The Reciprocal Role of Egr-1 and Sp Family Proteins in Regulation of the PTP1B Promoter in Response to the p210 Bcr-Abl Oncoprotein-tyrosine Kinase*

Protein-tyrosine phosphatase 1B (PTP1B) is an important regulator of protein-tyrosine kinase-dependent signaling pathways. Changes in expression and activity of PTP1B have been associated with various human diseases; however, the mechanisms by which PTP1B expression is regulated have yet to be characterized. Previously, we have shown that the expression of PTP1B is enhanced by p210 Bcr-Abl and that PTP1B is a specific antagonist of transformation induced by this oncoprotein protein-tyrosine kinase. Here we have characterized the PTP1B promoter and demonstrate that a motif with features of a stress-response element acts as ap210 Bcr-Abl-responsive sequence, termed PRS. We have shown that three C2H2 zinc finger proteins, namely Sp1, Sp3, and Egr-1, bind to PRS. Whereas binding of either Sp1 or Sp3 induced promoter function, Egr-1 repressed Sp3-mediated PTP1B promoter activation. The binding of Egr-1 to PRS is suppressed by p210 Bcr-Abl due to the inhibition of Egr-1 expression, resulting in the enhancement of PTP1Bpromoter activity. Our data indicate that Egr-1 and Sp family proteins play a reciprocal role in the control of expression from thePTP1B promoter.

The actions of protein-tyrosine phosphatases (PTPs) 1 and kinases (PTKs) are coordinated in vivo to regulate the reversible phosphorylation of tyrosine residues in proteins and thus control such fundamental physiological processes as growth and proliferation, differentiation, migration, metabolism, and cytoskeletal function. Furthermore, disturbance of the delicate balance between the actions of PTKs and PTPs has been shown to be the cause of human diseases, including cancer, diabetes, and inflammation. The PTPs are now known to consist of a large family of receptor-like and nontransmembrane enzymes that rival the PTKs in structural diversity and complexity (1). Furthermore, PTPs can both antagonize PTK-induced signaling events and cooperate with PTKs to promote signal transduction processes in vivo, thus highlighting further their importance as an integral component of the cellular response to environmental cues. With this realization that PTPs are critical regulators of cellular signaling events, they have become a focus of attention for their potential to identify novel targets for therapeutic intervention in human disease.
The prototypic member of the PTP family is PTP1B. It was first purified from human placenta as a catalytic domain of 37 kDa (2) and was subsequently shown to occur in vivo as a full-length protein of ϳ50 kDa (3)(4)(5). PTP1B is composed of an N-terminal catalytic domain fused to a non-catalytic, C-terminal segment that serves a regulatory function, targeting the protein to the cytoplasmic face of membranes of the endoplasmic reticulum (6). Recently, a number of insights have been gained into the physiological function of PTP1B that have emphasized its potential importance in various human disease states. For example, PTP1B has been reported to act as a negative regulator of signaling events initiated by several growth factor/hormone receptor PTKs, including the epidermal growth factor and insulin receptors, as well as signaling events induced by cytokines (1). Several reports noted enhanced expression levels of PTP1B in diabetic and insulin-resistant patients and animal models that may affect insulin signaling (7)(8)(9). Furthermore, it was also reported that there might be a negative feedback loop, governed by insulin and insulin-like growth factor stimulation, which could lead to up-regulation of PTP1B gene expression (10,11). Of particular interest has been the demonstration that PTP1B knockout mice display hypersensitivity to insulin and resistance to diet-induced obesity (12,13). Recent structural studies have defined how PTP1B recognizes the activation loop of the insulin receptor as a substrate and thus can modulate signaling in response to stimulation by the hormone (14). These observations suggest the exciting possibility that an inhibitor of PTP1B may be of therapeutic use in the treatment of diabetes and obesity (15). Furthermore, changes in the levels of PTP1B have been noted in several human diseases, particularly those associated with disruption of the normal patterns of tyrosine phosphorylation. Work from our own laboratory (16) has shown that the expression of PTP1B is induced specifically by the p210 Bcr-Abl oncoprotein, a PTK that is directly responsible for the initial manifestations of chronic myelogenous leukemia. We have also observed that PTP1B, but not TC-PTP, its closest relative, suppresses p210 Bcr-Abl-mediated signaling and transformation (17). Therefore, PTP1B may also function as an antagonist of the p210 Bcr-Abl oncoprotein PTK in vivo. In addition, changes in PTP1B levels have been associated with other human cancers associated with oncoprotein PTKs (18,19). However, the exact molecular mechanism by which PTP1B gene expression is regulated in these contexts remains to be determined.
We have observed that the effects of p210 Bcr-Abl on expression of the PTP1B gene are manifested at the transcriptional level. In order to define the mechanism by which PTP1B gene expression is regulated, we cloned the 5Ј-flanking region of the human gene. We found two elements that are important for expression from the human PTP1B promoter. A sequence motif that possesses features of a site of interaction with GATAbinding proteins was identified at Ϫ167 to Ϫ151 bp from the transcription start site. Disruption of this site inhibited promoter activity in the presence of p210 Bcr-Abl, but responsiveness to the oncoprotein PTK was maintained. However, a p210 Bcr-Abl-responsive sequence, termed PRS, which was important for stimulation of activity in response to the PTK, was identified at Ϫ49 to Ϫ37 bp from the transcription start site. The PRS is contained in a sequence that displays features of a stress-response element (STRE), a feature originally identified in Saccharomyces cerevisiae (20 -22) and that functions as a binding site for C 2 H 2 zinc finger proteins (23). In our study, we have shown that three mammalian C 2 H 2 zinc finger proteins, Egr-1, Sp1, and Sp3, bind to PRS. Of these, both of Sp1 and Sp3 function as positive regulators, whereas Egr-1 represses Sp3mediated transactivation of the PTP1B gene. Furthermore, expression of p210 Bcr-Abl results in down-regulation of the levels of Egr-1. Our data illustrate that the reciprocal actions of the Sp1/Sp3 and Egr-1 C 2 H 2 zinc finger transcription factors are an important aspect of the regulation of PTP1B expression in response to the p210 Bcr-Abl oncoprotein.

Isolation and Cloning of the 5Ј-Flanking Region of the Human PTP1B
Gene and Plasmid Construction-The 5Ј-flanking segment of the human PTP1B gene was isolated by PCR using the GenomeWalker kit (CLONTECH). Two gene-specific primers (GSPs), 5Ј-GATAATGGC-CGCCCAGCTCCCGGACTTGTC-3Ј as GSP1 (31-60 bp relative to the translation start site) and 5Ј-CCCAAGCTTGAGATCTCTCGAGG-ATCTGCTCGAACTCCTTTTCCATCTCCAT-3Ј as GSP2 (1-30 bp relative to the translation start site and including an additional HindIII site), were selected on the basis of the sequence of the human PTP1B cDNA reported by Brown-Shimer et al. (5). Primary PCR and secondary nested PCR were performed on adapter-ligated genomic libraries using the GSP1 and GSP2 gene-specific primers and the AP1 and AP2 adapter sequence primers. The resulting 2-kb PCR fragment was subcloned into the TA cloning vector pCR2.1-TOPO (Invitrogen) and sequenced.
Determination of 5Ј-Untranslated Region (UTR) Sequence-In order to identify the transcriptional start site, we used the CapSite 5Ј end cDNA library (Eurogenetec) as a template for PCR, using two 5Ј adapter sequences (RC1 and RC2) as 5Ј primers together with an internal sequence in PTP1B (241-270 bp from the translation start site, 5Ј-GTT-AGGCAAAGGGCCCTGGGTAAGAATGTA-3Ј) as the 3Ј primer. The other internal sequence in PTP1B (1-30 bp relative to the translation start site, 5Ј-ATGGAGATGGAAAAGGAGTTCGAGCAGATC-3Ј) was used as the 5Ј control primer.
Transient Transfection-Cells were transfected using Lipo-fectAMINE Reagent (Life Technologies, Inc.) according to the supplier's protocols. Typically, we used 1 g of the reporter plasmid for expression of firefly luciferase and 1 g of pRL-TK (Promega), an expression vector containing cDNA encoding Renilla luciferase, as an internal control of transfection efficiency. For Drosophila SL2 cells, 2.0 ϫ 10 6 cells were used for each transfection with LipofectAMINE Reagent (Life Technologies, Inc.). One g of the reporter plasmid for expression of firefly luciferase was used. Expression plasmids for either human Sp1 (pPac Sp1), Sp3 (pPacU Sp3), or control plasmid without insert (pPac) were cotransfected with different amounts of Egr-1 expression plasmid (pPac Egr-1) in the amounts indicated in the legend to Fig. 6. Cells were incubated with DNA-lipid complex for 24 h and washed with phosphatebuffered saline, and luciferase activity was assayed using the Dual-Luciferase Reporter Assay System (Promega).

Definition of Promoter Elements in the Human PTP1B
Gene-In order to begin to elucidate the molecular mechanisms underlying regulation of expression of the human PTP1B gene, we isolated a 2-kb fragment of the 5Ј-flanking region of the gene. The sequence of the proximal region up to Ϫ838 bp from the transcription initiation site is shown in Fig. 1A. In order to identify the transcriptional start site, we used the CapSite 5Ј end cDNA library (Eurogenetec) as a template for PCR, using two 5Ј adapter sequences (RC1 and RC2) as 5Ј primers together with an internal sequence in PTP1B (241-270 bp from the translation start site) as the 3Ј primer. Another internal sequence in PTP1B (1-30 bp relative to the translation start site) was used as the 5Ј control primer. By this approach, we cloned a 0.45-kb fragment (Fig. 1B), sequenced it, and identified 145 bp as the 5Ј-UTR in human PTP1B mRNA (Fig. 1A).
A search for potential sites for interaction with DNA-binding proteins in the 5Ј-flanking region of the human PTP1B gene, using TFSEARCH, revealed several candidates. These include potential sites of interaction with the following: the AML-1 DNA-binding protein, which has been implicated in hematopoietic cell proliferation and differentiation and is the product of a gene that is disrupted in certain myelogenous leukemias (25); the AP-1 transcription factor complex, which is regulated in response to various growth, differentiation, and stress stimuli (26); CHOP, which is induced under conditions of cellular stress (27); Ets-1, the founding member of a family of transcription factors that display homology to viral oncogenes and have been implicated in the control of growth, differentiation, and development (28); Ikaros 2, from the Ikaros family of zinc finger transcription factors that have been implicated in the regulation of lymphocyte development (29); MyoD, the muscle differentiation factor (30); c-Myb, which is essential for normal control of hematopoiesis (31); and NF-B, which is a critical regulator of immune and inflammatory gene expression in cell proliferation and apoptosis (32). In addition, there are putative binding sites for GATA-binding proteins, which are important regulators of diverse cell differentiation pathways during development (33)(34)(35)(36) and clusters of GC-rich regions, consistent with sites of interaction with C 2 H 2 zinc finger proteins, such as the specificity protein (Sp) family (37). Finally, we observed that the human PTP1B gene lacks a TATA box, an initiator, or a downstream promoter element (38). It has been reported that promoters that lack either a TATA box or an initiator element often contain multiple binding sites for Sp1, as we have noted here for the human PTP1B gene. However, unlike many such genes that are often associated with "housekeeping" functions, PTP1B expression varies in response to a variety of signals, and the enzyme serves a critical function as a regulator of tyrosine phosphorylation-dependent signaling pathways (1).
The Activity of the PTP1B Promoter Is Enhanced by Expression of p210 Bcr-Abl-Our laboratory has demonstrated previously (16) that the levels of PTP1B mRNA and protein are FIG. 1. A, nucleotide sequence of the promoter region of the human PTP1B gene, highlighting potential recognition motifs for DNA-binding proteins. B, determination of 5Ј-UTR of human PTP1B gene. The DNA sequence of the 5Ј-UTR upstream from the translation start site of human PTP1B was determined by PCR using CapSite cDNA (Eurogentec) as a template. Primary PCR and secondary nested PCR were performed using the 1RC and 2RC primers corresponding to the adapter sequences and adapter-ligated CapSite cDNA derived from human placenta. The 3Ј primer consists of 241-270 bp from the translation start site of human PTP1B and the control 3Ј primer consists of 1-30 bp from the translation start site. The resulting 0.45-kb PCR fragment (Product 1) was subcloned and sequenced. The 5Ј-UTR is shown as a gray box in A.
FIG. 2. The activity of the human PTP1B promoter is enhanced by p210 Bcr-Abl in Rat1 fibroblasts. We constructed a luciferase reporter plasmid containing the intact PTP1B promoter. Reporter activity was assessed in parental Rat1 fibroblasts (Rat1) and Rat1 fibroblasts expressing p210 Bcr-Abl (Rat1 p210), a catalytically inactive mutant form of p210 Bcr-Abl (Rat1 p210KN) and v-Abl (Rat1 v-Abl). The cells were transiently transfected with 1 g of the intact human PTP1B promoter luciferase construct, containing the 5Ј-flanking region of the human PTP1B gene, (pGL3-Basic:Ϫ2k/ϩ145), control plasmid, which contains SV40 promoter and enhancer sequence, (pGL3-Control), or empty vector (pGL3-Basic), and 1 g of pRL-TK, to normalize the transfection efficiency. Cells were harvested, lysed, and assayed for luciferase activity. Values were normalized for transfection efficiency and represent the means Ϯ S.D. of three separate experiments. enhanced specifically in various cells expressing p210 Bcr-Abl and that the PTK activity of the p210 Bcr-Abl oncoprotein is required for this effect. We have now generated a reporter construct in which expression of luciferase is driven by potential PTP1B promoter elements. This reporter construct, which contains 5Ј-flanking sequence from the PTP1B gene extending from Ϫ2 kb to ϩ145 bp relative to the transcription initiation site, was transfected into various Rat-1 fibroblast lines, and production of firefly luciferase was monitored as a measure of promoter activity. Transfection efficiency was standardized according to expression of Renilla luciferase from a control plasmid. We observed 4-fold higher promoter activity in p210 Bcr-Abl-transformed Rat-1 cells (Rat1 p210 Bcr-Abl) compared with the parental Rat1 cells (Fig. 2). In addition, we observed that optimal expression from the PTP1B promoter required the kinase activity of p210 Bcr-Abl. Furthermore, expression of v-Abl, which shares the same catalytic domain as p210 Bcr-Abl, stimulated promoter activity only weakly (Fig. 2), in agreement with our previous observations of the effects of these PTKs on the levels of PTP1B protein (16). These data suggest that p210 Bcr-Abl triggers signaling events that culminate specifically in stimulation of the PTP1B promoter.
Regulation of PTP1B Promoter Activity by a p210 Bcr-Ablresponsive Sequence (PRS)-To define the cis-acting element(s) in the PTP1B promoter, we made sequential truncations in the 5Ј-flanking region of the PTP1B gene and incorporated these deletion mutants into constructs where they drove expression of a luciferase reporter gene. The promoter activity of these constructs was compared following transfection into parental and p210 Bcr-Abl-transformed Rat1 fibroblasts. By using this approach, we identified a transcriptional enhancing element and a potential p210 Bcr-Abl-responsive sequence, which we have termed PRS (Fig. 3). Truncation as far as Ϫ167 bp from transcription initiation site did not affect total promoter activity or responsiveness to p210 Bcr-Abl. However, we identified a transcriptional enhancing element, between Ϫ167 and Ϫ151 bp upstream from the transcription start site (Fig. 3A), that displays features of a binding site for GATA family binding proteins (33). Deletion of this element by truncation to Ϫ151 bp suppressed promoter activity by ϳ60%, but the responsiveness of promoter activity to p210 Bcr-Abl was retained (Fig. 3A). Truncation to Ϫ49 bp upstream from the transcription start site resulted in no further alteration in promoter activity, whereas promoter function was severely attenuated by truncation to Ϫ37 bp. Importantly, the residual activity was no longer responsive to p210 Bcr-Abl (Fig. 3A). This latter element, spanning Ϫ49 to Ϫ37 bp upstream from the transcription start site, is termed PRS and corresponds to a segment possessing features of an STRE (Fig. 3B). STREs were originally identified in the promoter of the DDR2 and CTT1 gene in S. cerevisiae and shown to bind C 2 H 2 zinc finger proteins following stimulation by heat or osmotic stress (23). We introduced nucleotide substitution mutations to assess the importance of the putative STREs in PRS, and we observed that disruption of STRE-1 abolished p210 Bcr-Abl responsiveness, whereas mutation of STRE-2 was without effect (Fig. 3A). These data suggest that PRS may have a critical role in p210 Bcr-Abl-mediated upregulation of PTP1B expression.
The C 2 H 2 Zinc Finger Proteins Egr-1, Sp1, and Sp3 Bind PRS-Since we had noted that the sequence of the PRS motif possessed the features of an STRE, we assessed whether mammalian C 2 H 2 zinc finger proteins would bind to this motif from the PTP1B promoter. We performed DNA "pull-down" assays, using biotinylated DNA probes derived from the sequence of PRS and nuclear extracts of Rat-1 fibroblast lines, to determine which proteins may bind to this motif. We observed that three FIG. 3. Identification of a p210 Bcr-Abl response element in the human PTP1B promoter. A, luciferase constructs bearing various deletions within the sequences upstream, on the 5Ј side, of the human PTP1B gene were generated. Rat1 (stippled bar) and Rat1 p210 Bcr-Abl-transformed cells (black bar) were transiently transfected with 1 g of human PTP1B promoter luciferase construct containing segments of the human PTP1B promoter including PRS-mutated sequences (pGL3-Basic:-49(M1)/ϩ145 for STRE1 and pGL3-Basic:-49(M2)/ϩ145 for STRE2), control plasmid, which contains SV40 promoter and enhancer sequence, (pGL3-Control), empty vector (pGL3-Basic), and 1 g of pRL-TK to normalize transfection efficiency. Cells were harvested, lysed, and assayed for luciferase activity. Values were normalized to transfection efficiency and represent the means Ϯ S.D. of three separate experiments. B, schematic profile of human PTP1B promoter. Two elements were identified as important for the human PTP1B promoter-driven luciferase activity in response to p210 Bcr-Abl. A p210 Bcr-Abl-responsive sequence (PRS), which possesses features of an STRE that interacts with C 2 H 2 zinc finger proteins, was identified at Ϫ49 to Ϫ37 bp from the transcription start site. In addition, an element possessing a conserved sequence motif for recognition of GATA family proteins, such as GATA-1, GATA-2, was identified at Ϫ167 to Ϫ151 bp from the transcription start site and functioned as a transcriptional enhancing element. C 2 H 2 zinc finger family transcription factors, Egr-1, Sp1 and Sp3, from nuclear extracts of parental Rat-1 fibroblasts, associated with PRS probes (Fig. 4A). A mutant PRS probe, in which STRE-1 was disrupted (PRSM), displayed significantly reduced affinity for these proteins (Fig. 4A). Furthermore, we have used nuclear extracts of parental, p210 Bcr-Abl, and catalytically inactive p210 Bcr-Abl-expressing Rat-1 fibroblasts to determine the effects of expression of p210 Bcr-Abl on the interaction of these transcription factors with the PRS probe. We observed that expression of p210 Bcr-Abl led to a decrease in the level of Egr-1 protein (Fig. 4, A and B) and mRNA (Fig.  4C), compared with that in parental Rat1 cells, resulting in a decrease in the interaction detected between Egr-1 and the PRS DNA probe (Fig. 4A). In contrast, the presence of p210 Bcr-Abl enhanced the interaction of Sp1 protein with PRS (Fig.  4A) without affecting the expression level of this transcription factor (Fig. 4, B and C). These effects were dependent upon the catalytic activity of p210 Bcr-Abl. The expression levels of Sp3 and its association with the PRS probe were unaltered by p210 Bcr-Abl (Fig. 4). These data indicate that PRS, which is a regulator of the activity of the PTP1B promoter, binds Sp1, Sp3, and Egr-1 and that expression of p210 Bcr-Abl controls the activity and expression of Sp1 and Egr-1, respectively.
Regulation of PTP1B Promoter Function by the Sp1, Sp3, and Egr-1 Transcription Factors-In order to test directly whether the C 2 H 2 zinc finger family transcription factor Egr-1 may regulate PTP1B promoter activity, we performed cotransfection experiments in which this protein was expressed together with the reporter constructs in which expression of the luciferase gene was driven by the PTP1B promoter. As shown in Fig. 4, expression of p210 Bcr-Abl in Rat-1 fibroblasts suppressed the levels of endogenous Egr-1 compared with those observed in the parental cells. We observed that when we expressed Egr-1 ectopically in p210 Bcr-Abl-transformed Rat-1 fibroblasts, up to levels that approximated those seen in the parental cells (Fig. 5, B and C), the luciferase activity driven by the PTP1B promoter was inhibited in a dose-dependent manner (Fig. 5A). Similar results were observed with constructs containing either the full-length promoter or PRS alone (Fig.  5A). These effects occurred without changes in the levels of Sp1 or Sp3 (data not shown). Furthermore, inhibition of promoter function by Egr-1 was not observed in the reporter construct containing the mutant PRS sequence (PRSM) (Fig. 5A). These data suggest that PTP1B gene expression may be negatively regulated by the transcription factor Egr-1 acting on the PRS motif in the promoter.
We also conducted similar luciferase reporter assays in Drosophila SL2 cells, which lack endogenous Sp factors (39), to assess the effect of Sp1 and Sp3 proteins on the activity of the human PTP1B promoter. The reporter plasmid containing the intact PTP1B promoter was cotransfected with expression vectors for Sp1 and/or Sp3, in the presence or absence of the Egr-1 expression vector. We observed that the intact PTP1B promoter-induced luciferase activity was up-regulated, in a dose-dependent manner, by either Sp1 or Sp3, suggesting that PTP1B gene expression may be positively regulated by either of these transcription factors (Fig. 6A). Interestingly, whereas Egr-1 did not affect Sp1-mediated up-regulation of PTP1B promoter ac- FIG. 4. Characterization of proteins that associate with the PRS element in the human PTP1B promoter. A, effect of p210 Bcr-Abl on the association of Egr-1, Sp1, and Sp3 with the PRS element in the PTP1B promoter. Nuclear extracts were prepared from parental (Rat1), p210 Bcr-Abl-transformed (Rat1 p210), and p210 Bcr-Abl-inactive mutant (Rat1 p210KN)-expressing Rat1 cells. The extracts were incubated with biotinylated PRS or mutated PRS (PRSM) doublestranded DNA probe and streptavidin-agarose for 6 h at 4°C. The protein-DNA complexes were subjected to SDS-PAGE followed by immunoblotting with either anti-Egr-1, Sp1, or Sp3 antibodies. The level of each transcription factor was assessed by immunoprecipitation (IP) and immunoblotting (IB) using the appropriate antibodies, followed by visualization by ECL. B, the expression level of Egr-1 was suppressed by p210 Bcr-Abl. Nuclear extract (20 g) of each sample was resolved by SDS-PAGE, transferred to PVDF membrane, and immunoblotted with antibodies to Egr-1, Sp1, or Sp3 to assess changes in the expression levels of the transcription factors. The blots were visualized by ECL. C, the expression of egr-1 mRNA was suppressed by p210 Bcr-Abl. Total RNA (20 g) from each sample was resolved by electrophoresis and transferred to nylon membrane. Membranes were hybridized with 32 Plabeled cDNA probe and subjected to autoradiography. The probes used were rat egr-1 and human sp1 and sp3. Equal loading and integrity of the RNA samples was confirmed by ethidium bromide staining of 28 S and 18 S rRNAs in the gel. tivity, Sp3-mediated transactivation of the PTP1B promoter was suppressed by Egr-1 in a dose-dependent manner (Fig. 6A). Egr-1 also suppressed PTP1B promoter transactivation mediated by the expression of both Sp1 and Sp3 in SL2 cells (Fig.  6A). Similar effects were observed with constructs containing either the full-length promoter or PRS alone (Fig. 6B). Furthermore, there were no effects of Sp1 or Sp3 on promoter activity in constructs containing the mutated PRS sequence (PRSM) (Fig. 6C). These data suggest that Sp1 and Sp3 may act as positive regulators of PTP1B gene expression, whereas Egr-1 may function to down-regulate the PTP1B promoter by suppressing Sp3-mediated transactivation and that these effects are mediated through the PRS motif of the promoter. DISCUSSION In this study, we have cloned 5Ј-flanking sequences and determined the transcription initiation site of the human PTP1B gene. In addition, we have identified two promoter elements that are responsible for the regulation of PTP1B expression. A transcriptional enhancing element, which is located between Ϫ167 and Ϫ151 bp upstream of the transcription start site, features the consensus sequence for recognition of GATA-binding proteins. GATA-binding proteins are a family of transcription factors that recognize the consensus sequence T/A(GATA)A/G and are regulators of developmental differentiation pathways. GATA-1 is the founder of a family that now comprises six members. GATA-1, -2, and -3 are crucial for hematopoiesis (33), GATA-2 and -3 have been implicated in the development of the central nervous system (34,36), and GATA-4, -5, and -6 are involved in the development of the heart and viscera (35). However, further studies will be required to identify and characterize the proteins that bind to this site in the PTP1B promoter. Although this element is important for maximal promoter activity (Fig. 3A), comparison of the promoter sequences in human and mouse revealed that the highest level of conservation was in the region surrounding the second element, PRS (40). Therefore, since this sequence sim-  Fig. 6A was separated by SDS-PAGE, transferred to PVDF membrane, and immunoblotted with antibodies to Egr-1. The blots were visualized by ECL. C, ectopically expressed Egr-1 binds to the PRS element of the PTP1B promoter. Parental (Rat1) and p210 Bcr-Abl-transformed (Rat1 p210) Rat1 cells were cotransfected with Egr-1 expression plasmid (2 or 5 g) and control empty vector to a total of 5 g of DNA. Nuclear extracts were prepared and DNA binding assays performed using the PRS-containing probe. The levels of PRS-bound Egr-1 and total Egr-1 in the extract were assessed by immunoblotting (IB) with anti-Egr-1 antibodies and visualized by ECL.
ilarity is consistent with a function of PRS as a general regulator of PTP1B expression, at this time we have focused our attention on this latter motif.
PRS is located between Ϫ49 and Ϫ37 bp upstream of the transcription start site in the human PTP1B and is composed of a palindromic sequence (AGGGG and in the reverse orientation CCCCT) containing STREs. STREs were first identified in S. cerevisiae and shown to bind C 2 H 2 zinc finger proteins following stress stimuli (23). Several C 2 H 2 zinc finger transcription factors have been identified in mammals, including the specificity protein (Sp) family (Sp1-4), the early growth response family (Egr-1-4), and Wilm's tumor proteins (WT1) (37,41). Sp family members recognize GC-rich promoter elements such as the GC box, which appears frequently in the regulatory regions of genes that respond to signaling cues (37). Sp family proteins are structurally conserved but mediate distinct functions. For example, Sp1 has been implicated in promoting transcriptional activation, whereas Sp3 functions either as a repressor (42) or an activator (43,44). Egr-1 is the prototypic member of the Egr family of transcription factors, which contain a DNA binding domain that is homologous to that found in Sp1 and recognize the consensus GC-rich DNA sequence (45,46). Similarly to Sp3, Egr-1 behaves both as a positive (47) and negative (45) regulator of gene transcription, depending upon the cell type.
We have demonstrated that three mammalian C 2 H 2 zinc finger proteins, Sp1, Sp3, and Egr-1, are bound to PRS (Fig.  4A). Both Sp1 and Sp3 acted as positive regulators of PTP1B promoter function (Fig. 6, A and B), whereas Egr-1 functioned negatively (Fig. 5A) to antagonize Sp3-mediated transactivation of the PTP1B promoter (Fig. 6, A and B). In parental Rat1 fibroblasts in which promoter activity was low (Fig. 3A), the PRS element recognized Sp1, Sp3, and Egr-1 (Fig. 4A). In contrast, expression of p210 Bcr-Abl resulted in decreased levels of Egr-1 protein and consequently a reduction in PRS-bound Egr-1 (Fig. 4), whereas there was no detectable effect of the oncoprotein PTK on the levels of Sp1 or Sp3. Although the levels of Sp1 were unchanged, its association with PRS was enhanced (Fig. 4A). However, the mechanism by which Sp1 binding to the PRS element of PTP1B promoter is regulated remains to be determined. Interestingly, it has been reported that Sp1 may be phosphorylated in vivo, raising the possibility that such covalent modification would be of regulatory significance. Furthermore, enhanced phosphorylation of Sp1 following serum stimulation has been associated with Sp1-dependent activation of the DHFR promoter (48), and cAMP-dependent protein kinase phosphorylates Sp1 in vitro, and this phosphorylation leads to increased Sp1 DNA binding (49). Since p210 Bcr-Abl activates multiple signaling pathways, and its effects on PTP1B expression are dependent upon its PTK activity, expression of this oncoprotein PTK may induce the phosphorylation and activation of Sp1 in vivo. It will be important to define which of the signaling pathways initiated by p210 Bcr -FIG. 6. Egr-1 suppresses Sp1/Sp3-mediated activation of human PTP1B promoter. Drosophila SL2 cells were cotransfected with luciferase reporter constructs bearing the intact 5Ј upstream sequence of the human PTP1B gene (pGL3:Ϫ2k/ϩ145) (A), 5Ј end-deleted sequence (pGL3:Ϫ49/ϩ145) (B), or PRS-mutated sequence (pGL3: Ϫ49(M1)/ϩ145) (C) and indicated quantities of expression plasmid for Sp1 (pPac Sp1), Sp3 (pPacU Sp3), or Egr-1 (pPac Egr-1). The total quantity of pPac plasmid DNA was made up to 5 g by inclusion of the control plasmid lacking the insert (pPac). After incubation with the DNA-lipid complex for 24 h to facilitate transfection, the cells were lysed, and luciferase activity was assayed. Values were normalized to transfection efficiency and represent the means Ϯ S.D. of three separate experiments. White bar, experiments using pPac vector and Egr-1; diagonal-lined bar, vector and Sp1; heavy diagonal-lined bar, vector, Sp1, and Egr-1; cross-hatched-lined bar (white and black background), vector and Sp3; cross-hatched line (black and white background), vector, Sp3, and Egr-1; black bar, vector, Sp1, Sp3, and Egr-1.
Abl affect both the activity of the Sp family of transcription factors and the levels of Egr-1.
In many promoters containing GC-rich elements, Sp1-and Egr-1-binding sites have been shown to overlap, and Egr-1 has been shown to modulate Sp1-mediated transactivation both negatively and positively (45, 50 -52). For example, in the case of the platelet-derived growth factor-A (PDGF-A) and -B (PDGF-B) genes, Egr-1 competes with Sp1 protein for an overlapping region in the promoter and functions as a positive regulator (51,52). Thus, competition for promoter-binding sites between Sp and Egr family proteins may be of regulatory significance. We have observed that ectopic expression of Egr-1 in p210 Bcr-Abl-transformed Rat-1 fibroblasts was sufficient to antagonize transcription from the PTP1B promoter (Fig. 5A), consistent with an important role for Egr-1 as a suppressor of PTP1B expression. Precedent exists for such a function of Egr-1, which has been shown to act as a negative regulator of expression from the adenosine deaminase gene promoter (45). The data from our studies in SL2 cells (Fig. 6) revealed that although expression of either Sp1 or Sp3 induced PTP1B promoter activity, this inhibitory effect of Egr-1 appeared to be directed toward Sp3-mediated transactivation. Thus, the down-regulation of Egr-1 expression, alleviating an inhibitory effect on Sp3-mediated transactivation, appears to be an important component of the mechanism of enhancement of PTP1B expression in response to p210 Bcr-Abl. However, it remains to be determined whether there is competition for binding sites in the promoter.
It is now becoming apparent that members of the PTP family display exquisite specificity in their function as regulators of signal transduction processes in vivo. There are several aspects to the establishment and maintenance of such specificity. For example, instances of selectivity in PTP substrate recognition have been identified and are being characterized at the molecular level (14). This current study illustrates another aspect of specificity, that of control of PTP expression. The expression of PTP1B is induced following aberrant activation of multiple PTK-dependent signaling pathways in human disease states. For example, PTP1B is induced specifically in response to p210 Bcr-Abl and functions as an antagonist of this oncoprotein PTK (16,17). The results of this study provide the first mechanistic insights into the control of PTP1B promoter function and will form the basis for further analyses of the expression of this important signaling molecule under normal and other pathophysiological conditions. In particular, although these data illustrate a reciprocal role of Egr-1 and Sp family members in controlling PTP1B promoter activity in transfection assays, it will be important to investigate further which proteins interact with the promoter under physiological conditions and to examine the impact of Sp family members on the activity of chromosomal template.