Effect of Mutational Inactivation of Tyrosine Kinase Activity on BCR/ABL-Induced Abnormalities in Cell Growth and Adhesion in Human Hematopoietic Progenitors

Home , FER (gene)

[CANCER RESEARCH 64, 5322–5331, August 1, 2004] Effect of Mutational Inactivation of Tyrosine Kinase Activity on BCR/ABL-Induced Abnormalities in Cell Growth and Adhesion in Human Hematopoietic Progenitors

Pandurangan Ramaraj,1 Harjeet Singh,1 Ning Niu,1 Su Chu,1 Melissa Holtz,1 Jiing Kuan Yee,2 and Ravi Bhatia1 Divisions of 1Hematology and Bone Marrow Transplantation and 2Virology Research, City of Hope National Medical Center, Duarte, California

ABSTRACT (8–11). Primary CML hematopoietic progenitors have also been reported to be GF independent for proliferation and survival, although Chronic myelogenous leukemia (CML) results from transformation of this has not been observed consistently (12–15). In addition, CML a primitive hematopoietic cell by the BCR/ABL gene. The specific BCR/ progenitors are insensitive to inhibition by chemokines that inhibit ABL signaling mechanisms responsible for transformation of primitive human hematopoietic cells are not well defined. Previous studies have normal progenitor proliferation (16, 17). CML progenitors demon- suggested that constitutively activated tyrosine kinase activity plays an strate deficient integrin-mediated adhesion and impaired integrin- important role for in abnormal proliferation of CML progenitors but has mediated inhibition of proliferation, although normal levels of ␣4, ␣5, not clearly defined its role in abnormal adhesion and migration. We and ␤1-integrin receptors are expressed (18–23). CML progenitors established a human progenitor model of CML by ectopic expression of also demonstrate increased spontaneous motility on fibronectin (23, BCR/ABL in normal CD34؉ cells using retrovirus-mediated gene trans- 24). However, directed migration to a gradient of the chemokine fer. CD34؉ cells expressing BCR/ABL demonstrated several features stromal cell-derived factor 1 (SDF-1␣) is reduced (23, 25, 26). These characteristic of primary CML progenitors including increased prolifer- abnormalities may explain in part some of the clinical features of ation in committed and primitive progenitor culture, reduced adhesion to CML, including abnormal myeloproliferation, peripheral circulation fibronectin, and reduced chemotaxis to stroma-derived factor-1␣.We expressed a kinase-inactive BCR/ABL gene to directly investigate the role of hematopoietic progenitors, and extramedullary hematopoiesis. of kinase activity in abnormal progenitor function. Abnormalities in Improved understanding of the molecular mechanisms that lead to proliferation were completely reversed, whereas defects in adhesion and human progenitor transformation is required for a better understand- migration were significantly improved but not completely reversed in cells ing of CML pathogenesis and to assist development of additional expressing a kinase-inactive BCR/ABL. Furthermore, the BCR/ABL ki- mechanism-based therapies. Studies performed in immortalized cell nase inhibitor imatinib mesylate markedly inhibited proliferation of BCR/ lines have identified several potentially important BCR/ABL signal- ABL-expressing progenitors but did not fully correct the adhesion and ing domains and downstream signaling pathways (27). However, the migration defects. Expression of BCR/ABL genes with deletions of either contribution of different mechanisms to transformation can vary con- the COOH-terminal actin binding or proline-rich domains resulted in siderably from one cell type to the other, and the role of these enhanced adhesion and chemotaxis compared with wild-type BCR/ABL but did not affect progenitor proliferation. We conclude that abnormal mechanisms in transformation of primitive human hematopoietic cells kinase activity is essential for abnormal proliferation and survival of CML in which CML arises is unclear. One approach has been to express the progenitors but that abnormal adhesion and migration result from both BCR/ABL gene in murine stem cells using retroviral transduction kinase-dependent and -independent mechanisms. followed by transplantation to irradiated hosts, resulting in induction of a myeloproliferative disorder (5, 6, 28). This murine CML model has been used to study of the role of different BCR/ABL signaling INTRODUCTION domains in hematopoietic cell transformation (29–32). However, Chronic myelogenous leukemia (CML) is a lethal hematological murine disease differs from clinical CML in being highly aggressive disorder resulting from transformation of a primitive hematopoietic and fulminant and in often progressing to acute T-cell leukemia. cell. Malignant cells in CML are characterized by a balanced trans- These differences may be related to species-specific differences be- location between chromosomes 9 and 22, fusing a truncated BCR gene tween human and murine stem and progenitor cells. An approach that to sequences upstream of the second exon of ABL (1, 2). The resulting is highly relevant to clinical disease is to study primary progenitor BCR/ABL fusion oncogene encodes a protein tyrosine kinase with cells from CML patients. However, studies of molecular mechanisms elevated and dysregulated enzymatic activity (3). The BCR/ABL gene are limited by the low numbers of such cells that can be obtained, has been shown to play a critical role in the pathogenesis of CML considerable intersample variability related to the acquisition of ad- (4–6). Clinically, CML presents with a myeloproliferative disorder ditional genetic abnormalities and prior therapeutic exposures, the Ϫ ϩ characterized by a vast expansion of hematopoietic progenitor, pre- presence of both Ph and Ph cell populations, and difficulty in cursor, and mature cells (7). Other important features of CML are the performing structure-function correlations. A human model of CML BCR/ABL presence of increased numbers of circulating progenitors and ex- based on expression of the p210 transgene in primary human ϩ tramedullary hematopoiesis. CD34 cells could be potentially very useful to investigate specific CML committed and primitive progenitors demonstrate increased mechanisms resulting in progenitor transformation relevant to clinical sensitivity to growth factor (GF)-induced proliferation and maturation disease and for preclinical evaluation of new therapeutic interventions for CML.

Received 11/21/03; revised 4/13/04; accepted 5/26/04. Several laboratory and clinical observations suggest that abnormal Grant support: Leukemia and Lymphoma Society Translational Research Grant 6468 kinase activity plays a critical role in BCR/ABL-induced transforma- and the American Cancer Society Grant RPG-99-202-01-LBC (R. Bhatia). R. Bhatia is a tion (3, 33, 34). However, recent reports indicate that inactivation of Clinical Scholar of the Leukemia and Lymphoma Society. Note: P. Ramaraj, H. Singh, and N. Niu contributed equally to this work. P. Ramaraj kinase activity does not reverse abnormal adhesion of BCR/ABL- is currently at the Division of Endocrinology, Metabolism & Molecular Medicine, Charles expressing cell lines and that this abnormality results from kinase- R. Drew University, Los Angeles, CA. independent mechanisms. In contrast, in previous studies we found 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 that treatment of primary CML progenitors with a pharmacological 18 U.S.C. Section 1734 solely to indicate this fact. BCR/ABL kinase inhibitor led to significant reversal of the adhesion Requests for reprints: Ravi Bhatia, Division of Hematology and Bone Marrow Transplantation, City of Hope National Medical Center, Duarte, CA 91010. Phone: defect. However pharmacological inhibitors could potentially affect (626) 359-8111, extension 62683; Fax: (626) 301-8973; E-mail: [email protected]. cellular adhesion through effects on targets other than the BCR/ABL 5322

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2004 American Association for Cancer Research. BCR/ABL KINASE INACTIVATION IN HUMAN PROGENITORS kinase (33, 35). In the present study we established a human progen- were lysed in buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1 mM itor model of CML based on retroviral expression of the p210BCR/ABL EDTA, 0.5% NP40, and 0.5% sodium deoxycholate, supplemented with pro- transgene in primary human CD34ϩ cells and used this model to tease and phosphatase inhibitors. Primary CML CD34ϩ cells were also ana- directly investigate the role of kinase activity as well as kinase- lyzed for comparison. Protein extracts were resolved on 5% SDS-PAGE gels, independent mechanisms in BCR/ABL-induced abnormalities in pro- transferred to nitrocellulose membranes, blocked in 10% nonfat milk in PBS with 0.1% Tween, and labeled with an appropriate dilution of primary antibody liferation and adhesion of human hematopoietic progenitors. [anti-Abl (Ab-3), anti-BCR (Ab-2; both from Oncogene Science, Cambridge, MA), antiactin (AC-15; Sigma), and antiphosphotyrosine (4G10; a kind gift MATERIALS AND METHODS from Dr. Brian Druker)], followed by horseradish peroxidase-conjugated sec- ondary antibody (1:8000; Jackson). Antibody detection was performed using Cells the Superfemto kit (Pierce Biotechnology, Rockford, IL). Human umbilical cord blood samples were obtained using protocols ap- proved by the Institutional Review Board of the City of Hope National Medical Evaluation of Progenitor Growth Center. Mononuclear cells were isolated using Ficoll-Hypaque density gradient Colony Forming Cell (CFC) Assays. CD34ϩGFPϩ cells were plated in ϩ separation and CD34 cell enriched populations selected using immuno- methylcellulose progenitor culture and assessed for the presence of CFU-Mix, magnetic column separation as described previously (11). CFU-GM, and BFU-E colonies as described previously (11). Where indicated, increasing concentrations of imatinib mesylate were added to the culture Vectors medium. ϩ ϩ The MIGR1 and MIG-210 vectors were gifts from Dr. Warren Pear (Uni- Liquid Culture with GFs. CD34 GFP cells were cultured in IMDM versity of Pennsylvania, Philadelphia, PA; Ref. 5). The control vector, MIGR1, with 30% FCS, and identical GF conditions to those used in CFC assays and hasa5Ј murine stem cell virus-based long terminal repeat, an internal ribo- the number of cells generated were counted after 14 days. Where indicated, some entry site sequence downstream of a multiple cloning site, and the green increasing concentrations of imatinib mesylate were added to the culture fluorescent protein (GFP) gene downstream of the internal ribosome entry site. medium. ϩ ϩ The MIG-210 vector has the p210 BCR-ABL oncogene (7.1 Kb) inserted into Single Cell Analysis. Single CD34 GFP cells were sorted into individ- the EcoRI site in the multiple cloning site, upstream of the internal ribosome ual wells of a 96-well plate and cultured as described above. After 14 days, entry site. A vector containing a kinase inactive p210BCR-ABL gene (MIG- plates were scanned to identify wells containing cells, and the number of cells 210KI) was generated by excision of the K1176R p210BCR-ABL mutant from was counted in individual wells. ϩ ϩ a pGDP210KI vector (a kind gift from Dr. Brian Druker, Oregon Health Long-Term Bone Marrow Culture. CD34 GFP cells were plated in Sciences University, Portland, OR; Ref. 36) and cloning into the MIG R1 triplicate in long-term bone marrow culture medium on M2–10B4 murine vector. Vectors containing p210BCR/ABL genes with deletions of the COOH- fibroblast feeders subcultured in 24-well plates as described previously (11). terminal actin-binding domain (MIG-210dA) and proline-rich region were After culture for 2, 4, and 6 weeks, cells were harvested, pooled, and plated in generated by excising the respective mutant genes from pBabe-BCR-ABL-AD CFC culture and the number of colonies evaluated. (a kind gift from Dr. Ruibao Ren, Brandeis University, Waltham, MA; Ref. 37) and a pGDP210⌬P1P2 (a kind gift from Dr. Brian Druker; Ref. 38) vectors and Adhesion Assays ligation into the EcoRI site of the MIG R1 vector. Plates (96-well) were adsorbed with Fibronectin CH-296 at a concentration Infectious virus particles were produced by transient transfection of 293 of 8 ␮g/cm2. Control wells were adsorbed with BSA (Sigma). CD34ϩGFPϩ cells with retroviral vector plasmid and pCL-ampho plasmid (a kind gift from cells were incubated in serum-free medium with low concentrations of GF as Dr. Martin Haas, University of California San Diego, San Diego, CA; Refs. 39, described above for 24 h. Where indicated, increasing concentrations of 40). Supernatants were collected 24–48 h after transfection and the number of imatinib mesylate were added to the culture medium. Cells were then washed infectious particles titrated by measuring efficiency of transduction of HT1080 and resuspended in IMDMϩBSA and plated in fibronectin-coated wells for cell lines. 2 h. Subsequently, nonadherent and adherent fractions were separated as Retroviral Transduction described previously (42). Both fractions were plated in methylcellulose progenitor culture, and the percentage of adherent CFC was calculated. CD34ϩ cells were cultured in serum-free medium (Stem Cell Technologies, Vancouver, British Columbia, Canada) supplemented with the following GFs: Migration Assays Flt-3 ligand (100 ng/ml), stem cell factor (50 ng/ml), thrombopoietin (100 ng/ml), interleukin-6 (10 ng/ml), and interleukin-3 (25 ng/ml) at 37°C with 5% Progenitor migration was assayed by evaluating the movement of progenitors from the upper to the lower chamber of 6.5-mm Transwells with filters of 5-␮m CO2 for 48 h in plates that had been precoated with Fibronectin CH-296 (Retronectin; Pan Vera, Madison, WI). Subsequently, cells were resuspended diameter pore size, as described previously (43). The membranes were coated on 2 in virus containing supernatants (multiplicity of infection ϭ 10), in the pres- both sides using a Fibronectin CH-296 solution (8 ␮g/cm ). CD34ϩGFPϩ cells ence of the same GFs, and plated on fibronectin-coated plates pre-exposed to were incubated in serum-free medium with low concentrations of GF as described viral supernatants for 30 min to allow binding of virus particles and thereby above for 24 h. Where indicated, increasing concentrations of imatinib mesylate increase the cell exposure to virus. This was repeated after 24 h. Cells were were added to the culture medium. Cells were then washed and placed in the upper harvested 48 h after the second virus exposure, washed, and labeled with chamber of the transwell. Assays were performed with or without the addition of anti-CD34-APC antibodies or isotype controls (Becton Dickinson, San Jose, SDF-1␣ (100 ng/ml) to the lower chamber. Cells that migrated to the lower CA), and CD34ϩGFPϩ cells were collected using a MoFlo flow cytometer chamber of the transwell after incubation at 37°C for 6 h were assayed for CFC in (Cytomation Inc., Fort Collins, CO). methylcellulose progenitor culture. The percentage of migrating CFC was calculated as: (migrating CFC/total CFC) ϫ100%. Western Blotting Adhesion-Mediated Signaling Western blotting was performed to detect BCR/ABL expression and protein tyrosine phosphorylation. CD34ϩGFPϩ cells were cultured in medium sup- CFC proliferation was evaluated using thymidine suicide assays. plemented with low concentrations of GF similar to those present in stroma- CD34ϩGFPϩ cells were cultured in serum-free medium with low concentra- conditioned medium [granulocyte-macrophage colony stimulating factor (200 tions of GF for 48 h. Subsequently, cells were washed and resuspended in pg/ml), granulocyte colony stimulating factor (1 ng/ml), stem cell factor (200 IMDM with 0.3% BSA and incubated for4hinwells coated with fibronectin pg/ml), leukemia inhibitory factor (50 pg/ml), macrophage inhibitory protein CH-296 or control wells coated with BSA. The proliferation of CFC after 1␣ (200 pg/ml), and interleukin-6 (1 ng/ml; Ref. 41) ] for 16 h. For some culture with or without fibronectin was evaluated in thymidine suicide assays studies cell numbers were expanded by culture in GF containing medium. Cells as described previously (20). 5323

Statistics culture with virus-containing supernatants using an optimized protocol (Fig. 1A). The transduction results are summarized in Table 1 and Results of data obtained from multiple experiments were reported as the a representative experiment shown in Fig. 1B. Transduction efficiency mean Ϯ 1 SE. Significance levels were determined by Student’s t test analysis. was lower for MIG210 and MIG210KI vectors, possibly related to the presence of the large BCR/ABL insert upstream of the internal ribo- RESULTS some entry site and GFP. The efficiency of CFC transduction was also -CD34؉ Cell Transduction with BCR/ABL and Kinase-Inactive significantly lower for MIG210- and MIG210KI-exposed cells com BCR/ABL Expressing Murine Stem Cell Virus Vectors. Superna- pared with MIG (P Ͻ 0.05; n ϭ 5). Significantly greater cell expan- tants containing infectious murine stem cell virus retrovirus parti- sion was seen during transduction culture with MIG210 compared cles carrying the p210BCR/ABL gene (MIG210), a kinase-inactive with MIG vectors (P Ͻ 0.02; n ϭ 10). Expansion of MIG210KI p210BCR/ABL (MIG210KI), and controls containing GFP alone vector-exposed cells was less than that for MIG210-exposed cells (MIG) were generated by transient transfection of 293 cells. The (P ϭ 0.06) and did not differ significantly from cells exposed to MIG kinase-inactive BCR/ABL mutant has a point mutation changing the vectors. The yield of CD34ϩGFPϩ cells collected after flow cyto- lysine residue at position 1176 to arginine, referred to as K1176R, that metric sorting was lower for MIG210- and MIG210KI-exposed cells inactivates the tyrosine kinase activity of BCR/ABL This mutant has compared with MIG-exposed cells (P Ͻ 0.05; n ϭ 10). This notwith- been well characterized in previous in vitro studies and in the murine standing, this procedure resulted in consistent and reliable collection CML model (28, 36, 44). Human CD34ϩ cells were transduced by of sufficient numbers of transduced cells for subsequent analyses.

Fig. 1. Retroviral transduction of CD34ϩ cells with BCR/ABL genes. A, human CD34ϩ cells were transduced by culture in the presence of virus- containing supernatants as shown and described in “Materials and Methods.” B, the results of a representative experiment are shown. C, cell lysates were made from transduced CD34ϩgreen fluorescent protein (GFP)ϩ cells and primary chronic myelogenous leukemia (CML) CD34ϩ cells, and BCR/ABL expression was assessed by Western blotting (WB). Blots were reprobed with an antiactin antibody to confirm equal sample loading. D, assessment of protein tyrosine phosphorylation was performed by probing blots with antiphosphotyrosine (anti-PY) antibodies. FACS, fluorescence- activated cell sorter; IL, interleukin; SCF, stem cell factor; TPO, thrombopoietin; FL, Flt-3 ligand.

5324

Table 1 Results of transduction of human CD34ϩ cells with MIG, MIG210, and Proliferation and Survival of BCR/ABL and Kinase-Inactive MIG210KI vectors BCR/ABL-Expressing CD34؉ Cells. Selected CD34ϩGFPϩ cells The percentage of transduction of total cells and CD34ϩ cells was determined by flow cytometry analysis. The percentage of transduction of CFCa was determined by compar- were plated in methylcellulose progenitor assays to evaluate commit- ing the number of GFPϩ colonies detected by immunofluorescence microscopy with total ted progenitor growth. Cultures initiated with MIG210-transduced number of colonies detected by light microscopy. The fold cell expansion was determined by comparing the total number of cells present after transduction culture with the number cells were striking for the presence of very large colonies, which of cells initially placed in culture. The total number of CD34ϩGFPϩ cells obtained after included CFU-GM, BFU-E, and CFU-Mix colonies (Fig. 2, A and B). flow cytometry sorting is also shown. Results represent mean Ϯ SE of multiple experi- These large colonies were not observed in cultures initiated with ments. MIG210KI-transduced CD34ϩ cells. There was no significant differ- MIG MIG210 MIG210KI ence in total CFC cloning efficiency of CD34ϩ cells transduced with % transduction the different vectors (Fig. 2C). However, the numbers of CFU-Mix Total cells 61.5 Ϯ 7.6 18.9 Ϯ 3.7 17.1 Ϯ 3.0 CD34ϩ cells 62.1 Ϯ 8.0 17.4 Ϯ 3.6 17.3 Ϯ 3.3 were significantly increased in cells transduced with MIG210 vectors CFC 62.6 Ϯ 6.3 40.2 Ϯ 6.6 29.4 Ϯ 6.6 [CFU-Mix per 1000 cells (mean Ϯ SE) of 2.8 Ϯ 1.1 for MIG, Fold cell expansion 5.5 Ϯ 1.3 9.0 Ϯ 1.5 6.6 Ϯ 1.4 Ϯ Ϯ CD34ϩGFPϩ cells obtained (ϫ103) 1250 Ϯ 685 510 Ϯ 141 345 Ϯ 131 14.5 5.1% for MIG210, and 2.2 1.1 for MIG210KI-transduced ϩ Ͻ a CFC, colony forming cell; GFP, green fluorescent protein. CD34 cells, respectively; P 0.05 comparing MIG210 with MIG or MIG210KI; n ϭ 8]. The total number of cells generated during committed progenitor culture was assessed by growing CD34ϩ cells BCR/ABL expression in transduced CD34ϩGFPϩ cells was as- in similar serum and GF culture conditions as for methylcellulose sessed by Western blotting. The p210BCR/ABL protein was detected in progenitor culture, but substituting IMDM for methylcellulose. MIG210- and MIG210KI-transduced CD34ϩ cells at similar levels of CD34ϩ cells transduced with MIG210 vectors generated significantly expression and, as expected, was not seen in control MIG-transduced greater numbers of cells than MIG-transduced cells, and this abnor- cells (Fig. 1C). BCR/ABL expression levels in transduced CD34ϩ mality was reversed in cells expressing kinase-inactive BCR/ABL cells appeared to be increased compared with CD34ϩ cells obtained (Fig. 2D). These results confirm the observation of increased colony from CML patients. Phosphotyrosine blotting confirmed increased size (Fig. 2E) and indicate that BCR/ABL-induced increased clonal levels of protein tyrosine phosphorylation in MIG210-transduced proliferation of individual CD34ϩ cells was reversed with kinase CD34ϩ cells but not in MIG210KI-transduced cells, confirming inactivation. This was confirmed by analysis of single CD34ϩGFPϩ kinase inactivation in the mutant (Fig. 1D). cells sorted into individual wells of a 96-well plate and cultured as

Fig. 2. Growth of BCR/ABL-expressing CD34ϩ cells in progenitor culture. CD34ϩgreen fluorescent protein (GFP)ϩ cells selected by flow cytometry were plated in methylcellulose progenitor culture for 14 days at 37°C and 5% CO2 and assessed for the presence of colony forming cell (CFC) as described in “Materials and Methods.” A, cultures initiated with MIG210-transduced cells were striking for the presence of very large colonies. B, fluorescent microscopy confirmed GFP expression in cells composed of the colony. C, the total CFC frequency per 1000 CD34ϩ cells is shown (n ϭ 8). D, the total number of cells generated during committed progenitor culture was assessed by plating CD34ϩGFPϩ cells in liquid culture using identical serum and growth factor conditions as used in CFC assays. The number of cells generated per 1000 CD34ϩ cells is shown (n ϭ 5). E, the ratio of cells generated in liquid culture to colonies generated in semisolid methylcellulose culture from 1000 CD34ϩ cells is shown (n ϭ 5). Results of data obtained from multiple experiments are reported as the mean; bars, ϮSE. Significance levels determined by Student’s t test ,P Ͻ 0.01 ,ءء ;P Ͻ 0.05 ,ء :analysis are shown MIG versus MIG210 vector-transduced cells; †, P Ͻ 0.05; ††, P Ͻ 0.01, MIG210 versus MIG210KI vector-transduced cells.

5325

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2004 American Association for Cancer Research. BCR/ABL KINASE INACTIVATION IN HUMAN PROGENITORS described for bulk cultures. The number of cells/clone (mean Ϯ SE) Progenitor migration was measured in a transwell assay. Spon- after 14 days of culture was 925 Ϯ 381 for controls, 18,300 Ϯ 5,134 taneous migration through fibronectin-coated transwells appeared for wild-type BCR/ABL-expressing cells, and 1,757 Ϯ 505 for to be higher for MIG210-transduced CD34ϩ cells compared with kinase-inactive BCR/ABL-expressing cells. The percentage of clones MIG-transduced cells, but the difference did not reach statistical that generated Ͼ10,000 cells was 2%, 16%, and 3.5% for control, significance (3.6 Ϯ 0.8% for MIG versus 6.6 Ϯ 2.0% for MIG210- BCR/ABL-expressing, and kinase-inactive BCR/ABL-expressing transduced cells; P ϭ 0.08). Spontaneous migration was also not cells, respectively. significantly different between wild-type and kinase-inactive BCR/ The ability of committed progenitors cells to withstand GF depri- ABL-expressing cells (6.6 Ϯ 2.0% for MIG210 versus 8.0 Ϯ 2.7% vation was assessed by culture of CD34ϩGFPϩ cells in serum-free for MIG210KI-transduced cells). In contrast, directed migration to medium in the absence of GF for 96 h followed by assessment of CFC a gradient of the chemokine SDF-1␣ was significantly reduced in capacity. Increased survival of CFC was seen with p210BCR/ABL- MIG210-transduced cells compared with controls (Fig. 4B). Che- transduced cells, compared with controls (55.3 Ϯ 27.0% of input motaxis to SDF-1␣ was increased significantly in cells expressing number of CFC for BCR/ABL-expressing cells, compared with kinase-inactive BCR/ABL compared with MIG210-transduced 13.6 Ϯ 7.4% for controls; P ϭ 0.05; n ϭ 8). However, survival of cells. However, migration of kinase-inactive BCR/ABL-expressing CFC-expressing kinase-inactive p210BCR/ABL was significantly lower cells remained significantly less than for control cells. than BCR/ABL-expressing CFC and similar to those of control CFC Contact with fibronectin results in integrin-dependent inhibition of (15.1 Ϯ 4.0% of input number of CFC). proliferation of normal progenitors when studied in low, physiological The primitive progenitor potential of transduced CD34ϩ cells concentrations of GFs. This mechanism has been shown to be per- was assessed by evaluating their ability to generate CFC after turbed in CML progenitors. Using a thymidine suicide assay to culture on stromal adherent layers for 2– 6 weeks. p210BCR/ABL- measure the percentage of proliferating CFC, we observed that contact transduced cells generated significantly increased numbers of CFC with fibronectin resulted in significant inhibition of proliferation of after 2, 4, and 6 weeks of long-term bone marrow culture compared MIG-transduced cells (26.2 Ϯ 3.3% reduction in S-phase cells on with controls (Fig. 3). Enhanced growth in long-term bone marrow fibronectin compared with BSA controls; P Ͻ 0.001 for fibronectin culture was reversed in cells transduced with kinase-inactive compared with BSA controls; n ϭ 5) but did not inhibit proliferation p210BCR/ABL. of MIG210-transduced cells (3.2 Ϯ 1.3% reduction in S-phase cells The above studies indicate that increased proliferation and sur- on fibronectin compared with BSA controls; P Ͻ 0.005 compared vival of BCR/ABL-transformed committed and primitive progen- with MIG-transduced cells). In contrast, proliferation of MIG210KI- itors are completely reversed by inactivation of the BCR/ABL transduced cells was inhibited by fibronectin (19.3 Ϯ 3.6% reduction tyrosine kinase. in S-phase cells on fibronectin compared with BSA controls; P Ͻ 0.01 Adhesion and Migration of BCR/ABL and Kinase-Inactive compared with MIG210-transduced cells; P ϭ 0.08 compared with .(BCR/ABL-Expressing CD34؉ Cells. Adhesion assays were per- MIG-transduced cells formed to evaluate the effect of BCR/ABL expression on CD34ϩ The above results indicate that abnormalities in adhesion and mi- progenitor adhesion to the extracellular matrix protein fibronectin. gration in p210BCR/ABL-transduced CD34ϩ cells are significantly MIG210-transduced cells demonstrated significantly reduced adhe- reduced but are not completely reversed after kinase inactivation. sion to fibronectin compared with MIG-transduced controls (Fig. 4A). Effect of Imatinib Mesylate on BCR/ABL-Expressing Progen- In contrast, adhesion to fibronectin was significantly greater for itors. We investigated the effect of the BCR/ABL kinase inhibitor MIG210KI-transduced cells compared with MIG210-transduced cells. imatinib mesylate on BCR/ABL-induced proliferation, adhesion, Adhesion of MIG210KI-transduced cells was also less than for con- and migration abnormalities in MIG210-transduced CD34ϩ cells. trol MIG-transduced cells, although the difference did not reach The concentrations of imatinib used correspond to levels achieved statistical significance (P ϭ 0.07). in patients being treated with imatinib (34). Exposure to imatinib mesylate markedly reduced the number of CFC generated from BCR/ABL-expressing CD34ϩ cells and the numbers of cells generated after GF culture (Fig. 5, A and B). However, imatinib also significantly inhibited the growth of cells transduced with MIG vectors. Inhibition of MIG-transduced cells was significantly less than for MIG210-transduced cells (P Ͻ 0.05 for MIG210 compared with MIG-transduced cells treated with 0.2 and 1 ␮M imatinib). In contrast, imatinib exposure did not alter adhesion of MIG210-transduced progenitors to fibronectin (Fig. 5C). However, imatinib significantly reduced adhesion of MIG-transduced progenitors to fibronectin. Imatinib exposure resulted in significantly increased chemotaxis of MIG210-transduced progenitors to an SDF-1␣ gradient (Fig. 5D), but migration remained significantly less than that of MIG-transduced cells (P Ͻ 0.05 for MIG210- Fig. 3. Growth of BCR/ABL-expressing CD34ϩ cells in long-term bone marrow transduced cells treated with 0.2–5 ␮M imatinib compared with culture. CD34ϩgreen fluorescent proteinϩ cells were plated in triplicate in long-term MIG-transduced cells). Imatinib did not affect chemotaxis of MIG- bone marrow culture medium on M2–10B4 murine fibroblast feeders subcultured in 24-well plates. Cultures were maintained at 37°C in a humidified atmosphere with 5% transduced cells. These results indicate that imatinib treatment, CO2 and fed at weekly intervals by removal of half the medium from the wells and although profoundly reducing abnormally increased proliferation replacement with fresh medium. After culture for 2, 4, and 6 weeks, all of the nonadherent of BCR/ABL-expressing progenitors, does not completely correct and adherent cells were harvested, pooled, and plated in colony forming cell (CFC) culture and the number of colonies evaluated. Results of data obtained from six separate the defect in migration and does not alter progenitor adhesion. experiments are reported as the mean; bars, ϮSE. Significance levels determined by However, imatinib also significantly reduces proliferation and P ϭ 0.01, MIG versus MIG210 ,ءء ;P Ͻ 0.05 ,ء :Student’s t test analysis are shown vector-transduced cells; †, P Ͻ 0.05, MIG210 versus MIG210KI vector-transduced cells; adhesion of control progenitors expressing GFP alone, which must LTC, long-term culture. be considered while interpreting these results. 5326

Fig. 4. Adhesion and migration of BCR/ABL-expressing progenitors. A, progenitor adhesion was measured by incubating CD34ϩgreen fluorescent proteinϩ cells on fibronectin CH-296 (8 ␮g/cm2) or control BSA-adsorbed plates for 2 h. Nonadherent and adherent fractions were collected, and the percentage of adherent colony forming cell (CFC) calculated {% adherent CFC ϭ [adherent CFC Ϭ (adherent CFC ϩ nonadherent CFC)] ϫ100%}. B, progenitor migration was measured in a transwell assay by evaluating the movement of CD34ϩgreen fluorescent proteinϩ progenitors from the upper to the lower chamber of transwells with filters of 5-␮m diameter pore size. Transwell membranes were adsorbed with fibronectin CH-296 (8 ␮g/cm2). Stromal cell-derived factor 1 (SDF-1␣; 100 ng/ml) was added to the lower chamber. The percentage of CFC that migrated to the lower chamber over 6 h is shown. Results represent data obtained from five separate experiments and are reported as the mean; bars, Ϯ1 SE. Significance levels determined by Student’s t test analysis .P Ͻ 0.005, versus MIG-transduced cells; † P Ͻ 0.05, MIG210 versus MIG210KI vector-transduced cells ,ءء ;P Ͻ 0.05 ,ء :are shown

Proliferation, Adhesion, and Migration of Progenitors Express- itors was also observed. While these studies were in progress, two ing Actin-Binding Domain and Proline-Rich Domain-Deleted additional studies of human progenitor transduction with BCR/ABL BCR/ABL Gene Mutants. To additionally investigate potential have been reported (45, 46). Our results are consistent with those of mechanisms contributing to abnormal adhesion and migration of Zhao et al. (45), who also observed increased proliferation and sur- BCR/ABL-expressing progenitors, vectors containing BCR/ABL vival of BCR/ABL-transduced committed progenitors. Here we addi- genes with deletion of the COOH-terminal actin-binding domain tionally demonstrate that abnormal proliferation extends to more (MIG210dA) or the proline-rich domain (MIG210dP1P2) were primitive progenitors capable of generating hematopoiesis in long- generated and used to transduce CD34ϩ cells. These mutants have term bone marrow culture. In addition, BCR/ABL expressing human been well characterized in cell lines, but their role in transforma- CD34ϩ cells had significantly reduced adhesion to fibronectin and tion of primary human progenitors has not been studied. We significantly impaired inhibition of proliferation after coculture with confirmed that these mutants were expressed at similar levels as fibronectin. Zhao et al. (45) also observed reduced adhesion to fi- wild-type BCR/ABL-transduced cells by Western blotting using an bronectin in BCR/ABL-transduced human progenitors. We also dem- anti-BCR antibody (Fig. 6A). Phosphotyrosine blotting indicated onstrate that migration in response to the chemokine SDF-1␣ was that the deletion mutants retained tyrosine kinase activity (Fig. 6B). impaired significantly in BCR/ABL-expressing progenitors compared ϩ CD34 cells expressing either of the mutant genes proliferated in with controls. In contrast to our results and those of Zhao et al., CFC (data not shown) and liquid GF culture (Fig. 6C) to a similar Chalandon et al. (46) reported that BCR/ABL expression in human extent as cells expressing wild-type BCR/ABL. In contrast, ex- CD34ϩ cells was associated with increased erythroid differentiation pression of these mutants resulted in enhanced progenitor adhesion response, including granulocytic lineage programmed cells. Increased to fibronectin compared with wild-type BCR/ABL (Fig. 6D). In erythroid differentiation response is not a common feature of CML, addition, progenitors expressing either mutant demonstrated en- and the observed abnormalities in erythroid differentiation may be hanced migration toward an SDF-1␣ gradient compared with cells explained by much higher levels of BCR/ABL expression in trans- expressing wild-type BCR/ABL (Fig. 6E). These mutants did not duced CD34ϩ cells in the study by Chalandon et al. (46). affect spontaneous migration on fibronectin (data not shown). The abnormalities resulting from ectopic expression of BCR/ABL These results suggest that the COOH-terminal actin binding and in CD34ϩ cells are similar to those observed in committed and proline-rich domains of BCR/ABL contribute to abnormal adhe- primitive progenitors from CML patients. CML CD34ϩ progenitors sion and migration of BCR/ABL expressing human progenitors but do not contribute significantly to abnormal proliferation. also demonstrate increased proliferation in response to GF stimulation and resistance to apoptosis after GF withdrawal (8–12), reduced ␤1 integrin-mediated adhesion and signaling but increased mobility on DISCUSSION fibronectin (19, 20, 22), and impaired migration in response to We have established a human model of CML based on retrovirus- SDF-1␣ (23–26, 43). CD34ϩ progenitors ectopically expressing mediated transfer of the BCR/ABL gene into CD34ϩ progenitor cells BCR/ABL appeared to be more proliferative in progenitor culture and and have used this model to evaluate the role of specific BCR/ABL more resistant to apoptosis on GF withdrawal than we have observed signaling mechanisms in abnormal cellular regulation in CML pro- previously for primary CML progenitors. The use of the murine stem genitors. cell virus promoter results in enhanced BCR/ABL expression in Ectopic expression of p210BCR/ABL in human hematopoietic pro- transduced CD34ϩ cells that is not subject to normal regulatory genitors resulted in significantly increased progenitor proliferation. influences, which may result in enhanced cellular proliferation during Although CFC frequency was unchanged, very large colonies with in vitro culture (47). It is also possible that cord blood CD34ϩ significantly increased numbers of cells per colony were seen indi- progenitors, used for their ease of transduction, may be more prolif- cating increased clonal proliferation of individual progenitor cells. erative than their adult counterparts in their response to BCR/ABL Increased GF-independent survival of BCR/ABL-transduced progen- expression. Despite these limitations, the model described here 5327

Fig. 5. Effect of imatinib mesylate on BCR/ABL-expressing progenitors. The effect of increasing concentrations of imatinib on MIG210 and control MIG vector-transduced CD34ϩ cells was evaluated. A, CD34ϩgreen fluorescent proteinϩ cells were plated in methylcellulose progenitor culture with or without addition of imatinib to the culture medium for 14 days and assessed for colony forming cell (CFC) growth. B, CD34 green fluorescent proteinϩ cells were cultured in growth factor-containing medium as described for Fig. 2D, with or without addition of imatinib, and assessed for the number of cells generated after 14 days. C, adhesion of progenitors to fibronectin CH-296 (8 ␮g/cm2) was evaluated after prior exposure of cells to imatinib for 24 h in serum-free medium containing low concentrations of growth factors. The total number of CFC obtained from 1000 CD34ϩ cells after 24-h exposure to imatinib for MIG210-transduced cells was 157 Ϯ 16 without imatinib, 107 Ϯ 16 with 0.2 ␮M,94Ϯ 14 with 1 ␮M, and 80 Ϯ 11 with 5 ␮M imatinib; and for MIG transduced cells was 127 Ϯ 23 without imatinib, 131 Ϯ 21 with 0.2 ␮M, 132 Ϯ 24 with 1 ␮M, and 135 Ϯ 24 with 5 ␮M imatinib. D, migration of progenitors toward an stromal cell- derived factor 1 gradient (100 ng/ml) through fibronectin CH-296-coated transwells was evaluated after prior exposure of cells to imatinib for 24 h in serum-free medium containing low concentrations of growth factors. Results represent mean of four separate experiments; bars, Ϯ1 SE. ,ء :Significance levels were determined by Student’s t test analysis P Ͻ 0.0005, versus cells not exposed to ,ءءء ;P Ͻ 0.005 ,ءء ;P Ͻ 0.05 imatinib.

closely recapitulates abnormalities of cell growth and adhesion char- regulation of adhesion and migration (7, 27). Although previous acteristic of CML progenitors and has a major advantage of allowing studies have shown that constitutively activated tyrosine kinase ac- investigation of molecular mechanisms underlying BCR/ABL trans- tivity plays an important role for in abnormal proliferation and anti- formation of primary human progenitors using a mutational approach apoptotic signaling, these studies have not clearly defined its role in not possible with patient-derived cells. abnormal adhesion and migration of BCR/ABL-expressing cells. We Experimental and clinical evidence support a critical role for BCR/ have shown that the kinase inhibitor tyrphostin AG957 resulted in ABL kinase activity in cellular transformation by BCR/ABL (34, 48). significant improvement in abnormal adhesion and adhesion-mediated Abnormal kinase activity results in activation of downstream signal- signaling in CML progenitors (49). In contrast, Wertheim et al. (44) ing pathways implicated in mitogenic signaling and survival as well as found that abnormalities in adhesion of BCR/ABL-transformed 32D 5328

Fig. 6. Proliferation, adhesion, and migration of progenitors expressing ABD and proline-rich domain-deleted BCR/ABL genes. CD34ϩ cells were transduced with vectors expressing BCR/ABL genes with deletion of the actin binding (MIG210dA) or proline- rich (MIG210dP1P2) domains, as well as wild-type BCR/ABL (MIG210) and control vectors expressing the green fluorescent protein (GFP) alone (MIG), and CD34ϩGFPϩ cells were selected by flow cytometry. A, lysates were made from CD34ϩGFPϩ cells expanded in liquid growth factor culture, and BCR/ABL expression was assessed by Western blotting (WB) using an anti-BCR antibody. To confirm equal sample loading blots were reprobed with an antiactin antibody. B, protein tyrosine phosphorylation was assessed by probing blots with antiphosphotyrosine (anti-PY) antibodies. C, CD34ϩGFPϩ cells were cultured in growth factor containing medium as described for Fig. 2D, with or without addition of imatinib. The number of cells generated per 1000 CD34ϩ cells is shown (n ϭ 7). D, adhesion of progenitors to fibronectin CH-296 (8 ␮g/cm2) -coated plates was evaluated (n ϭ 6–7). E, migration of progenitors toward an stromal cell-derived factor 1 (SDF-1␣) gradient (100 ng/ml) through fibronectin CH-296-coated transwells was evaluated (n ϭ 6). Significance levels were determined by Student’s P Ͻ 0.005 versus MIG-transduced ,ءء ;P Ͻ 0.05 ,ء :t test analysis cells, †, P Ͻ 0.05 versus MIG210-transduced cells.

cells were not affected by expression of a kinase-inactive BCR/ABL genitor adhesion to fibronectin and chemotaxis to SDF-1␣ were mutant or by exposure to imatinib. It is possible that variances significantly corrected in cells after kinase inactivation indicate an between these studies are related to differences in effects of BCR/ important role for kinase-dependent signaling mechanisms in adhe- ABL expression and kinase inactivation in cell lines compared with sion and migration defects. However, lack of complete correction of primary progenitor cells. Unlike in primary progenitors, BCR/ABL adhesion and chemotactic defects in cells expressing the kinase- transformation resulted in increased adhesion in the 32D cell line. inactive BCR/ABL gene suggests that kinase-independent mecha- Differences in cell culture and assay conditions could also play a role, nisms may also contribute to these abnormalities. because GF stimulation up-regulates integrin-mediated adhesion in The role of tyrosine kinase activity in functional abnormalities in normal but not BCR/ABL-transformed cells. BCR/ABL-expressing BCR/ABL-expressing cells was additionally evaluated by exposing cells tested in the absence of GFs may demonstrate relatively in- cells to the kinase inhibitor imatinib mesylate. Imatinib resulted in a creased adhesion (43, 50). The present study demonstrates that BCR/ profound reduction in abnormally increased proliferation of BCR/ ABL-induced enhancement of proliferation and survival was com- ABL-expressing progenitors. However, proliferation of control GFP- pletely reversed in cells expressing a kinase-inactive BCR/ABL gene expressing cells was also inhibited, indicating that mechanisms other and confirms that abnormal proliferation of CML progenitors and than BCR/ABL kinase inhibition can contribute to inhibition of pro- abnormal survival in GF-deprived conditions is absolutely kinase genitor growth. Consistent with results obtained using kinase-inactive dependent. Our observations that BCR/ABL-induced defects in pro- BCR/ABL, imatinib exposure improved but did not fully correct the 5329

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2004 American Association for Cancer Research. BCR/ABL KINASE INACTIVATION IN HUMAN PROGENITORS chemotactic defect in BCR/ABL-expressing progenitors. In contrast defects in progenitor adhesion and migration. We additionally dem- to results obtained with kinase-inactive mutants, no improvement in onstrate that the BCR/ABL actin-binding and proline-rich domains adhesion of BCR/ABL-expressing progenitors was observed after significantly contribute to defects in progenitor adhesion and migra- imatinib exposure. However, imatinib significantly reduced adhesion tion. These results indicate that both kinase-dependent and kinase- of control progenitors expressing GFP alone. These additional non- independent mechanisms contribute to abnormal adhesion and migra- BCR/ABL-mediated effects of imatinib may account for differences tion. These studies demonstrate the effectiveness and utility of this in results obtained using kinase-inactive mutants compared with those approach for direct investigation of the role of specific BCR/ABL obtained after imatinib treatment. The lack of increase in adhesion of signaling activities in human progenitor cell transformation. BCR/ABL-expressing progenitors with imatinib may represent a bal- ance between adhesion-enhancing effects of BCR/ABL kinase inhi- ACKNOWLEDGMENTS bition and “nonspecific” adhesion-inhibitory effects. Our studies support a role for the BCR/ABL COOH-terminal We thank Khristine van Heijzen and Tinisha McDonald for assistance with actin-binding and proline-rich domains in abnormal adhesion and progenitor assays and Lucy Brown and Claudio Spalla of the Analytical migration of BCR/ABL-expressing progenitors. Deletion of these Cytometry Core. We also thank Dr. Warren Pear, University of Pennsylvania domains did not reduce BCR/ABL-induced increased progenitor pro- (Philadelphia, PA), Dr. Brian Druker, Oregon Health Sciences University liferation. The actin-binding domain is necessary for colocalization of (Portland, OR), Dr. Ruibao Ren, Brandeis University (Waltham, MA), and Dr. BCR/ABL with actin cytoskeleton (51, 52). Actin-binding domain Martin Haas, University of California, San Diego (San Diego, CA) for plasmid deletions reduced Rat-1 fibroblast transformation and reversed abnor- constructs and Priscilla Yam (Department of Virology Research) for assistance with the virus production protocol. mal adhesion of hematopoietic cell lines (32, 53). However, actin- binding domain deletions have had varying effects in murine CML models with reduced oncogenicity or no effect seen in different REFERENCES studies (32, 54). The proline-rich region of BCR/ABL can mediate 1. Rowley JD. A new consistent chromosome abnormality in chronic myelogenous interactions with SH3 domains in other proteins including CRKL and leukemia identified by quinacrine fluorescence and Giemsa staining. Nature 1973; the ABL interactor protein (38, 55). CRKL is a major phosphoprotein 243:209–13. 2. DeKlein A, Van Kessel AG, Grosveld G, et al. A cellular oncogene is translocated to in primary CML cells (56). Signaling through CRKL could potentially the Philadelphia chromosome in chronic myelocytic leukemia. Nature 1982;300: affect adhesion and migration via interactions with paxillin, CAS, and 765–7. CBL, and formation of complexes with focal adhesion proteins (57). 3. Lugo T, Pendergast A, Muller A, Witte O. Tyrosine kinase activity and transformation potency of bcr/abl oncogene products. Science 1990;237:1079–82. The ABL interactor protein has been shown to regulate cytoskeletal 4. Daley GQ, Baltimore D. Transformation of an interleukin 3-dependent hematopoietic function in cell lines (58). Combined SH3 and proline-rich region cell line by the chronic leukemia-specific P210 bcr/abl protein. Proc Natl Acad Sci USA 1988;85:9312–6. deletions abrogate spontaneous cell migration on fibronectin and 5. Pear WS, Miller JP, Xu L, et al. Efficient and rapid induction of a chronic myelog- impair the ability to induce CML-like disease in mice (55). The full enous leukemia-like myeloproliferative disease in mice receiving P210 bcr/abl- range of adhesion and migration defects in CML progenitors may transduced bone marrow. Blood 1998;92:3780–92. 6. Li S, Ilaria RL, Million RP, Daley GQ, Van Etten RA. The P190, P210, and P230 reflect a combination of signaling abnormalities, including cyto- forms of the BCR/ABL oncogene induce a similar chronic myeloid leukemia-like skeletal alteration mediated through the actin-binding and proline-rich syndrome in mice but have different lymphoid leukemogenic activity. J Exp Med domains and/or other kinase-independent interactions, and alteration 1999;189:1399–412. 7. Sawyers CL. Chronic myeloid leukemia. N Engl J Med 1999;340:1330–40. in signaling mechanisms regulating cytoskeletal and focal complex 8. Bedi A, Griffin CA, Barber JP, et al. Growth factor-mediated terminal differentiation assembly through kinase-dependent mechanisms. The model devel- of chronic myeloid leukemia. Cancer Res 1994;54:5535–8. 9. Traycoff CM, Halstead B, Rice S, McMahel J, Srour EF, Cornetta K. Chronic oped here will allow exploration of the role of these and other myelogenous leukaemia CD34ϩ cells exit G0/G1 phases of cell cycle more rapidly candidate signaling mechanisms in abnormal adhesion of CML pro- than normal marrow CD34ϩ cells. Br J Haematol 1998;102:759–67. genitor cells. 10. Petzer AL, Eaves CJ, Barnett MJ, Eaves AC. Selective expansion of primitive normal hematopoietic cells in cytokine-supplemented cultures of purified cells from patients Clinical studies with imatinib indicate that BCR/ABL kinase activ- with chronic myeloid leukemia. Blood 1997;90:64–9. ity is required to sustain the proliferative advantage of CML cells. 11. Bhatia R, Munthe HA, Williams AD, Zhang F, Forman SJ, Slovak ML. Increased However, it is not clear whether kinase inhibition is sufficient to sensitivity of chronic myelogenous leukemia primitive hematopoietic progenitors to growth factor induced cell division and maturation. Exp Hematol 2000;12:1401–12. eradicate leukemia cells. There is considerable evidence that residual 12. Bedi A, Zehnbauer BA, Barber JP, Sharkis SJ, Jones RJ. Inhibition of apoptosis by leukemia cells may persist and that malignant hematopoietic progen- BCR-ABL in chronic myeloid leukemia. Blood 1994;83:2038–44. 13. Maguer-Satta V, Burl S, Liu L, et al. BCR-ABL accelerates C2-ceramide-induced itors can be detected in CML patients responsive to imatinib treatment apoptosis. Oncogene 1998;16:237–48. (59, 60). The mechanisms underlying incomplete elimination of ma- 14. Amos TA, Lewis JL, Grand FH, Gooding RP, Goldman JM, Gordon MY. Apoptosis lignant progenitors are not clear. One possibility is that complete in chronic myeloid leukaemia: normal responses by progenitor cells to growth factor deprivation, X-irradiation and glucocorticoids. Br J Haematol 1995;91:387–93. inhibition of kinase activity is not achieved in all of the progenitor 15. Ghaffari S, Daley GQ, Lodish HF. Growth factor independence and BCR/ABL cells after imatinib exposure. However, in the current study, CD34ϩ transformation: promise and pitfalls of murine model systems and assays. Leukemia cells expressing kinase-inactive BCR/ABL retained capacity for nor- 1999;13:1200–6. 16. Eaves CJ, Casman JD, Wolpe SD, Eaves AC. Unresponsiveness of primitive chronic mal proliferation and survival. These results indicate that kinase myeloid leukemia cells to macrophage inflammatory protein 1 alpha, an inhibitor of inactivation in BCR/ABL-expressing cells does not by itself abrogate primitive normal hematopoietic cells. Proc Natl Acad Sci USA 1993;90:12015–9. 17. Cashman JD, Eaves CJ, Sarris AH, Eaves AC. MCP-1, not MIP-1alpha, is the responsiveness to normal proliferation and survival signals and, there- endogenous chemokine that cooperates with TGF-beta to inhibit the cycling of fore, may not lead to elimination of BCR/ABL-expressing cells under primitive normal but not leukemic (CML) progenitors in long-term human marrow physiological conditions. The failure of kinase inhibition to fully cultures. Blood 1998;92:2338–44. 18. Gordon MY, Dowding CR, Riley GP, Goldman JM, Greaves MF. Altered adhesive correct adhesion and migration defects in BCR/ABL-expressing pro- interactions with marrow stroma of hematopoietic progenitor cells in chronic my- genitors could also play a role in incomplete elimination of progenitor elogenous leukaemia. Nature 1984;328:342–4. cells in the context of the marrow microenvironment, and this possi- 19. Verfaillie CM, McCarthy JB, McGlave PB. Mechanisms underlying abnormal traf- ficking of malignant progenitors in chronic myelogenous leukemia: Decreased adhe- bility warrants investigation in future studies. sion to stroma and fibronectin but increased adhesion to the basement membrane In conclusion, we show that mutational inactivation and pharma- components laminin and collagen type IV. J Clin Investig 1992;90:1232–9. 20. Bhatia R, McCarthy JB, Verfaillie CM. Interferon-alpha restores normal beta 1 cological inhibition of BCR/ABL kinase activity completely reverses integrin-mediated inhibition of hematopoietic progenitor proliferation by the marrow abnormal progenitor proliferation but does not completely ameliorate microenvironment in chronic myelogenous leukemia. Blood 1996;87:3883–91. 5330

21. Hurley RW, McCarthy JB, Verfaillie CM. Direct adhesion to bone marrow stroma via 41. Bhatia R, McGlave PB, Dewald GW, Blazar BR, Verfaillie CM. Abnormal function fibronectin receptors inhibits hematopoietic progenitor proliferation. J Clin Investig of the bone marrow microenvironment in chronic myelogenous leukemia: role of 1995;96:511–9. malignant stromal macrophages. Blood 1995;85:3636–45. 22. Jiang Y, Zhao RC, Verfaillie CM. Abnormal integrin-mediated regulation of chronic 42. Bhatia R, Wayner EA, McGlave PB, Verfaillie CM. Interferon-alpha restores normal myelogenous leukemia CD34ϩ cell proliferation: BCR/ABL up-regulates the cyclin- adhesion of chronic myelogenous leukemia hematopoietic progenitors to bone mar- dependent kinase inhibitor, p27Kip, which is relocated to the cell cytoplasm and row stroma by correcting impaired beta 1 integrin receptor function. J Clin Investig incapable of regulating cdk2 activity. Proc Natl Acad Sci USA 2000;97:10538–43. 1994;94:384–91. ␤ 23. Peled A, Hardan I, Trakhtenbrot L, et al. Immature leukemic CD34ϩCXCR4ϩ cells 43. Bhatia R, Munthe HA, Forman SJ. Abnormal growth factor modulation of 1-integrin from CML patients have lower integrin-dependent migration and adhesion in re- mediated adhesion in chronic myelogenous leukemia hematopoietic progenitors. Br J sponse to the chemokine SDF-1. Stem Cells 2002;20:259–66. Hematol 2001;115:845–53. 24. Salgia R, Li J-L, Ewaniuk DS, et al. BCR/ABL induces multiple abnormalities of 44. Wertheim JA, Forsythe K, Druker BJ, Hammer D, Boettiger D, Pear WS. BCR-ABL- cytoskeletal function. J Clin Invest 1997;100:46–57. induced adhesion defects are tyrosine kinase-independent. Blood 2002;99:4122–30. 45. Zhao RC, Jiang Y, Verfaillie CM. A model of human p210(bcr/ABL)-mediated 25. Salgia R, Quackenbush E, Lin J, et al. The BCR/ABL oncogene alters the chemotactic chronic myelogenous leukemia by transduction of primary normal human CD34(ϩ) response to stromal-derived factor-1alpha. Blood 1999;94:4233–46. cells with a BCR/ABL-containing retroviral vector. Blood 2001;97:2406–12. 26. Ptasznik A, Urbanowska E, Chinta S, et al. Crosstalk between BCR/ABL oncoprotein 46. Chalandon Y, Jiang X, Hazlewood G, et al. Modulation of p210(BCR-ABL) activity and CXCR4 signaling through a Src family kinase in human leukemia cells. J Exp in transduced primary human hematopoietic cells controls lineage programming. Med 2002;196:667–78. Blood 2002;99:3197–204. 27. Deininger MW, Goldman JM, Melo JV. The molecular biology of chronic myeloid 47. Cambier N, Chopra R, Strasser A, Metcalf D, Elefanty AG. BCR-ABL activates leukemia. Blood 2000;96:3343–56. pathways mediating cytokine independence and protection against apoptosis in mu- 28. Zhang X, Ren R. Bcr-Abl efficiently induces a myeloproliferative disease and rine hematopoietic cells in a dose-dependent manner. Oncogene 1998;16:335–48. production of excess interleukin-3 and granulocyte-macrophage colony-stimulating 48. O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and factor in mice: a novel model for chronic myelogenous leukemia. Blood 1998;92: low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. 3829–40. N Engl J Med 2003;348:994–1004. 29. Zhang X, Wong R, Hao SX, Pear WS, Ren R. The SH2 domain of bcr-Abl is not 49. Bhatia R, Munthe HA, Verfaillie CM. Tyrphosin AG 957, a tyrosine kinase inhibitor required to induce a murine myeloproliferative disease; however, SH2 signaling with anti-BCR/ABL tyrosine kinase activity restores ␤1 integrin mediated adhesion influences disease latency and phenotype. Blood 2001;97:277–87. and inhibitory signaling in chronic myelogenous leukemia hematopoietic progenitors. 30. Roumiantsev S, de Aos IE, Varticovski L, Ilaria RL, Van Etten RA. The src Leukemia 1998;12:1708–17. homology 2 domain of Bcr/Abl is required for efficient induction of chronic myeloid 50. Bazzoni G, Carlesso N, Griffin JD, Hemler ME. Bcr/Abl expression stimulates leukemia-like disease in mice but not for lymphoid leukemogenesis or activation of integrin function in hematopoietic cell lines. J Clin Investig 1996;98:521–8. phosphatidylinositol 3-kinase. Blood 2001;97:4–13. 51. McWhirter JR, Wang JYJ. An actin-binding function contributes to transformation by 31. He Y, Wertheim JA, Xu L, et al. The coiled-coil domain and Tyr177 of bcr are the BCR-ABL oncoprotein of Philadelphia chromosome-positive human leukemias. required to induce a murine chronic myelogenous leukemia-like disease by bcr/abl. EMBO J 1993;12:1533–46. Blood 2002;99:2957–68. 52. Van Etten R, Jackson P, Baltimore D, Sanders M, Matsudaira P, Janmey P. The 32. Wertheim JA, Perera SA, Hammer DA, Ren R, Boettiger D, Pear WS. Localization COOH terminus of c-Abl tyrosine kinase contains distinct F- and G-actin binding of BCR-ABL to F-actin regulates cell adhesion but does not attenuate CML devel- domains with bundling activity. J Cell Biol 1994;124:325–40. opment. Blood 2003;102:2220–8. 53. McWhirter JR, Wang JY. Effect of Bcr sequences on the cellular function of the 33. Druker B, Tamura S, Buchdunger E, et al. Effects of a selective inhibitor of the Abl Bcr-Abl oncoprotein. Oncogene 1997;15:1625–34. tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 1996;2:561–6. 54. Heisterkamp N, Voncken JW, Senadheera D, et al. Reduced oncogenicity of p190 34. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the Bcr/Abl F-actin-binding domain mutants. Blood 2000;96:2226–32. BCR-ABL kinase in chronic myeloid leukemia. N Engl J Med 2001;344:1031–7. 55. Dai Z, Kerzic P, Schroeder WG, McNiece IK. Deletion of the Src homology 3 domain and C-terminal proline-rich sequences in Bcr-Abl prevents Abl interactor 2 degrada- 35. Kaur G, Gazit A, Levitzki A, Stowe E, Cooney DA, Sausville EA. Tyrphostin induced tion and spontaneous cell migration and impairs leukemogenesis. J Biol Chem growth inhibition: correlation with effect on p210bcr-abl autokinase activity in K562 2001;276:28954–60. chronic myelogenous leukemia cells. Anti-Cancer Drugs 1994;5:213–22. 56. Oda T, Heaney C, Hagopian JR, Okuda K, Griffin JD, Druker BJ. Crkl is the major 36. Druker B, Okuda K, Matulonis U, Salgia R, Roberts T, Griffin JD. Tyrosine tyrosine-phosphorylated protein in neutrophils from patients with chronic myeloge- phosphorylation of rasGAP and associated proteins in chronic myelogenous leukemia nous leukemia. J Biol Chem 1994;269:22925–8. cell lines. Blood 1992;79:2215–20. 57. Salgia R, Uemera N, Okuda K, et al. CRKL links p210BCR/ABL with paxillin in 37. Skourides PA, Perera SA, Ren R. Polarized distribution of Bcr-Abl in migrating chronic myelogenous leukemia cells. J Biol Chem 1995;270:29145–50. myeloid cells and colocalization of Bcr-Abl and its target proteins. Oncogene 58. Stradal T, Courtney KD, Rottner K, Hahne P, Small JV, Pendergast AM. The Abl 1999;18:1165–76. interactor proteins localize to sites of actin polymerization at the tips of lamellipodia 38. Heaney C, Kolibaba K, Bhat A, et al. Direct binding of CRKL to BCR-ABL is not and filopodia. Curr Biol 2001;11:891–5. required for BCR-ABL transformation. Blood 1997;89:297–306. 59. Hughes T, Kaeda J, Branford S, et al. Molecular responses to Imatinib (STI571) or 39. Pear WS, Nolan G, Scott M, Baltimore D. Production of high titer helper free Interferon ϩ Ara-C as initial therapy for CML: Results in the IRIS study. Blood retroviruses by transient transfection. Proc Natl Acad Sci USA 1993;90:8392–6. 2002;100:93A. 40. Norris PS, Jepsen K, Haas M. High-titer MSCV-based retrovirus generated in the 60. Bhatia R, Holtz M, Niu N, et al. Persistence of malignant hematopoietic progenitors pCL acute virus packaging system confers sustained gene expression in vivo. J Virol in chronic myelogenous leukemia patients in complete cytogenetic remission follow- Methods 1998;75:161–7. ing imatinib mesylate treatment. Blood 2003;101:4701–7.

5331

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2004 American Association for Cancer Research. Effect of Mutational Inactivation of Tyrosine Kinase Activity on BCR/ABL-Induced Abnormalities in Cell Growth and Adhesion in Human Hematopoietic Progenitors

Pandurangan Ramaraj, Harjeet Singh, Ning Niu, et al.

Cancer Res 2004;64:5322-5331.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/64/15/5322

Cited articles This article cites 60 articles, 31 of which you can access for free at: http://cancerres.aacrjournals.org/content/64/15/5322.full#ref-list-1

Citing articles This article has been cited by 15 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/64/15/5322.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/64/15/5322. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.