Introduction

Several lines of evidence have revealed that thrombopoietin (TPO) plays a central role in megakaryocytopoiesis and thrombopoiesis, in vivo and in vitro.^1,2,3,4,5 Recently, it was found that c-Mpl, which is the receptor for TPO, was expressed in platelets,⁶ and that TPO enhanced the normal human platelet aggregation induced by aggregating agonists, although TPO alone was unable to induce the aggregation of normal platelets.^7,8,9,10 In addition, TPO was proven to ameliorate thrombocytopenia in mice treated with chemotherapeutic agents or radiation.^11,12,13 In clinical trials, several studies revealed that TPO stimulated the recovery from thrombocytopenia in cancer patients who were receiving myelosuppressive chemotherapy.^11,12,13 These results indicate the clinical utility of TPO for treating patients with thrombocytopenia induced by chemotherapy and/or irradiation, and patients with hematologic disorders characterized by thrombocytopenia and/or platelet dysfunction.

Qualitative abnormalities of platelets, including acquired storage pool disease, abnormal arachidonic acid metabolism, and morphological and membrane abnormalities, are commonly found in patients with myeloproliferative disorders (MPDs).¹⁴ However, no causal relationship between any of these specific abnormalities and either bleeding or thrombosis has been clearly established.¹⁴ Notably, many studies have failed to assess the differences in intrinsic abnormalities of platelets among MPDs. Oda et al¹⁵ demonstrated that Crkl was constitutively tyrosine-phosphorylated in platelets from patients with chronic myelogenous leukemia in the chronic phase (CML-CP) but not in those from patients with other MPDs, although its biological function is unclear. Recently, Moliterno et al¹⁶ reported that expression of c-Mpl was markedly reduced or absent in platelets from patients with polycythemia vera (PV) and myelofibrosis (MF) but not in those from patients with CML-CP or essential thrombocythemia (ET). In contrast, other groups reported that expression of c-Mpl was heterogeneous in platelets from patients with PV^17,18 and that its expression was reduced in platelets from patients with ET.¹⁹ Kiladjian et al²⁰ demonstrated that c-mpl mRNA expression in patients with ET and normal subjects was identical. These studies on c-Mpl expression in platelets from patients with PV and ET show conflicting results.

Protein tyrosine phosphorylation in platelets plays an important role in the signal transduction involved in platelet activation induced by diverse aggregating agonists. Binding of TPO to c-Mpl leads to tyrosine phosphorylation of a set of proteins in normal human platelets, although c-Mpl has no consensus sequence of the tyrosine kinase domain. Given that TPO induced tyrosine phosphorylation of Jak2 and Tyk2 followed by their activation in normal platelets,^21,22,23 protein tyrosine kinases (PTKs) such as Jak2 and Tyk2 must mediate the signals from c-Mpl. The finding that genistein, an inhibitor of tyrosine kinases, inhibited TPO-enhanced platelet aggregation in a dose-dependent manner indicates that PTKs and TPO-induced tyrosine phosphorylation of proteins downstream of the PTKs are involved in the signal transduction for platelet aggregation enhanced by TPO.⁸ The activation of the Jak-Stat pathway, which is one of the major signaling cascades, was found to be impaired in platelets from PV and MF patients.¹⁷ However, other signaling pathway(s) may be involved in TPO-induced platelet activation, based on the fact that TPO enhanced the aggregation of platelets from patients with MPDs induced by aggregating agonists.²⁴

Phosphatidylinositol 3-kinase (PI3-kinase) is a heterodimeric enzyme that converts the three phosphoinositides of the canonical PI pathway [PI, PI(4)P, and PI(4, 5)P₂] to produce PI(3)P, PI(3, 4)P₂, and PI(3, 4, 5)P₃, respectively.²⁵ It consists of a regulatory subunit, p85, which has two SH2 domains and an SH3 domain, and a catalytic subunit, p110.²⁵ By binding of p85 to a phosphorylated tyrosine residue on PTKs, PI3-kinase is recruited to the membrane and activated. Interaction of Ras with p110 also leads to PI3-kinase activation.²⁶ In addition, the PI3-kinase pathway is activated by dysfunction of negative regulators of PI3-kinase, such as phosphatase and tensin homologue deleted on chromosome 10 (PTEN) and SH2-containing inositol 5-phosphatases (SHIP).^27,28 Given that PI3-kinase in platelets was activated by aggregating agonists such as thrombin, it appeared to play a role in regulating platelet aggregation.²⁹ Recently, TPO was found to induce tyrosine phosphorylation of p85 and activation of PI3-kinase in normal human platelets.³⁰ Moreover, it was reported that PI3-kinase mediated the signals implicated in the priming action of TPO on collagen receptor signaling in normal platelets,³¹ and that PI3-kinase played an important role in the TPO-enhanced activation of GpIIb/IIIa in both primary human megakaryocytic cells derived from CD34-positive cells in peripheral blood and HEL human leukemia cells.³²

In the present study, we examined the effect of TPO on the function of MPDs platelets. We found that TPO alone induced aggregation of platelets from patients with CML-CP but not of those from patients with PV, ET, or MF. We also found that PI3-kinase was constitutively activated in platelets from patients with CML-CP but not in those from patients with any other MPDs. Pretreatment of CML-CP platelets with inhibitors of PI3-kinase (wortmannin and LY294002) suppressed the TPO-induced aggregation. These findings indicate that the constitutive activation of PI3-kinase may prime CML-CP platelets for the aggregation induced by TPO. In addition, the suppressive effect of the Abl kinase inhibitor imatinib mesylate on the activity of PI3-kinase in CML-CP platelets was consistent with the reduction of TPO-induced platelet aggregation by the inhibitor.

Materials and methods

Patients

We studied 40 patients with MPDs after obtaining their informed consent. Diagnoses of MPDs were mainly based on routine laboratory tests, complete blood cell counts, bone marrow examinations of aspiration and/or biopsy specimens, the determination of leukocyte alkaline phosphatase and erythrocyte mass by ⁵¹Cr technique, measurement of Vitamin B₁₂ levels in sera, and cytogenetic studies. The patients included 17 with CML-CP, 10 with PV, 10 with ET, and three with MF. Philadelphia chromosome (Ph) was detected in 100% of cells examined in all patients with CML-CP before treatment. The percentage of Ph-positive cells was reduced to 0–90% in nine patients with CML-CP after treatment with interferon-, hydroxyurea, and allogeneic bone marrow transplantation. Platelet aggregation was monitored before and after treatment with hydroxyurea in one case of CML-CP and with bone marrow transplantation in another case of CML-CP. One patient with PV and three with ET received chemotherapy before platelet aggregation was measured. None of the patients had received therapy with cytotoxic agents, anticoagulants, or antiplatelet agents for at least 2 weeks before the measurement of platelet aggregation. Platelet aggregation studies in each patient were carried out twice or three times; the results were found to be reproducible.

Reagents

Wortmannin, LY294002, and AG490³³ were purchased from Wako (Tokyo, Japan). ADP, epinephrine, and collagen were purchased from Sigma (St Louis, MO, USA). Imatinib mesylate^34,35,36 and human recombinant TPO were kindly provided by Novartis (Basel, Switzerland) and Kirin Brewery (Tokyo, Japan). Other chemicals were purchased from commercial sources.

Antibodies

FITC-conjugated anti-CD63 monoclonal antibody and FITC-conjugated mouse IgG as a control were purchased from Coulter-Immunotech (Miami, FL, USA). Anti-Jak2 and anti-p85 polyclonal antibodies and anti-phosphotyrosine (PY) monoclonal antibody (4G10) were purchased from Upstate Biotechnology (Lake Placid, NY, USA).

Platelet aggregation

Platelet aggregation was monitored as described previously.⁸ In brief, after blood from patients with MPDs was collected into a 0.1 volume of 3.8% citrate, platelet-rich plasma (PRP) was prepared by centrifugation at 200 g for 10 min at room temperature. In order to avoid contamination of leukocytes, only the upper one-third of PRP was used in the following experiments. Under these conditions, the number of leukocytes contaminating PRP was less than 0.05% of that of platelets. Platelet-poor plasma (PPP) was also prepared, and we adjusted the platelet count to that of PRP (2 10⁵/l). In the experiments using platelets, platelets were isolated from PRP by centrifugation at 1000 g for 10 min in the presence of 0.5 M prostaglandin E. The PRP was transferred into an aggregometer cuvette and incubated at 37°C for 2 min with stirring before adding TPO and/or chemicals, after which the platelet aggregation was monitored by a turbidimetric method (NBS Hema Tracer, MC Medical, Tokyo, Japan). PPP was used as a reference. In the experiments using inhibitors, platelets were preincubated with inhibitors at 37°C for 10 min before TPO treatment. Spontaneous platelet aggregation was not observed in any cases under this condition. In some experiments, a particle counting method using light scattering (AG-10 aggregometer, Kowa, Ibaraki, Japan) developed by Ozaki et al³⁷ was applied to monitor platelet aggregation. The particle counting method is based on measuring the intensity of light scattered by particles passing through a limited area of PRP. As light scattering intensity increases in proportion to the size of the particles in suspension, the degree of intensity provides information on the number and size of aggregates in the suspension. This method is particularly sensitive to small aggregates such as those formed in platelets activated by agonists at low concentrations and is, therefore, more suitable than the turbidimetric method. In addition, the size distribution of aggregates and the extent of aggregation were estimated by this method.

Flow cytometric analysis

Blood from patients with MPDs was collected into a 0.1 volume of 100 mM EDTA, and then PRP was prepared as described above. After treatment with 100 ng/ml TPO at 37°C for various periods, platelets were fixed with 1% paraformaldehyde (final dilution 1:10) at room temperature for 30 min. The fixed platelets were incubated with 0.2% EDTA solution containing human -globulin to inhibit nonspecific binding of specific antibodies to the platelets, and then the platelets were incubated with FITC-conjugated anti-CD63 monoclonal antibody or mouse IgG as a control on ice for 30 min. After being washed with PBS containing 0.2% EDTA twice, the platelets were resuspended in the same buffer and analyzed by EPICS XL (Coulter, Miami, FL, USA).

Immunoprecipitation and immunoblotting

Immunoprecipitation and immunoblot analyses were performed as described previously.^8,38 In brief, after incubation with inhibitors at 37°C for 10 min, platelets (2 10⁸) were treated with 200 ng/ml TPO at 37°C for 5 min. Control platelets were treated with the same volume of DMSO (0.1% v/v), the solvent of inhibitors, then lysed in the lysis buffer.³⁸ The cell lysates were subjected to immunoprecipitation. Then, immunoprecipitated proteins were separated by SDS-PAGE and electrophoretically transferred to a PVDF membrane.

For immunoblot analysis, the membranes preincubated with 5% non-fat milk powder solution were probed with specific antibodies followed by detection using an enhanced chemiluminescence system (Amersham Pharmacia, Buckinghamshire, England).

PI3-kinase assay

An in vitro PI3-kinase assay was carried out as described previously.³⁹ The radiolabeled phospholipids were detected and quantified using a Bioimaging Analyzer BAS2000 (Fuji Film, Tokyo, Japan). Since the number of leukocytes contaminating the samples was low, contamination of leukocytes did not affect the results of PI3-kinase assay using platelets prepared as described above.

Results

Response of platelets from MPDs patients to aggregating agonists and TPO

Platelet aggregation induced by aggregating agonists was impaired in the majority of patients with MPDs (data not shown), as described previously.²⁴ To determine whether TPO modulates the function of platelets freshly isolated from patients with MPDs, we monitored platelet aggregation induced by TPO using a lumiaggregometer, as described in the Materials and methods. As shown in Figures 1a and b, TPO induced aggregation of platelets from 10 patients with CML-CP with Ph in 100% of metaphases (CML-CP with 100% Ph), and of those from one patient with CML-CP with 90% Ph in a dose-dependent manner. On the other hand, aggregation of platelets from patients with PV, ET, MF, or CML-CP with less than 30% Ph was not detected by the lumiaggregometer after treatment with TPO alone (Figures 1b and c). TPO, however, enhanced the secondary wave of platelet aggregation induced by ADP in a dose-dependent manner, excepting platelets from two patients with ET (Figure 1d), as demonstrated by experiments using normal human platelets.^7,8,9,10 TPO also enhanced the platelet aggregation induced by epinephrine and collagen (data not shown). The number of leukocytes contaminating PRP was low when PRP was prepared as described in the Materials and methods. Furthermore, even if the number of leukocytes contaminating PRP from patients with PV, ET, and MF was similar to that in PRP from those with CML-CP, TPO alone could induce aggregation of platelets from patients with CML-CP but not of those with other MPDs. Therefore, leukocytes contaminating PRP are very unlikely to affect TPO-induced aggregation of CML-CP platelets.

Figure 1.

Effects of TPO on aggregation of platelets from patients with MPDs. PRPs were prepared from patients with MPDs as described in 'Materials and methods.' After adding TPO (a, b, c) or TPO and ADP (d) to the samples at various concentrations, platelet aggregation was monitored with a lumiaggregometer. (a) The representative pattern of TPO-induced aggregation of platelets from a patient with CML-CP with 100% Ph. (b) The data are shown as maximum aggregation rates. Left, CML-CP patients with 100% Ph. Right, open circles show the data of CML-CP patients with 10–90% Ph. Closed circles show the data of CML-CP patients in cytogenetical complete remission. (c) Platelets from patients with PV, ET, and MF were treated with 10 and 200 ng/ml TPO, and then platelet aggregation was measured with a lumiaggregometer. (d) Platelets from patients with PV, ET, and MF were preincubated with TPO (200 ng/ml) at 37°C for 2 min before adding ADP (closed circles, 2 M ADP; open circles, 5 or 10 M ADP). The data are shown as maximum aggregation rates.

Full figure and legend (113K)

TPO induced the aggregation of platelets from patients with CML-CP suspended in normal human plasma in a dose-dependent manner. TPO did not induce the aggregation of normal platelets suspended in plasma from patients with CML-CP (data not shown). These findings indicate that defects in platelets from patients with CML-CP, rather than factors in their plasma, are responsible for TPO-induced platelet aggregation.

TPO-induced platelet aggregation in the mixture of PRPs from a normal subject and from a patient with CML-CP with 100% Ph

TPO-induced aggregation was detected in platelets from CML-CP patients with more than 90% Ph, but not from those with less than 30% Ph, or from patients with PV, ET, or MF by a lumiaggregometer detecting changes in light transmission (turbidimetric changes in platelet suspensions). While the turbidimetric method has the advantage of being relatively simple, changes in optical density are unlikely to quantitatively represent the intensity of platelet activation. This method cannot sensitively detect the formation of small aggregates of platelets. A reduction in optical density reflects the formation of large aggregates from preexisting small aggregates, as described previously.³⁷ Therefore, to determine the sensitivity of the turbidimetric method, we measured the platelet aggregation in a mixture of PRPs from a normal subject and from a patient with CML-CP with 100% Ph in response to TPO using a lumiaggregometer. As shown in Figure 2, TPO-induced platelet aggregation was detected by this method when the percentage of PRP from CML-CP in the mixture was more than 50%. However, the aggregation of the mixed platelets, in which CML-CP platelets accounted for 25%, was hardly detected by this method.

Figure 2.

Thrombopoietin-induced platelet aggregation in the mixture of PRPs from a normal subject and from a patient with CML-CP with 100% Ph. PRPs were prepared from a healthy volunteer and from a patient with CML-CP with 100% Ph. After both PRPs were mixed at various concentrations, platelet aggregation was measured with a lumiaggregometer. The results are representative of three independent experiments.

Full figure and legend (16K)

Detection of platelet aggregation by a particle counting method

A particle counting method can more sensitively detect the formation of small aggregates of platelets than can the turbidimetric method.³⁷ Therefore, to determine whether TPO alone induces the aggregation of platelets from patients with PV, ET, MF, or CML-CP with less than 30% Ph, we monitored platelet aggregation using a particle counting method.

In the experiment using platelets from a patient with CML-CP with 100% Ph, small aggregates were formed soon (1–5 min) after adding TPO (Figure 3a). At this time, platelet aggregation was not detected by the turbidimetric method. Next, mid-sized aggregates were formed from small aggregates, and finally, large aggregates were formed. As shown in Figure 3a in the lower panel, small aggregates, but not mid-sized or large ones, were formed when platelets from a patient with CML-CP with 15% Ph were treated with TPO. The formation of small aggregates induced by TPO was confirmed by morphological examination using cytospine followed by May-Grünwald-Giemsa staining (data not shown). In contrast, no aggregates were detected in patients with PV, ET, MF, or CML-CP in cytogenetical complete remission (Figure 3b). These findings indicate that TPO alone induces the aggregation of platelets from patients with CML-CP but not from those patients with PV, ET, or MF.

Figure 3.

Monitoring of TPO-induced aggregation of platelets from patients with MPDs by the particle counting method. PRPs were prepared from patients with MPDs. After adding TPO (200 ng/ml) to the samples, platelet aggregation was monitored by the particle counting method. (a) CML-CP patients with 100% (upper) or 15% (lower) Ph. (b) Patients with PV, ET, MF, and CML-CP in cytogenetical complete remission (CR). T, monitoring by the turbidimetric method; S, small aggregates; M, mid-sized aggregates; and L, large aggregates.

Full figure and legend (72K)

Flow cytometric analysis of CD63 expression in platelets from patients with CML-CP

To ascertain whether TPO activates the platelets from patients with CML-CP, we examined the effects of TPO on the expression of CD 63 in platelets from the patients. TPO did induce the expression of CD63 (data not shown). The maximal expression was achieved at 3 min after TPO treatment.

Constitutive activation of PI3-kinase in platelets from patients with CML-CP

The response of platelets from patients with CML-CP to TPO may be characteristic of the intrinsic abnormalities of CML-CP platelets. It was recently reported that TPO induced tyrosine phosphorylation of p85, a regulatory subunit of PI3-kinase, and activation of PI3-kinase.³⁰ In addition, the expression of bcr-abl fusion gene, which is essential for the pathogenesis of CML-CP, constitutively activates PI3-kinase.⁴⁰ Therefore, to ascertain whether PI3-kinase plays a role in the signal transduction involved in TPO-induced aggregation of platelets from patients with CML-CP, we examined the activity of PI3-kinase in platelets from patients with MPDs. As shown in Figure 4a, PI3-kinase was constitutively activated in platelets from patients with CML-CP but not in those from patients with other MPDs. TPO, however, induced the activation of PI3-kinase in platelets from patients with PV, ET, and MF. Since the number of leukocytes contaminating the samples subject to in vitro PI3-kinase assay was low, PI3-kinase activity in leukocytes did not affect the results of the assay.

Figure 4.

Involvement of PI3-kinase in TPO-induced aggregation of platelets from patients with MPDs. (a) PI3-kinase was constitutively activated in CML-CP platelets but not in platelets from patients with other MPDs. Lysates were prepared from MPDs platelets before and after treatment with TPO (200 ng/ml) at 37°C for 5 min. After immunoprecipitation from the lysates with PY antibody (4G10), PI3-kinase activity in PY immunoprecipitates was assayed in vitro as described in 'Materials and methods.' PIP, phosphatidylinositol 3-monophosphate; Ori., origin. (b, c) PI3-kinase inhibitors dose-dependently suppressed TPO-induced aggregation of CML-CP platelets. Platelets from CML-CP patients with 100% Ph were preincubated with various concentrations of each PI3-kinase inhibitor (LY294002 or wortmannin) at 37°C for 10 min before TPO treatment. After adding TPO (200 ng/ml), platelet aggregation was monitored by the turbidimetric (b) and particle counting (c) methods. The results are representative of three independent experiments. (d) Inhibition of PI3-kinase activity in CML-CP platelets by selective inhibitors of PI3-kinase. Platelets were incubated with each PI3-kinase inhibitor (50 M LY294002 or 100 nM wortmannin) at 37°C for 10 min. PI3-kinase activity in PY immunoprecipitates was then assayed in vitro. PIP, phosphatidylinositol 3-monophosphate; Ori., origin; LY, LY294002; WT, wortmannin.

Full figure and legend (228K)

Effects of PI3-kinase inhibitors on TPO-induced aggregation of platelets from patients with CML-CP

The constitutive activation of PI3-kinase is likely to reflect one of the intrinsic abnormalities of platelets from patients with CML-CP and to prime the platelets for aggregation induced by TPO. To confirm this, we examined the effects of PI3-kinase inhibitors on TPO-induced aggregation of platelets from patients with CML-CP. As shown in Figure 4b, pretreatment of the platelets with wortmannin or LY294002 inhibited the TPO-induced platelet aggregation in a dose-dependent manner as measured by the turbidimetric method. The analysis by the particle counting method showed that LY294002 inhibited the formation of large aggregates at lower concentrations, and that higher concentrations of LY294002 were required to inhibit the formation of small aggregates (Figure 4c).

To confirm the inhibitory effects of wortmannin and LY294002 on PI3-kinase activity in platelets from patients with CML-CP, we carried out an in vitro PI3-kinase assay using PY immunoprecipitates from lysates of the platelets. PI3-kinase activity was inhibited by each inhibitor (Figure 4d). Suppression of the activation of PI3-kinase in CML-CP platelets by the inhibitors was correlated with their inhibitory effects on TPO-induced platelet aggregation.

Effects of PTK inhibitors on TPO-induced aggregation of platelets from patients with CML-CP

Protein tyrosine phosphorylation plays a crucial role in c-Mpl-mediated signal transduction in normal human platelets.^7,8,9 Bcr-Abl is a pivotal PTK in the pathogenesis of CML-CP and is expressed in hematopoietic cells derived from a CML clone. To elucidate the role of Bcr-Abl in TPO-induced aggregation of platelets from patients with CML-CP, we examined the effect of imatinib mesylate on the TPO-induced platelet aggregation. Imatinib mesylate dose-dependently inhibited the TPO-induced platelet aggregation (Figure 5a). The same concentrations of DMSO, a solvent of the inhibitor, used as a control did not affect the TPO-induced platelet aggregation (data not shown).

Figure 5.

Effects of PTK inhibitors on TPO-induced activation of CML-CP platelets. (a) PTK inhibitors dose-dependently inhibited TPO-induced aggregation of CML-CP platelets. Platelets were preincubated with various concentrations of PTK inhibitors (imatinib mesylate or AG490) at 37°C for 10 min before TPO treatment. After adding TPO (200 ng/ml), platelet aggregation was measured with a lumiaggregometer. The results are representative of three independent experiments. (b) Inhibitory effects of PTK inhibitors on PI3-kinase activity in CML-CP platelets. Platelets were incubated with each PTK inhibitor (5 M imatinib mesylate or 50 M AG490) at 37°C for 10 min. PI3-kinase activity in PY immunoprecipitates was then assayed in vitro. PIP, phosphatidylinositol 3-monophosphate; Ori., origin. (c) Inhibitory effects of PTK inhibitors on TPO-induced tyrosine phosphorylation of Jak2 in CML-CP platelets. Platelets were preincubated with each PTK inhibitor (5 M imatinib mesylate or 50 M AG490) at 37°C for 10 min before treatment with TPO (200 ng/ml). Then the lysates were immunoprecipitated with anti-Jak2 antibody or normal rabbit serum as a control, followed by immunoblot analysis using PY antibody (4G10) (upper panel). The membrane was stripped and reprobed with anti-Jak2 antibody (lower panel).

Full figure and legend (212K)

Considering that the Jak family PTKs play an important role in c-Mpl-mediated signaling, we then examined the effect of AG490, a selective inhibitor of the Jak PTKs, on TPO-induced aggregation of platelets from patients with CML-CP. The platelet aggregation induced by TPO was inhibited by AG490 in a dose-dependent manner (Figure 5a).

Effects of PTK inhibitors on the constitutive activation of PI3-kinase in platelets from patients with CML-CP

Based on the findings that imatinib mesylate inhibited TPO-induced aggregation of platelets from patients with CML-CP, we believe that Bcr-Abl is likely to be involved in the activation of PI3-kinase in the platelets. Therefore, we examined the effects of the inhibitor on PI3-kinase activity in the platelets. As shown in Figure 5b, imatinib mesylate, but not AG490, suppressed the activity of PI3-kinase constitutively activated in platelets from a patient with CML-CP. Therefore, Bcr-Abl is involved in the constitutive activation of PI3-kinase in platelets from patients with CML-CP. The Jak family PTKs, however, are unlikely to be involved in the constitutive activation of PI3-kinase in the platelets.

Tyrosine phosphorylation of Jak2 in platelets from patients with CML-CP

Next, we examined the effects of AG490, and imatinib mesylate, respectively, on tyrosine phosphorylation of Jak2 in CML-CP platelets treated with TPO. As shown in Figure 5c, Jak2 was not tyrosine-phosphorylated in platelets from patients with CML-CP before TPO treatment. AG490 but not imatinib mesylate inhibited TPO-induced tyrosine phosphorylation of Jak2. Tyrosine phosphorylation of Jak2 is indispensable for the activation of Jak2. These findings indicate that Jak2 is not involved in the priming activity mediated by PI3-kinase for TPO-induced CML-CP platelets aggregation, although it plays a role in the signal transduction for TPO-induced platelet aggregation.

Discussion

In the present study, we demonstrated the ability of TPO to induce the aggregation of platelets from patients with CML-CP but not of those from patients with PV, ET, or MF. Subsequently, PI3-kinase was constitutively activated in platelets from patients with CML-CP but not in those from patients with other MPDs. Pretreatment of the platelets with PI3-kinase inhibitors abolished TPO-induced aggregation. In addition, suppression of PI3-kinase activity by treatment of CML-CP platelets with imatinib mesylate is consistent with its inhibitory effect on TPO-induced aggregation of their platelets. Based on these findings, we conclude that the activation of PI3-kinase primes the CML-CP platelets for induction of aggregation by TPO and that Bcr-Abl is a PTK that may constitutively activate PI3-kinase in CML-CP platelets.

Constitutively activated PI3-kinase potentiates TPO-induced aggregation of platelets from CML-CP patients. CML-CP platelets, however, are known to response poorly to aggregating agonists.^14,15,29,41 This poor response is partly due to the deficiency of receptors for the agonists.¹⁴ Alternatively, the activation of PI3-kinase may be insufficient to prime CML-CP platelets for aggregation induced by aggregating agonists, although the precise mechanism is unknown. Another intrinsic defects may be involved in the poor response to aggregating agonists.

Among the PTKs, Bcr-Abl is the most likely to constitutively activate PI3-kinase in CML-CP platelets, based on the finding that imatinib mesylate inhibits the activity of PI3-kinase in the CML-CP platelets. Oda et al¹⁵ reported that Bcr-Abl was not detected in platelets from patients with CML-CP by immunoblot analysis. Recently, however, Bcr-Abl was found to be expressed in such platelets by immunoblot analysis using a different antibody against Abl.⁴² The expression of bcr-abl mRNA was also detected in these platelets by RT-PCR (data not shown), as described in a previous study,⁴² although contamination of leukocytes in CML-CP platelet preparations could not be entirely ruled out. Additionally, the finding that imatinib mesylate suppressed tyrosine phosphorylation of Crkl, one of the major substrates of Bcr-Abl, in CML-CP platelets³⁴ supports the idea that Bcr-Abl regulates the activation of PI3-kinase in the platelets.

Imatinib mesylate was found to inhibit the tyrosine kinase activity of c-Kit, platelet-derived growth factor -receptor (PDGFR) and PDGFR.^34,35,36 Activation of these PTKs leads to the recruitment of PI3-kinase to the plasma membrane through its binding to the receptors, thereby allowing PI3-kinase to be activated. Functionally active PDGFR but not PDGFR has been shown to be expressed in resting platelets.⁴³ PDGFR was found to negatively regulate platelet aggregation induced by aggregating agonists such as thrombin.⁴³ Therefore, it is very unlikely that PDGFR potentiates the CML-CP platelets for induction of aggregation by TPO through constitutive activation of PI3-kinase, although the physiological function and mutations of PDGFR in platelets from patients with MPDs remains to be elucidated. Functionally active c-Kit has been shown to be expressed in normal platelets only after stimulation with aggregating agonists, and thus, stem cell factor, a cognate ligand for c-Kit, dose-dependently enhanced the secondary wave of platelet aggregation induced by aggregating agonists.⁴⁴ Stem cell factor by itself was unable to induce aggregation of normal platelets⁴⁴ and platelets from patients with MPDs (data not shown). In previous studies, mutations of the juxtamembrane and extracellular domains of c-kit were detected in seven (8.8%) of 80 and one (8.3%) of 12 cases with CML, respectively.^45,46 Mutations of the extracellular domain of c-kit were also detected in two of six cases with MF but not in six cases with ET nor in one with PV.⁴⁶ These mutants were not constitutively activated. Therefore, the mutation of c-kit is unlikely to be involved in constitutive activation of PI3-kinase in CML-CP platelets. Hallek et al⁴⁷ reported that c-Kit was constitutively tyrosine-phosphorylated and activated through its association with Bcr-Abl in Mo7 and 32D cells expressing c-Kit and Bcr-Abl. The possibility that Bcr-Abl constitutively activates c-Kit in CML-CP platelets cannot be entirely ruled out.

Although the activation of Jak2 is a critical step in coupling c-Mpl-mediated signals to the downstream signaling molecules, the function of Jak2 in the signal transduction involved in TPO-induced platelet aggregation remains to be elucidated. As shown in the present study, Jak2 was likely to be involved in transducing the signals for TPO-induced aggregation of CML-CP platelets, because TPO induced tyrosine phosphorylation of Jak2 and a selective inhibitor of Jak2 suppressed their aggregation in response to TPO in a dose-dependent manner. Jak2 in CML-CP platelets, however, was not activated before TPO treatment. Therefore, Jak2 is unable to prime CML-CP platelets for TPO-induced aggregation through activation of PI3-kinase.

The expression of c-Mpl is not reduced in CML-CP platelets.¹⁶ However, findings on the levels of c-Mpl expression in the platelets from patients with PV and ET are controversial.^{16,17,18,19,20} Moliterno et al¹⁶ reported that expression of c-Mpl was markedly reduced or absent in platelets from all 34 patients with PV and from 13 of 14 patients with MF, and that its expression in megakaryocytes from patients with PV was also reduced. They also reported that TPO failed to induce tyrosine phosphorylation of signaling molecules, including Jak2 and Stat5, in the platelets from patients with PV and MF. However, effects of TPO on agonist-induced platelet aggregation were not described in their report. In the present study, however, TPO enhanced ADP-induced aggregation of platelets from all 10 patients with PV, all three patients with MF, and eight of 10 patients with ET. Usuki et al²⁴ also reported that TPO enhanced agonist-induced aggregation of platelets from patients with MPDs. Therefore, although the expression of c-Mpl was not analyzed in the present study, c-Mpl in platelets from patients with PV, MF, and ET may be sufficient to initiate the signals implicated in TPO-enhanced platelet aggregation even if its expression is decreased.

Recently, TPO was found to be able to prime collagen receptor signaling in normal platelets through a PI3-kinase-dependent pathway.³¹ The priming action of TPO on platelets activated by ADP was also found in this study.³¹ Herein, we have shown that TPO activates PI3-kinase in platelets from patients with PV, ET, and MF and that pretreatment with TPO promotes the secondary wave of aggregation of their platelets induced by aggregating agonists. Furthermore, pretreatment of these platelets with PI3-kinase inhibitors suppressed a TPO-enhanced secondary wave of platelet aggregation induced by ADP, epinephrine, and collagen, respectively (data not shown). Taking these previous findings together with the present results, it seems probable that the activation of PI3-kinase by TPO also plays an important role in priming agonist-induced aggregation of platelets from patients with PV, ET, and MF.

Although PI3-kinase in their platelets was activated after TPO treatment, TPO alone was unable to induce platelet aggregation. In contrast, TPO alone was able to induce aggregation of CML-CP platelets, in which PI3-kinase was constitutively activated. Therefore, activation of PI3-kinase is required but not sufficient for induction of platelet aggregation by TPO. Considering that Bcr-Abl activates multiple signaling pathways, including the Ras/MAP-kinase pathway,⁴⁸ a PI3-kinase-independent pathway, which is activated by Bcr-Abl, may be involved in TPO-induced aggregation of CML-CP platelets.

In conclusion, the constitutive activation of PI3-kinase in platelets from patients with CML-CP plays a crucial role in priming the platelets for aggregation induced by TPO. Its activation, however, is insufficient for induction of platelet aggregation by TPO. Bcr-Abl appears to play a central role in constitutive activation of PI3-kinase in CML-CP platelets. Other mechanisms of PI3-kinase activation in CML platelets, such as dysfunction of PTEN or SHIP, two negative regulators of PI3-kinase, may also play an important role. In addition, the PI3-kinase-independent mechanism that is involved in TPO-induced aggregation of CML-CP platelets should also be elucidated. The response of platelets to TPO may be helpful for the diagnosis of CML-CP. Moreover, by increasing the platelet counts and improving platelet functioning, TPO may be useful for treatment of the thrombocytopenia induced in CML-CP patients by interferon- and/or chemotherapy, and the thrombocytopenia caused by blastic crisis.

References

1. de Sauvage FJ, Hass PE, Spencer SD, Malloy BE, Gurney AL, Spencer SA et al. Stimulation of megakaryocytopoieis and thrombopoiesis by the c-Mpl ligand. Nature 1994; 369: 533–538. | Article | PubMed | ChemPort |
2. Lok S, Kaushansky K, Holly RD, Kuijper JL, Lofton-Day CE, Oort PJ et al. Cloning and expression of murine thrombopoietin cDNA and stimulation of platelet production in vivo. Nature 1994; 369: 565–568. | Article | PubMed | ISI | ChemPort |
3. Wendling F, Maraskovsky E, Debili N, Florindo C, Teepe M, Titeux M et al. c-Mpl ligand is a humoral regulator of megakaryopoiesis. Nature 1994; 369: 571–574. | Article | PubMed | ChemPort |
4. Bartley TD, Bogenberger J, Hunt P, Li YS, Lu HS, Martin F et al. Identification and cloning of a megakaryocyte growth and development factor that is a ligand for the cytokine receptor Mpl. Cell 1994; 77: 1117–1124. | Article | PubMed | ISI | ChemPort |
5. Sohma Y, Akahori H, Seki N, Hori T, Ogami K, Kato T et al. Molecular cloning and chromosomal localization of the human thrombopoietin gene. FEBS Lett 1995; 353: 57–61. | Article |
6. Mathia N, Louache F, Vainchenker W, Wendling F. Oligodeoxynucleotides antisense to the proto-oncogene c-mpl specifically inhibit in vitro megakaryocytopoiesis. Blood 1993; 82: 1395–1401. | PubMed | ISI | ChemPort |
7. Kojima H, Hamazaki Y, Nagata Y, Todokoro K, Nagasawa T, Abe T. Modulation of platelet activation in vitro by thrombopoietin. Thromb Haemost 1995; 74: 1541–1545. | PubMed |
8. Kubota Y, Arai T, Tanaka T, Yamaoka G, Kiuchi H, Kajikawa T et al. Thrombopoietin modulates platelet activation in vitro through protein-tyrosine phosphorylation. Stem Cells 1996; 14: 439–444. | PubMed |
9. Montrucchio G, Brizzi MF, Calosso G, Marengo S, Pegoraro L, Camussi G. Effects of recombinant human megakaryocyte growth and development factor on platelet activation. Blood 1996; 87: 2762–2768. | PubMed |
10. Oda A, Miyakawa Y, Druker BJ, Ozaki K, Yabusaki K, Shirasawa Y et al. Thrombopoietin primes human platelet aggregation induced by shear stress and by multiple agonists. Blood 1996; 87: 4664–4670. | PubMed |
11. Hofmann WK, Ottmann OG, Hoelzer D. Megakaryocytic growth factors: is there a new approach for management of thrombocytopenia in patients with malignancies? Leukemia 1999; 13: 14–18. | Article | PubMed |
12. Kaushansky K. Use of thrombopoietic growth factors in acute leukemia. Leukemia 2000; 14: 505–508. | Article | PubMed |
13. Kuter DJ, Begley CG. Recombinant human thrombopoietin: basic biology and evaluation of clinical studies. Blood 2002; 100: 3457–3469. | Article | PubMed | ISI | ChemPort |
14. Wehmeier A, Sudhoff T, Meierkord F. Relation of platelet abnormalities to thrombosis and hemorrhage in chronic myeloproliferative disorders. Semin Thromb Hemost 1997; 23: 391–402. | PubMed |
15. Oda A, Miyakawa Y, Druker BJ, Ishida A, Ozaki K, Ohashi H et al. Crkl is constitutively tyrosine phosphorylated in platelets from chronic myelogenous leukemia patients and inducibly phosphorylated in normal platelets stimulated by thrombopoietin. Blood 1996; 88: 4304–4313. | PubMed |
16. Moliterno AR, Hankins WD, Spivak JL. Impaired expression of the thrombopoietin receptor by platelets from patients with polycythemia vera. N Engl J Med 1998; 338: 572–580. | Article | PubMed |
17. Le Blanc K, Andersson P, Samuelsson J. Marked heterogeneity in protein levels and functional integrity of the thrombopoietin receptor c-mpl in polycythaemia vera. Br J Haematol 2000; 108: 80–85. | Article | PubMed |
18. Tefferi A, Yoon SY, Li CY. Immunohistochemical staining for megakaryocyte c-mpl may complement morphologic distinction between polycythemia vera and secondary erythrocytosis. Blood 2000; 96: 771–772. | PubMed |
19. Horikawa Y, Matsumura I, Hashimoto K, Shiraga M, Kosugi S, Tadokoro S et al. Markedly reduced expression of platelet c-mpl receptor in essential thrombocythemia. Blood 1997; 90: 4031–4038. | PubMed |
20. Kiladjian JJ, Elkassar N, Hetet G, Briere J, Grandchamp B, Gardin C. Study of the thrombopoietin receptor in essential thrombocythemia. Leukemia 1997; 11: 1821–1826. | Article | PubMed |
21. Ezumi Y, Takayama H, Okuma M. Thrombopoietin, a c-Mpl ligand, induces tyrosine phosphorylation of Tyk2, JAK2, and STAT3, and enhances agonists-induced aggregation in platelets in vitro. FEBS Lett 1995; 374: 48–52. | Article | PubMed |
22. Miyakawa Y, Oda A, Druker BJ, Kato T, Miyazaki H, Handa M et al. Recombinant thrombopoietin induces rapid protein tyrosine phosphorylation of Janus kinase 2 and Shc in human blood platelets. Blood 1995; 86: 23–27. | PubMed | ChemPort |
23. Rodriguez-Linares B, Watson SP. Thrombopoietin potentiates activation of human platelets in association with JAK2 and TYK2 phosphorylation. Biochem J 1996; 316: 93–98. | PubMed |
24. Usuki K, Iki S, Endo M, Izutsu K, Inoue K, Nishimura T et al. Influence of thrombopoietin on platelet activation in myeloproliferative disorders. Br J Haematol 1997; 97: 530–537. | Article | PubMed |
25. Chang F, Lee JT, Navolanic PM, Steelman LS, Shelton JG, Blalock WL et al. Involvement of PI3K/Akt pathway in cell cycle progression, apoptosis, and neoplastic transformation: a target for cancer chemotherapy. Leukemia 2003; 17: 590–603. | Article | PubMed | ISI | ChemPort |
26. Lee Jr JT, McCubrey JA. The Raf/MEK/ERK signal transduction cascade as a target for chemotherapeutic intervention in leukemia. Leukemia 2002; 16: 486–507. | Article | PubMed | ISI | ChemPort |
27. Luo JM, Yoshida H, Komura S, Ohishi N, Pan L, Shigeno K et al. Possible dominant-negative mutation of the SHIP gene in acute myeloid leukemia. Leukemia 2003; 17: 1–8. | Article | PubMed | ISI | ChemPort |
28. Vazquez F, Sellers WR. The PTEN tumor suppressor protein: an antagonist of phosphoinositide 3-kinase signaling. Biochim Biophys Acta 2000; 1470: M21–M35. | Article | PubMed | ISI | ChemPort |
29. Rittenhouse SE. Phosphoinositide 3-kinase activation and platelet function. Blood 1996; 88: 4401–4414. | PubMed |
30. Chen J, Herceg-Harjacek L, Groopman JE, Grabarek J. Regulation of platelet activation in vitro by the c-Mpl ligand, thrombopoietin. Blood 1995; 86: 4054–4062. | PubMed |
31. Pasquet JM, Gross BS, Gratacap MP, Quek L, Pasquet S, Payrastre B et al. Thrombopoietin potentiates collagen receptor signaling in platelets through a phosphatidylinositol 3-kinase-dependent pathway. Blood 2000; 95: 3429–3434. | PubMed |
32. Zauli G, Bassini A, Vitale M, Gibellini D, Celeghini C, Caramelli E et al. Thrombopoietin enhances the alpha IIb beta 3-dependent adhesion of megakaryocytic cells to fibrinogen or fibronectin through PI 3 kinase. Blood 1997; 89: 883–895. | PubMed |
33. Meydan N, Grunberger T, Dadi H, Shahar M, Arpaia E, Lapidot Z et al. Inhibition of acute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature 1996; 379: 645–648. | Article | PubMed | ISI | ChemPort |
34. Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 1996; 2: 561–566. | Article | PubMed | ISI | ChemPort |
35. Schindler T, Bornmann W, Pellicena P, Miller WT, Clarkson B, Kuriyan J. Structural mechanism for STI-571 inhibition of abelson tyrosine kinase. Science 2000; 289: 1938–1942. | Article | PubMed | ISI | ChemPort |
36. La Rosee P, O'Dwyer ME, Druker BJ. Insights from pre-clinical studies for new combination treatment regimens with the Bcr-Abl kinase inhibitor imatinib mesylate (Gleevec/Glivec) in chronic myelogenous leukemia: a translational perspective. Leukemia 2002; 16: 1213–1219. | Article | PubMed | ISI | ChemPort |
37. Ozaki Y, Satoh K, Yatomi Y, Yamamoto T, Shirasawa Y, Kume S. Detection of platelet aggregates with a particle counting method using light scattering. Anal Biochem 1994; 218: 284–294. | Article | PubMed | ISI | ChemPort |
38. Kubota Y, Tanaka T, Kitanaka A, Ohnishi H, Okutani Y, Waki M et al. Src transduces erythropoietin-induced differentiation signals through phosphatidylinositol 3-kinase. EMBO J 2001; 20: 5666–5677. | Article | PubMed | ChemPort |
39. Kubota Y, Tanaka T, Yamaoka G, Yamaguchi M, Ohnishi H, Kawanishi K et al. Wortmannin, a specific inhibitor of phosphatidylinositol 3-kinase, inhibits erythropoietin-induced erythroid differentiation of K562 cells. Leukemia 1996; 10: 720–726. | PubMed | ChemPort |
40. Skorski T, Bellacosa A, Nieborowska-Skorska M, Majewski M, Martinez R, Choi JK et al. Transformation of hematopoietic cells by BCR/ABL requires activation of a PI-3k/Akt-dependent pathway. EMBO J 1997; 16: 6151–6161. | Article | PubMed | ISI | ChemPort |
41. Best D, Pasquet S, Littlewood TJ, Brunskill SJ, Pallister CJ, Watson SP. Platelet activation via the collagen receptor GPVI is not altered in platelets from chronic myeloid leukaemia patients despite the presence of the constitutively phosphorylated adapter protein CrkL. Br J Haematol 2001; 112: 609–615. | Article | PubMed |
42. ten Bosch GJ, Kessler JH, Blom J, Joosten AM, Gambacorti-Passerini C, Melief CJ et al. BCR-ABL oncoprotein is expressed by platelets from CML patients and associated with a special pattern of CrkL phosphorylation. Br J Haematol 1998; 103: 1109–1115. | Article | PubMed |
43. Vassbotn FS, Havnen OK, Heldin CH, Holmsen H. Negative feedback regulation of human platelets via autocrine activation of the platelet-derived growth factor alpha-receptor. J Biol Chem 1994; 269: 13874–13879. | PubMed |
44. Grabarek J, Groopman JE, Lyles YR, Jiang S, Bennett L, Zsebo K et al. Human kit ligand (stem cell factor) modulates platelet activation in vitro. J Biol Chem 1994; 269: 21718–21724. | PubMed |
45. Inokuchi K, Yamaguchi H, Tarusawa M, Futaki M, Hanawa H, Tanosaki S et al. Abnormality of c-kit oncoprotein in certain patients with chronic myelogenous leukemia – potential clinical significance. Leukemia 2002; 16: 170–177. | Article | PubMed | ISI | ChemPort |
46. Nakata Y, Kimura A, Katoh O, Kawaishi K, Hyodo H, Abe K et al. c-kit point mutation of extracellular domain in patients with myeloproliferative disorders. Br J Haematol 1995; 91: 661–663. | PubMed | ISI | ChemPort |
47. Hallek M, Danhauser-Riedl S, Herbst R, Warmuth M, Winkler A, Kolb HJ et al. Interaction of the receptor tyrosine kinase p145c-kit with the p210bcr/abl kinase in myeloid cells. Br J Haematol 1996; 94: 5–16. | Article | PubMed | ISI | ChemPort |
48. Steelman LS, Pohnert SC, Shelton JG, Franklin RA, Bertrand FE, McCubrey JA. JAK/STAT, Raf/MEK/ERK, PI3K/Akt and BCR-ABL in cell cycle progression and leukemogenesis. Leukemia 2004; 18: 189–218. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Takeshi Arai for monitoring of the platelet aggregation, Novartis (Basel, Switzerland) for providing imatinib mesylate, and the Kirin Brewery Co. (Tokyo, Japan) for providing recombinant human thrombopoietin.