptor binding activates one or more signaling pathways, ultimately regulating gene expression. Whereas the signals are relayed by a cascade of biochemical reactions in most signaling pathways, forming a network of signals, the studies on interferon-regulated gene expression revealed a direct signaling pathway from receptor to nucleus, the so called JAK¹-STAT signaling pathway (1, 2).

STAT (signal transducer and activator of transcription) forms a family of latent signaling transcription factors (Stat1 to Stat6) that are tyrosine-phosphorylated and activated by JAK (Janus kinase), a family of tyrosine kinases (Jak1, Jak2, Jak3, and Tyk2). Many cytokines and growth factors have been known to signal through JAK-STAT pathway (3, 4). Jaks associate with cytokine receptors as inactive forms and become activated by ligand-induced receptor aggregation. The activated Jak proteins, in turn, phosphorylate receptors at multiple tyrosine residues, providing the docking sites for Stats. Since cytokine receptors lack intrinsic tyrosine kinase activity, the ability of the receptors to recruit and activate Jaks is critical for activating the downstream signaling pathways. The recruited Stats become accessible to and therefore phosphorylated by Jaks. Unlike cytokine receptors, growth factor receptors such as epidermal growth factor (EGF) and platelet-derived growth factor receptors possess intrinsic tyrosine kinase activity. There are reports suggesting the direct activation of Stats by receptor kinases (5). A single tyrosine residue (e.g. tyrosine 701 in Stat1 and tyrosine 705 in Stat3) has been assigned as the phosphorylation site (6). After phosphorylation, Stats homo- or heterodimerize and then translocate to the nucleus, where they form active transcription complexes and regulate gene expression (7).

Stat proteins share several conserved functional domains. The most conserved region is an SH2 (Src homology 2) domain near the C terminus. The SH2 domain of Stat proteins mediates the selective binding to phosphotyrosine motifs of receptors (8) as well as dimerization of Stats (7). Phosphorylation at a single tyrosine residue is required for DNA binding activity of Stats, suggesting that dimerization through phosphotyrosine-SH2 domain interaction is essential for DNA binding of Stats (7). DNA binding activity of Stats has been attributed to a highly conserved DNA-binding domain in the middle of the protein (9). The selection of the optimal DNA binding sequences with random oligonucleotides showed that Stats have very similar preferences for DNA binding (10). The N-terminal region is also quite conserved among Stat proteins. Some reports suggest that the N terminus of Stats may be crucial for the interaction of Stats with other proteins including tyrosine kinases and phosphatases. Stat1 mutants lacking the N-terminal 141 or 216 amino acids are no longer tyrosine phosphorylated, although their SH2 domain and tyrosine phosphorylation site (tyrosine 701) remain intact (11). Mutational analysis of Stat1 implies that the N-terminal region may also be involved in dephosphorylation of Stats (12). Another study has proposed that besides a role in Stat deactivation, the N terminus may also be important for the regulation of nuclear translocation (13). In addition, the N terminus has been implicated in Stat-receptor interaction, as suggested by Stat2 association with interferon- receptor (14). Other studies have found the N-terminal region important for cooperative DNA binding by Stat1-Stat1 dimers to two tandem DNA sites (15). Sequence alignment shows that the C-terminal region is the most divergent in sequence. Isoforms of Stat1 and Stat5 naturally occurring by alternative splicing of mRNA are short forms lacking the C-terminal region (16). The observations that the splice variants have no transcriptional activity indicate the function of the C-terminal region as a transactivation domain (6, 17).

Besides transcriptional activation, other properties appear to be regulated by the C-terminal region of Stats. Unlike the short forms of Stat1 and Stat5, Stat3, a short form of Stat3 that was discovered as a c-Jun-interacting protein, has C-terminal 55 amino acid residues of Stat3 replaced with a unique 7 amino acid residues (18). Studies comparing the properties of the two isoforms uncovered a marked quantitative difference in the DNA binding activities of phosphorylated Stat3 and -3 in vitro (5) and in vivo (19). In this report, our studies suggest that the difference in dimer stability is the basis of the differential DNA binding activities of Stat3 isoforms. In addition, the studies with C-terminal deletion mutants of Stat3 lead us to propose that the C-terminal region may play an important role in regulating differential properties of Stat3 isoforms.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Construction of Plasmids-- Plasmids used in transfection experiments expressing Stat3 or -3 have been previously described (18, 19). Plasmids used for the expression in Sf9 insect cells were prepared as follows. DNA fragments containing either murine Stat3 or -3 were cloned into the cloning sites, EheI and HindIII, of the baculovirus expression vector, pFASTBAC HTa (Life Technologies, Inc.). The DNAs encoding Stat1 or Stat3 SH2 domain were prepared by polymerase chain reaction (amino acids 570-712 for Stat1 and 577-715 for Stat3) and inserted into EheI and HindIII cloning sites of pFASTBAC HTa. Plasmids expressing 5, 12, 19, 26, 33, 40, 48, and 55 were made by polymerase chain reaction amplification in which a 3' primer introduced a stop codon after the indicated amino acid.

Protein Purification-- Stat3 proteins with N-terminal hexahistidine tag were purified from baculovirus-infected Sf9 cells, as described previously (5). The tyrosine-phosphorylated proteins were prepared from cells co-infected with Jak1- and Jak2-expressing baculoviruses. SH2 domains of Stat1 and Stat3 were prepared under denaturing condition. Sf9 cells (6 × 10⁷ cells) infected with SH2 domain-encoding baculoviruses were resuspended and lysed in 10 ml of lysis buffer (6 M guanidine HCl, 100 mM sodium phosphate, 10 mM Tris-HCl, pH 8.0) for 2 h at room temperature. Lysates were centrifuged for 15 min at 20,000 × g to remove insoluble material and incubated with Ni²⁺ Probond resin (Invitrogen) for 1 h. The mixture of lysate and Ni²⁺ resin was packed in a column and extensively washed with lysis buffer, followed by buffer A (8 M urea, 100 mM sodium phosphate, 10 mM Tris-HCl, pH 8.0). Ni²⁺-bound proteins were eluted with a pH step gradient of pH 6.0, 5.0, and 3.5 in buffer A. SH2 proteins were eluted at pH 5.0. The purified SH2 proteins were incubated with 10 mM EDTA and serially dialyzed against a buffer (10 mM Hepes, pH 7.4, 100 mM NaCl, 0.5 mM dithiothreitol) containing 4, 2, 1, 0.5, and 0.1 M, and finally no urea.

Immunoblot Analysis-- Purified Stat proteins or nuclear extracts were electrophoresed in SDS-10% polyacrylamide gels, transferred to nitrocellulose, and detected as previously reported (18). The proteins labeled using anti-Stat3 (Trasduction Laboratories) or anti-phosphotyrosine (4G10, Upstate Biotechnology) monoclonal antibody and ¹²⁵I-labeled goat anti-mouse IgG were quantitated on a PhosphorImager (Molecular Dynamics). The determined protein concentrations were confirmed by quantitating autoradiograms with a densitometer (Molecular Dynamics). The amount of protein used through the experiments were corrected for tyrosine-phosphorylated Stat3 proteins, relative to phosphorylated Stat3 protein. Therefore, the protein concentrations indicate same tyrosine phosphorylation per unit protein.

Electrophoretic Mobility Shift Assay (EMSA)-- Purified Stat proteins or nuclear extracts were incubated with ³²P-labeled high affinity c-fos sis-inducible element (hSIE) (Santa Cruz) or IL-6-responsive element of ₂-macroglobulin gene promoter (IL-6 RE) (5'-CCTTAATCCTTCTGGGAATTCTGGCTAACG-3'; Stat3 binding sequence is underlined) in 24 µl of a buffer containing 10 mM HEPES, pH 7.8, 50 mM KCl, 1 mM EDTA, 5 mM MgCl₂, 10% glycerol, 5 mM dithiothreitol, 1 mg/ml bovine serum albumin, 10 µg/ml leupeptin, 5 µg/ml pepstatin A, 0.5 mM phenylmethylsulfonyl fluoride, 1 mM Na₃VO₄, and 1 µg of poly(dI-dC). After 15 min incubation at room temperature, the samples were electrophoresed on 4% polyacrylamide gels. For competition analyses, Stat proteins were incubated with the indicated amounts of each competitior for 15 min at room temperature before addition of the labeled oligonucleotides.

Cell Culture-- COS-7 cells were maintained in Dulbecco's modified Eagle's medium plus 10% fetal bovine serum at 37 °C and 5% CO₂. One day prior to transfection cells were plated at a density of 3 × 10⁵ per 5-cm dish or 1 × 10⁶ per 10-cm dish. Two hours before transfection the cells were fed with fresh Dulbecco's modified Eagle's medium, 10% fetal bovine serum. Transfections were performed using the calcium phosphate method (20) using the amounts of DNA indicated in the figure legends. After 16 h cells were washed twice and fed Dulbecco's modified Eagle's medium containing no serum. Forty hours post-transfection cells were stimulated with human recombinant EGF (100 ng/ml) (Upstate Biotechnology) for the indicated times and harvested for preparation of nuclear extracts or for chloramphenicol acetyltransferase and -galactosidase assays. Nuclear extracts were prepared by the method of Andrews (21) with the inclusion of Na₃VO₄ (1 mM), 10 µg/ml leupeptin, and 5 µg/ml pepstatin A. Chloramphenicol acetyltransferase activity was normalized to -galactosidase plasmid YN3214-lacZ. Experiments performed to determine the half-lives of the Stat3 deletion mutants used the previously described EGF receptor kinase inhibitor PD157655 (0.2 µM) (David Fry, Parke Davis).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Differential DNA Binding Activities of Stat3 and Stat3-- We previously reported that Stat3 activated by phosphorylation of tyrosine 705 was more active for DNA binding per unit phosphorylated protein than similarly activated Stat3 (5). Whether phosphorylated by JAK in Sf9 insect cells or by the EGF receptor kinase in vitro, phosphorylated Stat3 was approximately 20-fold more active for DNA binding than phosphorylated Stat3. To study the underlying mechanism of this quantitative difference in DNA binding, tyrosine 705-phosphorylated forms of Stat3 and -3 containing N-terminal polyhistidine were prepared from Sf9 cells co-infected with Stat3-, Jak1-, and Jak2-expressing baculoviruses as described previously (5). The purified Stat3 and -3 were tyrosine phosphorylated at equivalent levels, as determined by quantifying immunoblots of the protein preparations using monoclonal anti-phosphotyrosine or anti-Stat3 antibodies (data not shown). Fig. 1A shows a comparison of DNA binding activities of the preparations of Stat3 and -3 used in this and subsequent experiments. Two binding sites were used: the high affinity c-fos sis-inducible element (hSIE), a strong binding site for Stat3 (22); and the weaker IL-6-responsive element (IL-6RE) of the rat ₂-macroglobulin gene promoter (23). Stat3 had approximately 25-fold (hSIE) or 15-fold (IL-6RE) greater DNA binding activity than Stat3 on a molar basis. Similar results were obtained with several other preparations of the Stat3 proteins (data not shown).

View larger version (32K): [in this window] [in a new window]   Fig. 1.   Differential DNA binding activities of in vitro purified Stat3 and Stat3 proteins. A, EMSAs carried out with the indicated amounts of Stat3 and -3. B, proposed equilibrium processes for DNA binding of Stat proteins. See the text for details.

Rates of Formation and Dissociation of Stat3- and Stat3-hSIE Complexes-- Stat proteins bind to DNA as dimers formed by interaction of the SH2 domain of each monomer with a proximal phosphotyrosine (tyrosine 705 in the case of Stat3) of the other monomeric partner (7). Therefore, differential DNA binding activities of Stat proteins could be explained by differences in two distinct equilibrium processes (Fig. 1B): formation and dissociation of dimers (K1), and formation and dissociation of Stat dimer-DNA complexes (K2).

To begin to understand the mechanism underlying the differences in DNA binding activity between Stat3 and -3 homodimers, we first examined the relative rates of formation and dissociation of Stat3- and 3-hSIE complexes (Fig. 2). Stat3 and -3 were used at concentrations that gave similar hSIE-binding activities (25 and 1 ng, respectively, see Fig. 1); for each isoform the amount chosen was in the linear range of binding activity. To monitor association rates, Stat protein was incubated with labeled hSIE oligonucleotide and reaction aliquots were loaded onto a pre-running gel at various time points during the incubation (Fig. 2A). At the concentrations used, the two isoforms associated with DNA at very similar initial rates. We should note that the association of Stat protein with the hSIE is initially rapid and then continues at slower rate (Fig. 2A). A plausible interpretation of this observation is that the rapid phase of the association curve reflects the association of complexes between preformed dimer and oligonucleotide, whereas the slower phase represents the binding of dimers formed from monomers (Fig. 2). If this interpretation is correct, the formation of dimers from monomers is considerably slower than the association of dimer with DNA.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Stat3 and Stat3 have similar DNA binding strength (K2). A, time course for association rate. Purified Stat3 (25 ng) and -3 (1 ng) were mixed with labeled hSIE probe for EMSA. The samples incubated for the indicated times were loaded onto the running gels. Since the gels are running through the experiment, Stat-hSIE complex migrates higher at the later time points. B, time course for dissociation rate. Stat3 (25 ng) and -3 (1 ng) were preincubated with labeled hSIE probe for 15 min at room temperature. Excess of unlabeled hSIE oligonucleotide (100-fold excess) was then added at time 0, and the samples were removed and loaded onto the running gels at the indicated times. The complexes were quantitated using a PhosphorImager and the data were plotted in the right panels of A and B.

The measurement of dissociation rate was assessed by the addition of 100-fold molar excess of unlabeled hSIE to the preformed complex of Stat/hSIE and aliquots were removed at various times for gel analysis (Fig. 2B). As seen in Fig. 2B, the rates of dissociation of the Stat3- and 3-hSIE complexes were essentially identical (t_1/2 of 13 and 12 min, respectively). In preliminary experiments not presented, when lower levels of competitor were used no appreciable differences in the dissociation rates of Stat3 and -3 were observed although a longer incubation time was neccessary for them to dissociate from the DNA. These results lead to the conclusion that Stat3 and -3 have similar binding strength, excluding the possibility that differential DNA binding activity results from differential binding affinity for DNA.

Effect of Reagents That React with the Dimerization Domains of Stat3 on DNA Binding Activity-- Reagents that react with the SH2 domain or phosphotyrosine 705 of Stat3 and -3 should act as competitive inhibitors of dimerization and thereby cause dimer dissociation. To compare the effect of such competitors on the two Stat isoforms, we used three different reagents: an anti-phosphotyrosine antibody (Fig. 3A); the purified SH2 domain of Stat1 (Fig. 3B), which was more effective than the SH2 domain of Stat3 (data not shown); and a synthetic phosphopeptide corresponding to the phosphotyrosine 705 motif of Stat3 (Fig. 3C). Each reagent was incubated with Stat3 or -3 at various molar ratios and residual dimer was assessed by measuring DNA binding activity. With Stat3 there was a marked drop in DNA binding as the ratio of competitor increased. On a molar basis, the SH2 domain was a more active competitor than the anti-phosphotyrosine antibody or phosphopeptide. However, Stat3 showed little or no change in oligonucleotide binding activity even at molar ratios that almost completely abolished the binding activity of Stat3. Only at much higher ratios of competitor to Stat3 was there a noticeable effect of the competitors (Fig. 3D). Whereas phosphopeptide corresponding to the tyrosine 705 motif was effective, a nonphosphorylated version of the peptide had little or no effect (Fig. 3, C and D), demonstrating the specificity of the competitive effect, as anticipated for an interaction with the SH2 domain. We also observed that the dissociation of Stat3 dimer by the addition of competitors reached equilibrium within 15 min, during the incubation time of competition assay, whereas dissociation of Stat3 dimer occurred slowly, taking about 1 h for equilibrium (data not shown). Taken together, these results lead to the conclusion that Stat3 dimers are more stable than Stat3 dimers and that the difference in dimer stability is at least partly responsible for differential DNA binding activity.

View larger version (40K): [in this window] [in a new window]   Fig. 3.   Stat3 and Stat3 show significant difference in dimer stability (K1). Competition assays were performed with increasing amounts of anti-phosphotyrosine antibody (A), SH2 domain of Stat1 (B), and phosphopeptide (C), as described under "Experimental Procedures." The protein amounts of Stat3 and -3 used in the experiments were 25 ng (12 nM) and 1 ng (0.5 nM), respectively. The complexes were quantitated and plotted in the right panels as percentage of initial DNA binding activity measured in the absence of competitors (lanes 1 and 6). A, competition with anti-phosphotyrosine antibody. The molar ratios of (-pY Ab)/(Stat) were, 0.6 (lanes 2 and 7), 2.4 (lanes 3 and 8), 12 (lanes 4 and 9), and 120 (lanes 5 and 10). B, competition with SH2 domain. The molar ratios of (SH2)/(Stat) were, 1.5 (lanes 2 and 7), 7.5 (lanes 3 and 8), 30 (lanes 4 and 9), and 90 (lanes 5 and 10). C, competition with phosphopeptide. Unphosphorylated version was also included in the experiment as a control. The sequences were as follows; p-Peptide, EADPGSAAPYPLKTKF; Peptide, EADPGSAAPYLKTKF. The molar ratios of (p-Peptide or Peptide)/(Stat) were 200 (lanes 2 and 7), 800 (lanes 3 and 8), 2,000 (lanes 4 and 9), and 8,000 (lanes 5 and 10). D, dissociation of Stat3 dimers. The molar ratios of (Competitor)/(Stat) are shown in the figure.

Effect of C-terminal Deletions of Stat3 on Its DNA Binding Properties and Stability-- Stat3 and -3 differ only in their C-terminal sequence, the 55 terminal amino acid residues of Stat3 (containing 8 acidic residues) replaced by 7 residues specific to Stat3 (depicted as a cartoon in Fig. 4). They are otherwise identical, including domains involved in DNA binding and dimerization (SH2 and Tyr-705). Therefore, differences in DNA binding activity and dimer stability of Stat3 and -3 could be due to the presence of the C-terminal sequence of Stat3 and/or the presence of the specific C-terminal residues of Stat3. To examine this, we prepared mutant proteins with serial deletions (12, 19, 33, 40, 48, and 55) in the Stat3 C-terminal region. 48 mutant lacks most of the negatively charged C-terminal 48 amino acids of Stat3 (Fig. 4), resulting in a molecule with the same number of amino acid residues as Stat3. 55 mutant ends at the amino acid residue where Stat3 and -3 diverge in sequence. We first compared the DNA binding activities by EMSA (Fig. 4), using equimolar concentrations of the tyrosine 705-phosphorylated forms of each Stat protein, as described in the figure legend. DNA binding activity per unit of phosphorylated protein increased with successive deletion of the C terminus. We next determined whether the increase in DNA binding activity correlated with an increase in dimer stability assessed by use of dimerization competitors as above. As shown in Fig. 5, all of the C-terminal deletion mutants tested showed reduced sensitivity to competitors compared with Stat3, 33, 48, and 55 being approximately equivalent to Stat3. We conclude that the presence or absence of the C-terminal residues of Stat3 determines the differences in DNA binding activity and dimer stability between Stat3 and -3.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   DNA binding activities of in vitro purified Stat3 deletion mutants. EMSA analyses performed with Stat3 deletion mutant proteins prepared from Sf9 cells. The amounts of proteins were as follows; no proteins (lane 1), 1 ng (lane 2), 2.5 ng (lane 3), 5 ng (lane 4), 10 ng (lane 5), 25 ng (lane 6), and 50 ng (lane 7). Schematic diagrams of the deletion mutant proteins compared with Stat3 and -3 wild type are shown in the right panels.

View larger version (74K): [in this window] [in a new window]   Fig. 5.   Stat3 deletion mutants show enhanced dimer stability compared with Stat3 wild type. The competition assays were performed as described in the legend to Fig. 3. The amounts of proteins were as follows; Stat3, 25 ng (12 nM); 19, 6 ng (3 nM); 33, 3 ng (1.5 nM); 48, 1 ng (0.5 nM); 55, 1.5 ng (0.75 nM); Stat3, 1 ng (0.5 nM). The molar ratios of (Competitor)/(Stat) were same as in Fig. 3; (SH2)/(Stat) were, 1.5 (lanes 2, 7, 12, 17, 22, and 27), 7.5 (lanes 3, 8, 13, 18, 23, and 28), 30 (lanes 4, 9, 14, 19, 24, and 29) and 90 (lanes 5, 10, 15, 20, 25, and 30); (-pY Ab)/(Stat) were 0.6 (lanes 2, 7, 12, 17, 22, and 27), 2.4 (lanes 3, 8, 13, 18, 23, and 28), 24 (lanes 4, 9, 14, 19, 24, and 29), and 120 (lanes 5, 10, 15, 20, 25, and 30); (p-Peptide or Peptide)/(Stat) were 200 (lanes 2, 7, 12, 17, 22, and 27), 800 (lanes 3, 8, 13, 18, 23, and 28), 2,000 (lanes 4, 9, 14, 19, 24, and 29), and 8,000 (lanes 5, 10, 15, 20, 25, and 30).

DNA Binding Activity and Dimer Stability of Activated Stat Isoforms in Transfected Cells-- Expanding a previous study comparing the in vivo dimer stability of the two C-terminal deletion forms of Stat3 to that of Stat3 and -3 (19), we next determined the stability of activated forms of the entire set of deletion mutants (Fig. 6). COS-7 cells transfected with each expression plasmid and stimulated with EGF were treated with PD157655, a specific inhibitor of EGF receptor kinase. Nuclear extracts were then prepared at various times and assessed for hSIE binding, total Stat3 protein, and tyrosine 705 phosphorylation. There was a good agreement between hSIE binding activity and tyrosine 705 phosphorylation state, and little change in total Stat3 proteins during the course of the experiment. Comparison of the decay of DNA binding activity and tyrosine 705 phosphorylation revealed a marked difference in stability of Stat3 and -3; the half-life of active Stat3 was approximately 20 min (comparable to the 10 min seen in a previous report (5)) compared with 2 h or more for Stat3. Similar to results obtained in vitro with recombinant proteins, the truncation of C-terminal tail enhanced the half-life of Stat3. The half-lives of 40 and 48 deletion mutants were similar to that of Stat3 (2 h). The results suggest that the stability of tyrosine 705 phosphorylation accounts for the in vivo stability of activated Stat3 proteins and that the rate of dephosphorylation of tyrosine 705 is a function of dimer stability of Stat3 isoforms and is correlated with the DNA binding of the proteins.

View larger version (84K): [in this window] [in a new window]   Fig. 6.   In vivo stability of Stat3 wild type and deletion mutants. COS-7 cells in 5-cm dishes were transfected with 12.5 µg of Stat3 expression plasmid as described under "Experimental Procedures." Forty hours after transfection including 24 h in medium containing no serum, the cells were stimulated for 20 min with EGF (100 ng/ml) at which time the EGF receptor kinase inhibitor PD 157655 (0.2 µM) was added. Cells were harvested for nuclear extract preparation at the times (min) indicated. U indicates no EGF stimulation. Each sample was assessed for SIE binding (2.5 µg) (top panel of each set). Immunoblot analysis to determine the amount of tyrosine 705 phosphorylation was performed using 20 µg of nuclear extract and the Stat3 phosphotyrosine 705 specific antibody (middle panel of each set). The blot was stripped and re-probed with the Stat3 monoclonal antibody to determine the amount of total Stat3 protein. Approximate half-lives of the EGF inducible DNA binding activity and tyrosine 705 phosphorylation of the Stat3 wild type and deletions were calculated by quantitating the SIF A/A* complex using a PhosphorImager, and using a densitometer to quantitate the amount of tyrosine 705-phosphorylated protein and total Stat protein in each sample. Measurements were corrected for constitutive DNA binding and tyrosine 705 phosphorylation, as well as total Stat protein (accounting for viability in transfection efficiency and sample preparation). In all cases there was a good correlation between DNA binding and level of tyrosine 705 phosphorylation and the half-life determined for each.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Stat3 and -3 are the alternatively spliced products of the same gene, differing only at their C termini. Despite the minor difference in primary sequence, we report here that there is a marked difference in the DNA binding activities of recombinant Stat3 and -3 in vitro (Fig. 1A). This is consistent with a previous study that indicated DNA binding of Stat3 is significantly higher than Stat3 in COS-7 cells transfected with expression plasmids (19). In this report, we sought to understand these differences by individually examining the binding processes shown in Fig. 1B. We conclude that the differences in DNA binding activity between the Stat3 isoforms are determined by dimer stability (K1) but not by differential DNA binding strength (K2) (Figs. 2 and 3). Another important finding was that the deletions in the C terminus of Stat3 enhanced both DNA binding activity and dimer stability of Stat3 (Figs. 4 and 5). This indicates that the C-terminal region unique to Stat3 exerts a negative effect on dimer stability and, concomitantly, its DNA binding activity. Although it has yet to be determined how the C-terminal region of Stat3 plays an inhibitory role, one notable characteristic of C-terminal 55 amino acids is net negative charge (8 Asp and Glu versus 1 Arg; net charged 7). In fact, the estimated isoelectric points of Stat3 and -3 are 5.9 and 6.5, respectively. Consistent with this, a Stat3-hSIE oligonucleotide complex migrates faster than Stat3-hSIE complex in electrophoretic mobility shift experiments using native gels. Moreover, Stat3 mutant proteins with progressive C-terminal deletions displayed a proportional decrease in mobility in EMSA experiments, indicating the decrease of net-negative charge (Fig. 5). Although it is conceivable that the negatively charged C-terminal region of Stat3 mediates an interaction with and therefore masks its own dimerization domain, the addition of a GST fusion protein containing the C-terminal 55 amino acids of Stat3 did not result in the dissociation of DNA-bound Stat3 dimers (data not shown). Alternatively, electrostatic repulsion or a particular structure adopted by the C-terminal segment of Stat3 might be unfavorable for dimer formation. Additional mutational and structural analyses on the C terminus are needed to address this possibility.

We observed a good correlation between in vitro and in vivo dimer stability of both Stat3 wild type and deletion mutants. Results obtained with recombinant proteins in vitro that demonstrated an increase in DNA binding by C-terminal truncations were consistent with those observed previously in vivo with Stat3 (19, 24) and similar mutations of Stat5 (16). Assuming that the same phosphatase is responsible for dephosphorylating tyrosine 705 of Stat3 and -3, the C-terminal sequence of Stat3 may provide for a more favorable association with the phosphatase. Another possibility, due to the strong dimers formed by Stat3, tyrosine 705 of Stat3 may not be readily accessible to the phosphatase, in contrast to Stat3. It has been reported that there are naturally occurring short forms of Stat5 and -5, and that the carboxyl-truncated Stat5 short forms are more stably tyrosine-phosphorylated than wild type Stat5 in transfected cells (17). We found similar results with Stat3 truncations assessing the EGF-induced DNA binding in the presence of an EGF kinase inhibitor (19). Stat3 and the shorter deletion forms of Stat3 retained DNA binding activity (and phosphorylation of tyrosine 705) much longer than Stat3. These results imply that differential dimer stability may be a general property shared by pairs of Stat isoforms.

Stat3 and -3 differ in other properties. In transfected culture cells, Stat3 transcriptionally cooperates with c-Jun in the activation of an interleukin 6-responsive promoter that also contains a proximal AP-1 (c-Jun)-binding site, whereas Stat3 shows only an additive effect (18, 25) despite directly interacting with two regions common to both proteins (25). Another striking difference in transfected cells is that only Stat3 can constitutively bind DNA and activate transcription from promoters in the absence of added extracellular cytokines or growth factors (18). It remains to be determined if these properties are also attributable to the difference in dimer stability of the two isoforms.

Stat3 seems to play dual roles, depending on the promoter. There are also reports that Stat3 can function as a dominant negative regulator on certain promoters (26, 27) although the precise mechanism is not known. Our results presented here suggest that this could occur by the occupancy of Stat3-binding sites in promoters by the much more stable Stat3 homodimers. We have shown that despite having a higher dimerization potential than Stat3, Stat3 transcriptional activity is relatively weak (19), compared with Stat3. Stat3 expression level is low compared with Stat3 in many cell lines and tissues.² The unique properties of Stat3 such as constitutive activation and delayed dephosphorylation might have physiological significance when the expression level of Stat3 is elevated (e.g. Stat3 mRNA is produced as a major spliced form).

Proteins immunologically related to Stat3 are constitutively activated in cells transformed by the v-Src oncoprotein (28). Subsequent studies have shown that Stat3 is required for v-Src-mediated transformation (29, 30) and that a mutant Stat3 that can dimerize without tyrosine phosphorylation transforms fibroblasts and forms tumors in nude mice (31). In addition, constitutively activated Stat3 was observed in a number of tumors and tumor-derived cells (17, 27, 32-36). Stat3 also appears to block apoptosis by directly activating the bcl_xl gene (27). Thus, the notion that genes activated by Stat3 play a role in aberrant cell growth and preventing programmed cell death makes this protein a reasonable target for molecules designed to interfere with dimerization/DNA binding.

	ACKNOWLEDGEMENTS

We thank J. N. Ihle for Jak1 and Jak2-expressing baculoviruses, D. Fry for EGF kinase inhibitor PD157655, and Y. Nakabeppu for expression plasmids.

	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.

We dedicate this manuscript to the memory of Daniel Nathans (1928-1999) in whose laboratory this work was initiated.

^§ To whom correspondence should be addressed: Kumho Life and Environmental Science Laboratory (KLESL), 1 Oryong-dong, Puk-gu, Kwangju 500-480, Korea. Tel.: 82-62-970-2625; Fax: 82-62-972-5085; E-mail: omkim@hotmail.com.

Published, JBC Papers in Press, July 27, 2000, DOI 10.1074/jbc.M005082200

² L. K. Schaefer and T. S. Schaefer, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: JAK, Janus kinase; STAT, signal transducer and activator of transcription; EGF, epidermal growth factor; SH2, Src homology domain 2; EMSA, electrophoretic mobility shift assay; hSIE, high affinity c-fos sis-inducible element; IL, interleukin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Huang, J. Qiu, S. Dong, M. S. Redell, V. Poli, M. A. Mancini, and D. J. Tweardy Stat3 Isoforms, {alpha} and , Demonstrate Distinct Intracellular Dynamics with Prolonged Nuclear Retention of Stat3 Mapping to Its Unique C-terminal End J. Biol. Chem., November 30, 2007; 282(48): 34958 - 34967. [Abstract] [Full Text] [PDF]

				M. S. Redell, A. Tsimelzon, S. G. Hilsenbeck, and D. J. Tweardy Conditional overexpression of Stat3{alpha} in differentiating myeloid cells results in neutrophil expansion and induces a distinct, antiapoptotic and pro-oncogenic gene expression pattern J. Leukoc. Biol., October 1, 2007; 82(4): 975 - 985. [Abstract] [Full Text] [PDF]

				I. Lodige, A. Marg, B. Wiesner, B. Malecova, T. Oelgeschlager, and U. Vinkemeier Nuclear Export Determines the Cytokine Sensitivity of STAT Transcription Factors J. Biol. Chem., December 30, 2005; 280(52): 43087 - 43099. [Abstract] [Full Text] [PDF]

				H. Shao, A. J. Quintero, and D. J. Tweardy Identification and characterization of cis elements in the STAT3 gene regulating STAT3alpha and STAT3beta messenger RNA splicing Blood, December 15, 2001; 98(13): 3853 - 3856. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/32244    most recent M005082200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services