| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 47, 46378-46386, November 21, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶¶¶
||||
From the
Departments of Biochemistry and ||Pediatrics and Cancer Biology and Vanderbilt-Ingram Cancer Center, Vanderbilt University School of Medicine, Nashville, Tennessee 37232, **Cancer Research UK Oncology Unit, University of Southampton School of Medicine, Southampton General Hospital, Southampton SO16 6YD, United Kingdom, and Department of Basic Medical Sciences, Purdue University, West Lafayette, Indiana 47907
Received for publication, May 17, 2003 , and in revised form, September 2, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
or the tyrosine kinase ABL, respectively (1, 2). The Pointed domain located within the N terminus of TEL induces dimerization and subsequent activation of the tyrosine kinases (3–5). In these cases, aberrant regulation of the kinase activity of the fusion proteins is a contributing factor to leukemogenesis. On the other hand, fusions of TEL with transcription factors can either retain or eliminate the DNA binding activity of TEL. The t(12;22) fuses the transcription factor MN1 to the C-terminal DNA binding domain of TEL, which converts TEL from a repressor to an activator (6, 7). Alternatively, in 25% of pediatric acute lymphoid leukemias the t(12;21) fuses the N-terminal Pointed domain and repressor domains of TEL to nearly all of the AML11 transcription factor (8, 9). The TEL/AML1 fusion protein loses the DNA binding activity of TEL and functions as a constitutive inhibitor of AML1-dependent gene targets (10, 11). In addition, in 90% of patients with t(12;21) the second allele of TEL is deleted (12–14). Loss of heterozygosity of the TEL allele is also observed in other leukemias and some solid tumors (15–18). This loss of heterozygosity suggests a role for TEL as a tumor suppressor.
Over 20 mammalian ETS family members have been identified, and they participate in the regulation of cellular adhesion, proliferation, apoptosis, and differentiation (19–22). TEL is a widely expressed and essential gene. Targeted disruptions of both Tel alleles in mice resulted in early embryonic lethality, suggesting a role for TEL in multiple developmental pathways (23). These mice have a severe defect in the developing vascular network of the yolk sac, and in chimeric mice the TEL-null embryonic cells failed to contribute to adult bone marrow (23, 24). TEL heterodimerizes with a closely related family member, TEL2 or TEL B, but in contrast to TEL, the expression of TEL2 is variable among tissues (25–27). Therefore, TEL is hypothesized to have a non-redundant role in either migration or homing of hematopoietic progenitors.
All Ets family members share a homologous DNA binding domain (the ets domain) and bind to heterogeneous sequences centered around a core GGA sequence. TEL preferentially binds to the sequence T(G/T)(A/C)GGAAGT (28) and functions as a transcriptional repressor via interactions with the mSin3A, N-CoR, and SMRT corepressors, as well as histone deacetylase 3 (29–32). The biological role of TEL in normal cell growth is still unclear, but its repressor activity is required for TEL to inhibit the growth of Ras-transformed NIH 3T3 cells and to induce cellular aggregation (33–36). One target gene directly repressed via TEL is the matrix metalloproteinase 3 gene, Stromelysin-1 (36). TEL-mediated repression of Stromelysin-1 may affect cell-cell and cell-extracellular matrix (ECM) interactions leading to inhibition of cell growth.
The forced expression of TEL has growth inhibitory activity, and we observed increased apoptosis of TEL-expressing NIH 3T3 cells that was exacerbated when the cells were cultured in low serum concentrations. A key regulatory point in the apoptotic pathway in response to serum deprivation is the ratio of pro- and anti-apoptotic BCL-2 family members. One anti-apoptotic member, Bcl-XL, contains multiple potential ets factor binding sites (EBS) within its transcriptional control regions (37). Therefore, we investigated TEL-mediated regulation of Bcl-XL expression and its effect on cell survival.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ETS) used have been described elsewhere (10, 38). The pBabe retroviral constructs TEL wild-type (WT) and TELETS have been described elsewhere (36). A 1-kb EcoRI fragment containing the mouse Bcl-XL cDNA was isolated from pW3L-Bcl-X and was subcloned into the EcoRI site of the murine stem cell virus internal ribosome entry site green fluorescent protein retroviral vector. The proper orientation was confirmed with a restriction digest. The –2361-luc reporter gene plasmid was described previously (39). The –876-luc reporter gene plasmid was constructed by PCR amplification of the Bcl-X promoter region from Jurkat genomic DNA with the following primers: 5'-GCTAGCTTGAACCCCATTGAGAAGTCCC-3' and 5'-AAGCTTTCTCGTCTCTGGTTAGTGATTC-3' (–876 and –61 relative to the start ATG in Exon 2). The primers had flanking restriction enzyme sites, NheI and HindIII (shown in bold). The PCR fragment was subcloned into the pGL2 basic luciferase plasmid (Promega, Madison, WI). The reporter plasmid –793-luc, –565-luc, and –436-luc were made by PCR amplification of the –876-luc plasmid with the following 5' primers: 5'-CCCGCTCGAGCAAAAACCAACTAAATCCATAC-3', 5'-CAATCTCGAGTGCCGGGTCGCATGATCC-3', or 5'-CAATCTCGAGGCACCTGCCTGCCTTTGC-3' (respectively) and the 3' primer 5'-GATCAAGCTTCTAAGATCCAAAGCCAAGAT-3' (–84 relative to the start ATG of Exon 2). The primers had flanking restriction enzyme sites, XhoI and HindIII (shown in bold), and the PCR products were digested and then ligated into pGL2 basic luciferase plasmid. These plasmids were verified by sequencing reactions. The RSV Renilla plasmid (Promega, Madison, WI) was used as a control for transfection efficiency.
Cell Culture and Infections—NIH 3T3 cells were maintained in DMEM-CS (DMEM (BioWhittaker, Walkersville, MD) containing 10% calf bovine serum (Hyclone, Logan, UT) and 2 mM L-glutamine). Retroviral packaging 
NX cells were maintained in DMEM-FBS (DMEM containing 10% fetal bovine serum (Sigma) and 2 mM L-glutamine). Stably infected 3T3 cells were generated as described elsewhere (36) with the following modification: 100-cm2 plates containing 60–70% confluent cells were transfected with 5 µg of DNA and 8 µl of LipofectAMINE 2000 (Invitrogen). At least three independent batches of stably infected NIH 3T3 cells were used for analysis, and similar results were obtained. Heterozygous BCL-XL mice were mated to generate BCL-XL wild-type and knockout littermate embryos (40). Pregnant mothers were sacrificed at E12, and mouse embryonic fibroblasts were plated in DMEM supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 units of penicillin/streptomycin per ml, and non-essential amino acids (BioWhittaker, Walkersville, MD). Mouse embryonic fibroblasts were cultured for 10–12 passages before being immortalized on a standard 3T3 protocol and switched to DMEM-CS (41).
Transfections and Transcription Assays—NIH 3T3 were transfected as follows: 2 x 105 cells were plated per well in 6-well plates. The cells were transfected the next day with 6 µl of Superfect reagent (Qiagen, Valencia, CA), 1 µg of reporter plasmid, 200 µg of internal control plasmid, 50 ng of either pCMV5 empty vector or TEL (WT, ETS, or K99R) cDNA pCMV5 plasmid, and 250 ng of pCMV5. Cells were harvested 44 to 48 h post-transfection. The cells were washed with PBS and lysed in 150 µl of passive lysis buffer (Promega, Madison, WI). Luciferase activities were determined from 20-µl aliquots of lysate according to the manufacturer's instructions for the Promega Dual Luciferase Assay kit. The firefly luciferase assays were then normalized with respect to Renilla luciferase activity. Experiments were performed in triplicate and repeated to confirm results.
Cell Viability Assays, RNA Analysis, and Western Blots—Stably infected NIH 3T3 cells or BCL-XL wild-type, heterozygous, or null 3T3 cells were plated at a density of 5 x 104 cells per well into 6-well plates in triplicate. Twelve h after plating the cells were rinsed once with PBS, and DMEM containing 0.1% serum was added. Cells were harvested at the times indicated in the figure legends. The numbers of live and dead cells were determined using trypan blue exclusion. The 6-well plates were pre-treated with either 0.05 M HCL or 2 µg/cm2 collagen type IV (BD Biosciences) for 4 h and rinsed extensively with PBS prior to the plating of NIH 3T3 cells. Total RNA was extracted from 107 stably infected NIH 3T3 cells using the Promega RNA Mini kit. Ten µgofRNA was separated by electrophoresis on a 2.5 M formaldehyde-1% agarose gel and then transferred to Hybond N+ membrane (Amersham Biosciences) with 20x SSC and cross-linked using the Stratagene 1800 UV cross-linker. Blots were hybridized with the indicated [
-32P]dATP-labeled cDNA probes (Random Priming Kit; Stratagene, La Jolla, CA) according to the Rapid-hyb (Amersham Biosciences) protocol. Blots were then washed in 2x SSC/0.1% SDS at room temperature for 20 min, followed by two washes in 0.1x SSC/0.1% SDS at 65° for 15 min. Blots were exposed to film for visualization. Cellular extracts were prepared by trypsinizing cells from 100-mm dishes, washing two times with PBS, and then lysing with PBS containing 0.5% Triton, 0.5% SDS, 0.5% deoxycholate, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 8.5 µg/ml aprotinin. One hundred µg of protein were separated by SDS-PAGE and then transferred to polyvinylidene difluoride membrane (Millipore, Bedford, MA). The membrane was blocked in 5% milk-Tris-buffered saline with 0.2% Tween 20, probed with the following primary antibodies: actin (Sigma), Bcl-XL M125 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), GAPDH (Novus Biologicals, Littleton, CO), and mouse BCL-2 3F11 (BD Biosciences) and peroxidase-conjugated secondary antibodies anti-mouse IgG, anti-rabbit IgG (Sigma), or goat anti-hamster, followed by detections with enhanced chemiluminescence (Pierce).
Microscopy and Immunofluorescence—Stably infected NIH 3T3 cells were plated on coverslips in 10% serum containing DMEM overnight. The cells were then rinsed twice with PBS, and they were grown in either 10 or 0.1% serum-containing media for 24 h. The cells were fixed in 3.7% formaldehyde solution, extracted in 0.5% Triton X-100, and stained with 5 ng/ml Hoescht dye for 5 min. The images were visualized using a Zeiss Axiopot microscope using 40x 1.4 oil immersion objectives. The paxillin immunostaining was performed as described previously (14). Briefly, the cells were fixed in 3.7% formaldehyde solution, extracted in 0.5% Triton X-100, and stained. Focal adhesions were detected with anti-paxillin antibodies. Immunostaining was visualized with rhodamine-conjugated donkey anti-mouse (Chemicon, Temecula, CA) and fluorescein isothiocyanate-conjugated donkey anti-rabbit antibodies (Chemicon, Temecula, CA) using epi-fluorescence microscopy (Olympus BX60). Images were recorded onto T-Max 400 film (Eastman Kodak Co., Rochester, NY).
Chromatin Immunoprecipitation—The following procedure was modified from the Upstate ChIPs Assay protocol and is described briefly. NIH 3T3 cells were treated with 1% formaldehyde/PBS, lysed, and sonicated to generate fragments between 500 and 1000 bp. The extracts were immunoprecipitated with either a HA (Covance, Princeton, NJ) or an N-terminal TEL-specific antisera and washed extensively. The protein-DNA cross-linking was reversed, and the DNA was eluted. PCR was performed with the following primers: 5'-TCCCTTAGAACCCGGACTCAGACCTTCA-3' and 5'-GCTCCCGGTTGCTCTGAGACATTTT-3', which encompass –250 to +22 of the murine Bcl-X promoter (271-nt product); 5'-ACCCATACCTCCGGGAGAGTTCTCC-3' and 5'-ATCTATCTCCGGCGACAGCAAGCAGGTA-3', which encompass –759 to –402 of the ms Bcl-X promoter (357-bp product); 5'-TGTGGTTCTGCTGGGCTCACTCTTCAGT-3' and 5'-TTAGGGTCCCACCAACAAGACAGGCTCT-3', which flank Exon 3 of Bcl-XL (373 nt); and 5'-CACACGTGTATATTAGGTTGCCTGCCC-3' and 5'-CACGTAGCCTGTTCGTAGCAGACAGCC-3', located within the ETO-2 gene (279 nt).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To test whether TEL expression induced apoptosis, NIH 3T3 cells were infected with recombinant retroviruses expressing wild-type TEL or an inactive mutant of TEL that contains a small deletion within the DNA binding (ETS) domain (pBabe, TEL, or TELETS, respectively). The cells were fixed, and the chromosomal DNA was stained with Hoescht dye to detect the presence of micronuclei that are characteristic of the nuclear blebbing that occurs in cells undergoing apoptosis. TEL expression caused little or no increase in the number of apoptotic cells when grown in 10% serum (Fig. 1A, upper panel). However, these stably infected cultures were selected while growing in 10% serum, and therefore, these cells were selected for a level of TEL that allows for normal growth under these conditions. To circumvent this problem, we cultured the cells in media containing only 0.1% serum for 24 h. Under these conditions, the majority of the control cells arrest in the G1 phase of the cell cycle and survive for several days with a modest level of apoptosis (Fig. 1, A and B, and data not shown). By contrast, there was a marked increase in the percent of micronuclei present in cells expressing TEL (Fig. 1, A and B).
|
ETS; see Fig. 1C). TEL Impairs the Formation of Focal Adhesions—ETS family members are regulators of ECM remodeling, and TEL influences the expression of several ECM genes including fibronectin and the matrix metalloproteinase Stromelysin-1 (33, 36). In addition, the cellular aggregation phenotype observed upon TEL expression can be reverted when the cells are cultured on dishes coated with either fibronectin or the extracellular matrix component type IV collagen (33). Likewise, we have observed that laminin or collagen complements the cell adhesion defect in TEL-expressing Ras transformed cells (data not shown). Given that integrin-mediated signaling is a key apoptotic control pathway (43), we tested whether TEL expression affects the formation of focal adhesions. Monoclonal antibodies directed to the focal adhesion component paxillin (44) were used to detect focal adhesions by immunofluorescence. Paxillin was organized into multiple distinct focal adhesions in control cells (Fig. 2A, left panel). By contrast, in the TEL-expressing cells, paxillin was diffusely cytoplasmic with few distinct focal adhesions (Fig. 2A, right panel). Therefore, the TEL-mediated reduction in the number of focal adhesions occurs in non-transformed NIH 3T3 cells, which do not aggregate upon TEL expression (36), and is independent of the general changes in cellular morphology seen in TEL-expressing Ras-transformed NIH 3T3 cells (Fig. 2A, and data not shown).
|
Given the dramatic reduction in the number of focal adhesions in TEL-expressing cells, we tested whether the lack of focal adhesions was responsible for TEL-mediated apoptosis. TEL-expressing cells were plated on dishes coated with type IV collagen, and the growth of these cells was measured after culture in media containing 0.1% serum (Fig. 2B). Collagen failed to block the increased level of apoptosis observed in the TEL-expressing NIH 3T3 cells. Therefore, TEL-mediated apoptosis appears to be independent of its effects on the formation of focal adhesions or cellular aggregation.
TEL Represses the Bcl-XL Promoter—Given that TEL-induced apoptosis is independent of the state of cellular aggregation and integrin-mediated signaling, we assessed the effects of TEL expression on regulators of apoptosis. Of special interest was the anti-apoptotic Bcl-2 family member Bcl-XL. Stable overexpression of ETS2 in BAC1.2F5 macrophages up-regulates Bcl-XL expression and protects cells from apoptosis induced by removal of survival factors from the growth medium (37).
Transcriptional regulation of Bcl-X is complex with at least three different promoters driving tissue-specific transcription (39, 45). The translation initiation codon for Bcl-X is located within Exon 2, and transcription initiation can occur close to the start of Exon 2 or from two alternative noncoding Exons 1B or 1A (Fig. 3A). For the Exon 1A and Exon 2 promoters, most of the regulation was mapped to the region 900 bp upstream of the translational start site (45). NF
B, STATs and ETS factor binding sites identified in this region were demonstrated to regulate transactivation of Bcl-X (37, 46–48). Nine potential EBS were identified in the 800 nucleotides 5' to the ATG start site of human Bcl-X, and transient expression of ETS2 activated expression of a reporter gene linked to this region by 9-fold (37).
|
Given that the Exon 1A/Exon 2 promoter region contains nine potential EBS, as well as NFB and Stat5 binding sites (Fig. 3A), we focused on the TEL-mediated repression of this promoter region. TEL expression in NIH 3T3 cells consistently repressed transcription of the –876-luc promoter, as compared with the inactive TELETS (Fig. 3C), but in some of the experiments the level of repression was modest. TEL is regulated by nuclear export that is dependent on post-translational modification of lysine 99 with a small ubiquitin-like modifier molecule (38, 49). Thus, a TEL protein in which this lysine is mutated to arginine TEL(K99R) displays enhanced nuclear localization (38). Therefore, we transiently co-transfected NIH 3T3 cells with –876-luc and TEL(K99R) and found that whereas TEL repressed the Bcl-XL promoter-linked reporter gene 4-fold, TEL(K99R) repressed transcription even better (9-fold) (Fig. 3C). This increased level of repression was not because of differences in protein expression as all TEL proteins were expressed to equivalent levels (data not shown).
To determine whether any or all of the previously noted nine potential EBS were important for TEL-mediated repression of the Bcl-X promoter, we made a series of promoter deletions that contained all nine sites, four sites, or one EBS (–793-luc, –565-luc, or –436-luc respectively; see Fig. 3A). NIH 3T3 cells were transiently transfected with the indicated Bcl-X promoter deletion linked to luciferase and empty vector, TEL, or TELETS expression vectors (Fig. 3D). The –793-luc reporter gene had decreased basal activity relative to –876-luc, which may reflect the removal of the NFB activation site. The basal transcriptional activity further diminished with promoter deletions–565-luc and –436-luc. This decrease may reflect loss of potential activation sites or the loss of the upstream transcriptional start site (Exon 1A). The sequences within the –436-luc promoter contain a major transcriptional start site and maps to a minimal promoter region originally identified in the mouse bcl-X promoter (45). In all cases TEL, but not TEL ETS, significantly repressed the transcription from each Bcl-X promoter deletion-driven reporter gene.
Although the –436-luc promoter construct contains only one of the previously noted EBS (37), it contains an additional 16 GGA motifs, which comprises the core ETS factor consensus sequence. Rather than attempt to individually mutate each of the 17 GGA sequences within the minimal Bcl-X promoter, we used chromatin immunoprecipitation assays to determine whether TEL associates with Bcl-X in vivo. The 5' regulatory regions of the human and murine Bcl-X gene are highly homologous. All nine previously noted EBS within Exon 1A/Exon 2 promoter regions are conserved, and the murine sequence contains two additional potential EBS (37). NIH 3T3 cells express detectable amounts of endogenous TEL; therefore we used this system to determine whether TEL associates with the sequences just 5' to the transcriptional start site located within Exon 2 of Bcl-X. NIH 3T3 cells were cross-linked with formaldehyde, the cells were lysed, and the chromatin was sheared to yield genomic fragments between 500 and 1000 bp prior to immunoprecipitation with an antibody directed against the N terminus of TEL. PCR primers were designed to amplify the minimal Bcl-X promoter surrounding the EBS nearest to the Exon 2 transcriptional start site, to encompass the other eight EBS, as well as the promoter upstream of Exon 1B (see Fig. 3A and Fig. 4A). In principle, these primers could amplify TEL-associated genomic sequences within 1000 bp upstream or downstream of the distal primer sequence. Primers designed to detect an irrelevant gene (ETO2), as well as to detect a region 50-kbp downstream of the minimal promoter region (within Exon 3 of Bcl-X), were used as specificity controls (Fig. 4A). Amplification of the Bcl-X 1B promoter region was inconsistent, but in some experiments it was detected as slightly above background (data not shown). The lack of TEL-mediated repression of the –2361-luc promoter is consistent with little or no binding of TEL to the sequences upstream of Exon 1B. However, the regions containing EBSs 1–8, as well as EBS 9 upstream of the Exon 2, were specifically detected with the N-terminal TEL antiserum but not with anti-HA or anti-Gal4 antibodies (Fig. 4, B and C). The Bcl-XL promoter region with the EBS adjacent to Exon 2 (–142 nt relative to the translational start site) displayed the strongest association with TEL of the promoter regions analyzed, which is consistent with the ability of TEL to repress this promoter (see Fig. 3D and Fig. 4B). In addition, the PCR performed with samples immunoprecipitated with the N-terminal TEL antibody failed to amplify either the Exon 3 coding region of Bcl-XL or intronic sequences from ETO2 (Fig. 4, B and C).
|
ETS. The RNA was analyzed by Northern blot with a Bcl-XL-specific probe, as well as a GAPDH-specific probe as a loading control. Bcl-XL mRNA was significantly down-regulated in NIH 3T3 cells expressing TEL but not TELETS (Fig. 5A). This down-regulation of Bcl-XL mRNA was reflected in the levels of Bcl-XL protein expressed. The TEL-expressing NIH 3T3 cells contained 2–4-fold less Bcl-XL in comparison with controls (Fig. 5B). This decrease in Bcl-XL protein does not reflect an alteration in Bcl-X splicing given that we did not detect expression of Bcl-XS protein in the samples. This level of Bcl-XL protein down-regulation was similar whether the cells were grown in 10% serum, 0.1% serum, or on plates coated with type IV collagen (Fig. 5C, and data not shown). Furthermore, this down-regulation was specific to Bcl-XL, as we observed no alteration in the levels of other BCL2 family members, including Bcl-2 or Bax (data not shown). Thus, the change in the ratio of Bcl-XL to pro-apoptotic molecules may contribute to the apoptosis induced by TEL (Fig. 1).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The modest level of apoptosis that is triggered by the expression of oncogenes or tumor suppressors is often difficult to capture in the presence of growth factors, either because of their survival functions that suppress apoptosis or because of the rapid growth of cells not undergoing apoptosis. For instance, c-Myc is a potent trigger for apoptosis, but this phenotype is most clearly observed when cells are grown in low serum or when survival factors are removed from the culture medium (50). Although c-Myc-induced apoptosis can be mediated by induction of the p19ARF/p53 tumor suppressor pathway, c-Myc expression also causes a drop in Bcl-XL expression (51). This repression was not because of c-Myc directly repressing Bcl-XL transcription but appears to be an indirect response of the cell to the expression of a potent stimulator of cell cycle progression.
Subtle changes in the cellular levels of Bcl-XL expression can have dramatic phenotypes. For example, 2–4-fold reduction in the expression of Bcl-XL is sufficient to induce apoptosis following withdrawal of thrombopoietin from UT-7/thrombopoietin and normal megakaryocytic cells (52). Similarly, a 2-fold change in Bcl-XL protein was related to a 10% increase in apoptosis of normal megakaryocytic cells 24 h after thrombopoietin withdrawal (52). We found that cells lacking one allele of Bcl-XL, with a concomitant 50% reduction in protein expression, displayed increased apoptosis when cultured in medium containing low serum (Fig. 6, A and B). The changes in Bcl-XL levels and apoptosis were similar to those that we observed following expression of TEL. Conversely, overexpression of ETS2 in BAC1.2F5 macrophages protected these cells from apoptosis induced upon withdrawal of colony-stimulating factor-1, even though Bcl-XL mRNA levels increased only 4-fold (37). Thus, the level of repression of Bcl-XL that we observed upon TEL overexpression (Fig. 5) is likely sufficient to contribute to apoptosis upon withdrawal of survival factors where the scales are tipped to a pro-apoptotic versus an anti-apoptotic response.
Although TEL-mediated repression of Bcl-XL likely contributes to TEL-induced apoptosis, the repression of other genes may also contribute to apoptosis. The regulation of the stromelysin gene is critical for tissue homeostasis. Repression of Stromelysin-1 and other extracellular matrix components likely regulates the alterations in cellular morphology and cell adhesion observed when TEL is expressed in transformed or partially transformed fibroblasts. However, the changes in cell adhesion appear unlinked to TEL-induced alterations in focal adhesions, because TEL expression affects the formation of focal adhesions in the absence of changes in cellular morphology (Fig. 2). Also, collagen complemented adhesion phenotype but did not dramatically affect TEL-induced apoptosis (Fig. 2B). Therefore, if the changes in the numbers of focal adhesions or alterations in cell adhesion contribute to TEL-mediated apoptosis, the effect is subtle and was not observed in our assays.
Bcl-XL is an important integrator of diverse signaling pathways mediating cellular survival and death. These signaling cascades activate multiple transcription factors including NFB, STAT5, ETS1, and ETS2, which in turn activate Bcl-XL to promote cell survival (46–48). These signaling events are critical to disease processes including chronic myeloid leukemia in which the BCR-ABL fusion protein inappropriately activates STAT5 to induce Bcl-XL expression and increase cellular survival (53–55). Additionally, three fusion proteins involving TEL (TEL-platelet-derived growth factor receptor, TEL-JAK2, and TEL-ABL) strongly induce activation of STAT5 (56, 57). Although the ability of TEL-platelet-derived growth factor receptor , TEL-JAK2, or TEL-ABL to regulate Bcl-XL expression has not been established, all of these fusion proteins are sufficient for transformation of hematopoietic cells. Based on our data we speculate that the loss of one TEL allele through translocation may cooperate with the tyrosine kinase-mediated activation of STAT signaling and contribute to activation of Bcl-XL.
In contrast to the above fusion proteins, expression of TEL/AML1 is not sufficient for inducing transformation and appears to require additional mutations (58). Given that both alleles of TEL are disrupted in up to 90% of t(12;21)-containing cases of B-cell acute lymphoblastic leukemia, it is possible that loss of TEL contributes to cell survival. Consistent with this interpretation, we have found that Bcl-XL is slightly higher in t(12;21)-containing cases of B-cell acute lymphoblastic leukemia.2 However, loss of TEL would be predicted to only de-repress and allow activation by other transcription factors, and we have not been able to mechanistically link this small (2-fold) increase in Bcl-XL levels to loss of TEL. In this regard, the t(12;22) fusion protein would be predicted to stimulate Bcl-XL expression as it contains the DNA binding domain of TEL but activates rather than represses transcription (6). Thus, it is possible that reversing TEL-mediated repression of Bcl-XL contributes to t(12; 22)-mediated leukemogenesis.
|
FOOTNOTES |
|---|
Present address: Dept. of Immunology, Astral Inc., 3550 General Atomics Ct., San Diego, CA 92121.
¶ Present address: Ligand Pharmaceuticals, Inc., San Diego, CA 92121.
¶¶ Present address: MedImmune, Inc., 35 West Watkins Mill Rd., Gaithersburg, MD 20878.
|||| To whom correspondence should addressed: Dept. of Biochemistry, Vanderbilt-Ingram Cancer Center, Vanderbilt University School of Medicine, Preston Research Bldg., Rm. 512, 23rd and Pierce, Nashville, TN 37232. Tel.: 615-936-3582; Fax: 615-936-1790; E-mail: scott.hiebert{at}mcmail.vanderbilt.edu.
1 The abbreviations used are: AML, acute myeloid leukemia; ECM, extracellular matrix; EBS, ets factor binding sites; CMV, cytomegalovirus; WT, wild-type; RSV, Rous sarcoma virus; DMEM, Dulbecco's modified Eagle's medium; PBS, phosphate-buffered saline; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HA, hemagglutinin; nt, nucleotide; NFB, nuclear factor B; STAT, signal transducers and activators of transcription.
2 B. J. Irvin and S. W. Hiebert, unpublished data.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
), TEL (
), or TEL
) were grown in 0.1% serum for 36 h. The numbers of live and dead cells were determined using trypan blue exclusion assay at 0, 6, 12, 24, and 36 h. Each data point is the average from triplicate samples, and in some cases the error bars are too small to be seen.
, •) and then grown in 0.1% serum for 36 h. The numbers of live and dead cells were determined using trypan blue exclusion assay at 0, 6, 12, 24, and 36 h. Each data point is the average from triplicate samples, and in some cases the error bars are too small to be seen.
