FAK silencing inhibits leukemogenesis in BCR/ABL-transformedhematopoietic cells

Yi Le,1* Luhong Xu,1,2 Jiayun Lu,1 Jianpei Fang,2 Valentina Nardi,3 Li Chai,1 and Leslie E. Silberstein1*

Focal adhesion kinase (FAK) is constitutively activated and tyrosine phosphorylated in BCR/ABL-trans-formed hematopoietic cells, but the role it plays during leukemogenesis remains unclear. Here, we exam-ined the effects of RNA interference-mediated FAK silencing on leukemogenesis induced by a BCR/ABL-transformed cell line. Transduction of BCR/ABL-BaF3 cells with FAK shRNA inhibited FAK expression andreduced STAT5 phosphorylation, but induced caspase-3 activation. In vitro studies showed that treatmentwith FAK shRNA resulted in impaired cell proliferation and colony formation, while increasing cell apopto-sis. Mice that received transplants of BCR/ABL-BaF3 cells with FAK shRNA displayed significantly pro-longed survival time and diminished leukemia progression. In addition, FAK silencing enhanced in vitroand in vivo efficacy of ABL tyrosine kinase inhibitor imatinib in BCR/ABL-BaF3 cells. Our results suggestthat FAK is critical for leukemogenesis and might be a potential target for leukemia therapy. Am. J. Hematol.84:273–278, 2009. VVC 2009 Wiley-Liss, Inc.

IntroductionThe Philadelphia (Ph) chromosome arises from a translo-

cation between chromosomes 9 and 22 and results information of a chimeric and constitutively activated tyrosinekinase known as BCR/ABL, which induces acute B-celllymphoblastic leukemia (B-ALL) and chronic myeloid leuke-mia (CML) [1,2]. Approximately, up to 20% of adults and5% of children with ALL have the BCR/ABL fusion protein[3]. The protein encoded by the BCR/ABL fusion is pro-duced in a 210 or 190 kDa form, depending on the specificbreakpoint within the BCR gene [4]. Whereas the ABL tyro-sine kinase inhibitor imatinib induces a complete hemato-logic response in patients with CML, it is much less effec-tive in treating Ph1 B-ALL due to acquired resistance [5–7].Thus, targeting signaling pathways downstream of BCR/ABL may have potential therapeutic relevance in treatmentand/or prevention of ALL.Focal adhesion kinase (FAK) is a nonreceptor tyrosine

kinase playing an important role in cell survival and motility[8,9]. Furthermore, there is accumulating evidence thatFAK is expressed and activated at higher levels in malig-nant cells compared to normal cells [10–12]. Our previousstudies have shown that FAK is required for chemotacticand adhesive responses in hematopoietic precursor cells[13,14]. In addition, BCR/ABL-transformed leukemic cellsdisplay constitutive tyrosine phosphorylation and kinaseactivation of FAK [15]. However, the importance of FAKmediated malignant events in leukemic cells is not welldefined. Recently, screened by RNA interference (RNAi)technology, FAK has been demonstrated to be crucial tothe survival of acute myeloid leukemia [16].In the present study, we sought to assess the potential

role of FAK in BCR/ABL expressing leukemogenesis. Usinga p210 form of BCR/ABL-transformed pro-B lymphoid cellline (BCR/ABL-BaF3 cells) [17], we found that FAK silencingresulted in decreased cell proliferation and colony formation,while increasing cell apoptosis. Moreover, FAK silencing pro-longed median survival and diminished leukemia progressionin mice transplanted with BCR/ABL-BaF3 cells. In addition,FAK silencing enhanced in vitro and in vivo efficacy of imati-nib in BCR/ABL-BaF3 cells. Taken together, our results

suggest that FAK is critical for leukemogenesis and might bea potential target for Ph1 B-ALL therapy.

Results

Efficient silencing of FAK in BCR/ABL-BaF3 cellsTo knock down FAK in BCR/ABL-transformed hematopoi-

etic cells, BCR/ABL-BaF3 cells were transduced with lenti-viral-GFP-FAK-shRNA. The infection efficiency in FAKknock-down cells varied between 4 and 10%, while thetransduction efficiency in vector control cells variedbetween 20 and 30% as assessed by measuring the GFP1

fraction by flow cytometry (data not shown). The GFP1

sorted cells were used for Western blotting. As shown inFig. 1a, a significant inhibition of FAK expression wasobserved in FAK shRNA-transduced cells. The intensity ofprotein expression was decreased by 75% relative to con-trol vector in the leukemic cells as calculated using Image-Quant densitometer (data not shown). The specificity ofthis FAK shRNA was demonstrated in a previous studyusing REH cells, a human pro-B ALL cell line [13]. The

1Department of Pathology, Joint Program in Transfusion Medicine, Children’sHospital Boston, Boston, Massachusetts; 2Department of Pediatrics, SecondAffiliated Hospital of Zhongshan University, Guangzhou, People’s Republicof China; 3Division of Hematology/Oncology, Children’s Hospital Boston,Boston, Massachusetts

Y.L. and L.X. contributed equally to this work.

Conflict of interest: Nothing to report.

*Correspondence to: Yi Le, Department of Pathology, Joint Program inTransfusion Medicine, Karp Research Building, Room 10217, One BlackfanCircle, Boston, MA 02115. E-mail: yi.le@med.navy.mil or Leslie E. Silber-stein, Department of Pathology, Joint Program in Transfusion Medicine, KarpResearch Building, Room 10217, One Blackfan Circle, Boston, MA 02115.E-mail: leslie.silberstein@childrens.harvard.edu.

Received for publication 22 November 2008; Revised 30 January 2009;Accepted 9 February 2009

Am. J. Hematol. 84:273–278, 2009.

Published online 19 February 2009 in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/ajh.21381

Research Article

VVC 2009 Wiley-Liss, Inc.

American Journal of Hematology 273 http://www3.interscience.wiley.com/cgi-bin/jhome/35105

target sequence of this shRNA is similar in murine andhuman FAK.Notably, BaF3 cells with FAK shRNA had significantly

reduced STAT5 phosphorylation compared to BaF3 withvector control, while the levels of total STAT5 were similarbetween these two cell lines (Fig. 1b). In addition, FAKshRNA induced caspase-3 activation in leukemic cells, butdid not affect the activity of caspase-9 (Fig. 1c), indicatingthat FAK silencing might increase cell apoptosis throughthe caspase-3 pathway.

FAK shRNA impairs leukemia growth and proliferationFAK has been previously demonstrated to be critical for

cell proliferation and migration [8,9]. To study the require-ment of FAK in leukemia growth and proliferation, trans-duced BCR/ABL BaF3 cells were cultured in liquid or meth-ylcellulose medium in vitro. By trypan blue exclusionmethod, cell growth was diminished in the FAK knock-downcells compared to the vector control cells at 48 and 72 hrafter culture (Fig. 2a). The number of colonies in vectorcontrol and FAK shRNA groups was (215.6 ± 13.01) and(125 ± 9.06), respectively (Fig. 2b), and the differenceswere statistically significant (P < 0.01). We next testedapoptosis in these transduced cells and, as shown inFig. 2c, the percentage of apoptotic cells in vector controland FAK shRNA groups was (3.46 ± 0.56)% and (7.3 ±0.79)%, respectively, and there were significant differencesbetween these two groups (P < 0.01).

FAK silencing inhibits leukemogenesis in vivoTo study the effects of FAK silencing on leukemogenesis

in vivo, one million transduced BCR/ABL-BaF3 cells wereinjected into BALB/c mice on Day 0, and the survivals weremonitored daily. As shown in Fig. 3a, all the mice died afterinjection of leukemic cells. The mice in the vector controlgroup died between Day 21 and Day 27, with a median of24 days. These data were consistent with previous report[18]. The mice in FAK shRNA group died between Day 52and Day 60, with a median of 55 days. By log-rank analysisthere was a significant difference between the vector groupand FAK shRNA group (P < 0.01), indicating that FAKsilencing prolongs the survival time. Before the animal died,

blood smears were analyzed by Wright-Giemsa staining.We found that all the mice died of leukemia as the leukemicblasts were observed in the peripheral blood of the diseasemice (Fig. 3b).To assess leukemia progression in vivo, some of the

mice were sacrificed on Day 25. As shown in Fig. 3c, allthe leukemic mice developed splenomegaly, and FAKsilencing reduced the size of spleen compared with that ofvector group. Similarly, the leukemic mice had increasednumbers of white blood cells in the peripheral blood. Asshown in Fig. 3d, the leukocyte count in the vector controlgroup was sixfold higher than that in the normal mice, whilethe leukocyte count in FAK shRNA group was twofoldhigher than that in the normal mice, indicating that FAKsilencing reduces the leukemia burden in mice. Moreover,the percentage of GFP1 cells in bone marrow and spleenof leukemic mice was tested by a flow cytometer (Fig. 3e).We found that the percentage of GFP1 cells in bone

Figure 1. Western blot analysis of gene expression and caspases activity. BCR/ABL-BaF3 cells were infected with lentiviral-GFP-FAK-shRNA or empty vector, andwere sorted by GFP expression on a flow cytometer. Cell lysates were subjectedto SDS-PAGE, transferred to nitrocellulose membrane, and probed with differentantibodies. (a) FAK shRNA inhibited FAK expression in the leukemic cells. (b) FAKshRNA reduced STAT5 phosphorylation, but had no effect on STAT5 expression.(c) FAK shRNA induced caspases-3 activation, but did not affect caspase-9 activ-ity. The blots are representative of three independent experiments.

Figure 2. FAK is critical for leukemia cell proliferation and survival. (a) Trans-duced BCR/ABL-BaF3 cells were cultured in vitro for 72 hr. Cell proliferation wasdiminished in the FAK shRNA group compared to the vector control group. (b)Five hundred transduced leukemic cells were plated in methylcellulose mediumand incubated for 7 days. Colony formation was inhibited in the FAK shRNA group.(c) The transduced leukemic cells were stained with APC-Annexin V and analyzedby a flow cytometer. The rate of cell apoptosis was increased in the FAK shRNAgroup. Data are shown as mean ± SEM of three experiments. *P < 0.05, **P <0.01 the FAK shRNA group vs. the vector control group.

274 American Journal of Hematology

research article

Figure 3. FAK silencing inhibits leukemogenesis in vivo. One million transduced BCR/ABL-BaF3 cells were injected into BALB/c mice on Day 0. The survivors weremonitored daily, and leukemia progression was assayed on Day 25. (a) Kaplan–Meier survival analysis showing that FAK shRNA prolonged the mice survival. (b) Bloodsmears were performed on the dying mice, and leukemic blasts were found in the peripheral blood. (c) Leukemic mice developed splenomegaly on Day 25, butFAK silencing reduced the size of spleen compared with the vector. (d) The white blood cells were counted in the peripheral blood of mice, and the results showed thatFAK silencing reduced the leukemia burden. (e) Percentage of GFP1 cells in bone marrow and spleen of leukemic mice was tested by a flow cytometer, and the resultsshowed that FAK silencing significantly inhibited the leukemia progression. *P < 0.05 the FAK shRNA group vs. the vector control group. [Color figure can be viewed inthe online issue which is available at www.interscience.wiley.com.]

American Journal of Hematology 275

research article

marrow of mice in the vector control group and FAK shRNAgroup was (69 ± 5.1)% and (7.9 ± 2.7)%, respectively, andthe differences were statistically significant (P < 0.05). Asexpected, FAK shRNA also reduced the percentages ofGFP1 cells in the spleens of leukemic mice. Last, the histo-pathology was analyzed by hematoxylin and eosin staining.As shown in Fig. 4, the leukemic infiltration was observedin the spleen and bone marrow of leukemic mice, and FAKsilencing reduced the damage compared with the vectorcontrol. Taken together, FAK silencing inhibits leukemogen-esis in this murine model of leukemia.

FAK silencing promotes the efficacy of imatinibin BCR/ABL-BaF3 cellsTo further investigate the effects of FAK silencing on the

efficacy of imatinib in BCR/ABL-BaF3cells, the transducedcells were treated with different concentrations of imatiniband cell apoptosis was analyzed in vitro. As shown inFig. 5a, with 5 lM imatinib treatment, the percentage ofcells undergoing apoptosis in the vector control and FAKshRNA groups was (9.76 ± 1.97)% and (21.9 ± 3.2)%,respectively, and there were significant differences betweenthese two groups (P < 0.05). Similarly, with 50 lM imatinibtreatment, the percentage of apoptotic cells in FAK shRNAgroup was much higher than that in the vector control group,and the differences were statistically significant (P < 0.05).We next evaluated the effects of FAK downregulation on

drug efficacy in vivo. One million transduced BCR/ABL-BaF3 cells were injected into BALB/c mice on Day 0, andthe mice were treated with imatinib for 7 days starting fromDay 10. As shown in Fig. 5b, 40% of the mice in vectorgroup (4/10) died between Day 25 and Day 27, and theothers (6/10) died between Day 60 and Day 68. In the FAKshRNA group, 20% of the mice (2/10) died at Day 50 andDay 55, while the other mice (8/10) died at Day 80 andDay 95. By log-rank analysis, there was a significant differ-ence between these two groups (P < 0.01), indicating thatFAK silencing promotes the in vivo efficacy of imatinib inBCR/ABL leukemic cells.

Figure 4. Histology of spleen and femur in normal and leukemic mice. Twenty-five days after injection of leukemic cells, three mice in both vector group and FAK shRNAgroup were sacrificed for histological analysis. Normal mice without any treatment were used as controls. Histology sections were stained with hematoxylin and eosin andimaged at 403 magnification. FAK silencing reduced damage in spleens and bone marrow of leukemic mice. (a) Normal control group, (b) Vector control group, (c) FAKshRNA group. The histology sections are representative of three mice in each group. [Color figure can be viewed in the online issue which is available at www.interscience.wiley.com.]

Figure 5. FAK silencing promotes the in vitro and in vivo efficacy of imatinib inBCR/ABL-BaF3 cells. (a) Transduced cells were treated with different concentra-tions of imatinib (5 and 50 lM) at 378C for 24 hr, and then they were stained withAPC-Annexin V and analyzed by a flow cytometer. Increased cell apoptosis wasobserved in the FAK knock-down cells. All experiments were performed in tripli-cate. *P < 0.05 the FAK shRNA group vs. the vector control group. (b) One milliontransduced leukemic cells were injected into BALB/c mice on Day 0, and the micewere treated with imatinib (25 mg/kg/day) for 7 days starting from Day 10. Theresults showed that FAK silencing prolonged the survival time of mice comparedto the vector control.

276 American Journal of Hematology

research article

DiscussionFAK is a 125-kDa nonreceptor protein tyrosine kinase

that plays an important role in cell survival, migration, andinvasion. Moreover, FAK is overexpressed in a large num-ber of cancer cells. Recently, Halder et al. [19] found thatFAK short interference RNA (siRNA) diminished ovarian tu-mor formation. In addition, using FAK inhibitor (TAE226),they found that FAK blockade inhibited the tumor cellgrowth in a time- and dose-dependent manner, and signifi-cantly reduced the tumor burden in mice [20]. Therefore,FAK is an attractive target for therapeutic intervention.Here, we sought to assess the potential role of FAK inBCR/ABL-mediated leukemogenesis. By Western blotting,the results showed that lentiviral RNAi mediated FAKsilencing successfully inhibited FAK expression in BCR/ABL-BaF3 cells. In vitro studies showed that FAK silencingresulted in decreased cell proliferation and colony formationwhile increasing cell apoptosis, indicating that FAK is criti-cal for leukemia cell proliferation and survival.Using a murine model of leukemia induced by BCR/ABL-

BaF3 cells, we studied the effects of FAK silencing on leu-kemogenesis in vivo. All the mice died after injection ofleukemic cells, but FAK silencing prolonged the median sur-vival. Of note, all the mice in FAK shRNA group developedhind-limb paralysis due to infiltration of the cells into thespinal cord. To further assess the effects of FAK silencingon leukemia progression, some of the mice were sacrificedon Day 25. We found that FAK shRNA reduced the size ofspleen compared with that of the vector control group, andFAK shRNA diminished the leukocytes in peripheral bloodof leukemic mice. By flow cytometric analysis of GFP1 leu-kemic cells, we found that FAK silencing significantlyreduced the leukemia burden in bone marrow and spleenof leukemic mice. Last, the histopathology analysis demon-strated that FAK silencing inhibited leukemia progression inthe spleens and bone marrow of leukemic mice. Thesefindings suggest that FAK silencing inhibits leukemogenesisin this murine model of leukemia.The mechanisms of the inhibition effects of FAK silencing

on leukemogenesis have not been clearly defined. It hasbeen reported that FAK is one of the signaling moleculesdownstream of BCR/ABL [21], thus FAK silencing might in-hibit BCR/ABL-induced leukemia. Recently, STAT5 hasbeen demonstrated to be an essential signaling pathwaythat mediates survival and proliferation of Ph1 leukemia[22]. In our studies, we found that STAT5 phosphorylationwas significantly inhibited in BCR/ABL-BaF3 cells treatedwith FAK shRNA, while no effect were found on totalSTAT5 expression. These results confirm that STAT5 sig-naling molecule is downstream of FAK in Ph1 leukemiccells [21].Our results also showed that FAK silencing enhanced

cell apoptosis, which is an important mechanism of inhibi-ting leukemogenesis [23,24]. Recently, FAK silencing hasbeen found to be induced caspase-8-dependent apoptosisin human acute myeloid leukemia cells [25]. We found thatFAK silencing in BCR/ABL-BaF3 cells induced caspase-3activation, implicating this pathway in cell apoptosis. Alter-natively, whether FAK silencing affects homing and engraft-ment of BCR/ABL-BaF3 cells remains unknown. Our previ-ous studies have found that FAK downregulation signifi-cantly inhibits SDF-1-induced responses in mouse pro-Bcells [13]. More recently, the hyaluronan receptor CD44 hasbeen demonstrated to be required in homing and engraft-ment of BCR/ABL-expressing leukemic cells [26]. In ourexperiments, we found the BCR/ABL-BaF3 cells expressedCD44, but FAK silencing had no effect on the CD44expression (data not shown). A crosstalk has been found

between CD44 and SDF-1 signaling [27], while the role ofFAK in the crosstalk remains further investigation.Since the ABL tyrosine kinase inhibitor imatinib is much

less effective in treating Ph1 B-ALL owing to acquired re-sistance [5–7], it is of interest to target essential signalingmolecules downstream of BCR/ABL, which may help pre-vent imatinib resistance. FAK siRNA has been shown toenhance pancreatic adenocarcinoma chemosensitivity togemcitabine [28], and FAK siRNA can augment docetaxel-mediated apoptosis in ovarian cancer cells [19]. In ourpresent studies, we found that FAK silencing increased therate of cell apoptosis in BCR/ABL-BaF3 cells treated withimatinib. Compared to the control vector, FAK silencingprolonged the survival time of mice treated with leukemiccells and imatinib. These results indicate that FAK silenc-ing promotes the in vitro and in vivo efficacy of imatinib inBCR/ABL-transformed hematopoietic cells. In conclusion,our studies suggest that FAK plays an important role inleukemogenesis and might be a potential target for Ph1

B-ALL therapy.

MethodsCell line. BCR/ABL-mediated transformation of the interleukin-3-de-

pendent murine pro-B cell line BaF3 has been described previously[17]. Stable transformants (BCR/ABL-BaF3) were maintained in RPMIsupplemented with 10% fetal calf serum, 2 mM glutamine, and 50 U/mlpenicillin and streptomycin.

Lentiviral RNAi-mediated FAK silencing. Short-hairpin RNA (shRNA)-expressing lentivirus-vector delivery system was employed asdescribed previously [29]. Briefly, 63-nt DNA oligomers (IDT, Coralville,IA) containing the sense sequence, a 9-nt loop (TTCAAGAGA) and theantisense sequence were cloned into a lentiviral-GFP vector [14]. TheLentiviral-GFP-FAK-shRNA vector or empty vector was cotransfectedinto phoenix-293T cells with the pCMV-VSVg plasmid. The resultinglentivirus-containing supernatant from the transfected cells was usedfor the spininfection of BCR/ABL-BaF3 cells. Forty-eight hours after invitro infection, the leukemic cells were sorted by GFP expression on aflow cytometer (FACSAria, Becton Dickinson, CA) and were chosen forfurther experiments. Western blot analysis revealed that the bestsilencing efficiency was achieved by shRNA designated as FAK3-RNAi;FAK target sequence: 50-GGAATGCTTCAAGTGTGCT-30.

Western blot analysis. Western blotting was performed as describedpreviously [30]. Rabbit anti-FAK, rabbit anti-b-actin, rabbit anti-STAT5,phospho anti-STAT5, rabbit anti-caspase-3, rabbit anti-caspase-9 werepurchased from Cell Signaling Technology (Danvers, MA). Secondaryhorseradish peroxidase-conjugated antibodies were from Bio-Rad (Her-cules, CA). Enhanced Chemiluminescence reagent was from Amer-sham Biosciences (Piscataway, NJ).

Colony-forming assay. Transduced BCR/ABL-BaF3 cells were platedin methylcellulose (MethoCult1 M3630, Stem Cells Technologies, Van-couver, Canada) containing recombinant human IL-7. Briefly, 500 BCR/ABL-BaF3 cells (0.1 ml) were added to 35-mm dishes containing 1 mlmethylcellulose medium, and were incubated at 378C under 5% CO2 ina humidified atmosphere. Each experiment was performed in triplicates.Colonies were counted under an inverted microscope after 7 days ofincubation. Clusters of 40 or more cells were scored as a colony.

Apoptosis detection. Transduced BCR/ABL-BaF3 cells werestained with APC-Annexin V (BD Biosciences, CA) following manufac-turer’s instructions, and analyzed using a flow cytometer (FACSCanto,Becton Dickinson, CA). In some cases, the transduced cells were incu-bated with different concentrations of imatinib (5 and 50 lM) at 378Cfor 24 hr before analysis.

In vivo studies. The murine model of leukemia was established asdescribed [18]. Briefly, 1 3 106 transduced BCR/ABL-BaF3 cells wereinjected intravenously into the lateral tail veins of young BALB/c mice (6-weeks-old; Jackson Laboratory, Bar Harbor, ME). For in vivo assessmentof drug effects, 10 days postleukemic cells transfer, the mice were intra-peritoneally treated with imatinib (25 mg/kg/day) for 7 days. Mortality wasscored daily. Complete blood counts and blood smears were performedand leukemia progression was assayed at indicated time. All the animalswere handled and housed in accordance with the guidelines of the Child-ren’s Hospital Boston Animal Care and Use Committee.

Histopathologic analysis. Mice injected with BCR/ABL-BaF3 cells weresacrificed at indicated time point, spleens and femurs were fixed in 10%

American Journal of Hematology 277

research article

formaldehyde. Pathologic analysis was performed by the histopathologydepartment in Brigham Women’s Hospital according to standard protocols.Histology sections were stained with hematoxylin and eosin and imaged at403 magnification.

Statistical analysis. Results are expressed as mean ± standard errorof the mean, statistical analysis was carried out using Student’s t-test.Values of log-rank P were determined using the Kaplan–Meier methodcomparing survival curves. A P value � 0.05 was considered statisti-cally significant.

References1. Harb JG, Chyla BI, Huettner CS. Loss of Bcl-x in Ph1 B-ALL increases cellular

proliferation and does not inhibit leukemogenesis. Blood 2008;111:3760–3769.2. Perez-Losada J, Gutierrez-Cianca N, Sanchez-Garcia I. Philadelphia-positive

B-cell acute lymphoblastic leukemia is initiated in an uncommitted progenitorcell. Leuk Lymphoma 2001;42:569–576.

3. Druker BJ, Sawyers CL, Kantarjian H, et al. Activity of a specific inhibitor ofthe BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemiaand acute lymphoblastic leukemia with the Philadelphia chromosome. N EnglJ Med 2001;344:1038–1042.

4. Kurzrock R, Shtalrid M, Romero P, et al. A novel c-abl protein product inPhiladelphia-positive acute lymphoblastic leukaemia. Nature 1987;325:631–635.

5. Branford S, Rudzki Z, Walsh S, et al. High frequency of point mutations clus-tered within the adenosine triphosphate-binding region of BCR/ABL inpatients with chronic myeloid leukemia or Ph-positive acute lymphoblastic leu-kemia who develop imatinib (STI571) resistance. Blood 2002;99:3472–3475.

6. Gorre ME, Mohammed M, Ellwood K, et al. Clinical resistance to STI-571cancer therapy caused by BCR-ABL gene mutation or amplification. Science2001;293:876–880.

7. Kujawski L, Talpaz M. Strategies for overcoming imatinib resistance in chronicmyeloid leukemia. Leuk Lymphoma 2007;48:2310–2322.

8. Wozniak MA, Modzelewska K, Kwong L, Keely PJ. Focal adhesion regulationof cell behavior. Biochim Biophys Acta 2004;1692:103–119.

9. Cox BD, Natarajan M, Stettner MR, Gladson CL. New concepts regardingfocal adhesion kinase promotion of cell migration and proliferation. J Cell Bio-chem 2006;99:35–52.

10. Tilghman RW, Parsons JT. Focal adhesion kinase as a regulator of cell ten-sion in the progression of cancer. Semin Cancer Biol 2008;18:45–52.

11. van Nimwegen MJ, van de Water B. Focal adhesion kinase: A potential targetin cancer therapy. Biochem Pharmacol 2007;73:597–609.

12. Chatzizacharias NA, Kouraklis GP, Theocharis SE. Focal adhesion kinase: Apromising target for anticancer therapy. Expert Opin Ther Targets 2007;11:1315–1328.

13. Glodek AM, Le Y, Dykxhoorn DM, et al. Focal adhesion kinase is required forCXCL12-induced chemotactic and pro-adhesive responses in hematopoieticprecursor cells. Leukemia 2007;21:1723–1732.

14. Le Y, Zhu BM, Harley B, et al. SOCS3 protein developmentally regulates thechemokine receptor CXCR4-FAK signaling pathway during B lymphopoiesis.Immunity 2007;27:811–823.

15. Gotoh A, Miyazawa K, Ohyashiki K, et al. Tyrosine phosphorylation and acti-vation of focal adhesion kinase (p125FAK) by BCR-ABL oncoprotein. ExpHematol 1995;23:1153–1159.

16. Tyner JW, Walters DK, Willis SG, et al. RNAi screening of the tyrosinekinome identifies therapeutic targets in acute myeloid leukemia. Blood 2008;111:2238–2245.

17. Daley GQ, Baltimore D. Transformation of an interleukin 3-dependent hemato-poietic cell line by the chronic myelogenous leukemia-specific P210bcr/ablprotein. Proc Natl Acad Sci USA 1988;85:9312–9316.

18. Peters DG, Hoover RR, Gerlach MJ, et al. Activity of the farnesyl proteintransferase inhibitor SCH66336 against BCR/ABL-induced murine leukemiaand primary cells from patients with chronic myeloid leukemia. Blood 2001;97:1404–1412.

19. Halder J, Kamat AA, Landen CN Jr, et al. Focal adhesion kinase targetingusing in vivo short interfering RNA delivery in neutral liposomes for ovariancarcinoma therapy. Clin Cancer Res 2006;12:4916–4924.

20. Halder J, Lin YG, Merritt WM, et al. Therapeutic efficacy of a novel focal ad-hesion kinase inhibitor TAE226 in ovarian carcinoma. Cancer Res 2007;67:10976–10983.

21. Nardi V, Azam M, Daley GQ. Mechanisms and implications of imatinib resist-ance mutations in BCR-ABL. Curr Opin Hematol 2004;11:35–43.

22. Hoover RR, Gerlach MJ, Koh EY, Daley GQ. Cooperative and redundanteffects of STAT5 and Ras signaling in BCR/ABL transformed hematopoieticcells. Oncogene 2001;20:5826–5835.

23. Schimmer AD. Apoptosis in leukemia: From molecular pathways to targetedtherapies. Best Pract Res Clin Haematol 2008;21:5–11.

24. Pui CH, Robison LL, Look AT. Acute lymphoblastic leukaemia. Lancet 2008;371:1030–1043.

25. Casanova I, Bosch R, Lasa A, et al. A celecoxib derivative inhibits focal adhe-sion signaling and induces caspase-8-dependent apoptosis in human acutemyeloid leukemia cells. Int J Cancer 2008;123:217–226.

26. Krause DS, Lazarides K, von Andrian UH, Van Etten RA. Requirement forCD44 in homing and engraftment of BCR-ABL-expressing leukemic stemcells. Nat Med 2006;12:1175–1180.

27. Avigdor A, Goichberg P, Shivtiel S, et al. CD44 and hyaluronic acid cooperatewith SDF-1 in the trafficking of human CD341 stem/progenitor cells to bonemarrow. Blood 2004;103:2981–2989.

28. Duxbury MS, Ito H, Benoit E, et al. RNA interference targeting focal adhesionkinase enhances pancreatic adenocarcinoma gemcitabine chemosensitivity.Biochem Biophys Res Commun 2003;311:786–792.

29. Stewart SA, Dykxhoorn DM, Palliser D, et al. Lentivirus-delivered stable genesilencing by RNAi in primary cells. RNA 2003;9:493–501.

30. Le Y, Honczarenko M, Glodek AM, et al. CXC chemokine ligand 12-inducedfocal adhesion kinase activation and segregation into membrane domains ismodulated by regulator of G protein signaling 1 in pro-B cells. J Immunol2005;174:2582–2590.

278 American Journal of Hematology

research article