SEARCH

SEARCH BY CITATION

Keywords:

  • anagrelide;
  • megakaryocyte;
  • signal transduction;
  • thrombocythemia;
  • thrombopoietin;
  • transcription factor

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Summary. Background: Anagrelide is a selective inhibitor of megakaryocytopoiesis used to treat thrombocytosis in patients with chronic myeloproliferative disorders. The effectiveness of anagrelide in lowering platelet counts is firmly established, but its primary mechanism of action remains elusive. Objectives and Methods: Here, we have evaluated whether anagrelide interferes with the major signal transduction cascades stimulated by thrombopoietin in the hematopoietic cell line UT-7/mpl and in cultured CD34+-derived human hematopoietic cells. In addition, we have used quantitative mRNA expression analysis to assess whether the drug affects the levels of known transcription factors that control megakaryocytopoiesis. Results: In UT-7/mpl cells, anagrelide (1 μm) did not interfere with MPL-mediated signaling as monitored by its lack of effect on JAK2 phosphorylation. Similarly, the drug did not affect the phosphorylation of STAT3, ERK1/2 or AKT in either UT-7/mpl cells or primary hematopoietic cells. In contrast, during thrombopoietin-induced megakaryocytic differentiation of normal hematopoietic cultures, anagrelide (0.3 μm) reduced the rise in the mRNA levels of the transcription factors GATA-1 and FOG-1 as well as those of the downstream genes encoding FLI-1, NF-E2, glycoprotein IIb and MPL. However, the drug showed no effect on GATA-2 or RUNX-1 mRNA expression. Furthermore, anagrelide did not diminish the rise in GATA-1 and FOG-1 expression during erythropoietin-stimulated erythroid differentiation. Cilostamide, an exclusive and equipotent phosphodiesterase III (PDEIII) inhibitor, did not alter the expression of these genes. Conclusions: Anagrelide suppresses megakaryocytopoiesis by reducing the expression levels of GATA-1 and FOG-1 via a PDEIII-independent mechanism that is differentiation context-specific and does not involve inhibition of MPL-mediated early signal transduction events.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Anagrelide is an imidazoquinazoline derivative that inhibits bone marrow megakaryocytopoiesis [1–3]. This activity underlies the effectiveness of the drug in reducing the high platelet count in essential thrombocythemia (ET) and other myeloproliferative disorders [4]. Anagrelide also inhibits a type III phosphodiesterase [phosphodiesterase III (PDEIII)] found in platelets and the myocardium, but this effect does not appear to be related to its antimegakaryocytopoietic properties [5]. Using primary umbilical cord-derived human hematopoietic progenitor cells stimulated to proliferate and/or differentiate along different lineages, we have previously demonstrated that anagrelide acts as a selective inhibitor of thrombopoietin (TPO)-induced megakaryocyte development [5,6]. These findings suggested that the drug might interfere with one or more molecular events required for the action of the hormone.

TPO drives megakaryocyte development by promoting the proliferation, survival and lineage-specific differentiation of a hierarchy of increasingly restricted hematopoietic progenitors [7,8]. These biological effects are mediated by binding of TPO to MPL, its cognate receptor. This interaction induces receptor dimerization and a conformational change that results in activation of the non-receptor tyrosine kinase JAK2 by transphosphorylation. Activated JAK2, in turn, phosphorylates tyrosine residues on the distal cytoplasmic portion of MPL, and this modification promotes the recruitment of additional scaffolding and signaling proteins to the receptor. In primary megakaryocyte progenitors and TPO-responsive cell lines, these initial events lead to the activation of several signal transduction cascades, namely the JAK–STAT, the RAS–mitogen-activated protein kinase (MAPK), the phosphoinositide-3-kinase (PI3K)–AKT and protein kinase C (PKC) pathways [9,10].

At the gene expression level, megakaryocyte development is regulated by changes in the activity and/or expression of a complex network of transcription factors, for which functional cis-acting regulatory elements have been identified in the promoter regions of many megakaryocyte-expressed genes (reviewed by Shivdasani [11] and Goldfarb [12]). Prominent among these regulators are GATA-1 [13], FLI-1 [14] and p45 NF-E2 [15]. Together with FOG-1 [16], a transcriptional coregulator that binds to GATA-1, these hematopoietic transcription factors have been shown to play critical cooperative roles in the execution of the megakaryocyte differentiation program [16–18].

In an attempt to elucidate further the molecular mechanism underlying the antimegakaryocytopoietic activity of anagrelide, we have carried out a detailed examination of the effects of the drug on the immediate signal transduction events initiated by binding of TPO to MPL and on the expression of transcription factors associated with megakaryocyte development. Here, we present evidence demonstrating that anagrelide suppresses the mRNA and protein levels of GATA-1, as well as the mRNA levels of FOG-1, FLI-1 and NF-E2. In addition, we show that although GATA-1, FOG-1 and NF-E2 are also upregulated in the erythroid lineage, in this context the drug does not affect their expression. Furthermore, we demonstrate that the primary mechanism of transcription factor downregulation by anagrelide does not involve inhibition of MPL-mediated signaling, including activation of JAK2 or its other major immediate downstream effectors. The results from this study suggest that anagrelide targets a pathway upstream of GATA-1 that is only active in the megakaryocytic lineage.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Drugs

Anagrelide hydrochloride was obtained from Cambridge Major Laboratories (Germantown, WI, USA). All other inhibitors were purchased from Calbiochem (Nottingham, UK). Stock solutions were made in dimethylsulfoxide, stored in aliquots at − 20 °C and, when required, diluted in culture medium immediately before use.

Cells

UT-7/mpl cells (clone 5.1) [19] were grown in Iscove’s modified Dulbecco’s medium (IMDM) (GIBCO, Invitrogen, Paisley, UK) supplemented with 10% fetal bovine serum (HyClone, Thermo Fisher Scientific, Cramlington, UK), 2.5 ng mL−1 granulocyte–macrophage colony-stimulating factor (GM-CSF) (R&D Systems, Abingdon, UK), 2 mm glutamine and 0.5 mg mL−1 G418, as previously described [20]. CD34+ cells isolated from human umbilical cord blood (Stem Cell Technologies, London, UK) were seeded at a density of 1.5 × 105 cells mL−1 and cultured for 4 days in Stemspan medium (Stem Cell Technologies) supplemented with 100 U mL−1 penicillin, 0.1 mg mL−1 streptomycin, 0.25 μg mL−1 amphotericin B, 2% human umbilical cord blood plasma and a combination of growth factors that promotes the expansion of hematopoietic progenitors, consisting of 40 ng mL−1 recombinant human TPO (Insight, Biotechnology, Wembley, UK), 50 ng mL−1 SCF, 100 ng mL−1 Flt3 ligand and 10 ng mL−1 interleukin (IL)-3 (all from R&D Systems) [21]. All cultures were maintained at 37 °C in a humidified incubator under 5% CO2/95% air.

Protein phosphorylation studies and western blot analysis

To examine the activation state of phosphorylation cascades, cells were washed free of growth factors and rendered quiescent by incubation for 18 h in IMDM containing 0.5% bovine serum albumin (BSA, further purified fraction V; Sigma-Aldrich, Poole, UK). Cells were judged to be quiescent as previously described [20]. Quiescent cells were resuspended in fresh BSA-containing IMDM at a density of 0.5–1.0 × 106 cells mL−1 and then pretreated with the indicated drugs or an equivalent amount of vehicle for 15 min at 37 °C; this was followed by stimulation for a further 30 min with 100 ng mL−1 TPO. Aliquots of 2 × 106 cells were subsequently washed with ice-cold phosphate-buffered saline, resuspended by adding 70 μL of ice-cold lysis buffer [10 mm Tris–HCl, pH 7.4, 100 mm NaCl, 10% glycerol, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% sodium dodecylsulfate (SDS), 5 mm EDTA, 1 mm EGTA, 20 mm Na4P2O7, 2 mm Na3VO4, 1 mm NaF, 100 nm calyculin A, 1 mm phenylmethylsulfonyl fluoride and Halt Protease Inhibitor Cocktail (Thermo-Fisher Scientific, Basingstoke, UK)], and disrupted by sonication for 10 s at an amplitude of 35% with a VCX500 Ultrasonic Processor (Sonics & Materials, Newtown, CT, USA). The resultant lysates were centrifuged at 14 000 × g for 5 min at 4 °C to remove particulate material. Proteins (approximately 25–50 μg) in samples of the supernatant fraction were separated by SDS–10% polyacrylamide gel electrophoresis under reducing conditions and transferred to nitrocellulose membranes (Invitrogen, Paisley, UK). Phosphorylated proteins of interest were identified by immunoblotting for 18 h at 4 °C with anti-phospho-specific antibodies. To control for variations in protein loading, membranes were stripped and reprobed with the corresponding pan-protein antibodies or with a rabbit anti-actin IgG (Sigma-Aldrich). The following rabbit phospho-specific and pan-protein primary antibodies (all from Cell Signaling, Danvers, MA, USA) were used at 1 : 1000 dilution, unless otherwise indicated: anti-phospho-JAK2(Tyr1007/1008) (clone C80C3, 1 : 500), anti-JAK2 (clone 24B11), anti-phospho-STAT3(Tyr705) (clone D3A7), anti-STAT3 (clone 79D7), polyclonal anti-phospho-ERK1/2(Thr202/Tyr204), anti-ERK1/2 (clone 137F5), polyclonal anti-phospho-AKT(Ser473) and polyclonal anti-AKT. GATA-1 detection was carried out with an anti-GATA-1 primary antibody (clone D24E4), following identical procedures. Immunoreactive bands were detected with a horseradish peroxidase-labeled anti-rabbit IgG (Cell Signaling) and SuperSignal West Pico Chemiluminescent Substrate (Thermo-Fisher Scientific). The relative intensity of the bands was quantified by scanning densitometric analysis, using the Scion Image program (available at http://www.scioncorp.com).

RNA expression studies and quantitative polymerase chain reaction (Q-PCR) analysis

Four-day-expanded primary hematopoietic cells were subcultured in fresh medium supplemented with 2% human umbilical cord blood plasma and either 40 ng mL−1 TPO or 8 U mL−1 erythropoietin (EPO) in the presence of anagrelide, cilostamide or an equivalent amount of vehicle for various lengths of time. Cells were then harvested and processed for RNA extraction. Cellular RNA was extracted with RNeasy reagent (Qiagen, Crawley, UK), according to the manufacturer’s instructions, and cDNA was prepared from 0.3–0.5 μg of total RNA template with a High Capacity Reverse Transcription kit (Applied Biosystems, Warrington, UK). cDNA aliquots (equivalent to 5–10 ng of input RNA) were analyzed in duplicate by Q-PCR, using gene-specific TaqMan probes on an ABI Prism 7500 Fast Sequence Detection System (Applied Biosystems). Probes from the Assay on Demand gene expression collection (Applied Biosystems) for the following genes were used: GATA1 (Hs00231112_m1), GATA2 (Hs00231119_m1), FOG1 (Hs00419119_m1), RUNX1 (Hs00231079_m1), NFE2 (Hs00232351_m1), FLI1 (Hs00231107_m1), platelet glycoprotein (GP) IIb (ITGA2B, Hs00166246_m1), MPL (Hs00180489_m1), glycophorin A (GYPA, Hs00266777_m1), glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 4326317E), β-glucuronidase (GUSB, Hs99999908_m1) and TATA box-binding protein (TBP, Hs99999910_m1). Thermal conditions were as follows: 20 s at 95 °C, followed by 40 cycles of 3 s at 95 °C and 30 s at 60 °C. Reactions without cDNA were included as controls. Relative mRNA expression levels were calculated by the comparative cycle threshold (CT) method [22], using the CT values obtained for GUSB or TBP as internal references. For each target gene to be quantified, validation experiments were performed to confirm that the efficiency of amplification of target and reference genes was approximately equal.

Statistical analysis

Results were evaluated by t-test or by one-way or two-way anova, followed by Dunnet’s or Bonferroni’s post hoc tests as appropriate. A P-value < 0.05 was considered to denote statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Anagrelide does not inhibit MPL receptor-mediated signal transduction

To assess whether anagrelide interferes with TPO/MPL signaling, we initially studied the effects of the drug in UT7-7/mpl cells, a growth factor-dependent pluripotent cell line expressing the human form of the TPO receptor [19,20]. Figure 1 shows that incubation of these cells with 1 μm anagrelide, a saturating concentration for the inhibition of megakaryocyte development in culture [5], did not prevent the TPO-stimulated increase in the phosphorylation levels of JAK2, STAT3, ERK1/2 and AKT. This lack of effect of anagrelide contrasted with the reduction in the phosphorylation levels of the above signaling proteins caused by known inhibitors of the corresponding signaling pathways, and also with the increase in AKT phosphorylation caused by the PKC inhibitor GF109203X (Fig. 1).

image

Figure 1.  Lack of effect of anagrelide on the signaling pathways activated by thrombopoietin (TPO) in UT-7/mpl cells. Cells were left untreated (control) or were stimulated with 100 ng mL−1 TPO for 30 min in the presence of either 1 μm anagrelide (ANA), 10 μm WP1066, 20 μm Stattic, 20 μm PD98059, 20 μm LY294002 or 3 μm GF109203X. Phosphoproteins were detected by immunoblotting and quantified by scanning densitometry. A representative immunoblot is shown on the left; quantification is shown on the right. Relative phosphorylation levels were calculated as the ratio of the intensity of the phosphorylated polypeptide bands to the intensity of the corresponding total polypeptide bands. Results are expressed as a percentage of the level of phosphorylation measured in the sample incubated with TPO and no inhibitors. Error bars denote standard deviations; *< 0.05 and ***< 0.001 vs. vehicle (n = 3–6).

Download figure to PowerPoint

To validate the above findings in a more physiologic model, we examined the effects of anagrelide in primary CD34+-derived hematopoietic cells. Figure 2 shows that in these cells also, the drug failed to significantly affect the phosphorylation of ERK1/2, AKT and STAT3. Taken together, these results indicated that anagrelide does not inhibit the engagement and activation of the known major MPL downstream signaling pathways.

image

Figure 2.  Lack of effect of anagrelide on signaling pathways activated by thrombopoietin (TPO) in primary human hematopoietic cells. Cells were treated with 100 ng mL−1 TPO for 30 min in the presence of 1 μm anagrelide (ANA) or an equivalent amount of vehicle. Phosphorylation of STAT3, ERK1/2 and AKT was detected by immunoblotting and quantified as described in Fig. 1. A representative immunoblot is shown at the top; quantification is shown at the bottom. Error bars denote standard deviations; = non-significant vs. vehicle (n = 3–4).

Download figure to PowerPoint

Differential effects of anagrelide on the expression of transcription factors involved in megakaryocyte development

The lack of effect of anagrelide on early phosphorylation events prompted us to examine whether the drug affected other downstream processes associated with the regulation of megakaryocytopoiesis. As shown in Fig. 3, incubation of primary hematopoietic cells with 0.3 μm anagrelide under megakaryocyte differentiation conditions caused a marked reduction in the time-dependent raise in the mRNA levels of MPL and the megakaryocyte differentiation marker GPIIb. Furthermore, anagrelide also reduced the increase in the mRNA levels of GATA-1 and FOG-1, two transcription factors known to regulate the expression of both MPL and GPIIb [17,23]. In contrast, the mRNA levels of RUNX1 and GATA-2, two additional transcription factors implicated in megakaryocytopoiesis [24,25], remained unaffected. Similar differential effects were observed when the concentration of anagrelide was raised to 1 μm (data not shown). As shown in Fig. 4, anagrelide also suppressed the rise in the mRNA levels of FLI-1 and p45 NF-E2, in agreement with the notion that the genes encoding both of these are responsive to GATA-1 regulation [16,18,26]. Downregulation of GPIIb, MPL, GATA-1, FOG-1, FLI-1 and NF-E2 was also observed when incubation with anagrelide was extended beyond the early stages of differentiation or if anagrelide was present only during the latter stages of this process (Fig. 4). In all of these cases, comparable results were obtained when mRNA expression levels were quantified using GUSB or TBP as the internal references (data not shown). Taken together, these results indicate that anagrelide inhibits the expression of a subset of transcription factors associated with megakaryocyte development, and that this effect is not restricted to the initial stages of the differentiation program.

image

Figure 3.  Effect of anagrelide on the time courses of mRNA expression corresponding to GPIIb, MPL, GATA-1, GATA-2, FOG-1 and RUNX1. Primary hematopoietic cells were cultured with thrombopoietin under megakaryocyte differentiation conditions in the absence or presence of 0.3 μm anagrelide (ANA). Cells were harvested at various time points, and mRNA levels for the indicated target genes were determined by quantitative polymerase chain reaction (Q-PCR). Experiments were repeated three times, each time using a different batch of cells. Results from representative time courses are expressed relative to the respective mRNA levels at the initiation of the culture period. Error bars denote standard deviations of replicate Q-PCR assays. *< 0.05 and **< 0.01 by two-way anova.

Download figure to PowerPoint

image

Figure 4.  Comparison between the effects of early and late addition of anagrelide on gene expression during megakaryocyte differentiation. Cells were cultured for 4 or 8 days in the absence (white bars) or presence (black bars) of 0.3 μm anagrelide as described in Fig. 3. Alternatively, cells were differentiated for 4 days before addition of anagrelide for another 4 days (indicated as 4 + 4). mRNA levels for the indicated target genes was determined by quantitative polymerase chain reaction (Q-PCR). Experiments were repeated three times, each time with a different batch of cells. Results from a representative experiment are expressed relative to the respective mRNA levels at the initiation of the culture period. Error bars denote standard deviations of replicate Q-PCR assays. *< 0.05, **< 0.01 and ***< 0.001 by t-test.

Download figure to PowerPoint

Anagrelide reduces GATA-1 protein levels

To substantiate the above findings, we also examined GATA-1 expression by immunoblotting. As shown in Fig. 5, anagrelide reduced the level of this protein by approximately 50%, indicating that the inhibition of transcription factor expression seen at the mRNA level results in comparable changes at the protein level.

image

Figure 5.  Downregulation of GATA-1 protein levels by anagrelide during megakaryocyte differentiation. Cells were cultured with anagrelide or an equivalent amount of vehicle for 4 days as described in Fig. 3. GATA-1 protein levels were quantified by immunoblotting and densitometry. Experiments were repeated three times. Results from a representative immunoblot are shown. The values under the image represent the ratio of the intensity of the GATA-1 polypeptide band to the intensity of the corresponding β-actin band.

Download figure to PowerPoint

Anagrelide does not suppress the expression of GATA-1, FOG-1 and NF-E2 during erythroid differentiation

GATA-1, FOG-1 and NF-E2 are also key positive modulators of erythropoiesis [16,27,28]. In agreement with this notion, as shown in Fig. 6, incubation of CD34+-derived hematopoietic cells with EPO for 4 days resulted in a 2.5-fold to 4-fold increase in the levels of these transcripts. In contrast, consistent with its role as a suppressor of erythroid differentiation [29], levels of FLI-1 mRNA were reduced by approximately 6-fold. Under these conditions, mRNA levels for the erythroid marker GYPA increased by > 100-fold, whereas the expression of the housekeeping gene GAPDH remained unaltered. Importantly, anagrelide did not affect the levels of any of these transcripts, indicating that the suppression of transcription factor expression induced by this drug does not occur in the context of erythroid differentiation.

image

Figure 6.  Lack of effect of anagrelide on gene expression during erythroid differentiation. Cells were cultured with erythropoietin for 4 days in the absence or presence of 0.3 μm anagrelide. mRNA levels for the indicated target genes were determined by quantitative polymerase chain reaction. Results are expressed as described in Fig. 3. Error bars denote standard deviations of pooled data from two separate experiments; = non-significant vs. vehicle (n = 4).

Download figure to PowerPoint

The downregulation of GATA-1 and FOG-1 transcripts is not related to the PDEIII inhibitory activity of anagrelide

In order to ascertain whether the reduction in FOG-1 and GATA-1 mRNA levels caused by anagrelide was related to its anti-PDEIII activity, we evaluated the effects of cilostamide, a compound that has a similar potency to anagrelide for this particular effect [5]. Figure 7 shows that, whereas treatment with anagrelide led to a significant reduction in the levels of these transcripts, cilostamide, at a concentration approximately 10-fold above its IC50 for PDEIII [5], was completely ineffective. Thus, these results ruled out the possibility that PDEIII inhibition was the cause of the reduction in the accumulation of GATA-1 and FOG-1 transcripts.

image

Figure 7.  Comparison between the effects of anagrelide and cilostamide on the mRNA levels of GPIIb, GATA-1 and FOG-1 during megakaryocyte differentiation. Cells were cultured with thrombopoietin for 4 days in the presence of the indicated compounds (0.3 μm). mRNA levels for the indicated target genes were determined by quantitative polymerase chain reaction. Results are expressed as described in Fig. 3. Error bars denote standard deviations of pooled data from four separate experiments; **< 0.01 vs. vehicle (n = 10).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

In previous investigations, we found that although anagrelide showed potent inhibitory activity against TPO-induced megakaryocyte development, it did not impair the effects of the hormone on the ex vivo expansion and survival of primary hematopoietic cells [5,6]. We also found that anagrelide did not suppress the growth and/or differentiation processes induced in culture by stem cell factor, IL-3, Flt-3 ligand, EPO or GM-CSF [5,6]. Altogether, these findings suggested that anagrelide does not interfere with the binding of TPO to its receptor or with the activation of major downstream signal transduction pathways that are common to MPL and other cytokine receptors. In contrast, other investigators had previously argued that anagrelide does indeed inhibit TPO receptor-mediated signaling [30]. Thus, in order to resolve this apparent controversy, in the present study we first investigated whether anagrelide interfered with the early signaling events initiated by TPO in UT-7/mpl cells.

UT-7/mpl is a growth factor-dependent pluripotent human cell line transduced with human c-MPL. Like parental UT-7 cells, UT-7/mpl cells depend on GM-CSF or EPO for growth and survival, but, in addition, they proliferate and differentiate in the presence of TPO [19,20]. Hence, these cells have been found to be a convenient and biologically relevant model system for the characterization of intracellular signaling events stimulated upon binding of TPO to its receptor [20,31–34]. The results obtained here with these cells provide direct evidence that anagrelide does not inhibit the TPO/MPL-mediated activation of JAK2 or the three major signaling cascades engaged by JAK2 activation, namely the JAK–STAT, RAS–MAPK and PI3K–AKT pathways.

TPO is also known to activate PKC-mediated signaling (reviewed by Redig et al. [35]). In the present study, we found that the bisindolylmaleimide GF109203X, a pan-PKC inhibitor, increased TPO-stimulated AKT phosphorylation in UT-7/mpl cells. This finding is in agreement with studies in other biological systems showing that members of the PKC family negatively regulate AKT [35–37]. Thus, the striking increase in AKT phosphorylation observed in the presence of GF109203X, as compared with the lack of significant change in phospho-AKT levels in the presence of anagrelide, also provided indirect evidence that the latter does not inhibit TPO-induced activation of PKC pathways.

In apparent disagreement with our findings, in the murine leukemic cell line BAF3, anagrelide was reported to inhibit TPO-stimulated tyrosine phosphorylation when the cells carried human c-MPL but not if they carried the murine ortholog [30]. On the basis of those results, it was concluded that anagrelide interferes with human MPL-mediated signal transduction. However, as the identity of the affected phosphorylated proteins had not been established, and given that the human and murine forms of MPL are remarkably similar, displaying 82% amino acid sequence identity [38], the mechanism underlying the reported inhibition in BAF3 cells remains to be ascertained. Nevertheless, a likely explanation for the lack of agreement between the results in BAF3 cells and the present findings is that the concentrations at which anagrelide was tested in the respective studies were not comparable. For our phosphorylation experiments, UT-7/mpl cells were treated with 1 μm anagrelide, a dose that causes maximal inhibition of megakaryocyte development in cell culture [5]. In contrast, in BAF3/mpl cells, the dose of drug shown to inhibit tyrosine phosphorylation was equivalent to approximately 17 μm, a concentration that may have broader effects, unrelated to the inhibition of megakaryocyte development.

In order to completely rule out the possibility that the lack of effect of anagrelide in UT-7/mpl cells was attributable to some unaccounted characteristic of this model system, its effect on TPO signaling was also investigated in normal hematopoietic cells. Results obtained with these cells confirmed that, indeed, anagrelide does not interfere with signaling through the three major phosphorylation cascades engaged by MPL activation.

A salient finding of the present work is that anagrelide suppressed the TPO-stimulated increase in the mRNA expression levels of GATA-1 and FOG-1. As these interacting transcription factors are essential for complete development of the megakaryocytic lineage, being required from the progenitor stage through to the last stages of differentiation [13,16,18], this result provides an important clue for understanding the molecular basis for the platelet-lowering action of anagrelide. We also observed that the drug suppressed the increase in the mRNA levels of FLI-1 and p45 NF-E2, two additional hematopoietic transcription factors whose disruption affects the later stages of megakaryocyte development. FLI-1 is a member of the Ets family of transcription factors that acts primarily in the early maturing megakaryocyte [11,14,39]. p45 NF-E2 is essential for terminal megakaryocyte maturation and pro-platelet formation [11,15,28]. Expression of the genes encoding these proteins is known to be positively regulated by GATA-1 [16,18,26], raising the possibility that their downregulation by anagrelide results from the effect of the drug on the expression of the latter. However, an alternative mechanism in which anagrelide suppresses the expression of FLI-1 and NF-E2 via GATA-1-independent means cannot be entirely discounted at this time. Importantly, we also found that anagrelide did not affect the mRNA levels of GATA-2 and RUNX1, even though the expression and function of the genes encoding these proteins are known to be strongly associated with those of GATA-1 [24,25,40]. Maintenance of GATA-2 levels and the upregulation of RUNX1 represent normal transcriptional events during the earlier stages of megakaryocytopoiesis that are necessary to promote this process [24,25,40,41]. Therefore, the facts that anagrelide did not reduce GATA-2 expression and did not counteract the increase in RUNX1 demonstrate that the downregulation of GATA-1 and FOG-1 did not result from complete shutdown of the megakaryocyte developmental program or from a generalized non-specific inhibition of gene expression. It should be noted that although anagrelide did not alter the mRNA expression levels of GATA-2 and RUNX1, the present study does not rule out the possibility that it could have acted at the protein level or by interfering with their transcriptional activity.

GATA-1 and FOG-1 are also essential for erythropoiesis [16,27]. Indeed, GATA-binding sites exist in all erythroid-specific genes. GATA-1 expression levels are low in CD34+ multipotent hematopoietic progenitors, and increase during both erythroid and megakaryocytic differentiation, attesting to the relatedness of the two lineages [42]. Crucially, the present study shows that, in contrast to its inhibitory effects on the expression of GATA-1, FOG-1, NF-E2 and FLI-1 during megakaryocyte development, anagrelide did not alter the expression of these factors during EPO-stimulated erythroid differentiation. These results are consistent with previous studies demonstrating that in patients who responded to anagrelide there was no decrease in the number of erythroid progenitors [43], and also that in culture the drug did not affect the development of erythroid cells from hematopoietic progenitors [5]. Nevertheless, long-term therapy with anagrelide has been reported to induce anemia in some patients [44,45]. Although the origin of this side effect is still unclear, our findings suggest that it is not attributable to inhibition of the erythroid gene expression program. Taken together, our findings suggest that the drug-sensitive step may lie upstream of GATA-1 and FOG-1 in a pathway that controls their expression and that is predominantly active in the context of megakaryocyte development. The present study also emphasizes the value of anagrelide as an experimental tool with which to dissect the mechanisms that control hematopoietic developmental decisions, and in particular the divergence between the erythroid and megakaryocytic lineages.

A recent study has shown that GATA-1 mRNA is overexpressed in ET and in polycythemia vera patients, independently of the presence of the JAK2 mutation [46]. Therefore, our findings demonstrating that anagrelide suppresses GATA-1 expression in hematopoietic cell cultures may warrant further studies to determine whether the drug also reduces GATA-1 levels in the setting of ET and, if so, whether this effect correlates with the reduction in platelet counts. Such studies could be important to determine whether monitoring of levels of GATA-1 in ET may be used as a tool in the clinical setting to determine treatment modalities for this condition.

In summary, our findings demonstrate that anagrelide lowers GATA-1 and FOG-1 expression levels via a mechanism that does not involve inhibition of MPL-mediated early signal transduction events, and that is dispensable for the expression of these genes during erythroid differentiation. The present work suggests that anagrelide suppresses megakaryocyte development and lowers platelet counts by interfering with the transcriptional control of the megakaryocyte gene expression program.

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

M. Ahluwalia designed and performed experiments and analyzed the data. H. Donovan: performed experiments and analyzed the data. N. Singh and L. Butcher performed experiments. J. D. Erusalimsky conceived and supervised the study, analyzed the data and wrote the article. All the authors interpreted the data and revised the manuscript.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

We thank R. Franklin and S. Damment for their valuable advice during the course of this work.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

J. D. Erusalimsky has received consulting fees from Shire Pharmaceuticals. This work was supported by a research grant from Shire Pharmaceuticals.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References