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

J. Biol. Chem., Vol. 279, Issue 34, 35604-35615, August 20, 2004

This Article Abstract Full Text (PDF) All Versions of this Article: 279/34/35604    most recent M401851200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ray, S. Articles by Brasier, A. R. Search for Related Content PubMed PubMed Citation Articles by Ray, S. Articles by Brasier, A. R. Social Bookmarking                      What's this?

Genomic Mechanisms of p210^BCR-ABL Signaling

INDUCTION OF HEAT SHOCK PROTEIN 70 THROUGH THE GATA RESPONSE ELEMENT CONFERS RESISTANCE TO PACLITAXEL-INDUCED APOPTOSIS^*

Sutapa Ray, Ying Lu, Scott H. Kaufmann, W. Clay Gustafson¶, Judith E. Karp||, Istvan Boldogh**, Alan P. Fields, and Allan R. Brasier

From the Department of Internal Medicine, University of Texas Medical Branch, Galveston, Texas 77555-1060, the Division of Oncology Research, Mayo Clinic, Rochester, Minnesota 55905, the ^¶Sealy Center for Cancer Cell Biology, University of Texas Medical Branch, Galveston, Texas 77555-1048, the ^||Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, Maryland 21231, the ^**Department of Microbiology and Immunology, University of Texas Medical Branch, Galveston, Texas 77555-1019, and the Mayo Clinic Comprehensive Cancer Center, Jacksonville, Florida 32224

Received for publication, February 19, 2004 , and in revised form, April 28, 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Chronic myelogenous leukemia (CML) results from a t(9,22) translocation, producing the p210^BCR-ABL oncoprotein, a tyrosine kinase that causes transformation and chemotherapy resistance. To further understand mechanisms mediating chemotherapy resistance, we identified 556 differentially regulated genes in HL-60 cells stably expressing p210^BCR-ABL versus those expressing an empty vector using cDNA macro- and oligonucleotide microarrays. These BCR-ABL-regulated gene products play diverse roles in cellular function including apoptosis, cell cycle regulation, intracellular signaling, transcription, and cellular adhesion. In particular, we identified up-regulation of the inducible form of heat shock protein 70 (Hsp70), and further explored the mechanism for its up-regulation. In HL-60/BCR-ABL and K562 cells (expressing p210^BCR-ABL), abundant cytoplasmic Hsp70 expression was detected by immunoblot analysis. Moreover, cells isolated from bone marrow aspirates of patients in different stages of CML (chronic, aggressive, and blast crisis) express Hsp70. Expression of p210^BCR-ABL in BCR-ABL negative cells induced transcription of the proximal Hsp70 promoter. Mutational analysis mapped the major p210^BCR-ABL responsive element to a high affinity 5'(A/T)GATA(A/G)-3' "GATA" response element (GATA-RE) that binds GATA-1 in CML cells. The GATA-RE was sufficient to confer p210^BCR-ABL- and p185^BCR-ABL-mediated trans-activation to an inert promoter. Short interfering RNA mediated "knockdown" of Hsp70 expression in K562 cells induced marked sensitivity to paclitaxel-induced apoptosis. Together these findings indicate that BCR-ABL confers chemotherapeutic resistance through intracellular signaling to the GATA-RE element found in the promoter region of the anti-apoptotic Hsp70 protein. We suggest that down-regulation of the GATA-Hsp70 pathway may be useful in the treatment of chemotherapy-resistant CML.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Chronic myelogenous leukemia (CML),¹ a diagnosis given for 15% of adult leukemias, represents a clonal expansion of granulocytic progenitor cells (1). CML is caused by a reciprocal t(9,22) translocation of c-abl from chromosome 9 to the break-point cluster region (bcr) gene on chromosome 22 (1, 2). In 95% of CML cases, translocation involves the major bcr, leading to generation of the 210-kDa BCR-ABL fusion oncoprotein (p210^BCR-ABL). A similar BCR-ABL translocation is also be found in 20–30% of cases of adult acute lymphoblastic leukemia and 5% of childhood acute lymphoblastic leukemia (3, 4). In both cases, the serine-threonine kinase domain from BCR partially replaces the inhibitory NH₂-terminal Src homology 3 domain of ABL, producing a fusion protein with constitutive ABL tyrosine kinase activity (3, 5, 6).

Dysregulated BCR-ABL signaling induces pleiotropic phenotypic changes in granulocytic cells, including resistance to chemotherapy-induced apoptosis, disruption of cell cycle check-points, induction of growth factor-independent proliferation (7), cellular transformation in vitro (3), and production of a CML-like leukemia in vivo (8, 9). Moreover, BCR-ABL signaling is required to actively maintain cellular viability. For example, down-regulation of BCR-ABL expression (10, 11) or inhibition of its ABL tyrosine kinase activity using the 2-phenlyaminopyridmidine inhibitor, imatinib (STI571), induces apoptosis in vitro and remissions during the chronic phase of CML in the clinic (10, 12). A body of work has shown that BCR-ABL regulates diverse signaling pathways including p21^ras, phosphatidylinositol 3-kinase, protein kinase C (PKC) (13), Jak-STAT (3, 14), and NF-B (15, 16). The relationships of these signaling pathways to specific cellular responses are still being elucidated. Perhaps best understood is the p21^ras/Raf pathway in mediating cellular transformation by BCR-ABL. Inhibition of p21^ras, either by administration of antisense oligonucleotides or microinjection of blocking antibodies, prevents expression of the c-myc proto-oncogene, thereby blocking cellular transformation (reviewed in Ref. 17).

A hallmark of BCR-ABL-expressing cells is the profound resistance to chemotherapeutic agent-induced apoptosis (13, 16, 18, 19). Chemotherapy-induced cell death involves proteolytic cleavage of cysteine proteases, called caspases, whose activation can be triggered by cytochrome c release from mitochondria (20). In the cytosol, cytochrome c binds apoptotic protease activation factor (Apaf)-1, inducing its oligomerization to form the "apoptosome" (21), an enzymatically competent Apaf-1·procaspase-9 complex that cleaves procaspase-3, thereby committing the cell to autolysis.

BCR-ABL causes chemotherapy resistance, at least in part, by interfering with the caspase activation pathway at multiple steps. Previous studies have shown that Ph+ cells isolated either during CML blast crisis or after ectopic BCR-ABL expression release diminished amounts of cytochrome c into the cytosol after exposure to high dose Ara-C or etoposide (22, 23). At the biochemical level, BCR-ABL-transformed cells express increased amounts of the outer mitochondrial membrane protein Bcl-x_L, which prevents Bax insertion into the outer mitochrondial membrane, reducing chemotherapy-induced cytochrome c release (23). In addition, we have recently shown that BCR-ABL signaling through the PKC-NF-B pathway is also required for resistance to paclitaxel (Taxol)-induced apoptosis (16). NF-B is an inducible transcription factor that activates expression of several inhibitors of apoptosis proteins, polypeptides that bind and inactivate caspases as well as inducing ubiquitin-mediated degradation of the RHG proteins (Reaper, HID, and Grim, Ref. 24). Together these findings suggest that BCR-ABL may induce chemotherapy resistance by controlling expression of proteins that antagonize caspase activation at multiple points.

Several studies have begun to identify BCR-ABL-activated genomic programs, focusing on BCR-ABL-activated genes important in cell cycle regulation or molecular signatures (25–27). However, whether BCR-ABL activates other genetic targets controlling anti-apoptosis has not been completely resolved. Here we apply discovery based tools to investigate the effects of BCR-ABL on gene expression. Differential expression of a variety of signaling kinases, transcription factors, cell surface receptors, and adhesion molecules was observed in HL-60 cells stably expressing p210^BCR-ABL versus empty vector controls. Further investigation focused on the BCR-ABL induced up-regulation of heat shock protein (Hsp)-70 kDa isoform, an anti-apoptotic protein known to bind and inhibit Apaf-1, apoptosis inducing factor, and inhibit the apoptosis signal-inducing kinase-1 (Refs. 28–30). After Hsp70 up-regulation was validated in K562 cells and bone marrow aspirates from CML patients, subsequent experiments mapped sequences responsible for BCR-ABL-mediated transactivation to a high affinity GATA response element (GATA-RE) located in the proximal Hsp70 promoter between –82 and –58 nt. Finally, short interfering RNA (siRNA)-mediated down-regulation of Hsp70 expression sensitized K562 cells to paclitaxel-induced apoptosis. Together these findings indicate that p210^BCR-ABL transactivates the GATA-RE in the Hsp70 promoter, leading to up-regulation of the anti-apoptotic Hsp70 gene.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Cell Culture and Treatment—Human K562 erythroleukemia (16), HepG2 hepatocellular carcinoma (31), and HL-60 acute myeloid leukemia cells stably transfected with BCR-ABL (HL-60/BCR-ABL) or empty vector (HL-60/Neo) were cultured as described (23). Where indicated, paclitaxel (Sigma) was resuspended in Me₂SO and added to a final concentration of 1 µM.

In some experiments, cells were isolated from leukemic bone marrow aspirates. Bone marrow aspirates were obtained after informed consent was obtained under an institutional review board-approved protocol at the time of diagnosis or treatment on the Adult Leukemia Service at the Johns Hopkins Hospital. Fractions of mononuclear cells or granulocytes were isolated on double Ficoll-Hypaque gradients (32), washed with RPMI 1640 medium containing 10 mM HEPES (pH 7.4), lysed in 6 M guanidine hydrochloride under reducing conditions, reacted with iodoacetamide, dialyzed into 0.1% (w/v) SDS, and lyophilized as described (33).

Membrane Based cDNA Macroarrays—Total RNA was extracted from control (HL-60/Neo) or p210^BCR-ABL-expressing (HL-60/BCR-ABL) cells by acid guanidium-phenol extraction (TRI Reagent, Sigma) and treated with DNase. Five micrograms were reverse transcribed in the presence of 35 µCi of [-³³P]dATP and cDNA was purified by column chromatography (Chroma spin-200, Clontech). Atlas 1.2 Arrays (Clontech) were hybridized with 10⁶ cpm/ml of probe at 68 °C and washed as recommended by the manufacturer. Membranes were exposed to a PhosphorImager cassette and relative changes in hybridization intensity were determined by AtlasImage 1.01 software (Clontech). Comparisons of mRNA populations between control and p210^BCR-ABL-expressing cells were performed with two different sets of Atlas Array membrane lots in two independent experiments.

For each gene, local background was subtracted from the total hybridization intensity and the average signal intensity was determined for duplicate spots. To compare differences in gene expression between arrays, background subtracted average intensity was normalized to that of housekeeping genes. Only those genes that showed an average 3.5-fold up-regulation or down-regulation across duplicate membranes were further considered.

High Density Oligonucleotide Arrays—Four independent RNA samples were prepared from the HL-60/Neo and the HL-60/BCR-ABL-transfected cells for hybridization to the Hu95A GeneChip (Affymetrix, Santa Clara, CA). First-strand cDNA synthesis was performed using total RNA (10–25 µg), a T7-(dT)₂₄ oligomer and SuperScript II reverse transcriptase (Invitrogen). Second strand synthesis, target RNA labeling, and hybridization were as previously described (34). Gene Chip arrays were scanned using a Gene Array Scanner (Hewlett-Packard) and analyzed using the Gene Chip Analysis Suite 4 software (Affymetrix Inc.). The average difference statistic was retrieved for quantification of mRNA abundance in those samples in which the absolute call indicated that the gene was present.

Oligonucleotide Microarray Data Analysis—Reproducibility of the four independent microarrays was determined by calculating the correlation coefficient for the log-transformed average difference values for the probe sets in each array (34). For each pairwise comparison, the mean correlation coefficient was 0.945 ± 0.024 (n = 6) in the HL-60/Neo data sets, and 0.977 ± 0.010 (n = 6) for the HL-60/BCR-ABL data sets, indicating that the measurements were highly reproducible (Table I). For comparison of the fluorescence intensity (average difference) values among multiple experiments, the average difference values for each "experimental" GeneChip were scaled to that of the "base" GeneChip (34). Genes differentially expressed were identified by one-way ANOVA with replicates comparing the average difference values of a probe sets in HL-60/Neo versus HL-60/BCR-ABL. Genes (probe sets) with a p value (Pr(F)) < 0.0001 were selected for further analysis. Agglomerative hierarchical clustering using the unweighted pair-group method with arithmetic mean (34) was performed on the indicated genes (Spotfire Array Explorer, version 7, Spotfire Inc., Cambridge, MA). Data are graphically presented as heat maps in which fluorescence intensity is represented by a color gradient.

Western Blot Analysis—Western blots were performed as described previously (13). Briefly, cells were counted and equal numbers were harvested and lysed directly into Laemmli sample buffer. Equal volumes of sample were then fractionated by a 5–20% SDS-PAGE gradient gel (Invitrogen), transferred to nitrocellulose membranes, and blocked with 5% nonfat dry milk in phosphate-buffered saline, 1% Tween 20. Blots were then probed with the indicated primary antibody; immune complexes were detected by enhanced chemiluminescence (ECL, Amersham) or, where indicated, near-infrared fluorescence (Odyssey Imaging System, LiCor BioSciences).

Plasmids—The –259/+37 Hsp70/LUC promoter-driven luciferase reporter plasmid was constructed by PCR of HL60 genomic DNA using the upstream primer –259 5'-ACGGATCCCACCGCCACTCCCCCTTC-3', and the Hsp70 downstream primer 5'-AAAAAGCTTGTGGACTGTCGCAGCAGCTC-3' (HindIII site underlined). The restricted gelpurified PCR product was ligated into pOLUC reporter digested with the same endonucleases (35). A series of 5' deletions were constructed by PCR using Hsp70 (–259/+37)/LUC as template with the upstream primers –200, 5'-GTGGATCCCAGAAGACTCTG-3'; –163, 5'-GCGGATCCCTGGCCTCTGATT-3'; –123, 5'-GGGGATCCACGGGAGGCGAAA-3'; –82, 5'-CCTGGATCCCTCATCGAGCTC-3'; –58, 5'-GATTGGATCCGAAGGGAAAAGG-3', and the Hsp70 downstream primer. Each 5' deletion was digested with BamHI (site is underlined) and HindIII, gel purified, and ligated into pOLUC. The site-directed mutation of the GATA binding element was constructed by PCR SOEing (36) using the sense primer 5'-GAGCTCGGTCTCAGGCTCAGGATA-3' and the antisense primer 5'-TCTGAGCCTGAGACCGAG-3' (mutations underlined) with the Hsp70 downstream primer.

The multimeric GATA (WT) and GATA (Mut)-driven reporter was constructed by annealing the sense and antisense oligonucleotides as: GATA (WT): sense, GATCGAGCTCGGTGATTGGCTCAGAA and anti-sense, CTCGAGCCACTAACCGAGTCTTCTAG; GATA (Mut): sense, GATCGAGCTCGGTCTCAGGCTCAGAA and antisense, CTCGAGCCAGAGTCCGAGTCTTCTAG. Duplex oligonucleotides were then phosphorylated, ligated with T4 DNA ligase, 3 copies were ligated into BamHI linearized –59/+22 rAGT/LUC, driven by the inert angiotensinogen TATA box (35). pCMV-FLAG BCR-ABL expression vectors were constructed by ligating the BCR-ABL coding sequences into pCMV-Tag plasmid (37). All plasmids were purified by ion exchange chromatography (Qiagen) and cloned inserts were sequenced to confirm authenticity.

Cell Transfection and Reporter Assays—Transient transfections in K562 cells were carried out using DMRIE-C reagent (Invitrogen). For reporter plasmids, 4 to 10 µg of plasmid DNA was resuspended into 0.5 ml of OPTI-MEM® I reduced serum medium (Invitrogen), mixed with an equal volume of medium containing 10 µl of DMRIE-C Reagent, and incubated at room temperature for 15 min to allow lipid-DNA complex formation. Logarithmically growing K562 cells (2 x 10⁶ cells) were centrifuged, resuspended in 0.5 ml of reduced serum medium, and added to the lipid-DNA complex and incubated at 37 °C in a CO₂ incubator for 5 h. After transfection, growth medium (containing 10% fetal bovine serum) was added and the cells were incubated overnight. For reporter assays, cells were harvested at the indicated time points and washed two times with cold phosphate-buffered saline. Cytoplasmic lysates were prepared and independently measured for luciferase and -galactosidase activity (Promega, Madison, WI) as described previously (35). Luciferase reporter activity was normalized to the internal control of -galactosidase activity to control for differences in transfection efficiency.

Electrophoretic Mobility Shift Assay (EMSA)—Nuclear proteins were purified over a sucrose cushion (31) normalized for protein concentration by the Coomassie G-250 assay (Bio-Rad). The GATA (WT) monomeric duplex was radiolabeled with Klenow polymerase and purified by gel filtration chromatography. DNA-binding reactions were carried out in a mixture of 20 µg of nuclear proteins, 12 mM HEPES (pH 7.9), 40 mM KCl, 120 mM NaCl, 0.2 mM EDTA, 0.2 mM EGTA, 0.4 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 8% glycerol, 2 µg of poly(dA-dT), and 20,000 cpm of -³²P-labeled double-stranded GATA (WT) probe in a total volume of 20 µl. The reaction mixture was incubated on ice for 15 min and then fractionated by 6% nondenaturing PAGE containing 2% glycerol. Gels were dried and subjected to autoradiography using Kodak X-AR film at –70 °C. Competition was performed by the addition of a 100-fold molar excess of nonradioactive double-stranded oligonucleotide competitor at the time of addition of the radioactive probe. For supershift, anti-GATA-1 antibody was added to the gel shift reaction and incubated on ice for 1 h prior to fractionation by nondenaturing PAGE.

Apoptosis—Cells were stained with annexin V-phycoerythrin using an Annexin V-Phycoerythrin Apoptosis Detection Kit I (BD Pharmingen). Briefly, 5 x 10⁵ washed cells were collected by centrifugation and resuspended in annexin V-binding buffer. Cell suspensions were then incubated with annexin V-phycoerythrin at 1 µg/ml and 7 aminoactinomycin D at a final concentration of 1 µg/ml for 25 min at room temperature in the dark. The percentage of apoptotic cells was determined by flow cytometry (BD Biosciences FACScan) analysis (38).

SiRNA-mediated Hsp-70 "Knockdown"—Various concentrations from 50 to 200 nM Hsp70 or lamin A/C siRNA (custom SMART pool, HSPA1L-NM,43005346, and siRNA CONTROl, Dharmacon Research Inc., Lafayette, CO) were substituted for the reporter plasmid and transfected into logarithmically growing K562 cells using DIMRIE-C as described above. After 5 h, growth medium was added and cells were returned to culture in the absence or presence of paclitaxel (1 µM final concentration) for the times indicated.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Identification of Genes Downstream of the BCR-ABL Pathway—In this study, we initially determined gene expression profiles in HL-60 cells stably expressing p210^BCR-ABL because these cells show enhanced Bcl-x_L expression and decreased cytochrome c release in response to chemotherapeutic agent administration (23). To confirm stable p210^BCR-ABL expression, Western immunoblot analysis was performed using an anti-c-Abl antibody. HL-60/BCR-ABL cells express amounts of p210^BCR-ABL comparable with Ph+ K562 cells, whereas the HL-60/Neo controls have no detectable BCR-ABL antigen (Fig. 1). To identify genomic targets of p210^BCR-ABL we systematically examined differences in mRNA expression by cDNA macroarrays and high density oligonucleotide microarrays. Using membrane-based cDNA macroarrays containing 1,176 sequenced human gene probes, we identified 25 genes whose expression was increased by 3.5-fold (or greater), and 34 genes whose expression was reduced by 3.5-fold (or greater) in duplicate pairwise experiments (Table II). The major biochemical functions of the differentially expressed genes were apoptosis (Bcl-x_L and Hsp70), cellular signaling (cell surface receptors (apoE and PAF) and intracellular signaling kinases (PKC-1, phospholipase , and RAP1 GAP)), DNA synthesis/repair, transcription (Id), cell adhesion/and immune recognition, and extracellular matrix turnover/regulation. Importantly, we noted that c-Abl expression (representing signal from ectopic p210^BCR-ABL expression) was 7.99 ± 2.11-fold up-regulated, and that Bcl-x_L was 4-fold up-regulated in the BCR-ABL transfected relative to control cells, confirming the findings of others (23). Together these indicated a valid data set was generated by the macroarray. The 23-fold up-regulation of inducible the Hsp 70-kDa isoform in the HL-60/BCR-ABL cells (Table II, Fig. 2) was particularly noteworthy because Hsp70 inhibits apoptosis by interfering with oligomerization of Apaf-1 (39, 40) and function of apoptosis inducing factor (28).

View larger version (30K): [in this window] [in a new window]   FIG. 1. p210BCR-ABL expression in HL-60/BCR-ABL cells. Western immunoblot was performed on cell extracts from Ph+ K562 (lane 1), HL-60/Neo (lane 2), and HL-60/BCR-ABL (lane 3) cells. Top panel, blot was probed with anti-Abl antibody and detected by enhanced chemiluminescence (ECL). Bottom panel, blot was probed with anti--actin antibody.

View larger version (70K): [in this window] [in a new window]   FIG. 2. Hsp70 up-regulation in cDNA macroarray. Nylon membranes containing spotted, immobilized cDNAs (Atlas 1.2 Array, Clontech) were hybridized with 33P-labeled cDNA from HL-60/Neo and HL-60/BCR-ABL stable transfectants. Shown is a section of the autoradiographic image containing the Hsp70 cDNA (arrow).

View larger version (41K): [in this window] [in a new window]   FIG. 3. Hiearchical clustering and heat map analysis of 42 genes differentially regulated by p210BCR-ABL. The normalized average difference values for 42 genes whose expression is changed by p210BCR-ABL (Pr(F) < 0.00001) were retrieved and subjected to hierarchical clustering (as described under "Experimental Procedures"). Green represents the minimum average difference value (100 scaled units), black represents the mid-average difference value (4,500 scaled units), and red represents the maximum average difference value (9,000 scaled units). Genes numbered 1–20 are undetectable in HL-60/Neo cells and highly up-regulated by p210BCR-ABL expression; those numbered 21–36 are highly expressed in HL-60/Neo and down-regulated by p210BCR-ABL (see Table III for gene identities).

View larger version (80K): [in this window] [in a new window]   FIG. 4. Validation of Hsp70 up-regulation in p210BCR-ABL-expressing cells. A, Northern blot. Total RNA from K562 (lane 1), HL-60/Neo (lane 2), and HL-60/BCR-ABL (lane 3) cells was fractionated by denaturing MOPS-formaldehyde-agarose gel electrophoresis. Shown is an autoradiogram after transfer and probing with radiolabeled Hsp70 cDNA. Relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and 18 S, Hsp70 is not detectable in HL-60/Neo and strongly expressed in p210BCR-ABL expressing cells. B, Western immunoblot. Cytoplasmic extracts (50 µg) were fractionated on SDS-PAGE and transferred to polyvinylidene difluoride membranes. Lane 1, HL-60/Neo; lane 2, HL-60/BCR-ABL; lane 3, K562. Top panel, the blot was probed with anti-Hsp-70 antibodies and immune complexes detected by ECL. Bottom panel, the membrane was re-probed with anti--actin to control for loading.

View larger version (65K): [in this window] [in a new window]   FIG. 5. Hsp70 expression in clinical CML samples. Marrow fractions (5 x 105 cells/aliquot) enriched in granulocytes (lanes 5 and 7) or mononuclear cells (lanes 4, 6, and 8–12) were subjected to Western immunoblot analysis with anti-Hsp70 (Stressgen SPA-810) or, as a loading control histone H1 (a kind gift from James Sorace, Veterans Affairs Medical Center, Baltimore, MD) using previously described techniques (60). A line above multiple lanes indicates multiple samples from a single patient. Samples came from patients with chronic phase CML, aggressive phase CML, or blast crisis (BC), which was either lymphoid (L), myeloid (M), or mixed. HL-60 cells loaded at 1.25 x 105, 2.5 x 105, or 5 x 105 (lanes 1–3, respectively) served as a positive control. The dotted line indicates juxtaposition of nonadjacent lanes from one Hsp70 blot and one histone H1 blot to create this figure.

View larger version (29K): [in this window] [in a new window]   FIG. 6. Identification of p210BCR-ABL response elements in the Hsp70 promoter. A, sequence of the –200/+37 Hsp70 promoter. Numbers are relative to the transcription start site. Location of the HSEs are indicated by the dotted underline. The GATA site is indicated in bold. The TATAA box is indicated by double underline. Location of various 5' deletions are indicated by vertical arrows. Horizontal arrow, primary transcript start site. Cloning sites are indicated by single underlines. B, cis elements in Ph+ K562 cells. A series of 5' Hsp70 promoter deletions were constructed and tested for luciferase reporter activity in K562 cells. Normalized luciferase activity is shown for each mutant tested. Deletion between –82 and –58 nt upstream of the transcription start site reduces promoter activity significantly. C, p210BCR-ABL response elements. Promoter mapping was performed using serial 5' deletions of the Hsp70 promoter in transient transfection experiments into p210BCR-ABL-deficient HepG2 cells. Where indicated, cells were cotransfected with the internal control reporter and CMV-BCR-ABL expression plasmids. Two response elements were identified; one between –200 and –163; the second between –82 and –58 nt containing the GATA-RE. D, Hsp70 –82/–58 "GATA" site is required for p210BCR-ABL induction. –123Hsp70/LUC WT or containing a site mutation in the GATA sequence (Mut) was transfected in the presence of empty or CMV-BCR-ABL expression plasmid. E, p210BCR-ABL activates Hsp70 independently of the heat shock response pathway. –123 Hsp70/LUC transfected in the presence of empty or CMV-BCR-ABL expression plasmids. Cells were subjected to heat shock at 42 °C for 3 h, and returned to 37 °C for 24 h prior to harvest and assay for reporter activity. Normalized luciferase activity is shown from a representative experiment (n = 3). -Fold induction values are indicated for each treatment condition.

A role for GATA in regulation of Hsp70 expression has not been described previously. To determine its function within the Hsp70 promoter in p210^BCR-ABL-dependent transactivation, a mutation was introduced into the core GATA sequence in the –123/+37 Hsp70 reporter (Fig. 6A); the resulting construct was tested for the BCR-ABL inducibility. As seen in Fig. 6D, the Hsp70 promoter containing the GATA site mutation was not induced by p210^BCR-ABL, indicating that this sequence was essential for BCR-ABL-mediated transactivation. Previous work has shown that the major inducible regulator of Hsp70 expression is the HSE, an upstream cis sequence that binds to oligomeric heat shock factor formed in response to elevated temperature (41). To determine whether the BCR-ABL-GATA pathway interacted with the heat shock response-HSE pathway, cells transfected with –123 Hsp70/LUC in the presence or absence of p210^BCR-ABL were subjected to heat shock. Heat shock induced the Hsp70 expression 9.5-fold in the absence of p210^BCR-ABL, and to a similar degree (6-fold) in its presence (Fig. 6E). These two stimuli together produced a 67-fold induction of reporter activity, resulting in an absolute magnitude that was clearly greater than that produced by the sum of either stimulus alone, representing a greater than additive response. Together, these data indicate that the GATA response element was essential for p210^BCR-ABL-induced transactivation of the Hsp70 promoter and functioned in a pathway independent from that induced by heat shock on the HSE.

BCR-ABL Transactivates the GATA-RE without Apparent Change in the Pattern of GATA-1 Binding—We next performed EMSA on HL-60/Neo and HL-60/BCR-ABL cells to determine how p210^BCR-ABL regulates factors binding to the GATA element (Fig. 7A). Somewhat surprisingly, there was no consistent difference in the pattern or abundance of DNA-binding proteins that bound the GATA sequences in cells expressing p210^BCR-ABL. Competition experiments using the transcriptionally active GATA WT duplex and an inert GATA Mut sequence in EMSA indicated that a single major specific band interacted with GATA response element in both cell types (Fig. 7A). To help identify the GATA family member interacting with the Hsp70 promoter, we examined the high density microarray data for expression of all GATA isoforms, notably GATA isoforms-1, -2, and -3, which have been implicated in hematopoiesis (43). The major isoform expressed in HL-60 cells was GATA-1, whose steady state mRNA levels was not different between those cells transfected with empty vector and those transfected with BCR-ABL (not shown). Supershift assays confirmed that GATA-1 was the major isoform binding the Hsp70 GATA element (Fig. 7B). Addition of anti-GATA-1 antibody produced a strong supershift, depleting over 50% of the specific GATA binding complex. Nuclear extracts from the Ph+ K562 cells exhibited a similar supershift indicating that GATA-1 is also the major GATA binding species of the Hsp70 promoter in CML cells (Fig. 7B, lanes 5 and 6). Finally, to establish that the GATA response element was a bona fide p210^BCR-ABL-inducible enhancer, multimeric copies of the GATA WT or GATA Mut DNA binding sequences were ligated upstream of an inert TATA-driven luciferase reporter and assessed for p210^BCR-ABL responsiveness. We found that the GATA WT sequences were potently induced by p210^BCR-ABL over a wide concentration range, whereas the GATA Mut was poorly induced (Fig. 7C).

View larger version (32K): [in this window] [in a new window]   FIG. 7. GATA-1 binds the Hsp70 –58 to –82 GATA element. A, comparison of DNA binding activities in HL-60 transfectants. Nuclear extracts (NE) from HL-60/Neo or HL-60/BCR-ABL transfectants were used to bind radiolabeled Hsp70 –58/–82 duplex oligonucleotide in EMSA. Where indicated, a 10-fold molar excess of GATA WT or Mut oligonucleotide was added prior to fractionation by nondenaturing PAGE. An autoradiogram is shown. NS, nonspecific binding. B, the GATA-1 isoform binds the Hsp70 –58 to –82 element in p210BCR-ABL transfectants. NE from HL-60/BCR-ABL-transfected cells were used in EMSA to bind radiolabeled Hsp70 –58/–82 duplex (lanes 1–4). Competition analysis was performed by addition of 10-fold molar excess nonradioactive double-stranded oligonucleotide competitor at the time of addition of radioactive probe. Lane 2 competitor was unlabeled Hsp70 –58/–82 WT; lane 3 was unlabeled Hsp70 –58/–82 mutant. Lane 4, antibody supershift. HL-60/BCR-ABL NE were pre-mixed with 1 µl of affinity purified anti-GATA-1 polyclonal antibodies for 1 h at 4 °C prior to EMSA analysis. Two supershifted complexes were produced (indicated by asterisk). Lane 5, 6 NE from K562 cells were subjected to EMSA with radiolabeled Hsp70 –58/–82 duplex. Lane 6, GATA-1 supershift. An identical supershifting pattern was observed. C, Hsp70 GATA response element is highly p210BCR-ABL inducible. Wild type and mutant multimeric sequences corresponding to the Hsp70 –82/–58 response element were ligated upstream of an inert promoter and transiently co-transfected with cytomegalovirus (CMV) expression plasmid with or without BCR-ABL coding sequences. BCR-ABL expression potently transactivated GATA WT driven reporter activity at all concentrations of plasmid used.

View larger version (10K): [in this window] [in a new window]   FIG. 8. Comparison of p185BCR-ABL and p210BCR-ABL oncoproteins on Hsp70 promoter transactivation. Wild type and mutant multimeric sequences corresponding to the Hsp70 –82/–58 response element were transiently co-transfected with cytomegalovirus expression plasmid expressing 185- or 210-kDa BCR-ABL coding sequences. Normalized luciferase activity is shown.

View larger version (12K): [in this window] [in a new window]   FIG. 9. SiRNA-mediated Hsp70 knockdown sensitizes K562 cells to chemotherapy-induced apoptosis. A, Western immunoblot. Mock or siRNA-transfected K562 cells were harvested and 100 µg of protein were analyzed for Hsp70 expression (top) or -actin (bottom) by Western immunoblot using IRD-labeled secondary antibody and detection by near infrared fluorescence using the Odyssey Imaging system (LiCor, Inc.). This method for protein quantitation is linear over log orders of magnitude, as determined by serial dilutions of input protein (not shown and Ref. 61). Shown is a digital image scanned at 700 nM. After normalization of Hsp70 signal to -actin control, Hsp70 abundance is 50% of mock transfected cells. B, down-regulation of Hsp70 sensitizes K562 cells to paclitaxel-induced apoptosis. K562 cells transfected with 0 or 100 nM siRNA to lamin A/C or Hsp70. Where indicated, cells were also treated with 1 µM paclitaxel for 48 h prior to harvest and annexin V staining ("Experimental Procedures"). Annexin V positive (apoptotic) cells are shown. In the absence of paclitaxel, the apoptosis rate in mock or siRNA-treated cells was identical at 2.2%. Paclitaxel induced apoptosis to a similar degree in control or lamin A/C-transfected cells, whereas 93% of Hsp70 siRNA-transfected cells were annexin V positive. *, p < 0.01 compared with lamin A/C siRNA-transfected cells. C, down-regulation of Hsp-70 sensitizes K562 cells to paclitaxel-induced apoptosis. K562 cells transfected with 0, 50, 100, or 200 nM siRNA were treated in the absence or presence of 1 µM paclitaxel for the indicated times (in h) prior to annexin V staining. *, p < 0.01 compared with non-siRNA-transfected cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A clinical hallmark of CML is the presence of circulating progenitor cells prematurely released from the bone marrow microenvironment (44). In normal hematopoiesis, cell surface adhesion molecules expressed by immature progenitor cells bind elements of the bone marrow stroma to inhibit proliferation and induce differentiation. A characteristic of CML cells is deficient adhesion to bone marrow stroma through the decreased expression of L-selectin and reduced activity of ₁-integrin, an effect reversed by treatment with BCR-ABL anti-sense oligonucleotides (45). Our microarray data in Table II extend these findings to indicate that p210^BCR-ABL reduces expression of cadherin-8, -11, CD44H, CD18, and CD33 cell surface molecules. In contrast, MUC18, a homolog of the neural cell adhesion molecule whose expression correlates with the metastatic potential in melanoma (46), is up-regulated 18-fold in p210^BCR-ABL expressing cells (Table II). Together these findings suggest that p210^BCR-ABL significantly alters the expression of cell surface adhesion molecules that may contribute to premature progenitor cell release from the bone marrow.

Key regulatory proteins that mediate intracellular signaling through cadherin complexes are coordinately controlled by p210^BCR-ABL. Our data indicate that catenin/plakoglobin, a cytosolic protein that couples the intracellular portion of cadherin to the cytoplasmic proteins catenin and actin, is up-regulated 8.7-fold by p210^BCR-ABL (Table II). Likewise, the binding target of catenin/plakoglobin, catenin, is also strongly up-regulated in both microarray experiments (Tables II and III). Further work is needed to determine whether p210^BCR-ABL disrupts catenin-cadherin complex formation, thereby inducing altered intracellular signaling to produce the differentiation block seen in CML.

p210^BCR-ABL is a potent activator of multiple signaling pathways, including the p21^ras/mitogen-activated protein kinase pathway. p210^BCR-ABL activates the p21^ras/mitogen-activated protein kinase pathway through phosphorylation of the Ras-GTPase activating protein (Ras-GAP), SHC, and GRB2 molecules (47). This signaling pathway is important in BCR-ABL-mediated cellular transformation because inhibition of p21^ras blocks BCR-ABL transformation (14). We find that p210^BCR-ABL also strongly induces expression of RAP-GAP, a p21^rap1-specific GTPase. Rap1 is a member of the Ras family of small GTPases that antagonizes the growth promoting activity of Ras by competing for binding to Ras effectors, such as Raf-1 and p120^Ras-GAP (48). In addition to altering the balance of Ras/Rap GTPase signaling, we note that p210^BCR-ABL reduces abundance of the phospholipase C isoform and up-regulates the phospholipase C isoform by 21-fold (Table II). Similarly, alterations in PKC isoforms are found, with a down-regulation of PKC and up-regulation of PKC and PKC (49). We have shown that specific PKC isozymes play distinct roles in cell cycle progression, chemotherapeutic drug resistance, and cellular transformation in CML cells (13, 50, 51). Together, these findings demonstrate that expression of many intracellular signaling molecules is dramatically altered by p210^BCR-ABL expression.

Moreover, p210^BCR-ABL is a potent inducer of transcription factors and their regulators. Our study confirms the finding that p210^BCR-ABL strongly up-regulates expression of the inhibitor of DNA binding (Id, Table II, Ref. 27), a dominant negative regulator of helix loop helix DNA binding factors, including D47/E12, SCL, and lyl-1. The action of these helix loop helix transcription factors is important at specific stages of differentiation in hematopoiesis (52). High expression of Id proteins occurs in normal proliferating uni- and bi-potential hematopoietic cells and declines as the cells mature (52). Ectopic Id-1 expression inhibits the differentiation of erythroleukemic (53) and granulocytic cells (54). It is interesting to speculate that Id expression induced by p210^BCR-ABL is partially responsible for the block in terminal differentiation observed in CML.

p210^BCR-ABL induces extreme resistance to apoptosis in response to serum deprivation, irradiation, and chemotherapeutic agents (13, 18, 19). Earlier work has shown the pivotal role played by Bcl-x_L, an outer mitochrondrial membrane channel protein that prevents chemotherapeutic agent-induced cytochrome c release (19). Our study confirms that expression of Bcl-x_L is induced by p210^BCR-ABL. In this study, we have extended the mechanism for p210^BCR-ABL-induced resistance to apoptosis with the surprising finding that Hsp70 expression is also up-regulated. Inducible Hsp70 expression was identified by cDNA- and oligonucleotide-based microarrays and independently confirmed by Northern and Western blot analysis. Heat shock proteins are a highly conserved family of proteins that prevent protein denaturation and promote inactivated protein disaggregation through a ATP-dependent mechanism (55). Hsp70 blocks apoptosis through multiple mechanisms (56). In particular, cytoplasmic Hsp70 binds the caspase recruitment domain of Apaf-1, preventing binding to and activation of procaspase 9 (40). Hsp70 also binds apoptosis inducing factor, a mitochondrial intermembrane flavoprotein that produces chromatin condensation and cell death in a caspase-independent manner (28). Finally, Hsp70 has recently been shown to be an endogenous inhibitor of the apoptosis signal-inducing kinase-1 kinase (30). Here, Hsp70 directly binds apoptosis signal-inducing kinase-1, preventing its oligomerization and interfering with apoptotic signaling. Together, these findings indicate that Hsp70 exerts multiple anti-apoptotic effects that could act in concert with those of Bcl-x_L.

Hsp70 induction is largely regulated at the transcriptional level in response to heat shock and mitogenic agents (41, 42). The major identified regulatory site controlling Hsp70 expression is the HSE, a series of inverted repeats of the sequence 5'-NGAAN-3' that bind to oligomeric HSF. Interestingly, the HSE does not play an important role in Hsp70 promoter induction by p210^BCR-ABL. Instead, our findings indicate that p210^BCR-ABL acts on several cis elements in the Hsp70 promoter, the most important of which is the core GATA-1 binding site located at nt –82 to –58. This site primarily binds the GATA-1 isoform in CML cells. The observation that mutation of the GATA-RE renders the Hsp70 promoter unresponsive to p210^BCR-ABL, along with the finding that a multimeric GATA-RE is sufficient to confer BCR-ABL inducibility to an inert promoter strongly support the conclusion that this regulatory site is a bona fide BCR-ABL target. GATA-1 is a zinc finger containing transcription factor that plays a central role in development of erythrocytes and megakaryocytes. This transcription factor is required for expression of globin, heme biosynthesis, and red blood cell transcription factor genes (43). In GATA-1^–/– mice, erythroid precursors undergo maturation arrest and apoptosis, indicating that GATA-1 also plays a role in cellular survival (57). Additional studies have demonstrated that GATA-1 promotes erythroid cell survival by regulating Bcl-x_L expression of, in conjunction with, the growth factor, erythropoietin (58). Our studies indicate that Hsp70 expression is also coordinated through the actions of GATA-1 in combination with p210^BCR-ABL in a manner that does not result in significant changes in DNA binding abundance or affinity. The transcriptional actions of GATA-1 are regulated by a variety of mechanisms, including association with coactivators, friend of GATA (59), and by changes in its acetylation/deacetylation state. Determining which of these mechanisms is altered by p210^BCR-ABL will require further study. The previous observations that GATA-1 plays a critical role in Bcl-x_L expression, coupled with the present demonstration that GATA-1 regulates Hsp70 expression provides further insight into how the actions of this transcription factor is critical for cellular survival. It should be noted that BCR-ABL appears to induce its anti-apoptotic program through activation of multiple pathways that impinge on many apparently distinct transcription factors. For example, we recently demonstrated that BCR-ABL-mediated induction of PKC expression (an event also important in conferring chemotherapy resistance) is regulated primarily that an ELK1-like element in the proximal PKC promoter (49). In separate studies, we have shown the actions of PKC on cellular survival are, in turn, dependent on the NF-B pathway (16).

In summary, our studies demonstrate that the BCR-ABL oncoprotein induces global genetic changes in target cells. Among these are regulatory proteins controlling apoptotic pathways, cell cycle regulation, cell signaling (receptors and intracellular kinases), transcription, and cellular adhesion. Genomic signaling initiated by BCR-ABL and mediated through the GATA-RE affects the expression of Hsp70 and Bcl-x_L, two anti-apoptotic proteins important in conferring cellular resistance to chemotherapeutic agents. Based on these findings, we hypothesize that targeting the BCR-ABL-GATA signaling pathway may enhance the chemotherapeutic treatment of drug-resistant Ph+ leukemias.

	FOOTNOTES

 
^* This work was supported in part by National Institutes of Health Grants CA56869 (to A. P. F.), AI40218 (to A. R. B.), and NIEHS ES06676 (to J. Halpert, University of Texas Medical Branch). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Division of Endocrinology, MRB 8.138, The University of Texas Medical Branch, 301 University Blvd., Galveston, TX 77555-1060. Tel.: 409-772-2824; Fax: 409-772-8709; E-mail: arbrasie{at}utmb.edu.

¹ The abbreviations used are: CML, chronic myelogenous leukemia; p210^BCR-ABL, 210-kDa BCR-ABL fusion oncoprotein; PKC, protein kinase C; Apaf-1, apoptotic protease activation factor-1; Hsp, heat shock protein; GATA-RE, GATA-response element; nt, nucleotide(s); siRNA, short interfering RNA; ANOVA, analysis of variance; EMSA, electrophoretic mobility shift assay; WT, wild type(s); HSE, heat shock element; MOPS, 4-morpholinepropanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Faderl, S., Talpaz, M., Estrov, Z., O'Brien, S., Kurzrock, R., and Kantarjian, H. (1999) N. Engl. J. Med. 341, 164–172[Free Full Text]
2. Rowley, J. D. (1973) Nature 243, 290–293[CrossRef][Medline] [Order article via Infotrieve]
3. Raitano, A. B., Whang, Y. E., and Sawyers, C. L. (1997) Biochim. Biophys. Acta 1333, F201–F216[Medline] [Order article via Infotrieve]
4. Melo, J. V. (1996) Blood 88, 2375[Free Full Text]
5. Kurzrock, R., Gutterman, J. U., and Talpaz, M. (1988) N. Engl. J. Med. 319, 990–998[Medline] [Order article via Infotrieve]
6. Konopka, J., Watanabe, S., and Witte, O. (1984) Cell 37, 1035[CrossRef][Medline] [Order article via Infotrieve]
7. Sirard, C., Laneuville, P., and Dick, J. E. (1994) Blood 83, 1575[Abstract/Free Full Text]
8. Daley, G. Q., Van Etten, R. A., and Baltimore, D. (1990) Science 247, 824–830[Abstract/Free Full Text]
9. Van Etten, R. A. (2002) Oncogene 21, 8643–8651[CrossRef][Medline] [Order article via Infotrieve]
10. Wilda, M., Fuchs, U., Wossmann, W., and Borkhardt, A. (2002) Oncogene 21, 5716–5724[CrossRef][Medline] [Order article via Infotrieve]
11. McGahon, A., Bissonnette, R., Schmitt, M., Cotter, K. M., Green, D. R., and Cotter, T. G. (1994) Blood 83, 1179–1187[Abstract/Free Full Text]
12. Goldman, J. M., and Druker, B. J. (2001) Blood 98, 2039–2042[Abstract/Free Full Text]
13. Jamieson, L., Carpenter, L., Biden, T. J., and Fields, A. P. (1999) J. Biol. Chem. 274, 3927–3930[Abstract/Free Full Text]
14. Zou, X., and Calame, K. (1999) J. Biol. Chem. 274, 18141[Free Full Text]
15. Reuther, J. Y., Reuther, G. W., Cortez, D., Pendergast, A. M., and Baldwin, A. S., Jr. (1998) Genes Dev. 12, 968–981[Abstract/Free Full Text]
16. Lu, Y., Jamieson, L., Brasier, A. R., and Fields, A. P. (2001) Oncogene 20, 4777–4792[CrossRef][Medline] [Order article via Infotrieve]
17. Zou, X., and Calame, K. (1999) J. Biol. Chem. 274, 18141–18144[Free Full Text]
18. Bedi, A., Barber, J. P., Bedi, G. C., el-Deiry, W. S., Sidransky, D., Vala, M. S., Akhtar, A. J., Hilton, J., and Jones, R. J. (1995) Blood 86, 1148–1158[Abstract/Free Full Text]
19. Amarante-Mendes, G. P., McGahon, A. J., Nishioka, W. K., Afar, D. E., Witte, O. N., and Green, D. R. (1998) Oncogene 16, 1383–1390[CrossRef][Medline] [Order article via Infotrieve]
20. Earnshaw, W. C., Martins, L. M., and Kaufmann, S. H. (1999) Annu. Rev. Biochem. 68, 383–424[CrossRef][Medline] [Order article via Infotrieve]
21. Adams, J. M., and Cory, S. (2002) Curr. Opin. Cell Biol. 14, 715–720[CrossRef][Medline] [Order article via Infotrieve]
22. Martins, L. M., Mesner, P. W., Kottke, T. J., Basi, G. S., Sinha, S., Tung, J. S., Svingen, P. A., Madden, B. J., Takahashi, A., McCormick, D. J., Earnshaw, W. C., and Kaufmann, S. H. (1997) Blood 90, 4283–4296[Abstract/Free Full Text]
23. Amarante-Mendes, G. P., Naekyung, K. C., Liu, L., Huang, Y., Perkins, C. L., Green, D. R., and Bhalla, K. (1998) Blood 91, 1700–1705[Abstract/Free Full Text]
24. Olson, M. R., Holley, C. L., Yoo, S. J., Huh, J. R., Hay, B. A., and Kornbluth, S