Inhibition of Abl tyrosine kinase enhances nerve growth factor-mediated signaling in Bcr–Abl transformed cells via the alteration of signaling complex and the receptor turnover

Article metrics


Receptor tyrosine kinase-mediated signaling is tightly regulated by a number of cytoplasmic signaling molecules. In this report, we show that Bcr–Abl transformed chronic myelogenous leukemia (CML) cell lines, K562 and Meg-01, express the receptor for nerve growth factor (NGF), TrkA, on the cell surface; however, the NGF-mediated signal is not particularly strong. Treatment with imatinib, a potent inhibitor of Bcr–Abl tyrosine kinase, downmodulates phosphorylation of downstream molecules. Upon stimulation with NGF, Erk and Akt are phosphorylated to a much greater degree in imatinib-treated cells than in untreated cells. Knockdown of expression of Bcr–Abl using small interfering RNA technique also enhanced NGF-mediated Akt phosphorylation, indicating that Bcr–Abl kinase modifies NGF signaling directly. Imatinib treatment also enhanced NGF signaling in rat adrenal pheochromocytoma cell line PC12 that expresses TrkA and c-Abl, suggesting that it is not only restoration of responsiveness to NGF after blocking oncoprotein activity, but also c-Abl tyrosine kinase per se may be a negative regulator of growth factor signaling. Furthermore, inhibition of Abl tyrosine kinase enhanced clearance of surface TrkA after NGF treatment and simultaneously enhanced NGF-mediated signaling, suggesting that as in neuronal cells ‘signaling endosomes’ are formed in hematopoietic cells. To examine the role of TrkA in CML cells, we studied cell growth or colony formation in the presence or absence of imatinib with or without NGF. We found that NGF treatment induces cell survival in imatinib-treated CML cell lines, as well as colony formation of primary CD34+ CML cells, strongly suggesting that NGF/TrkA signaling contributes to aberrant signaling in CML.


Nerve growth factor (NGF) is a member of the family of neurotrophins and is essential for the survival and differentiation of neurons in central and peripheral nerve systems (Snider 1994). The binding of NGF to its high affinity receptor, TrkA, causes activation of the receptor-associated tyrosine kinase leading to autophosphorylation of the cytoplasmic domain at multiple tyrosine residues, including Y490, Y670, Y674, Y675 and Y785. The newly formed phosphotyrosines constitute binding sites for the Src homology (SH2) domain or phosphotyrosine-binding domain containing proteins, which are thought to participate in the control of mitogenic, survival or differentiation pathways. Evidence has accumulated that neurotrophins and their receptors involve hematopoietic cell development and leukaemogenesis (Matsuda et al., 1988; Chevalier et al., 1994; Kaebisch et al., 1996; Mulloy et al., 2005). TrkA is expressed in acute myelogenous leukemia and chronic myelogenous leukemia (CML); however, NGF/TrkA signaling in leukemic cells has not yet been well studied.

Chronic myelogenous leukemia is a clonal stem cell disease characterized by the t(9;22) (q34;q11) translocation, the Philadelphia (Ph) chromosome, which is associated with the presence of Bcr–Abl fusion protein. It has been well documented that Bcr–Abl is constitutively activated in its tyrosine kinase, and plays an important role in preventing programmed cell death (Scheijen and Griffin, 2002). Imatinib mesylate, a potent inhibitor of Abl tyrosine kinase, emerged as the frontline therapy for early diseased CML patients, since it had the highest selectivity for growth inhibition of Bcr–Abl-expressing cells (Buchdunger et al., 1996; Druker et al., 1996, 2007; O’Brien et al., 2003; Deininger et al., 2005). Although imatinib is remarkably successful in treating patients with CML, sustained molecular remissions have only rarely been observed. Acquired resistance to imatinib is frequently caused by point mutations in Bcr–Abl or by expression of the fusion protein at high level (Druker et al., 1996, 2001; von Bubnoff et al., 2003).

Using Abl tyrosine kinase inhibitor, imatinib, and Bcr–Abl-specific small interfering RNA technique, we show here that the constitutively active Abl tyrosine kinase, Bcr–Abl, suppresses signaling mediated by growth factor NGF via modification of downstream signaling complexes as well as surface TrkA clearance. Furthermore, we have demonstrated that, for the first time, NGF/TrkA signaling enhanced colony formation of imatinib-treated primary CD34+ CML cells, suggesting that NGF/TrkA signaling may participate in leukemia development.


TrkA is expressed in leukemic cell lines on the cell surface

We first examined TrkA expression in several leukemia cell lines, including K562 (human CML in blast crisis), CML-T1 (human CML in lymphoid blast crisis), Meg-01 (human CML in megakaryocytic blast crisis), BV-173 (human CML in lymphoid blast crisis), the oncogenic c-Kit transformed human mastocytoma cell line HMC-1 (Butterfield et al., 1988), human promyelocytic leukemia cell line HL60 and human histocytic lymphoma U937. Trk tyrosine kinase could be clearly detected in K562, Meg-01 and HMC-1 by western blotting (Figure 1a). As the activated TrkA is found predominantly in intracellular membranes in some systems (Rajagopal et al., 2004), we next examined the cell surface expression of TrkA in leukemia cells by labeling the cell surface expressing TrkA with biotin. As shown in Figure 1b, TrkA was detected on the cell surface in K562 cells. As controls, we applied TrkA expressing HEK293 and HMC-1 for this experiment (Tam et al., 1997). We and others previously demonstrated the direct association of TrkA with Abl tyrosine kinase (Koch et al., 2000; Yano et al., 2000), raising the question of whether TrkA is phosphorylated by Bcr–Abl in the absence of NGF in CML cells. Therefore, we examined TrkA phosphorylation using an antibody specific for phosphorylated tyrosine residue 490 in TrkA before and after NGF stimulation. TrkA is not phosphorylated in the absence of NGF but is phosphorylated upon stimulation with NGF within 15 min (Figure 1c), indicating that TrkA binds to the NGF and is then activated.

Figure 1

TrkA is expressed in K562, Meg-01 and HMC-1 cells. (a) Cell extracts from 1 × 106 of K562, HL60, BV-173, CML-T1, Meg-01, HMC-1 and U937 cells were analysed by western blotting (WB) using anti-Trk, c-Kit, c-Abl or actin antibodies. HEK293 cells and HEK293 cells expressing human TrkA (HEK293/TrkA) were applied as a control. (b) K562, HMC-1, HEK293 and HEK293/TrkA cells (7.5 × 107) were labeled with a membrane impermeable biotinylation reagent. Cells were then lysed for precipitation. Cell extracts were precipitated by anti-Trk antibody/protein A Sepharose (IP: Trk) or streptavidin sepharose (P: streptavidin: cell surface Trk). Precipitates were analysed by Trk-specific western blotting (WB: Trk). (c) 2 × 106 of K562, HMC-1 and HEK293/TrkA cells were incubated for 24 h with serum-free medium and were then incubated with or without imatinib (10 μM) 1 h before NGF stimulation. Cells were stimulated with NGF (100 ng/ml) for 5 or 15 min in the presence or absence of imanitib. Cell lysates were analysed by western blotting (WB) using antibodies against phospho-Y490 Trk (pY490 Trk) or TrkA (Trk). CML, chronic myelogenous leukemia; NGF, nerve growth factor.

NGF-induced phosphorylation of Erk and Akt are impaired in CML cells

To determine whether TrkA-mediated signaling is induced upon stimulation with NGF in the leukemia cell lines, we examined phosphorylation of Erk and Akt. Phosphorylation of both kinases was only slightly upregulated in K562 and Meg-01 cells upon NGF stimulation (Figure 2a). Fifteen minutes after stimulation with NGF, both Erk and Akt were phosphorylated in these cells. As NGF-mediated signaling is very modest, we examined whether Bcr–Abl tyrosine kinase activity downmodulates the TrkA-signaling pathway directly. For this purpose, we utilized Abl inhibitor imatinib.

Figure 2

Bcr–Abl suppresses phosphorylation of Akt and Erk induced by TrkA signaling. K562 and Meg-01 cells (2 × 106) were incubated for 24 h with serum-free medium and were then incubated with or without 10 μM imanitib (a) and/or Trk inhibitor, 100 nM K252a (b) for 1 h before NGF stimulation. After NGF (100 ng/ml) stimulation for 15 or 30 min, cell lysates were analysed by western blotting (WB) using anti-phospho-Erk1/2 (pErk1/2), Erk1/2, phospho-Akt (pSer Akt) or Akt antibodies as indicated. Phospho-Erk (pErk2) and phospho-Akt (pAkt) bands were quantified by TINA2.0 software, normalized to unphosphorylated protein and depicted as relative increase compared to unstimulated lysate in the absence or presence of imatinib. Bar diagrams show mean values±s.d. of several experiments (n: number of independent experiments). Paired t-test was applied to describe the significance of differences in phosphorylation in the absence (black bars) or the presence of imatinib (light grey bars) at 15 min of NGF stimulation. P-values 0.05 were considered as significant and indicated by bold letters. NGF, nerve growth factor.

Imatinib treatment enhances Akt and Erk phosphorylation induced by NGF/TrkA in K562 cells

Serum starved cells were treated with 10 μM imatinib for 1 h before NGF stimulation. As expected, phosphorylation of Erk and Akt kinase is downmodulated by imatinib treatment. Interestingly, upon stimulation with NGF, phosphorylation of Akt kinase was upregulated about two times in imatinib-treated K562 (P=0.0004) and Meg-01 (P=0.006) cells than untreated cells (Figure 2a), indicating that imatinib treatment enhances NGF-mediated signaling in CML cells. As expected, the treatment with imatinib and TrkA inhibitor K252a downmodulated NGF-mediated Akt phosphorylation (Figure 2b).

Silencing of Bcr–Abl expression in K562 cells enhanced TrkA-mediated signaling

Imatinib is a potent-specific inhibitor for Abl tyrosine kinase; however, one cannot rule out the possibility that imatinib treatment has effects other than inhibiting tyrosine kinases, so that TrkA signaling was somewhat enhanced. To examine whether Bcr–Abl inhibition per se causes the enhancement, we utilized short hairpin RNA (shRNA) directed against Bcr–Abl, which was previously established in K562 cells (Scherr et al., 2003). K562 cells were transduced with Lentivirus carrying shRNA directed against Bcr–Abl or GL4 (control virus) for 48 h. The cells were then incubated for 17 h with medium containing 0.1% fetal calf serum (FCS) and finally NGF-mediated phosphorylation of Akt and Erk was examined. In agreement with previous data (Scherr et al., 2003), Bcr–Abl expression was reduced by about 75% compared to the control virus-infected cells. In these cells, the levels of phospho-Akt and phospho-Erk 15 min after NGF stimulation were about two- and fivefold greater, respectively, than in control cells (Figure 3), suggesting that Bcr–Abl per se downmodulates TrkA signaling.

Figure 3

Silencing of Bcr–Abl expression by Bcr–Abl-specific shRNA enhances Akt phosphorylation after NGF stimulation in K562 cells. Cells (9 × 106) transduced with lentivirus carrying Bcr–Abl-specific (white bars) or control shRNA (black bars) for 48 h were incubated for 17 h with medium containing 0.1% serum and were then stimulated with NGF (100 ng/ml) for 15 or 30 min. (a) Cell lysates were analysed by western blotting (WB) using anti-phospho-Akt (pSer Akt), Akt, anti phospho-Erk (pErk1/2), Erk, glyceraldehyde-3-phosphate dehydrogenase, c-Abl, phospho-STAT5 (pSTAT5) and STAT5 antibodies as indicated. (b) Relative increase of phospho-Akt and phospho-Erk2 after NGF stimulation was compared to unstimulated lysate and quantification and statistic analysis were performed as described in Figure 2. NGF, nerve growth factor.

SHC phosphorylation by NGF stimulation was restored by imatinib treatment

As K562 cells contain multiple copies of Bcr–Abl, the signaling pathway required to induce survival and growth is already maximally stimulated, and the data obtained here might be due to restoration of the responsiveness to NGF expected by the normal counterpart of these cells. We therefore determined whether a particular pathway is restored by the treatment with imatinib. It has been well documented that CrkL binds to Bcr–Abl directly (Feller, 2001) and SH2 domain-containing inositol 5-phosphatase (SHIP)2 forms a complex with CrkL and Bcr–Abl (Wisniewski et al., 1999). In agreement with previous data (Wisniewski et al., 1999; Sattler et al., 2002), our results show that Bcr–Abl, GAB2 and SHIP2 co-precipitated with CrkL in K562 cells (Figure 4a). These complexes were reduced in the presence of imatinib (Figure 4a). We then examined phosphorylation of several downstream signaling molecules for 15 min to 6 h in the presence or absence of imatinib and/or NGF. In untreated K562 cells, STAT5, SHIP2 and CrkL were highly phosphorylated. After treatment with imatinib, these molecules were dephosphorylated. Upon stimulation with NGF, STAT5, SHIP2 and CrkL were not re-phosphorylated within 15 min in imatinib-treated K562 cells (Figure 4b), suggesting that NGF-mediated signaling is not via these signaling molecules. We next examined the adaptor protein Src homologous and collagen gene (SHC) that was identified as one of the two major binding proteins to the TrkA receptor and upstream signaling molecule for Erk and Akt activation (Obermeier et al., 1993; Stephens et al., 1994; Huang and Reichardt, 2003). In agreement with a previous report (Bonati et al., 2000), SHC is constitutively phosphorylated in Bcr–Abl transformed cells. Upon stimulation with NGF, phosphorylation of SHC was upregulated only 1.3 times (P=0.05); however, after imatinib treatment, SHC was 3.6-fold (P=0.03) more phosphorylated upon stimulation with NGF (Figure 4c), indicating that TrkA/SHC pathway is restored by imatinib treatment.

Figure 4

Imatinib treatment restores TrkA-mediated phosphorylation of SHC. (a) The complex of CrkL with Bcr–Abl, GAB2 and SHIP2 in K562 cells is not detectable after imatinib treatment. K562 cells were incubated for 24 h with serum-free medium and were then incubated with and without imatinib for 1 h. After NGF stimulation (100 ng/ml) for 15 min, cell lysates were precipitated with polyclonal antibodies against CrkL/ protein A Sepharose (IP: CrkL) then analysed by western blotting (WB) using anti-c-Abl, GAB2, SHIP2 or CrkL antibodies. As control, cell lysates were precipitated with rabbit IgG (IP: Ig)/ protein A Sepharose. (b) Imatinib treatment enhances TrkA-mediated phosphorylation of Erk and Akt, but not of CrkL, STAT5 and SHIP2. K562 cells were incubated for 24 h in the absence of serum and then treated with and without imatinib (10 μM) for 1 h. After imatinib treatment, NGF (50 ng/ml) was added for 15 min to 6 h in the presence or absence of imatinib. Cell lysates were analysed by western blotting (WB) using antibodies against phospho-Akt (pSer Akt), Akt, phospho-Erk (pErk1/2), Erk, phospho-SHIP2 (pSHIP2), SHIP2, phospho-STAT5 (pSTAT5), STAT5, phospho-CrkL (pCrkL) or CrkL. (c) Imatinib treatment restores TrkA-mediated phosphorylation of SHC. Cells were treated as in (a) and then cell lysates were precipitated by SHC-specific antibody. Immunoprecipitates were analysed by phosphotyrosine-specific immunoblot. Phosphorylation of the SHC isoform p52 or Erk2 was quantified, normalized to unphosphorylated protein and depicted as relative increase compared to unstimulated lysate in the absence or presence of imatinib. Bar diagrams (black: −NGF; light grey: +NGF) show mean values±s.d. of 3 independent experiments. Statistic analysis was performed as described in Figure 2. NGF, nerve growth factor; SHIP, SH2 domain-containing inositol 5-phosphatase.

Proto-oncogene product c-Abl kinase also influences Erk phosphorylation induced by NGF

To determine whether these results are simply due to a restoration of the responsiveness to NGF, or whether c-Abl also influences TrkA-mediated signaling, we applied the rat adrenal pheochromocytoma cell line PC12 cells that express TrkA and c-Abl (Figure 5a). In the PC12 cell system, it has been clearly demonstrated that Erk1/2 phosphorylation is one of the major TrkA signaling molecules (Huang and Reichardt, 2003). NGF-mediated Erk1 phosphorylation was about two times (P=0.03) greater than in the absence of imatinib (Figure 5b), indicating that c-Abl has a similar effect on NGF signaling. Furthermore, c-Abl activity also suppresses signaling induced by other growth factors, such as basic fibroblast growth factor (Figure 5c), or epidermal growth factor (EGF) (Figure 5d), suggesting that Abl tyrosine kinase may regulate growth factor signaling.

Figure 5

Imatinib treatment enhances Erk phosphorylation induced by NGF in PC12 cells. (a) PC12 cells express TrkA and c-Abl. Cell lysates from 1.5 × 106 PC12 and K562 cells were analysed by western blot (WB) using anti-c-Abl, Trk or actin antibodies. (bd) PC12 cells (5 × 106) were incubated for 24 h with serum-free medium and were then incubated with or without 10 μM imanitib for 1 h before growth factor stimulation. After NGF (100 ng/ml) (b), basic fibroblast growth factor (100 ng/ml) (c) or EGF (100 ng/ml) (d) stimulation for the indicated times, cell lysates were analysed by immunoblotting using anti-phospho-Akt (pSer Akt), Akt, phospho-Erk (pErk) or Erk antibodies as indicated. Phospho-Erk (pErk1 or 2) and phospho-Akt (pAkt) bands were quantified by TINA2.0 software, normalized to unphosphorylated protein and depicted as relative increase compared to unstimulated lysate in the absence of imatinib. Statistic analyses were performed as described in Figure 2. EGF, epidermal growth factor; NGF, nerve growth factor.

Imatinib treatment enhances clearance of surface receptor after NGF treatment

Abl tyrosine kinase has been shown to influence receptor degradation via c-Cbl-mediated ubiquitination (Tanos and Pendergast, 2006). Therefore, we first examined the half-life of TrkA after NGF stimulation in the presence or absence of imatinib with cyclohexamide. In both cases, about 50% of the TrkA was degraded within 40–45 min, indicating that imatinib treatment does not influence the receptor half-life (data not shown). Next, we examined cell surface TrkA after NGF treatment. Cell surface proteins were labeled with biotin and then isolated by streptavidin. Bound materials were analysed by Trk-specific immunoblot. In imatinib-treated cells, 40% of the receptors on the cell surface disappeared within 15 min after NGF treatment. In the absence of imatinib, only 16% (P=0.05) of the molecules disappeared from the cell surface within 15 min (Figure 6). In the presence of Monensin, a potent inhibitor of receptor recycling (Saxena et al., 2005), the level of surface expression of TrkA decreases more than 40% within 15 min in the presence or absence of imatinib, suggesting that Abl tyrosine kinase suppresses receptor internalization and/or enhances receptor recycling (Chen et al., 2005). As NGF signaling from clathrin-coated vesicles was demonstrated (Howe et al., 2001), we examined the correlation between phosphorylation of Akt and clearance of TrkA receptor from the cell surface. Upon stimulation with NGF, within 15 min the phosphorylation of Akt is enhanced about fourfold (P=0.02) in imatinib-treated cells than untreated cells. Interestingly, Monensin treatment alone also enhances the level of phospho-Akt about twofold (Figures 6c and d), suggesting that TrkA mediates signaling mainly from intracellular compartments. These data suggest that Bcr–Abl suppresses TrkA receptor internalization, thus TrkA-induced intracellular signaling is reduced.

Figure 6

Imatinib treatment enhances clearance of surface TrkA receptor after NGF treatment. K562 cells (6 × 107) were starved for 24 h with serum-free medium and were then incubated with (grey circles) or without 10 μM imanitib (black squares) and 9 μM Monensin (dotted lines and open symbols) for 1 h before growth factor stimulation. After NGF (200 ng/ml) stimulation for 5, 15 and 30 min, cell surface proteins were labelled with biotin. (a) Cell surface proteins were precipitated from cell extracts using streptavidin sepharose (P: streptavidin). Precipitates were analysed by Trk-specific western Blot (WB). The membrane was then stripped and reprobed with a transferrin receptor (TfR)-specific antibody. (b) Intensity of TrkA bands after NGF-induced internalization was quantified by TINA2.0 software and depicted as relative amount of surface protein (% cell surface Trk) compared to unstimulated lysate in the absence or presence of inhibitors. Trk levels were standardized using TfR as loading control. Data points represents the mean of at least four independent experiments±s.d. P-values were obtained as described in Figure 2 to indicate the significance of the difference between cell surface Trk in the absence and presence of imatinib at 15 and 30 min after NGF stimulation. (c) Lysates of the experiments described in (a) were subjected to phospho-Akt (pSer Akt), Akt, Trk and glyceraldehyde-3-phosphate dehydrogenase-specific western blotting (WB). (d) Intensity of phospho-Akt bands after 15 min NGF stimulation was analysed by TINA2.0 software, standardized with the level of Akt, and induction level in the presence of imatinib and Monensin was compared to phospho-Akt without inhibitor and NGF treatment (pSer Akt relative increase). The increase in phospho-Akt levels is depicted as mean of three independent experiments±s.d. Statistic analysis was carried out as described in Figure 2. NGF, nerve growth factor.

Zhang et al. (2000) reported that cell surface Trk receptors mediate NGF-induced survival signaling via Akt pathways. We next examined cell growth and survival in imatinib- and NGF-treated cells.

NGF treatment induces cell survival of TrkA expressing CML cell lines in the presence of imatinib

To examine whether NGF-TrkA signaling may rescue cells from imatinib-mediated cell death, we incubated K562, Meg-01, HMC-1 and U937 cells with or without NGF in the presence of imatinib for 4 days and the number of living cells was then counted. After treatment with 0.5 or 5 μM imatinib, more than 50% of all K562 cells died within 4 days. In the presence of NGF, the number of living K562 cells treated with imatinib increased by 1.5-fold (Figure 7). These surviving K562 cells can be cultivated over 6 months in the presence of imatinib and NGF; however, they are still NGF dependent. Although the effect of NGF on 5 μM imatinib-treated Meg-01 cells was more modest, eightfold more cells survived in the presence of NGF than in its absence (Figure 7, Meg-01). Next, we examined the effect of TrkA inhibitor K252a on these cells. NGF did not rescue cells treated with both K252a and imatinib from cell death in either cell line (Figure 7). As a control cell line, we utilized U937 cells that grow nearly normally in the presence of imatinib; however, treatment of U937 with K252a suppresses cell growth (Figure 7; U937). This may be explained by the fact that K252a also inhibits several serine/threonine kinases, such as PKC. Indeed, PKC is phosphorylated in U937 cells (data not shown). The effect of NGF on imatinib-treated HMC-1 was more drastic. In the presence of 0.5 or 5 μM imatinib that is also a strong inhibitor of c-Kit tyrosine kinase (Buchdunger et al., 2000; Heinrich et al., 2000; Wang et al., 2000; Tuveson et al., 2001), most HMC-1 cells died within 24 h. In the presence of both imatinib and NGF, however, cells continue to grow nearly normally. In the presence of K252a and imatinib, all cells died within 2 days (Figure 7; HMC-1).

Figure 7

NGF treatment rescues TrkA expressing cells from imatinib-induced rapid cell death. Aliquots of 1 × 105 of K562, Meg-01, HMC-1 and U937 cells were grown in medium containing 10% FCS with or without NGF (60 ng/ml) in the presence or absence of imatinib (0.5 or 5 μM) and/or K252a (200 nM) for 4 days. Cells were stained with 0.1% Trypan Blue and living cells were counted using a Neubauer cell counting chamber. Cell numbers are presented as the mean±s.d. for three independent experiments performed in triplicate. FCS, fetal calf serum; NGF, nerve growth factor.

NGF treatment enhances colony formation of imatinib-treated primary CD34+ CML cells

To examine whether the present findings are relevant to primary CML, we isolated CD34+ cells from three newly diagnosed CML patients in chronic phase. We first established a TrkA detection method using an immunofluorescence technique. We utilized K562 and HMC-1 cells as positive controls and CML-T1 cells as a negative control (Figure 8a), and then applied primary CD34+ cells for TrkA-specific staining. All primary CD34+ CML cells were clearly shown to be TrkA positive (Figure 8b). In agreement with previous work (Bracci-Laudiero et al., 2003), CD34+ cells from peripheral blood from healthy donors were stained weakly by TrkA-specific antibody (data not shown). We then performed a colony-forming assay in the presence of interleukin-3 (10 ng/ml) and granulocyte–macrophage colony-stimulating factor (20 ng/ml) with and without imatinib and/or NGF (50 or 200 ng/ml). In the absence of imatinib, NGF did not enhance the number of colonies in the presence of interleukin-3 and granulocyte–macrophage colony-stimulating factor (Table 1). In the presence of imatinib, however, NGF treatment enhanced the number of colony by about 1.5- (50 ng/ml NGF, P=0.07) and 2-fold (200 ng/ml NGF, P=0.03). NGF treatment had no significant effects on colony formation of CD34+ cells from healthy donors in the presence or absence of imatinib (Table 1). These data strongly suggest that NGF plays an important role in survival of CML cells.

Figure 8

CD34+ cells from CML patients express TrkA. (a) K562, HMC-1 and CML-T1 cells were stained by TrkA-specific immunofluorescence technique. (b) CD34+ cells from three individual patients (CML nos. 1–3) with CML in early chronic phase were isolated and were then specifically immunostained for TrkA (anti-TrkA). Isotype control: mouse IgG2a kappa isotype control antibody. Bars represent 25 μm. CML, chronic myelogenous leukemia.

Table 1 NGF treatment enhances colony formation of CD34+ CML cells in the presence of imatinib


Activation of receptor tyrosine kinase is tightly regulated by growth factors and effector molecules, and as a result induces cell proliferation, survival and/or differentiation signals. NGF was shown to be a potent growth factor that plays a key role in neuronal cell differentiation (Snider, 1994).

Here we report on TrkA signaling in Bcr–Abl transformed CML cells by dissecting both tyrosine kinase signal pathways using the Bcr–Abl inhibitor, imatinib. Several signaling molecules such as SHC, CrkL, Akt or Erk are shared by Bcr–Abl and TrkA, but when Bcr–Abl is active, TrkA signaling is not particularly strong. An explanation might be that active Bcr–Abl recruits signaling molecules such as SHC, CrkL or GAB2. Treatment with imatinib reduced the amount of the signaling complex and restored NGF-mediated signals. Indeed, NGF-induced phosphorylation of one of the major TrkA-binding proteins SHC was restored by imatinib treatment. It has been shown previously that SHIP2 is highly phosphorylated on tyrosine and associates with the SH3 domain of Abl tyrosine kinase and Crk, and PTB domain of SHC in K562 cells (Wisniewski et al., 1999). We show here that the SHIP2/CrkL/Bcr–Abl complex was not detected after imatinib treatment; however, TrkA activation did not cause SHIP2 phosphorylation. At present, the role of SHIP2 in K562 cells is not clear; however, in these cells, it has been shown that SHIP2 reduces the proliferation rate (Giuriato et al., 2002), suggesting that the dephosphorylation of SHIP2 may participate in imatinib-mediated enhancement of TrkA signaling.

We also present here a second mechanism, in which imatinib treatment influences receptor internalization. It has been demonstrated that the family of Abl tyrosine kinases plays a crucial role in Shigella uptake (Burton et al., 2003). In this system, downstream molecules of Abl, Crk, Rho family GTPases Rac and Cdc42 are required for bacterial entry. On the other hand, it has been demonstrated that highly activated Abl impaired EGF receptor internalization (Tanos and Pendergast, 2006). In this case, activated Abl inhibits the accumulation of the ubiquitin ligase c-Cbl at the plasma membrane, whereby, receptor degradation is inhibited (Tanos and Pendergast, 2006). These data suggested that activated Abl enhanced EGF signaling. Accordingly, inhibition of Bcr–Abl, enhanced TrkA clearance from the cell surface after stimulation with NGF. In contrast to the EGF receptor (Tanos and Pendergast, 2006), inhibition of Bcr–Abl enhanced TrkA signaling. In the neuronal system, the NGF signal was shown to be transmitted via endocytosis of complexes containing NGF bound to TrkA, followed by retrograde transport of the ‘signaling endosomes’ (Riccio et al., 1997; Howe et al., 2001). Here we show that the level of clearance of TrkA correlates with the level of phosphorylation of Akt, suggesting that ‘signaling endosomes’ with TrkA are present in hematopoietic cells.

The data presented here also provide evidence that NGF treatment rescues cells from imanitib-induced cell death. Although only two of the four CML cell lines tested were positive for TrkA, all primary CD34+ cells from three independent CML patients express TrkA. The difference may be explained by the fact that TrkA-negative cell lines, BV-173 and CML-T1 are CD34− and derived from patients in blast crisis, whereas all primary CD34+ cells were derived from patients with CML in early chronic phase. In addition, CD34+ cells from healthy donors also express TrkA (Bracci-Laudiero et al., 2003) and NGF is normally expressed by bone marrow stromal cells (Simone et al., 1999), suggesting that NGF plays a role in maintenance of CD34+ cells in bone marrow. Moreover, TrkA may play a role in survival of Bcr–Abl-positive cells followed by mutations in Bcr-Abl that inactivate imatinib (von Bubnoff et al., 2003) in bone marrow. TrkA may be a potential target for therapeutic approaches to TrkA-positive CML.

Materials and methods

Cell culture

All leukemia cell lines and PC12 cells were grown in RPMI1640 medium without any cytokine, supplemented with 10–20% (v/v) FCS or 10% horse serum and 5% FCS, respectively. HEK293 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% FCS.

Antibodies and reagents

Polyclonal antibodies against Akt, phospho-Akt (Ser473), phospho-Erk, c-Kit, phospho-CrkL (Tyr207), phospho-STAT5 (Tyr694) and phosphoY490TrkA were from New England Biolabs/Cell Signaling Technology (Frankfurt Germany), Erk1/2 and phosphoErk1/2 from Promega (Madison, WI, USA), and CrkL (C-20), STAT5 (C-17), SHC (H-108) and pan Trk (C-14), goat polyclonal antibody against Actin (I-19), GAB-2 (M-19), and mouse monoclonal antibody against glyceraldehyde-3-phosphate dehydrogenase and phosphotyrosine (PY99) were from Santa Cruz Biotechnology (Santa Cruz, CA, USA), monoclonal antibody against Abl (8E9) and SHC from BD Pharmingen and BD Transduction Laboratories (San Diego, CA, USA), respectively, monoclonal antibody against transferrin receptor (H68.4) from Invitrogen (Zymed laboratories Invitrogen immunodetection, San Francisco, CA, USA) and anti-goat IgG from rabbit was purchased from Sigma-Aldrich GmbH (Taufkirchen, Germany). Rabbit sera against SHIP2 (Muraille et al., 1999) and phospho-SHIP2 (Blero et al., 2001) were kindly provided by Ch. Erneux (Brussels).

β-NGF, EGF and basic fibroblast growth factor were purchased from PeproTech Inc. (Rocky Hill, NJ, USA). Imatinib mesylate was kindly provided by Novartis (Basel, Switzerland). Inhibitors K252a, and Monensin were from Calbiochem (San Diego, CA, USA).

Isolation of CD34+ cells and colony formation assay

CD34+ cells from patients with CML in early chronic phase were isolated at the time of initial diagnosis as described previously (Scherr et al., 2003). CD34+ cells were purified to at least 98% by magnetic cell sorting (Clini MACS Miltenyi Biotech, Bergisch Gladbach, Germany). Informed consent from the patients was obtained in accordance with the Declaration of Helsinki. Colony assays of CD34+ cells were described previously (Scherr et al., 2002, 2005). Briefly, 1 × 103 cells were plated per methylcellulose culture and then stimulated with human granulocyte–macrophage colony-stimulating factor (20 ng/ml) and interleukin-3 (10 ng/ml) in the presence or absence of NGF (50 or 200 ng/ml) and imatinib (1 μM). Nine days after plating the colonies were counted.

Cell proliferation assay

To assay cell survival, cells were grown in medium containing 10% FCS for 4 days in the absence or presence of β-NGF (PeproTech Inc.) and/or imatinib (Novartis), and K252a (Calbiochem). The number of living cells was counted in a Neubauer counting chamber using 0.1% Trypan Blue (Sigma-Aldrich).

Immunofluorescence staining

Chronic myelogenous leukemia cell lines were plated on poly-L-lysine (Sigma-Aldrich) coated Petri dishes, patient material was plated onto adhesion object slides (Superior, Paul Marienfeld GmbH, Lauda-Königshofen, Germany). Cells were washed with phosphate-buffered saline (PBS), then fixed for 15 min and permeabilized for 5 min using IntraPrep Leukocyte Permeabilization Reagent (Beckman Coulter/Immunotech SAS, Marseille, France). Blocking was performed once with human IgG (Octagam, Octapharma GmbH Langenfeld, Germany) 5 mg/ml in 10% FCS/PBS for 1 h and with 10% FCS in PBS for additional 40 min. Cells were incubated for 1 h with a monoclonal antibody against the C terminus of human TrkA (B-3, Santa Cruz Biotechnology) or mouse IgG2a kappa isotype control antibody (BD Pharmingen) in 10% FCS/PBS. Cells were washed three times with 10% FCS/PBS and then stained with fluorescein isothiocyanate-conjugated Fab-specific anti-mouse IgG (Sigma-Aldrich) for 1 h. Immunofluorescence staining comparing the TrkA antibody and the isotype control was photographed with identical exposure times on a Nikon Eclipse TE 300 microscope (Düsseldorf, Germany) using a Spot2 camera (Diagnostic Instruments, Ismaning/Munich, Germany).


Cells were serum starved for 24 h, treated with inhibitors for 1 h and then stimulated with growth factors in the absence or presence of inhibitors for the indicated times. After stimulation was stopped in ice-cold PBS, cells were extracted with lysis buffer containing 10 mM sodium phosphate pH 7.5, 50 mM NaF, 10 mM EDTA, 1% Triton-X 100, 1% Trasylol (Bayer Vital, Leverkusen, Germany) and 400 μM sodium orthovanadate. Details of immunoblotting have been described previously (Mancini et al., 2002; Koch et al., 2005). Corresponding proteins were visualized by incubation with peroxidase-conjugated anti-mouse, rabbit (Pierce, Rockford, IL, USA) or goat (Santa Cruz) immunoglobulin followed by incubation with SuperSignal West Femto Maximum Sensitivity Substrate (Pierce). Chemoluminescence was documented by a LAS3000 camera (Fujifilm, Kanagawa, Japan). For quantification, we also performed immunoblotting using 125I-anti rabbit IgG (Amerscham Biosciences, Freiburg, Germany) as a secondary antibody to compare data obtained by the two methods. Signal intensity of radioactivity as well as chemoluminescence was quantified using TINA 2.0 software (raytest Isotopenmessgeraete GmbH, Straubenhardt, Germany).

Before reprobing, membranes were stripped for 2 × 30 min using buffer containing 50 mM Tris/HCl pH 8, 3% sodium dodecyl sulfate and 100 mM β-mercapthoethanol.


Cells were extracted using the same lysis buffer as used for western blotting plus 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride and 1 mM benzamidine. Clarified lysate (1 × 107 cells) was incubated at 4 °C for 4 h with anti-CrkL or SHC antibody preabsorbed on Protein A Sepharose 6MB (Amerscham Bioscience/GE Healthcare, Uppsala, Sweden). After washing with precipitation buffer, materials were analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and by immunoblotting analysis using the appropriate antibodies.

Cell surface biotinylation and precipitation

Biotinylation was performed according to the manufacturer's protocol using 250 μg/ml membrane impermeable EZ-link Sulfo-NHS-LC-Biotin (Pierce) for 30 min on ice. For internalization assays, cells (1.5 × 106/time point) were serum starved and stimulated as described for immunoblotting, biotinylated and then lysed in radioimmunoprecipitation assay buffer (10 mM sodium phosphate pH 7.5, 50 mM NaF, 10 mM EDTA, 1% Triton-X 100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 1% Trasylol). Biotinylated proteins were precipitated from lysates with streptavidin sepharose (Amerscham Bioscience/GE Healthcare). Aliquots of cell lysates were applied for immunoprecipitation using anti-Trk antibody and protein A Sepharose 6MB (Amerscham Bioscience/GE Healthcare).


K562 cells were transduced with lentiviral shRNA directed against Bcr–Abl or an irrelevant shRNA (GL4) as estabilished previously (Scherr et al., 2005). Ninety-five percent of the cells were transduced as shown by expression of red fluorescencent protein. Forty-eight hours after transduction, cells were starved for 17 h in medium containing 0.1% FCS and were then stimulated with NGF for the indicated times. Cell lysates were prepared as described under immunoblotting.


  1. Blero D, De Smedt F, Pesesse X, Paternotte N, Moreau C, Payrastre B et al. (2001). The SH2 domain containing inositol 5-phosphatase SHIP2 controls phosphatidylinositol 3,4,5-trisphosphate levels in CHO-IR cells stimulated by insulin. Biochem Biophys Res Com 282: 839–843.

  2. Bonati A, Carlo-Stella C, Lunghi P, Albertini R, Pinelli S, Migliaccio E et al. (2000). Selective expression and constitutive phosphorylation of SHC proteins in the CD34+ fraction of chronic myelogenous leukemias. Cancer Res 60: 728–732.

  3. Bracci-Laudiero L, Celestino D, Starace G, Antonelli A, Lambiase A, Procoli A et al. (2003). CD34-positive cells in human umbilical cord blood express nerve growth factor and its specific receptor TrkA. J Neuroimmunol 136: 130–139.

  4. Buchdunger E, Cioffi CL, Law N, Stover D, Ohno-Jones S, Druker BJ et al. (2000). Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet-derived growth factor receptors. J Pharmacol Exp Ther 295: 139–145.

  5. Buchdunger E, Zimmermann J, Mett H, Meyer T, Müller M, Druker BJ et al. (1996). Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res 56: 100–104.

  6. Burton EA, Plattner R, Pendergast AM . (2003). Abl tyrosine kinases are required for infection by Shigella flexneri. EMBO J 22: 5471–5479.

  7. Butterfield JH, Weiler D, Dewald G, Gleich GJ . (1988). Establishment of an immature mast cell line from a patient with mast cell leukemia. Leuk Res 12: 345–355.

  8. Chen Z-Y, Ieraci A, Tanowitz M, Lee FS . (2005). A novel endocytic recycling signal distinguishes biological responses of Trk neurotrophin recptors. Mol Biol Cell 16: 5761–5772.

  9. Chevalier S, Praloran V, Smith C, MacGrogan D, Ip NY, Yancopoulos GD et al. (1994). Expression and functionality of the trkA proto-oncogene product/NGF receptor in undifferentiated hematopoietic cells. Blood 83: 1479–1485.

  10. Deininger M, Buchdunger E, Druker BJ . (2005). The development of imatinib as a therapeutic agent for chronic myeloid leukemia. Blood 105: 2640–2653.

  11. Druker BJ, Guilhot F, O’Brien SG, Gathmann I, Kantarjian H, Gattermann N et al. (2007). Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. New Eng J Med 355: 2408–2417.

  12. Druker BJ, Sawyers CL, Kantarjian H, Resta DJ, Reese SF, Ford JM et al. (2001). Activity of a specific inhibitor of the Bcr–Abltyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with Philadelphia chromosome. N Engl J Med 344: 1038–1042.

  13. Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S et al. (1996). Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr–Ablpositive cells. Nat Med 2: 561–566.

  14. Feller SM . (2001). Crk family adaptors-signalling complex formation and biological roles. Oncogene 20: 6348–6371.

  15. Giuriato S, Blero D, Robaye B, Bruyns C, Payrastre B, Erneux C . (2002). SHIP2 overexpression strongly reduces the proliferation rate of K562 erythroleukemia cell line. Biochem Biophys Res Commun 296: 106–110.

  16. Heinrich MC, Griffith DJ, Druker BJ, Wait CL, Ott KA, Zigler AJ . (2000). Inhibition of c-kit receptor tyrosine kinase activity by STI 571, a selective tyrosine kinase inhibitor. Blood 96: 925–932.

  17. Howe CL, Valletta JS, Rusnak AS, Mobley WC . (2001). NGF signaling from clathrin-coated vesicles: evidence that signaling endosomes serve as a platform for the Ras-MAPK pathway. Neuron 32: 801–814.

  18. Huang EJ, Reichardt LF . (2003). Trk receptors: roles in neuronal signal transduction. Annu Rev Biochem 72: 609–642.

  19. Kaebisch A, Brokt S, Seay U, Lohmeyer J, Jaeger U, Pralle H . (1996). Expression of the nerve growth factor receptor c-TRK in human myeloid leukaemia cells. Br J Haematol 95: 102–109.

  20. Koch A, Mancini A, El Bounkari O, Tamura T . (2005). The SH2-domian-containing inositol 5-phosphatase (SHIP)-2 binds to c-Met directly via tyrosine residue 1356 and involves hepatocyte growth factor (HGF)-induced lamellipodium formation, cell scattering and cell spreading. Oncogene 24: 3436–3447.

  21. Koch A, Mancini A, Stefan M, Niedenthal R, Niemann H, Tamura T . (2000). Direct interaction of nerve growth factor receptor, TrkA, with non-receptor tyrosine kinase, c-Abl, through the activation loop. FEBS letter 469: 72–76.

  22. Mancini A, Koch A, Wilms R, Tamura T . (2002). c-Cbl associates directly with the C-terminal tail of the receptor for the macrophage colony-stimulating factor, c-Fms, and down-modulates this receptor but not the viral oncogene v-Fms. J Biol Chem 277: 14635–14640.

  23. Matsuda H, Coughlin MD, Bienenstock J, Denburg JA . (1988). Nerve growth factor promotes human hemopoietic colony growth and differentiation. Proc Natl Acad Sci USA 85: 6508–6512.

  24. Mulloy JC, Jankovic V, Wunderlich M, Delwel R, Cammenga J, Krejci O et al. (2005). AML1-ETO fusion protein up-regulates TRKA mRNA expression in human CD34+ cells, allowing nerve growth factor-induced expansion. Proc Natl Acad Sci USA 102: 4016–4021.

  25. Muraille E, Pessese X, Kuntz C, Erneux C . (1999). Distribution of the src-homology-2-domain-containing inositol 5-phosphatase SHIP-2 in both non-haemopoietic and haemopoietic cells and possible involvement of SHIP-2 in negative signalling of B-cells. Biochem J 342: 697–705.

  26. Obermeier A, Lammers R, Wiesmüller KH, Jung G, Schlessinger J, Ullrich A . (1993). Identification of Trk binding site for SHC and phosphatidylinositol 3′-kinase and formation of a multimeric signaling complex. J Biol Chem 268: 22963–22966.

  27. O’Brien SG, Guilhot F, Larson RA, Gathmann I, Baccarani M, Cervantes F et al.. (2003). Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. New Eng J Med 348: 994–1004.

  28. Rajagopal R, Chen ZY, Lee FS, Chao MV . (2004). Transactivation of Trk neurotrophin receptors by G-protein-coupled receptor ligands occurs on intracellular membranes. J Neurosci 24: 6650–6658.

  29. Riccio A, Pierchala BA, Ciarallo CL, Ginty DD . (1997). An NGF-TrkA-mediated retrograde signal to transcription factor CREB in sympathetic neurons. Science 277: 1097–1100.

  30. Sattler M, Mohi MG, Pride YB, Quinnan LR, Malouf NA, Podar K et al. (2002). Critical role for Gab2 in transformation by BCR/ABL. Cancer Cell 1: 479–492.

  31. Saxena S, Howe CL, Cosgaya JM, Steiner P, Hirling H, Chan JR et al. (2005). Differential endocytic sorting of p75NTR and TrkA in response to NGF: a role for late endosomes in TrkA trafficking. Mol Cell Neurosci 28: 571–587.

  32. Scheijen B, Griffin JD . (2002). Tyrosine kinase oncogenes in normal hematopoiesis and hematological disease. Oncogene 21: 3314–3333.

  33. Scherr M, Battmer K, Blömer U, Schiedlmeier B, Ganser A, Grez M et al. (2002). Lentiviral gene transfer into peripheral blood-derived CD34+ NOD/SCID-repopulating cells. Blood 99: 709–712.

  34. Scherr M, Battmer K, Schultheis B, Ganser A, Eder M . (2005). Stable RNA interference (RNAi) as an option for anti-bcr–abl therapy. Gene Therapy 12: 12–21.

  35. Scherr M, Battmer K, Winkler T, Heidenreich O, Ganser A, Eder M . (2003). Specific inhibition of bcr–abl gene expression by small interfering RNA. Blood 101: 1566–1569.

  36. Simone MD, De Santis S, Vigneti E, Papa G, Amadori S, Aloe L . (1999). Nerve growth factor: a survey of activity on immune and hematopoietic cells. Hematol Oncol 17: 1–10.

  37. Snider WD . (1994). Functions of the neurotrophins during nervous system development: what the knockouts are teaching us. Cell 77: 627–638.

  38. Stephens RM, Loeb DM, Copeland TD, Pawson T, Greene LA, Kaplan D . (1994). Trk receptors use redundant signal transduction pathways involving SHC and PLC-γ1 to mediate NGF response. Neuron 12: 691–705.

  39. Tam S-Y, Tsai M, Yamaguchi M, Yano K, Butterfield JH, Galli SJ . (1997). Expression of functional TrkA receptor tyrosine kinase in the HMC-1 human mast cell line and in human mast cells. Blood 90: 1807–1820.

  40. Tanos B, Pendergast AM . (2006). Abl tyrosine kinase regulates endocytosis of the epidermal growth factor receptor. J Biol Chem 281: 32714–32723.

  41. Tuveson DA, Willis NA, Jacks T, Griffin JD, Singer S, Fletcher CD et al. (2001). STI571 inactivation of the gastrointestinal stromal tumor c-KIT oncoprotein: biological and clinical implications. Oncogene 20: 5054–5058.

  42. von Bubnoff N, Peschel C, Duyster J . (2003). Resistance of Philadelphia-chromosome positive leukemia towards the kinase inhibitor imatinib (STI571, Glivec): a targeted oncoprotein strikes back. Leukemia 17: 829–838.

  43. Wang WL, Healy ME, Sattler M, Verma S, Lin J, Maulik G et al. (2000). Growth inhibition and modulation of kinase pathways of small cell lung cancer cell lines by the novel tyrosine kinase inhibitor STI 571. Oncogene 19: 3521–3528.

  44. Wisniewski D, Strife A, Swendeman S, Erdjument-Bromage H, Geromanos S, Kavanaugh WM et al. (1999). A novel SH2-containing phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase (SHIP2) is constitutively tyrosine phosphorylated and associated with src homologous and collagen gene (SHC) in chronic myelogenous leukemia progenitor cells. Blood 93: 2707–2720.

  45. Yano H, Cong F, Birge RB, Goff SP, Chao MV . (2000). Association of the Abl tyrosine kinase with the Trk nerve growth factor receptor. J Neurosci Res 59: 356–364.

  46. Zhang Y, Moheban DB, Conway BR, Bhattacharyya A, Segal RA . (2000). Cell surface Trk receptors mediate NGF-induced survival while internalized receptors regulate NGF-induced differentiation. J Neurosci 20: 5671–5678.

Download references


We thank Regina Wilms and Sabine Klebba-Faerber for technical assistance, and Bruce Boschek for critically reading the paper. The research was supported by Sonderforschungsbereich 566 (B2 and A12), the Deutsche Forschungsgemeinschaft (Ta-111/8/-4, Ko-2249/3-1), MHH-HiLF program to AK, and HW and J Hector Stiftung to MS and ME.

Author information

Correspondence to T Tamura.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Koch, A., Scherr, M., Breyer, B. et al. Inhibition of Abl tyrosine kinase enhances nerve growth factor-mediated signaling in Bcr–Abl transformed cells via the alteration of signaling complex and the receptor turnover. Oncogene 27, 4678–4689 (2008) doi:10.1038/onc.2008.107

Download citation


  • TrkA and Bcr–Abl
  • CML
  • imatinib
  • receptor internalization
  • CD34+ cells

Further reading