Leukemia (1999) 13, 855–861  1999 Stockton Press All rights reserved 0887-6924/99 $12.00 http://www.stockton-press.co.uk/leu Therapeutic targeting of Src- Lyn in myeloid leukemic cell growth V Roginskaya1, S Zuo1, E Caudell1, G Nambudiri2, AJ Kraker3 and SJ Corey1,4

1Department of Pediatrics, Children’s Hospital of Pittsburgh; Department of 4Pharmacology, and 2Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA; and 3Cancer Research, Parke-Davis, Ann Arbor, MI, USA

cell leukemia-derived cell lines, we found constitutive Lyn play a major role in promoting cell activity in primary leukemic blast cells from patients with growth, and their activity in solid tumors is well established. Inhibitors of protein tyrosine kinases are now in advanced clini- AML. To determine the role of Lyn in promoting leukemic cal trials for the treatment of breast and brain . Because growth, we inhibited Lyn by antisense oligonucleotides, Src-related PTK have been shown to be activated in leukemic somatic cell targeting, and a novel Src kinase inhibitor. cell lines, we studied their activation in human myeloid leuke- This compound, the kinase inhibitor, mia. Blasts from the majority of patients with acute leukemia PD166285, is a member of a structural class, the 6-aryl-pyr- showed constitutive activity of the Src kinase Lyn. In contrast, ido[2,3-d]pyrimidines, which has the attractive feature of oral no patient samples showed constitutive activation of Jak2. Genetic and pharmacologic targeting of Lyn was used to deter- bioavailability. PD166285 is most potent against Src-related m 7 mine its contribution to leukemic cell growth. Antisense Lyn protein tyrosine kinases, with an IC50 of 8 n . Each of these oligonucleotide treatment resulted in the inhibition of tritiated therapeutic strategies resulted in inhibition of leukemic cell thymidine incorporation following GM-CSF stimulation of the growth, with the most promising approach being pharmacol- factor-dependent line MO7e. The Src kinase inhibitor PD166285 ogic. Thus, the addition of a tyrosine kinase inhibitor, which inhibited the growth of human leukemic cell lines and leukemic includes Src as a target, may improve outcome for patients blasts. When combined with doxorubicin, an additive effect on the inhibition of leukemic cell growth occurred. These studies with hematologic malignancies. demonstrate the importance of Src kinases in promoting leu- kemic cell growth and suggests that further development of agents which target Src kinases and their inclusion in multi- Materials and methods drug regimens are warranted for novel therapies of myeloid leukemia. Reagents Keywords: leukemia; Lyn; tyrosine kinase; chemotherapy Recombinant human GM-CSF was a gift of Genetics Institute (Cambridge, MA, USA). Rabbit polyclonal antibodies were Introduction developed specifically to immunoprecipitate and immunoblot against Btk, Fyn, , Lyn, Src, Syk, and ZAP-70 by Dr Joseph Although great improvements in clinical outcome have been Bolen8–12 and were used for the data presented in Figure 1a. achieved in childhood acute lymphoblastic leukemia, the Antibodies against Blk, Fyn Hck, Lck, and Lyn were purchased same is not true for acute myelogenous leukemia (AML). AML from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and occurs more frequently in adults. Despite intensive chemo- Jak2 and phosphotyrosine 4G10 were purchased from Upstate therapy and bone marrow transplantation, long-term disease Biotechnology (Lake Placid, NY, USA). The Src inhibitor, 1 survival is no better than 65%. In the elderly, AML typically PD166285 was a gift from Dr Alan Kraker of Parke-Davis (Ann follows a preleukemic syndrome, bone marrow transplan- Arbor, MI, USA). Wortmannin was purchased from Sigma (St tation is not a therapeutic option, and it is difficult to treat Louis, MO, USA). AG490 was a kind gift of Dr Alex Levitzki successfully. Greater understanding of myeloid leukemogen- (Hebrew University, Jerusalem, Israel) PD166285, wortman- esis may lead to identification of the biochemical and genetic nin, and AG490 were dissolved in dimethyl sulfoxide (DMSO) controls of myeloid cell proliferation and differentiation, as and stored at −20°C prior to use. well as to the development of better therapeutics. Hypothesizing that dysregulation of hematopoietic growth factor-induced proliferative signalling pathways contribute to Cell tissue leukemogenesis, we have been studying the immediate bio- chemical events following stimulation of factor-dependent Primary leukemic blasts were isolated by Ficoll–Hypaque sep- hematopoietic cells by G-CSF and GM-CSF. By changing acti- aration of bone marrow aspirates from newly diagnosed chil- vation states of or recruiting additional signalling dren with acute leukemia. Fresh myeloid blasts were obtained molecules, protein tyrosine kinases (PTK) transduce signals from diagnostic marrow aspirates of patients treated at Chil- that control cell proliferation and differentiation. One major dren’s Hospital of Pittsburgh. Preparations were Ͼ95% viable 2 family of PTK is related to the oncogene product, Src. Evi- leukemic blasts. Cryopreserved blast cells were obtained from dence has accumulated from a variety of experimental Dr Irwin Bernstein at the Children’s Group Myeloid approaches that the Src-related protein tyrosine kinase Lyn Leukemia Repository (Fred C Hutchinson Cancer Center, plays a critical role in mediating GM-CSF and G-CSF Seattle, WA, USA). The human myeloid leukemia cell lines 3–6 responses and myeloid leukemic cell proliferation. Having MO7e and TF-1 were grown in RPMI 1640 medium, sup- shown that the Lyn kinase is constitutively active in myeloid plemented with 10% fetal calf serum (Hyclone, Orham, UT, USA), 2 mml-glutamine, 10 U/ml penicillin, 10 U/ml strepto- mycin (Life Technologies, Grand Island, NY, USA) and Correspondence: S Corey, Division of Hematology-Oncology, Chil- dren’s Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh, PA 10 ng/ml of recombinant human GM-CSF (provided by Gen- ° 15213, USA; Fax: 412 692 6637 etics Institute) at 37 C with 5% CO2. The HL-60 cell line was Received 26 November 1998; accepted 26 February 1999 grown in RPMI and 10% FCS, and the DT40 cell line was Lyn in myeloid leukemia V Roginskaya et al

856 × 2 TP buffer (10 mm MgCl2,10mm MnCl2,40mm Tris-Cl, 5 ␮m ATP, and 10 ␮Ci/sample ␥-32P-ATP). After the 20-min reaction was stopped by the addition of 2× Laemmli buffer, the samples were boiled for 5 min and loaded on to a 7.5% polyacrylamide gel. The gel was fixed, dried and exposed to Dupont film for autoradiography. For the JAK2 activation assay, proteins immunoprecipitated with anti-Jak2 antibody and Staphylococcus protein A (Sigma) were washed three times with PBS/1%NP-40 and loaded on to a 7.5% SDS- PAGE. The gel was transferred on to PVDF filter paper and blotted with monoclonal antibody 4G10 to phosphotyrosine. Detection was by enhanced chemiluminescence (Amersham, Arlington Heights, IL, USA).

Oligonucleotide treatment of MO7e cells

Nineteen antisense and sense phosphorothioate (S-) oligodeoxynucleotides, which included the ATG translation start site of Lyn sequence, were synthesized and purified (Midland Certified Reagent, Midland, TX, USA). The antisense S-oligonucleotide sequence consisted of 5Ј-ATT TTA TAC ATC CCA TAT T-3Ј and the sense S-oligonucleotide sequence consisted of 5Ј-AAT ATG GGA TGT ATA AAA T-3Ј. MO7e cells were cultured in triplicate at 10 000 cells/well in flat bot- tom microtiter plates in the presence of 0.3 ng/ml GM-CSF. Multiple doses of the S-oligonucleotides were added at the indicated concentrations at 0, 18, and 24 h. The cultures were pulsed with 1 ␮Ci3H-thymidine (NEN Dupont, Boston, MA, USA) for the last 4 h of culture. The plates were harvested on to glass fiber filters with a semi-automated cell harvester, 1 ml Cytoscint scintillation fluid (ICN Biomedical, Irvine, CA, USA) was added, and the samples were counted on a Beta counter (Packard, Downers Grove, IL, USA).

Src kinase inhibition of myeloid leukemic cell lines, Figure 1 (a) Constitutive activation of tyrosine kinases in AML blasts, and bone marrow cells blast cells. Mononuclear cells from two patients with M1 type AML were assayed for constitutive kinase activity. Src-related PTK are: Src, Bone marrow-derived blast cells from patients with AML or ZAP-70, Fyn, Yes, Lyn, Blk, Hck, Fgr, and Lck. Syk and ZAP belong progenitor cells from healthy individuals were obtained by to a separate sub-family of cytosolic PTK. Syk is ubiquitously Ficoll–Hypaque separation. Mononuclear cells were plated in expressed while ZAP-70 is expressed primarily in lineage lym- phocytes. Btk is a member of the Tec family of cytosolic PTK, and it methylcellulose supplemented with GM-CSF, erythropoietin, is chiefly expressed in myeloid and B cells. (b) Activation of Jak2 in stem cell factor, and IL-3 (Methocult, Stemcell Technologies, AML blast cells. Mononuclear cells from a patient with M2 AML were Vancouver, BC, Canada), with DMSO or PD166285 and col- treated −/+ 100 ng/ml G-CSF for 10 min and then lysed. Jak2 was onies were counted at 17 days. Leukemic cells were also incu- immunoprecipitated with anti-Jak2 antibody and electrophoresed pro- bated in RPMI and 10% FCS with DMSO or PD166285, and teins were transferred and blotted with anti-phosphotyrosine antibody. short-term growth was measured.

grown in RPMI supplemented with 10% FCS, and 2% chicken Informed consent serum (Sigma). Analysis of primary leukemic blasts was performed on residual patient samples. Informed consent for bone marrow examin- In vitro assay ation was obtained, and permission to use residual tissue was given by Children’s Hospital of Pittsburgh Institutional Leukemia blasts were isolated by Ficoll–Hypaque separation, Review Board. and lysed with detergent buffer (1% NP40, 137 mm NaCl, 20 mm Tris-HCl, pH 8, the protease inhibitors 1 mm EDTA, 1mm PMSF, 10 ␮g/ml aprotinin, 10 ␮g/ml leupeptin, and the Results tyrosine inhibitor 2 mm sodium orthovanadate). Post-nuclear supernatants were obtained by microcentrifug- Expression of Src kinase activity in primary leukemic ation and were standardized by assaying 400 ␮g protein per blast cells immunoprecipitation. Immune complexes were washed three times with PBS/1% NP-40, twice with TNE (100 mm NaCl, Constitutive protein tyrosine kinase activity was determined 10 mm tris-Cl, 1 mm EDTA), and then incubated in 20 ␮m of by immune complex kinase assay, either on fresh or cryopre- Lyn in myeloid leukemia V Roginskaya et al 857 Table 1 Activation of protein tyrosine kinases in fresh human However, when blast cells were obtained from the bone mar- leukemic blasts row aspirate of a patient with M2 AML, the blast cells showed activation of Jak2 following exposure to G-CSF (Figure 1b). Blk Fyn Hck Lck Lyn Jak2 This indicates that functional Jak2 is present, but not constitut- ively active, in primary myeloid leukemia cells. AML-fresha 0/0 0/3 2/7 0/0 6/7 0/4 AML-cryopreserved 0/12 3/14 8/14 4/14 10/14 0/5 a Lyn antisense treatment results in inhibition of GM- Data are represented as No. positive/No. tested. CSF-induced proliferation

To determine if Lyn played a role in promoting myeloid leu- served samples. Kinase activity was determined by autophos- kemic cell growth, we targeted Lyn by genetic and pharmacol- phorylation, as exemplified in Figure 1a. We studied the Src ogic means. The expression and function of Lyn may be affec- kinases found predominantly in myeloid cells (Hck and Lyn) ted by treating cells with antisense oligonucleotides.5,13,14 We and compared them to those found predominantly in lymph- therefore studied the effects of antisense treatment on GM- oid tissues (Blk, Fyn, and Lck). The majority of patient samples CSF-induced proliferation. 19-mer antisense and sense phos- demonstrated constitutive Lyn activity (Table 1). Whenever phorothioate (S-) oligodeoxynucleotides, which included the Hck was activated, so was Lyn. In 4/14 patients, the lymphoid ATG translation start site of lyn sequence, were synthesized Src kinases Lck and Fyn were activated. Fyn was found to be and purified. Lyn kinase activity was inhibited by treatment activated in the presence of Lck. The FAB classification of with antisense, but not nonsense, oligonucleotides (Figure 2a). these four patients were M1, M2, and M4. Four of the 20 Incubation of MO7e cells with a lyn antisense oligonucleotide patients studied showed no kinase activity, and the FAB classi- resulted in a dose-dependent decrease in GM-CSF-induced fication comprised M1, M4, and M5. Due to sample size, stat- proliferation (Figure 2b). The apparent therapeutic window for istical significance of constitutive kinase activity cannot be lyn antisense S-oligonucleotide is narrow since greater than attributed to either prognosis or FAB classification. 2.5 ␮m concentrations of either antisense or sense S-oligonu- Because Jak2 plays a critical role in hematopoietic cell sig- cleotides resulted in poor mitogenic response. Although the nalling, we studied Jak2 activation by phosphotyrosine blot- toxic effects of the sense oligonucleotide develop at concen- ting of Jak2 immunoprecipitated protein. Constitutive acti- trations greater than 1.25 ␮m, a difference of 30% inhibition vation of Jak2 was found in none of the nine patients studied. remains at this concentration. To determine the efficacy of lyn antisense treatment in preventing Lyn protein synthesis, and hence kinase activity, MO7e cells were treated with 1.25 ␮m S-oligonucleotides for 24 h, harvested, and assayed for Lyn activity. Densitometric analysis showed that lyn antisense treatment resulted in a 42 ± 10% decrease in kinase activity (data not shown). The lack of complete inhibition of lyn anti-

Figure 2 (a) Antisense inhibition of Lyn kinase activity. 106 cells were treated with 1 ␮m of nonsense or antisense LYN oligonucleotides Figure 3 Dose-response of tyrosine kinase inhibitor PD166285 on for 6 h, then lysed, and proteins were immunoprecipitated with anti- Lyn in TF-1, HL60 cells, and DT40. TF-1 and Lyn antibody. In vitro kinase assay was performed, products elec- HL60 cells, which express multiple Src kinases, and DT40 cells, trophoresed, and an autoradiograph was made of the gel. (b) Antisense which express only Lyn, were treated during mid-log phase of growth inhibition of GM-CSF-induced DNA synthesis. The GM-CSF-depen- with varying concentrations of PD166285 or DMSO (control diluent) dent MO7e cell line was incubated with the indicated concentrations for 1 h. Cells were then lysed, and immunoprecipitated Lyn was of either the LYN antisense or sense oligonucleotides. Cells were lab- assayed for kinase activity. The gel was analyzed by Molecular eled for the last 4 h of a 24-h culture period. Results are expressed Dynamics Phosphorimager, and % of control kinase activity was cal- as % c.p.m. of untreated cells, ± standard error. culated. Results shown are the averages. Lyn in myeloid leukemia V Roginskaya et al 858

Figure 4 Inhibition of leukemic cell growth by PD166285. (a) HL60 and (b) TF-1 cells were incubated with different concentrations of PD166285 (PD166) (100 nm,1␮m,30␮m), doxorubicin (doxo) (2 ␮g/ml, 1 ␮g/ml, or 0.2 ␮g/ml), both (PD166285 30 ␮m and doxorubicin 0.2 ␮g/ml) or the diluent DMSO as a control. Cell growth was measured at 24 and 48 h. Data from the 48 h time-point are presented as the averages ± standard deviation. (c) Leukemic blasts from patients with AML were incubated in RPMI and 10% FCS and PD166285 and/or DMSO. The average number of viable cells were counted daily and are presented. Lyn in myeloid leukemia V Roginskaya et al 859 sense on GM-CSF-induced proliferation of MO7e cells can at Table 2 Effects of Src and PI 3-kinase inhibitors on DT40 cell least be explained by its partial inhibition of protein viability expression. Unfortunately, at greater concentrations the toxicity of S-oligonucleotides obscured differences between Treatment PD166 PD166 PD166 DMSO 0.1 ␮M 1 ␮M 10 ␮M antisense and sense treatment. DMSO 43a 0 0 100 Wortmannin 0.1 ␮M 38 — — 43 Pharmacologic inhibition of Lyn by PD166285 Wortmannin 1 ␮M 10 — — 17 Wortmannin 10 ␮M 4—— 13 PD166285 is a novel protein tyrosine kinase inhibitor, one of the most potent analogs of the pyrido[2,3-d] pyrimidine class, aExpressed as % of control cells. which acts as an ATP competitive inhibitor. PD166285 DT40 cells were treated with varying concentrations of PD166285 (PD166) and/or Wortmannin. Cell viability was determined by trypan inhibits Src activity in vitro with an IC50 of 8 nm and receptor- encoded tyrosine kinase (eg FGFR-1, EGFR, and PDGFR) blue exclusion at 24 h. was confirmed by flow cytometric 7 analysis of propidium iodine-stained cells (data not shown). activity in vitro with an IC50 of 40–100 nm. Its inhibition of PDGFR was long-lasting. As an inhibitor of an early signalling event, PD166285 also effects downstream biochemical events such as MAP kinase activity. PD166285 inhibited Lyn kinase stained with propidium iodine and analyzed for DNA content activity in HL-60 and TF-1 cells with an IC50 of 500 nm and 100 nm, respectively (Figure 3). In contrast to both HL60 and by flow cytometry. Both HL60 and TF-1 cells showed early TF-1 cells, which express multiple Src kinases, the DT40 B apoptotic changes to PD166285 and wortmannin, which were cell line expresses only Lyn. Thus, DT40 is a unique cell line greater than that observed in low-dose doxorubicin (Table 3). to study the effects of a single functioning Src kinase.15 The

IC50 of Lyn kinase inhibition in DT40 cells was 100 nm. PD166285 was specific for Src kinase activity as there was no Inhibition of hematopoietic colony growth effect on -induced activation of Jak2 in any of the cell lines tested (data not shown). Mononuclear bone marrow cells from healthy bone marrow transplant donors were obtained by Ficoll–Hypaque separ- ation. Varying concentrations of PD166285, wortmannin, or PD166285 inhibition of myeloid leukemic cell lines doxorubicin were added to the cultures at day 0. Colonies and myeloid leukemic blasts were then counted 17 days later. While there was a dose- response inhibition by the Src kinase inhibitor on the growth Varying concentrations of the Src kinase inhibitor PD166285 of BFU-E and CFU-GM derived from bone marrow cells, no and/or doxorubicin were added to freshly split HL60 and TF- such effect was seen with the PI 3-kinase inhibitor (Figure 5). ␮ 1 cells, and the number of viable cells were counted at 24 Doxorubin at 0.2 g/ml completely inhibited growth of hema- and 48 h. 50% growth inhibition occurred at approximately topoietic colonies. Of note, bone marrow cells cultured in 5 ␮m for both HL60 and TF-1 cells (Figure 4a and b). Treat- liquid medium and 10% FCS and treated with PD166285 ment of HL60 and TF-1 cells with AG490, a relatively specific failed to generate fibroblasts (data not shown). Fibroblasts inhibitor of Jak2, showed a 50% growth inhibition at 10 ␮m grew in both wortmannin and doxorubicin-treated cells. It is and 100 ␮m, respectively. possible that one mechanism of action for the tyrosine kinase When both PD166285 and a low-dose of doxorubicin were inhibitor is to block stromal growth. In contrast to the pro- added, an additive effect in growth inhibition was seen in both found inhibition on in vitro methylcellulose colony growth by cell lines. The DT40 cells, which express a single Src kinase, PD 166285, mice treated with PD166285 did not have were exquisitely sensitive to PD166285. No cells excluded hematologic toxicity (data not shown). trypan blue at 24 h at PD166285 concentrations of 1 or 10 ␮m. The enhanced sensitivity to the Src kinase inhibitor in DT40 cells may be due to the lack of redundant Src kinases. This possibility is suggested from the ability of 1 ␮m Table 3 Effects of inhibitors on the sub-G1 peak in DNA flow PD166285 to inhibit PI 3-kinase activity in DT40 cells but not cytometric analysis in TF-1 and HL60 cells (data not shown). An additive effect on growth inhibition was seen in cells treated with 0.1 ␮m Cell line Control PD166 Wortmannin Doxorubicin incubation 10 ␮M 10 ␮M 0.2 ␮g/ml PD166285 and the PI 3-kinase inhibitor wortmannin (Table 2). G-CSF treatment of DT40 cells which express the human HL60 G-CSF receptor did not rescue the effects of the Src or PI 3- 16 h 13a 28 21 13 kinase inhibitors (data not shown). 24 h 14 35 25 16 Fresh leukemia cells were isolated from a diagnostic bone marrow aspirate from a patient with AML, M2 subtype. Cell TF1 population, comprising Ͼ95% blasts, were suspended in 16 h 35 53 49 35 RPMI and 10% FCS and either PD166285, wortmannin, 24 h 25 66 31 46 doxorubicin or both PD166285 and doxorubicin. While each a agent resulted in a time-dependent decrease in viable cells, Expressed as percent of events. Freshly split cells were treated with either the Src kinase inhibitor as compared to DMSO-treated control cells, the combination PD166285 (PD166), Pi 3-kinase inhibitor wortmannin, or of PD166285 and doxorubicin resulted in no viable cells by doxorubicin. Cells were harvested at 16 or 24 h, fixed and stained 72 h (Figure 4c). with propidium iodide. Analysis was performed at the University of To determine the effects on apoptosis, treated cells were Pittsburgh Cancer Institute core flow cytometric facility. Lyn in myeloid leukemia V Roginskaya et al 860

Figure 5 Effects of inhibitors on hematopoietic colony growth. Mononuclear bone marrow cells were purified and placed in methylcellulose cultures with or without PD166285 (PD) (100 nm,1␮m, and 10 ␮m), wortmannin (wort) (100 nm,1␮m, and 10 ␮m), or doxorubicin (doxo) (0.02 ␮g/ml and 0.2 ␮g/ml). Colonies were counted on day 17. Data are presented as the means ± s.d.

Discussion Corey). In a complementary approach, we treated the GM- CSF-dependent cell line MO7e with antisense oligonucleo- Leukemia may result from autocrine or paracrine production tides to Lyn and observed a dose-dependent inhibition of of hematopoietic growth factors, such as GM-CSF. The critical GM-CSF-induced growth. dependency of acute myeloid leukemias (AML) for GM-CSF These experiments demonstrate that PTK contribute signifi- has been suggested by either the requirement of exogenous cantly to leukemic cell growth and suggest that therapy GM-CSF for in vitro culture of AML cells or the aberrant pro- directed against Src kinases, such as Lyn, may be effective in duction of either GM-CSF or IL-3 by leukemic cells.16–19 Leu- inhibiting myeloid cell growth. Because the activation of Lyn kemia may also result from mutation of the cytokine receptors. occurs rapidly and proximal to engage- For example, a subset of patients with severe congenital neu- ment, it activates downstream PTK such as Syk and Btk (a tropenia who develop myelodysplasia or AML have mutations member of the Tec family of PTK). Lyn is also associated with in their G-CSF receptors.20,21 These mutations introduce a other important growth-promoting signalling molecules such premature stop codon, which results in a receptor which trig- as PI 3-kinase, Cbl, and cdc2.27–30 While Lyn appears to be gers proliferative but not differentiative signalling cascades. most relevant Src-related PTK in myeloid cells, other Src kin- , such as GM-CSF and G-CSF, typically stimulate ases such as Hck or Fyn participate in myeloid cell signal- rapid changes in intracellular tyrosine phosphoproteins. One ling.31 Selective inhibition of Src-related PTK may offer a important class of PTK is a family related to the oncogene v- highly effective, but less toxic addition to the treatment of src. Several src-related PTK are found in myeloid cells: Fgr, myeloid leukemias. Furthermore, these studies should prompt Hck, and Lyn. We and others have reported that the GM-CSF the development of agents which inhibit Pl 3-kinase activity. or IL-3 increases the tyrosine kinase activity of Hck and GM- CSF, IL-3, and G-CSF increase the activity of Lyn in myeloid- derived cells.4–6,22,23 Constitutive Lyn activity has been found Acknowledgements in several factor-independent myeloid cell lines. Myeloid cells transformed with BcrAbl demonstrated an association This work was supported by grants from the Leukemia Society between BcrAbl and Lyn and greater Lyn activity, suggesting of America (Translational Award to SJC), American Cancer 24 a possible role for Lyn in chronic myelogenous leukemia. Society DHP-135 and DHP-168 (SJC), and the National Insti- Here, we report that constitutive Lyn activity is found in leu- tutes of Health (SJC), Dr Corey was an AMA Florence Carter kemic blasts freshly purified from the bone marrow aspirates Fellow in Leukemia. of newly diagnosed patients with leukemia. Because new forms of anti-leukemic therapy can be based on targeting growth-promoting events, investigators have stud- References ied inhibitors of tyrosine kinases. The non-specific inhibitor genistein is highly effective in inhibiting leukemic cell 1 Mayer RJ, Davis RB, Schiffer CA, Berg DT, Powell BL, Schulman 25,26 growth. We have asked whether Lyn specifically contrib- P, Omura GA, Moore JO, Mclntyre OR, Frei E 3rd. Intensive post- utes to G-CSF or GM-CSF-stimulated DNA synthesis. We remission chemotherapy in adults with acute myeloid leukemia. introduced components of the human G-CSF receptor or GM- Cancer and Leukemia Group B. New Engl J Med 1994; 331: CSF receptor complex into a bursal lymphoma cell line and its 896–903. mutagenized, lyn-deficient form. G-CSF and GM-CSF induced 2 Brown MT, Cooper JA. Regulation, substrates and functions of src. Biochim Biophys Acta 1996; 1287: 121–149. DNA synthesis was found only when the G-CSF or GM-CSF 3 Torigoe T, O’Connor R, Fagard R, Fischer S, Santoli D, Reed JC. receptors were expressed in the cell line which contains Lyn Regulation of SRC-family protein tyrosine kinases by interleukins, (Ref. 15 and unpublished observations, SM Anderson and SJ IL-2, and IL-3. Leukemia 1992; 3: 94S–97S. Lyn in myeloid leukemia V Roginskaya et al 861 4 Corey S, Eguinoa A, Puyana-Theall K, Bolen JB, Cantley L, Molli- JD. The effects of GM-CSF and G-CSF in promoting growth of nedo F, Jackson TR, Hawkins PT, Stephens LR. – clonogenic cells in acute myeloblastic leukemia. Blood 1987; 69: colony-stimulating factor activates Ptdlns 3OH-kin- 1771–1776. ase via src-related tyrosine kinase(s) in human myeloid-derived 17 Young DC, Griffin JD. Autocrine secretion of GM-CSF in acute cells. EMBO J 1993; 12: 2681–2690. myeloblastic leukemia. Blood 1986; 68: 1178–1181. 5 Wei S, Liu JH, Epling-Burnette PK, Gamero AM, Ussery D, Pearson 18 Griffin JD, Lowenberg B. Clonogenic cells in acute myeloblastic EW, Elkabani ME, Diaz JI, Djeu JY. Critical role of Lyn kinase in leukemia. Blood 1986; 68: 1185–1195. inhibition of apoptosis by granulocyte–macrophage 19 Young DC, Wagner K, Griffin JD. Constitutive expression of the colony-stimulating factor. J Immunol 1996; 157: 5155–5162. granulocyte–macrophage colony-stimulating factor gene in acute 6 Corey SJ, Burkhardt AL, Bolen JB, Geahlen RL, Tkatch LS, Tweardy myeloblastic leukemia. J Clin Invest 1987; 79: 100–106. DJ. G-CSF receptor signalling involves the formation of a three- 20 Dong F, Brynes RK, Tidow N, Welte K, Lowenberg B, Tuow IP. component complex with lyn and syk protein-tyrosine kinases. Mutations in the gene for the granulocyte colony-stimulating fac- Proc Natl Acad Sci USA 1994; 91: 4683–4687. tor receptor in patients with acute myeloid leukemia preceded by 7 Panek RL, Lu GH, Klutchko SR, Batley BL, Dahring TK, Hamby severe congenital neutropenia. New Engl J Med 1995; 333: JM, Hallak H, Doherty AM, Keiser JA. In vitro pharmacological 487–493. characterization of PD 166285, a new nanomolar potent and 21 Touw IP, Schelen A, Brynes RK, Tidow W, Welte K, Lowenberg B, broadly active protein tyrosine kinase inhibitor. J Pharm Exp Ther Dong F. Point mutations in the G-CSFR gene in severe congenital 1997; 283: 1433–1444. neutropenia terminating in acute myeloid leukemia. Blood 1994; 8 Burkhardt AL, Brunswick M, Bolen JB, Mond JJ. Anti-immunoglob- 84: 375. ulin stimulation of B lymphocytes activates src-related protein- 22 Anderson SM, Jorgensen B. Activation of src-related tyrosine kin- tyrosine kinases. Proc Natl Acad Sci USA 1991; 88: 7410–7414. ases by IL-3. J Immunol 1995; 155: 1660–1670. 9 Burkhardt AL, Bolen JB, Kieff E, Longnecker R. An Epstein–Barr 23 Torigoe T, O’Connor R, Santoli D, Reed JC. Interleukin-3 regulates virus transformation-associated membrane protein interacts with the activity of the LYN protein-tyrosine kinase in myeloid-commit- src family tyrosine kinases. J Virol 1992; 66: 5161–5167. ted leukemic cell lines. Blood 1992; 80: 617–624. 10 Bell GM, Bolen JB, Imboden JB. Association of Src-like protein 24 Danhauser-Riedl S, Warmuth M, Druker BJ, Emmerich B, Hallek M. Activation of Src kinases p53/56lyn and p59hck by tyrosine kinases with the CD2 cell surface molecule in rat T lym- p210bcr/ in myeloid cells. Cancer Res 1996; 56: 3589–3596. phocytes and natural killer cells. Mol Cell Biol 1992; 12: 5548– 25 Uckun FM, Evans WE, Forsyth CJ, Waddick KG, Ahlgren LT, Chel- 5554. strom LM, Burkhardt A, Bolen J, Myers DE. Biotherapy of B-cell 11 Saouaf SJ, Mahajan S, Rowley RB, Kut SA, Fargnoli J, Burkhardt precursor leukemia by targeting genistein to CD19-associated AL, Tsukada S, Witte ON, Bolen JB. Temporal differences in the tyrosine kinases. Science 1995; 267: 886–889. activation of three classes of non-transmembrane protein tyrosine 26 Finlay GJ, Holdaway KM, Baguley BC. Comparison of the effects kinases following B-cell receptor surface engagement. of genistein and amsacrine on leukemia cell proliferation. Oncol Proc Natl Acad Sci USA 1994; 91: 9524–9528. Res 1995; 6: 33–37. 12 Gauen LK, Zhu Y, Letourneur F, Hu Q, Bolen JB, Matis LA, 27 Dombrosky-Ferlan PM, Corey SJ. Yeast two-hybrid in vivo associ- Klausner RD, Shaw AS. Interactions of p59fyn and ZAP-70 with ation of src kinase Lyn with the proto-oncogene product Cbl but T-cell receptor activation motifs: defining the nature of a signalling not with the p85 subunit of PI 3-kinase. Oncogene 1997; 14: motif. Mol Cell Biol 1994; 14: 3729–3741. 2019–2024. 13 Katagiri K, Yokoyama KK, Yamamoto T, Omura S, Irie S, Katagiri 28 Uckun FM, Tuel-Ahlgren L, Waddick KG, Jun X, Jin J, Myers DE, T. Lyn and Fgr protein-tyrosine kinases prevent apoptosis during Rowley RB, Burkhardt AL, Bolen JB. Physical and functional inter- retinoic acid-induced granulocytic differentiation of HL-60 cells. actions between Lyn and p34cdc2 kinases in irradiated human B- J Biol Chem 1996; 271: 11557–11566. cell precursors. J Biol Chem 1996; 271: 6389–6397. 14 Scheuermann RH, Racila E, Tucker T, Yefenof E, Street NE, Vitetta 29 Kharbanda S, Saleem A, Yuan ZM, Kraeft S, Weichselbaum R, ES, Picker LJ, Uhr JW. Lyn tyrosine kinase signals cell cycle arrest Chen LB, Kufe D. Nuclear signaling induced by ionizing radiation but not apoptosis in B-lineage lymphoma cells. Proc Natl Acad involves colocalization of the activated p56/p53lyn tyrosine kin- Sci USA 1994; 91: 4048–4052. ase with p34cdc2. Cancer Res 1996; 56: 3617–3621. 15 Corey SJ, Dombrosky-Ferlan PM, Zuo S, Krohn E, Donnenberg 30 Pathan NI, Geahlen RL, Harrison ML. The protein-tyrosine kinase AD, Zorich P, Romero G, Takata M, Kurosaki T. Requirement of Lck associates with and is phosphorylated by Cdc2. J Biol Chem src kinase lyn for induction of DNA synthesis by granulocyte col- 1996; 271: 27517–27523. ony-stimulating factor. J Biol Chem 1998; 273: 3230–3235. 31 Corey SJ, Anderson SM. Src-related protein tyrosine kinases in 16 Vellenga E, Young DC, Wagner K, Wiper D, Ostapovicz D, Griffin hematopoiesis. Blood 1999; 93: 1–14.