The role of Sp1 in the differential expression of transforming growth factor-beta receptor type II in human breast adenocarcinoma MCF-7 cells.

Progression of MCF-7 cells from early passage (MCF-7E, <200 passage) to late passage (MCF-7L, >500 passage) correlates with a loss of sensitivity to exogenous TGFbeta1. We have previously shown that loss of TGFbeta sensitivity is due to decreased expression of the transforming growth factor receptor type II (TbetaRII) and is associated with increased tumorigenicity in nude mice. Reduced TbetaRII expression in MCF-7L cells is caused by decreased TbetaRII promoter activity in this cell line. Our previous studies using 5' deletion constructs of this promoter revealed that MCF-7L cells were unable to support transcription of the minimal promoter (-47 to +2) to the same levels as the MCF-7E cells. This region of the promoter contains an Sp1 element at position -25 from the major transcription start site. In this study, we investigated the role of Sp1 in TbetaRII transcription. Mutation of the Sp1 site resulted in decreased transcription of TbetaRII in MCF-7E and MCF-7L cells, indicating that this site played a role in transcription of this promoter. Gel shift assays using the proximal Sp1 site from the TbetaRII promoter showed enhanced DNA:protein complex formation with nuclear proteins isolated from MCF-7E cells compared with MCF-7L cells. Supershift analysis identified this binding activity as Sp1. Western blot analysis of Sp1 levels demonstrated that MCF-7E cells contain increased Sp1 protein compared with MCF-7L cells, paralleling the increased binding activity. Differential Sp1 activity was also demonstrated by higher levels of transcription of an Sp1-dependent insulin-like growth factor II promoter construct in MCF-7E cells compared with MCF-7L cells. Co-transfection of an Sp1 expression vector with a TbetaRII promoter construct in MCF-7L cells induced the expression from the promoter-CAT constructs and resulted in an increase of endogenous TbetaRII protein levels. These results demonstrate that the transcriptional repression of TbetaRII in MCF-7L cells is caused, in part, by lower Sp1 levels.

Progression of MCF-7 cells from early passage (MCF-7E, <200 passage) to late passage (MCF-7L, >500 passage) correlates with a loss of sensitivity to exogenous TGF␤1. We have previously shown that loss of TGF␤ sensitivity is due to decreased expression of the transforming growth factor receptor type II (T␤RII) and is associated with increased tumorigenicity in nude mice. Reduced T␤RII expression in MCF-7L cells is caused by decreased T␤RII promoter activity in this cell line. Our previous studies using 5 deletion constructs of this promoter revealed that MCF-7L cells were unable to support transcription of the minimal promoter (؊47 to ؉2) to the same levels as the MCF-7E cells. This region of the promoter contains an Sp1 element at position ؊25 from the major transcription start site. In this study, we investigated the role of Sp1 in T␤RII transcription. Mutation of the Sp1 site resulted in decreased transcription of T␤RII in MCF-7E and MCF-7L cells, indicating that this site played a role in transcription of this promoter. Gel shift assays using the proximal Sp1 site from the T␤RII promoter showed enhanced DNA:protein complex formation with nuclear proteins isolated from MCF-7E cells compared with MCF-7L cells. Supershift analysis identified this binding activity as Sp1. Western blot analysis of Sp1 levels demonstrated that MCF-7E cells contain increased Sp1 protein compared with MCF-7L cells, paralleling the increased binding activity. Differential Sp1 activity was also demonstrated by higher levels of transcription of an Sp1-dependent insulin-like growth factor II promoter construct in MCF-7E cells compared with MCF-7L cells. Co-transfection of an Sp1 expression vector with a T␤RII promoter construct in MCF-7L cells induced the expression from the promoter-CAT constructs and resulted in an increase of endogenous T␤RII protein levels. These results demonstrate that the transcriptional repression of T␤RII in MCF-7L cells is caused, in part, by lower Sp1 levels.
Transforming growth factor ␤ (TGF␤) 1 is a prototypical member of a superfamily of multifunctional cytokines. Activities of TGF␤ include the regulation of cell adhesion through modulation of extracellular matrix composition, guidance of the morphogenic events of embryogenesis, immunosuppression, and the lengthening or arresting of cells at the G 1 phase of the cell cycle (1,2). Of special interest is TGF␤'s function as a growth inhibitor of epithelial cells. TGF␤ elicits its effects by binding to specific cell surface receptors. Three major types of receptors, denoted receptor type I, type II, and type III have been identified and cloned (3)(4)(5). Type III receptor (T␤RIII), alternately known as betaglycan, is a large 280 -330-kDa transmembrane proteoglycan with no known signaling motif. T␤RIII is thought to play a role in presenting the ligand to the other receptors (6). Receptor type I (T␤RI), a 50 -60-kDa glycoprotein, and receptor type II (T␤RII), a 75-85-kDa glycoprotein, belong to a superfamily of receptor serine/threonine kinases. The presence of both T␤RI and T␤RII is necessary to effect a TGF␤ response, and both kinase activities must be functional for proper signal transduction (7)(8)(9)(10). Therefore, cells defective in one or the other receptor are refractory to the effects of TGF␤ and escape its autocrine-negative effects on growth.
The loss of TGF␤ sensitivity has been correlated to tumor progression. This loss has been most often associated with a loss of T␤RII expression (11). Replication error-positive colorectal cancer cells show a mutation in T␤RII coding region, which results in message instability and loss of protein (12)(13)(14). Deletion of the T␤RII gene, loss of T␤RII message, and expression of a truncated T␤RII message have also been observed in other cancer cell lines (15)(16)(17)(18)(19)(20).
The human breast cancer MCF-7 cells have no detectable cell surface T␤RII and are refractory to the effects of TGF␤ (18,19). The re-expression of T␤RII in these cells restores their sensitivity to the growth inhibitory effects of TGF␤ (19). In addition, the transfectants showed decreased anchorage-independent growth, and a decreased tumorigenicity in nude mice compared with vector-transfected cells. These results suggest that T␤RII behaves as a tumor suppressor in the MCF-7 cells.
Due to the significant role of T␤RII in determining tumorigenicity, it is likely that defects of T␤RII expression can lead to malignant progression. MCF-7 cells represent a well established model system for studying the biology of breast cancer cells. Our earlier work using MCF-7 early passage (MCF-7E) and MCF-7 late passage (MCF-7L) cells demonstrated that MCF-7L cells' resistance to exogenous TGF␤ correlated with the low, or undetectable level of T␤RII in this cell line (18).
These cells were tumorigenic in athymic nude mice, whereas the TGF␤-sensitive MCF-7E cells were not. Thus, the MCF-7 system can be viewed as a model system for progression. The early passage MCF-7 cells are nontumorigenic and have changed through time in culture to become a more tumorigenic cell line. This progression is accompanied by the loss of T␤RII. We have shown that the differential levels of T␤RII in the MCF-7E versus MCF-7L cells were due to the difference of T␤RII promoter activities in MCF-7E and MCF-7L cells (18). The minimal promoter, as well as other promoter constructs studied, were expressed at very low levels in the late passage cells compared with the early passage cells.
In this study, we use this MCF-7 system to understand the underlying mechanism of T␤RII transcriptional regulation of the minimal T␤RII promoter. We find that the transcription of the T␤RII minimal promoter in the MCF-7E and MCF-7L system is dependent on an Sp1 element at Ϫ25. The Sp1 DNA binding activity is decreased in MCF-7L cells compared with the MCF-7E cells. This decreased binding activity correlates with decreased levels of Sp1 protein. In addition, transfection with an Sp1 expression plasmid results in the induced transcription of a co-transfected T␤RII promoter construct and increased endogenous T␤RII protein levels.
Transient Transfection and CAT Assay-MCF-7E and MCF-7L cells were transiently transfected with 30 g of RII promoter-CAT constructs and 10 g of ␤-galactosidase by electroporation using a Bio-Rad Gene Pulser at 250 V and 960 microfarads. The electroporated cells were plated onto 100-mm culture dishes. Sp1 co-transfection experiments were conducted using Transfast (Promega) at a ratio of 1.5:1 lipid to DNA. Cell lysates were prepared 48 h following transfection by three cycles of freeze-thaw in 0.25 M Tris, pH 8. Transfection efficiencies of either condition was measured by overnight staining with 5-bromo-4chloro-3-indoyl ␤-D-galactopyranoside. The percentage of blue colonies for both electroporation and Transfast reagent was approximately 15-20%. Protein assays (Bio-Rad) and ␤-galactosidase activity assays were performed to normalize the amounts of extracts to use for the CAT assays. Lysates representing equal transfection equivalents were incubated with [ 14 C]chloramphenicol (0.25 Ci) and acetyl coenzyme A (40 mM) overnight at 37°C, extracted with ethyl acetate, and separated by thin layer chromatography. The radioactivity in the acetylated products were directly quantified using the Ambis system (San Diego, CA), or were analyzed by autoradiography.
Nuclear Protein Extraction-MCF-7E and MCF-7L cells were plated onto 100-mm dishes and grown to confluence. Nuclear proteins of MCF-7E and MCF-7L cells were isolated at 4°C as described with minor modifications (23). Briefly, cells were harvested by scraping, washed in cold phosphate-buffered saline, and incubated in 2 packed cell volumes of Buffer A (10 mM HEPES, pH 8.0, 1.5 mM MgCl 2 , 10 mM KC1, 0.5 mM dithiothreitol, 0.5 mM phenylmethanesulfonyl fluoride, 1 g/ml amounts of both leupeptin and aprotinin, and 0.1% Nonidet P-40) for 5 min at 4°C. The nuclei were collected by microcentrifugation for 5 min at 3,000 rpm, rinsed once in buffer A, then resuspended in twothirds packed cell volume of buffer C (20 mM HEPES, pH 2.9, 1.5 mM MgCl 2 , 420 mM NaCl, 0.2 mM EDTA, 0.5 mM phenylmethanesulfonyl fluoride, 1.0 mM dithiothreitol, and 1.0 g/ml amounts of both leupeptin and aprotinin). Nuclei were incubated with rotation at 4°C for 30 min and clarified by microcentrifugation for 5 min at 3,000 rpm. The resulting supernatants were dialyzed versus buffer D (20 mM HEPES, pH 2.9, 100 mM KCI, 0.2 mM EDTA, 20% glycerol, 1 mM dithiothreitol, 0.5 mM phenylmethanesulfonyl fluoride, and 1 g/ml amounts of both leupeptin and aprotinin) and frozen in aliquots at Ϫ70°C.
Gel Shift Assays-Gel shift assays were conducted as described previously (18,24). Double-stranded oligonucleotides representing the two Sp1 sites on RII promoter (Ϫ25 sense strand, GAGAAGGCTCTTC-CGGGCGGAGAGAGGTCCTG; Ϫ143 sense strand, AGTGGTGTGG-GAGGGCGGTGAGGGGCAGCAGCT) and a mutant Sp1 oligonucleotide (sense strand, GAGAAGGCTCTTCCGGGCccAGAGAGGTCCTG) were prepared (Genosys). Sp1 consensus oligonucleotide and recombinant human Sp1 protein were obtained from Promega. Oligonucleotides were annealed to their corresponding strand and labeled by end labeling using [␥-32 P]ATP (950 Ci, 3,000 Ci/mol) and T4 polynucleotide kinase (Promega). The labeled oligonucleotides were then purified by using probe quant G-50 microcolumns (Amersham Pharmacia Biotech). Binding reactions were conducted in a 20-l volume that contained 10 l of nuclear protein in a buffer containing 10 mM Tris, pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, and 5% glycerol, 2 g of poly(dI-dC), and 20,000 cpm of labeled oligonucleotides. Reactions were incubated at room temperature for 30 min. Competition reactions were performed by adding 100-fold excess, unlabeled double-stranded oligonucleotide to the reaction mixture, prior to the addition of radiolabeled probe. Reactions were separated by electrophoresis on a 4% polyacrylamide gel at 150 V for 2 h in a 100 mM Tris borate-EDTA buffer at 4°C. Gels were dried and analyzed by autoradiography. For the supershift assay, the 32 P-labeled oligonucleotide and nuclear protein reaction mixture were incubated for an additional 30 min with 2 g of Sp1-specific peptide antibody (Santa Cruz) at room temperature, prior to electrophoresis and autoradiography.
Western Blot Analysis-Thirty micrograms of total cellular protein or five micrograms of nuclear protein were separated by electrophoresis on 8% SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Amersham Pharmacia Biotech). The membrane was blocked in blocking buffer (920 mM Tris-HC1, pH 2.5, 500 mM NaCl, 0.05% Tween 20, 5% milk) at room temperature for 1 h, then the membrane was subjected to anti-Sp1 polyclonal antibody (Santa Cruz) at 4°C overnight, followed by 1 h incubation of secondary antibody at room temperature for 1 h. Proteins were visualized by using the enhanced chemiluminescence (ECL) detection system (Amersham Pharmacia Biotech).

RESULTS
TGF␤ RII Promoter Activity-Our previous results demonstrated that MCF-7E cells and MCF-7L cells represent TGF␤sensitive and -insensitive strains, respectively, of this well studied cell line (18). This differential sensitivity corresponded to a differential expression of T␤RII due to differences in T␤RII transcriptional activity. Use of T␤RII promoter-CAT constructs revealed that the differential expression in MCF-7L and MCF-7E cells occurred in all deletion constructs including the minimal promoter. This construct, Ϫ47T␤RII-CAT, contains an Sp1 site at Ϫ25 and the initiator sequence. It has been shown that mutation of this Sp1 site within the T␤RII minimal promoter decreased T␤RII promoter activity in the high T␤RIIexpressing HepG2 cells (22). We questioned whether Sp1 also played a role in regulating T␤RII transcription in the MCF-7E and MCF-7L cells. A T␤RII-CAT construct in which the Sp1 site at Ϫ25 was mutated was used in our transient transfection studies and compared its activity to that of the wild type Ϫ47T␤RII-CAT. Consistent with our previous results, MCF-7E cells displayed higher CAT activity when transfected with the wild-type construct than similarly transfected MCF-7L cells (Fig. 1). Transfection with the Sp1 mutated construct resulted in a decreased promoter activity in both MCF-7E and MCF-7L cells. The level of transcription of the Sp1 mutated T␤RII construct was essentially identical in both the cells lines, indicating that the Sp1 was responsible for the differential expression. Notably, the reduction of RII promoter activity in MCF-7E cells is more dramatic than in MCF-7L cells since its original activity is higher. This result demonstrates that the Sp1 element is an important determinant in the activity of the T␤RII promoter activity in both MCF-7E and MCF-7L cells.
Sp-1 Binding Activities Are Differentially Expressed in MCF-7E and MCF-7L Cells-Since the Sp1 element was found to participate in the regulation of the T␤RII promoter, the Sp1 element in this promoter should bind Sp1. Gel shift assays were performed to address whether the MCF-7 cells contained binding activities for these regions. Results using nuclear protein of MCF-7E and MCF-7L cells and a probe corresponding to the Sp1 element at position Ϫ25, show a DNA:protein complex with a mobility identical to that of the recombinant human Sp1-bound probe (Fig. 2). Nuclear protein from MCF-7E cells contains greater amounts of these binding activities than the nuclear protein from MCF-7L cells. These bands represent specific binding of protein to the Sp1 sequence element, since complex formation was diminished by the addition of 100 M excess of unlabeled identical competitor, but not by addition of the identical oligonucleotide in which the Sp1 site was mutated. As an additional control, the mutant oligonucleotide was used as the probe. This resulted in no shifted species using either MCF-7E or MCF-7L extracts, indicating that the protein binding was dependent on the intact SP1 sequence. Gel shift assays using the labeled Sp1 consensus sequence also resulted in increased complex formation using MCF-7E-derived nuclear extract compared with MCF-7L-derived extracts (Fig. 2B), identical to our results with the Sp1 sequence from the T␤RII minimal promoter.
To confirm that the enhanced protein-DNA complex in MCF-7E cells contains Sp1, "supershift" assays were carried out using an antibody specific for Sp1. Addition of this antibody in the binding reactions decreased the mobility of the DNA: protein complex, indicating that this complex contained Sp1 (Fig. 3). Nuclear protein from both MCF-7E and MCF-7L cells contained Sp1 protein; however, MCF-7E cells contain higher levels than the MCF-7L cells. Similar results were obtained using the Sp1 element found at Ϫ143 of the T␤RII promoter as a probe for the gel shift assays (data not shown). These results suggest that the higher T␤RII minimal promoter activity in MCF-7E cells correlates with enhanced Sp1 activity in these cells.
Sp1-dependent Promoter (Insulin-like Growth Factor II (IGF-II)) Activity-MCF-7E cells differentially regulate the T␤RII promoter construct containing the Sp1 binding site and contain increased levels of Sp1 binding activity compared with MCF-7L cells. We reasoned that the differential levels of the Sp1 binding activity in these two cell lines may result in the differential regulation of other Sp1-dependent promoters. A promoter con-struct representing the Ϫ58 to ϩ124 of the IGF-II Sp1 gene was used in transient transfection of the MCF-7E and MCF-7L cells. This promoter-CAT fusion construct contains two Sp1 sites and a TATA box sequence. The activity of this promoter has been shown to be Sp1-dependent (25). MCF-7E cells supported a dramatically higher IGF-II promoter activity than MCF-7L cells consistent with our results with the T␤RII promoter activity (Fig. 4).
Differential Expression Levels of Sp1-Increased Sp1 binding activity could result from modification of an existing form of the protein rendering it more active or from increased protein levels. We performed Western blot analyses to investigate whether enhanced Sp1 DNA binding activity is paralleled by increased levels of Sp1 in MCF-7E cells compared with the MCF-7L cells. Proteins from MCF-7E and MCF-7L nuclei were separated by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose for the immunoblot assay. The immunoblot results showed significant levels of the 95-and 105-kDa forms of Sp1 in the nuclear extracts from MCF-7E cells compared with lower levels from the MCF-7L cells (Fig.  5A). In contrast, no differences for c-Jun levels are seen in MCF-7E and MCF-7L cells. These results suggest that enhanced Sp1 binding activity is a result of increased levels of Sp1 protein in the nucleus.
One mechanism for increasing the levels of a specific protein in the nucleus is to import existing protein from the cytoplasm. If this were the case, then it is possible that the MCF-7E and MCF-7L cells contain equivalent amounts of Sp1 protein, but it is preferentially located in the nucleus of the MCF-7E cell. To test this, we repeated our immunoblot assays using whole cell lysates from MCF-7E and MCF-7L cells. Immunoblot assays reveal that the total Sp1 protein is higher in the MCF-7E then the MCF-7L cells, indicating that the regulation of Sp1 activity is at the level of protein rather than the level of nuclear localization (Fig. 5B).
Re-expression of Sp1 in MCF-7L Cells Increases the Tran- scriptional Activity of the T␤RII-CAT Construct and Leads to the Re-expression of Endogenous T␤RII-The differential level of Sp1 in MCF-7L and MCF-7E cells appears to contribute to the differential regulation of the T␤RII minimal promoter and other Sp1-dependent constructs. We examined whether re-expression of Sp1 in the MCF-7L cells would lead to increased expression of the T␤RII promoter. We obtained an Sp1-expressing plasmid, CMVSp1, from the laboratory of Dr. Sophia Tsai (Baylor College of Medicine, Houston, TX) (26). MCF-7L cells were co-transfected with CMVSp1 and the T␤RII constructs used above. In addition, the construct Ϫ274T␤RII-CAT was also used. This construct represents 274 base pairs of the proximal portion of the T␤RII promoter, which contains a putative AP1/CREB site at Ϫ195, and two Sp1 sites, one at position Ϫ143 and the other at Ϫ25. This construct had been previously shown to support high levels of transcription of a fused reporter gene (18,22). Results show that co-transfection with the Sp1 expression plasmid supports a higher transcriptional activity of the T␤RII-CAT constructs compared with cells co-transfected with T␤RII-CAT and the empty vector (Fig. 6). Co-transfection of CMVSP1 had no effect on the transcription of the Sp1 mutant T␤RII-CAT, confirming the necessity of the Sp1 binding site.
The increased Sp1 levels following transfection of the CMV-SP1 could support higher transcription of the T␤RII-CAT constructs, we hypothesized that the endogenous gene would be expressed as well. Increasing amounts of CMVSP1 was transiently transfected into MCF-7L cells, and cell lysates were prepared for Western blot analysis 48 h following the transfection. As expected, increasing amounts of SP1 protein were detected in the cells correlating with increasing amounts of CMVSp1 transfected (Fig. 7). In addition, increased amounts of T␤RII was also detected in the CMVSp1-transfected cells, indicating that Sp1 expression positively effects endogenous T␤RII expression. DISCUSSION Direct involvement of T␤RII in TGF␤ signal transduction suggests that loss of T␤RII expression contributes to the loss of TGF␤ response resulting in a growth advantage. Estrogen receptor-containing breast cancer cells (ERϩ) acquire resistance to TGF␤ due to lack or insufficient expression of T␤RII (18 -20). Our earlier work demonstrated that progression from MCF-7 early passage to late passage correlates with a decreased RII expression and increased tumorigenicity. Re-expression RII in MCF-7L cells restored the cell response to exogenous TGF␤ and reversed the tumorigenicity (19). This correlation between T␤RII expression and malignant progression demonstrates the significant tumor suppressing function of T␤RII in ERϩ breast cancers.
Our previous work demonstrated that the loss of T␤RII expression in MCF-7L cells is due to decreased activity of the T␤RII promoter (18). In this study, MCF-7E and MCF-7L cells were used to determine the mechanism of T␤RII transcriptional repression in MCF-7L cells. The results from transient transfecton of T␤RII promoter-CAT constructs which contained mutated Sp1 site showed Sp1 was responsible for the differential regulation of the T␤RII promoter in MCF-7E and MCF-7L cells. Mobility shift assays and supershift assays, utilizing oligonucleotides representing the T␤RII Sp1 DNA binding element and nuclear protein from MCF-7E or MCF-7L cells, indicated that MCF-7E contained enhanced Sp1 activity. This enhanced Sp1 binding activity was found to be due to higher levels of Sp1 protein in MCF-7E cells. These results demonstrate that suboptimal levels of Sp1 activity in MCF-7L cells contribute to the reduced T␤RII expression, leading to resistance to TGF␤ and lower tumorigenicity of MCF-7L cells.
Transfection of the MCF-7L cells with an Sp1 expression plasmid resulted in increased expression of the co-transfected T␤RII-CAT constructs. One of the constructs used in this assay contained two Sp1 sites as well as a AP1/CRE site necessary for maximal T␤RII transcription. While this construct is also transcribed to levels lower in MCF-7L cells compared with early cells, re-expression of the Sp1 site up-regulated expression of this larger promoter construct. This result suggests that levels of Sp1 can affect the transcription of larger segments of the promoter in addition to its effects on an Sp1-dependent minimal promoter, which contains no other transcriptional elements. Indeed, the expression of the endogenous T␤RII was also increased. Therefore, Sp1 may be a possible target for manipulation of T␤RII expression in T␤RII-deficient cells. In support of this suggestion, it had been shown that the DNA demethylating reagent 5-aza-2Ј-deoxycytidine results in an induction of T␤RII expression in MCF-7L breast cancer cells and other ERϩ breast cancer cell lines, leading to the re-establishment of the TGF␤ response. This restoration of T␤RII expression occurs through the stabilization of Sp1 protein and not through the demethylation of the T␤RII promoter (24).
In addition to the SP1 motifs in the T␤RII promoter, the region from Ϫ271 to Ϫ137 is necessary for maximal transcription of T␤RII in MCF-7 cells (18). This region has been shown to contain an AP1/CRE site (22), which is an important determinant during the differentiation of EC cells (27). Of related interest is the fact that some breast cancer cells have been shown to be deficient in the activity of AP1 (28). We are currently investigating the contribution of the AP1 site on the T␤RII promoter in MFC-7E and MCF-7L cells.
The increased Sp1 levels in MCF-7E nuclei correlated with increased transcription of the T␤RII gene in these cells compared with MCF-7L cells. This higher expression of Sp1 may be caused by increased transcription of the Sp1 gene, or by translational or post-translational mechanisms. Post-translational modifications of Sp1 can influence its activity or stability. Sp1 has been shown to be phosphorylated and O-glycosylated (29,30). The addition of N-acetylglucosamine residues has been shown to influence the transport of Sp1 into the nucleus, and reduced O-glycosylation of Sp1 has been shown to be associated with proteasome degradation during glucose starvation (31). MCF-7L cells treated with 5-aza-2Ј-deoxycytidine show an increased half-life of Sp1 nuclear protein (24). Current work is under way to investigate whether changes in Sp1 stability is responsible for the differences in Sp1 levels in MCF-7E and MCF-7L cells or whether other mechanisms result in the differential Sp1 levels in these cells.
Sp1 is necessary for the transcription of TATA-less genes through its interaction with the TFIID to direct the start site of transcription (32). In addition, Sp1 regulates the transcription of a number of TATA-containing genes, often coordinating with other cellular factors to regulate gene-specific transcription. A model has been proposed wherein the functional interaction between pRb and Sp1 in vivo results in the "superactivation" of Sp1-mediated transcription (33). Sp1 has also been shown to interact with Ets family of transcription factors (34,35) and the estrogen receptor (36,37). The specificity of the genes regulated by Sp1 is most probably determined by co-factors or activators that interact with Sp1 or the combinatorial effects of other regulating elements. A member of the Ets family of transcription factors has been found to bind to a region of the T␤RII promoter downstream from the start of transcription (ϩ13 to ϩ24) (38). Our constructs did not contain this site; therefore, it is not known what the interaction of this member of the Ets family has on T␤RII transcription in our system. Other putative Ets-like sites do exist in the minimal promoter region. These sites may participate in the transcription of this promoter in MCF-7E and MCF-7L cells.  7. Transfection of the CMVSp1 leads to increased expression of the endogenous T␤RII. MCF-7 late cells were transiently transfected with the CMVSP1 expression plasmid. Forty-eight hours following transfection, cell lysates were prepared for Western blot analysis. Fifty micrograms of total cellular protein were separated by electrophoresis and transferred to nitrocellulose membranes. Western blot analysis using Sp1 antibody and the T␤RII antibody (Santa Cruz) was performed. Proteins were visualized by ECL (Amersham Pharmacia Biotech).
Of interest is the fact that the promoters of other TGF␤ family members contain Sp1 sites. The TGF␤1 and T␤RI genes have been shown to contain functionally significant Sp1 motifs in their promoters (39,40). Since Sp1 is a general transcription factor, it is likely that other factors are involved to direct the specificity of its actions to certain promoters. We have previously shown that the RNA levels for TGF␤1 and T␤RI are similar in both the MCF-7L and MCF-7E cells (18). Therefore, other factors must participate in specifying transcription of these genes when Sp1 is deficient. It is possible that the transcription of T␤RII may be more dependent on Sp1 than the other family members.
Several TGF␤-responsive genes contain Sp1 motifs. TGF␤ itself signals through various cytoplasmic signal transducers, including those of the Smad family (41). The p21/Waf/Cip1 promoter has been shown to be regulated by TGF␤ through Sp1 via an interaction with Smad proteins (42). T␤RII has been shown in some cases to be induced by its own ligand (43,44). Since Sp1 is an important determinant in T␤RII transcription, it is possible that T␤RII is regulated in an autocrine fashion via Sp1 and Smad interactions.
Our earlier work demonstrated that the reduced autocrinenegative regulation by TGF␤ is a result of decreased T␤RII expression in MCF-7L cells and is associated with increased tumorigenicity. Cells that express T␤RII, the early passage MCF-7 and T␤RII transfectants, show a reduced tumorigenicity, suggesting that this loss of this receptor is involved in tumor progression. Our understanding of the mechanisms that control T␤RII expression could lead to therapeutics that target the expression of this gene. Reduced expression of T␤RII in the MCF-7L cells is due to suboptimal Sp1 levels. Manipulation of Sp1 activity may be a possible mechanism by which to induce T␤RII expression, leading to restored sensitivity to the negative growth effects of TGF␤.