¹Department of Microbiology/Immunology, Indiana University School of Medicine, 46202 Indianapolis, USA

²Department of Medicine (Hematology/Oncology), Indiana University School of Medicine, 46202 Indianapolis, USA

³Walther Oncology Center, Indiana University School of Medicine, 46202 Indianapolis, USA

⁴Walther Cancer Institute, 42208, Indianapolis, Indiana, USA

^aAuthor for correspondence

^bCurrent address: Department of Vascular Biology, The Scripps Research Institute 10666 N. Torrey Pines Rd, La Jolla, CA 92037, USA

Thrombopoietin is a cytokine with potent megakaryocytopoietic and thrombopoietic activities in vivo. Wild-type p53 is a conformationally flexible, anti-oncogenic transcription factor that plays a principal role in mediating growth factor withdrawal-induced apoptosis in factor-dependent hematopoietic cells. We recently reported that Tpo induces a conformational change in and functional inactivation of p53, coincident with its anti-apoptotic effects, in the human factor-dependent cell line M07e. In an effort to identify potential signaling cascades through which Tpo illicits these effects on p53, we report here that treating M07e cells with MAPK kinase inhibitor PD98059 dramatically suppressed Tpo-induced conformational change in p53 as well as Tpo-enhanced viability in M07e cells in a p53-dependent manner. Furthermore, the expression of constitutively active Raf1 in M07e cells induced conformational change in p53 independent of Tpo stimulation. Inhibition of the JAK/STAT pathway revealed that JAK/STAT signaling plays an insignificant role in conformational modulation of p53 and apoptosis suppression. Inhibition of phosphatidylinositol-3 kinase did not have a significant effect on p53 conformation but did have a weak but significant effect on Tpo-enhanced viability. Cytokine-induced activation of the MAPK pathway and the subsequent functional neutralization of p53, may be an event by which apoptosis is commonly suppressed in hematopoiesis.

MAPK; p53 conformation; thrombopoietin; apoptosis; M07e

Introduction

Thrombopoietin (Tpo), also known as the c-Mpl ligand or megakaryocyte growth and development factor (MGDF), has diverse biological effects on hematopoietic cells at various stages of maturation within the myeloid compartment (Carver-Moore et al., 1996; Cohen-Solal et al., 1997). Tpo can synergize with other hematopoietic cytokines and stimulates the survival and proliferation of primitive, multilineage progenitor cells (Broudy et al., 1995; Ramsfjell et al., 1997; Ritchie et al., 1996). Tpo's unique contribution to hematopoiesis is manifest in its ability to exert a highly positive influence on expanding the megakaryocyte pool and facilitating the maturation of those megakaryocytes into platelet shedding cells (Kaushansky et al., 1995; Sitnicka et al., 1996). Therefore, Tpo is not entirely lineage-specific but is still considered the primary megakaryocytopoietic and thrombopoietic regulator in vivo.

Lacking any intrinsic kinase or phosphatase activity, the Tpo receptor, c-Mpl, is a member of the hematopoietic cytokine receptor superfamily (Skoda et al., 1993; Vigon et al., 1992). Regions of conserved sequences found in other members of this family, termed box 1 and box 2, are present on the intracellular portion of this receptor (Murakami et al., 1991). It has been convincingly demonstrated that these two domains are highly critical for c-Mpl-mediated proliferation induction in a wide range of cell systems, and that deletion or mutation of either of these two domains results in an inability to activate members of the Janus kinase family, namely JAK2 and Tyk2 (Drachman and Kaushansky, 1997; Gurney et al., 1995; Morita et al., 1996). STAT (signal transducers and activators of transcription) proteins are also potently activated through these two receptor domains (Drachman and Kaushansky, 1997; Gurney et al., 1995; Morita et al., 1996).

A region less critical to proliferation induction, at the C-terminus of the receptor, has been shown to transmit the maturational effects of Tpo into the cell and mediates the phosphorylation/activation of c-Mpl itself as well as adapter proteins and other signaling molecules such as Shc, Grb2, Vav, SOS, and SHIP (Alexander et al., 1996; Drachman and Kaushansky, 1997; Porteu et al., 1996; Sasaki et al., 1995). This portion of the protein is most probably responsible for activating the Ras/Raf-1/Mitogen activated protein kinase (MAPK) kinase/MAPK pathway as well, potentially downstream of Shc/Grb2/SOS complexes (Brizzi et al., 1996; Dorsch et al., 1997; Nagata and Todokoro, 1995; Rouyez et al., 1997; Yamada et al., 1995). The p44/p42 MAPK pathway is activated by a wide range of hematopoietic cytokines and results in the upregulation of protein synthesis and immediate early gene expression, particularly that of c-fos (Miyazawa et al., 1991; Welham et al., 1992). Furthermore, The MAPK pathway may play a potent and dominant role in mediating the anti-apoptotic effect of hematopoietic growth factors (Kinoshita et al., 1997; Xia et al., 1995). The current dogma is that c-Mpl contains separate structural elements, which possess unique and overlapping functions with regard to proliferation and differentiation.

In the absence of cell cycle progression, Tpo behaves as a survival factor in the human leukemic cell line M07e by suppressing apoptosis following growth factor withdrawal (Ritchie et al., 1996; 1997). The M07e cell line originated from a child with acute megakaryoblastic leukemia, and displays surface markers characteristic of both early myeloid progenitors (CD34) and more committed members of the platelet lineage such as CD41a (Avanzi et al., 1988). M07e cells are factor-dependent, requiring either granulocyte-macrophage colony-stimulating factor (GM-CSF) or interleukin-3 (IL-3) for growth (Avanzi et al., 1988; Hendrie et al., 1991) and have served as a useful system for studying signal transduction events following cytokine receptor antagonism. Since our subline of M07e does not proliferate in response to Tpo and is differentiation-defective (Ritchie et al., 1996), Tpo/M07e together constitute a model system within which we have been able to study the biological and biochemical aspects of survival factor signaling.

In the M07e cell line, we recently demonstrated that both Tpo and GM-CSF can effect a functional inactivation of the p53 tumor suppressor protein (Ritchie et al., 1997), a critical regulatory component of the apoptotic process following growth factor withdrawal in factor-dependent hematopoietic cells (Blandino et al., 1995; Yonish-Rouach et al., 1991; Zhu et al., 1994). In accordance with the conformational hypothesis put forward by Jo Milner (Milner, 1991), it has been demonstrated that following growth factor stimulation wild-type p53 can switch from a conformation associated with an anti-proliferative, pro-apoptotic phenotype to a conformation that has a retarded ability to arrest cell cycle progression and induce apoptosis (Milner and Watson, 1990; Ritchie et al., 1997; Zerrahn et al., 1992; Zhang and Deisseroth, 1994; Zhang et al., 1992). These two conformations are termed `suppressor' and `promoter', respectively, in regard to their phenotypic effect on cell growth (Milner, 1991). The intracellular mechanisms involved in conformational modulation of p53 are not clear, but accumulating evidence suggests that redox regulation of three critical cysteine residues and divalent metal cation changes may play the most physiologically-relevant roles (Hainaut and Milner, 1993a,b; Rainwater et al., 1995). If these cysteine residues, which coordinate a zinc ion in the core DNA-binding domain of p53, are either mutated or oxidized, p53 adopts the promoter conformation. However, no report to date has provided a satisfying insight into which signal transduction cascades might bridge cytokine stimulation and the molecular machinery directly conferring conformational change and functional inactivation of tumor-suppressing p53.

In this report, we demonstrate that Tpo-induced conformational changes and Tpo-enhanced survival lie downstream of the MAPK pathway in the M07e cell line, while JAK/STAT and phosphatidylinositol-3 kinase (PI-3K) signaling do not appear to play a significant role in mediating this post-translational effect on p53. This data supports the notion that MAPK and JAK/STAT pathways in Tpo signaling play unique roles with regard to survival and proliferation induction, respectively. It also provides support for the argument that activation of the MAPK pathway may be the shared event by which cytokines maintain hematopoietic cell viability in lieu of proliferation induction.

Results

M07e cells contain genotypically wild-type p53

To demonstrate that Tpo is inducing a conformational and functional modulation of not only phenotypically wild-type p53 (Ritchie et al., 1997), but also genotypically wild-type p53 in M07e cells, M07e cell genomic DNA was isolated and exons 4 - 9, which include the conserved, or hot spot regions, were bi-directionally sequenced from four PCR products. The determined sequence (not shown) lacks nucleotide substitution, addition, or deletion when compared with the frequently referenced sequence reported by Harlow et al. (1985).

Time course of Tpo-induced upregulation of p53 in the promoter conformation

We initially determined the earliest time point at which Tpo-induced upregulation of p53 could be measured in the functionally inactivated, promoter conformation (Pab 240⁺) at or near maximal levels. After factor starving M07e cells for 18 h, 1´10⁶ viable cells/ml were stimulated with 50 ng/ml Tpo in serum free media for up to 24 h, with 1 ml samples taken at the times indicated in Figure 1. Samples collected at 0, 1, 2, 4 and 8 h were washed and fixed in a buffered paraformaldehyde solution, then kept at 4°C overnight and prepared for intracellular staining of p53 with the 24 h samples as described in Materials and methods. The flow cytometric data shown in Figure 1a reveals that Tpo-induced upregulation of Pab 240⁺ p53 is not clearly detectable for at least 2 h post-stimulation. However, at 8 h the level of Pab 240⁺ p53 has reached a maximum level. It was shown previously that the protein levels of promoter conformation p53 do not noticeably change between 24 and 48 h post-stimulation (Ritchie et al., 1997). This data is compared with data generated when p53 is stained with DO-1, a monoclonal antibody that recognizes all conformations of p53. The % change in mean channel fluorescence (MCF) between the murine IgG1a isotype control and the anti-p53 monoclonal is shown. The data shown in Figure 1b demonstrates that the ratio of promoter conformation p53 to total cellular p53 becomes increasingly smaller during Tpo stimulation, so that by 8 h post-stimulation virtually all of the total cellular p53 is in the promoter conformation. This data also suggests that Pab 240⁺ p53 levels can be measured as early as 8 h if so desired. It further suggests that cytokine-induced conformational change in p53 may or may not initiate immediately following receptor antagonism, but that detection of Pab 240⁺ p53 is very weak within the first 2 h following c-Mpl activation.

Tpo induces the activation of p44/p42 MAPK in M07e cells, which is suppressed with the inhibitor PD98059

Viability factors Tpo and GM-CSF induce a conformational change in and functional inactivation of p53 in M07e cells (Ritchie et al., 1997). This is highly relevant because p53 is the molecular mediator of growth factor withdrawal-induced apoptosis, which these cytokines effectively suppress. Activation of the p44/p42 MAPK pathway has been shown to play a role in apoptosis suppression in the hematopoietic system (Kinoshita et al., 1997; Xia et al., 1995). Since Tpo and GM-CSF have been shown to activate the MAPK pathway (Miyazawa et al., 1991; Rouyez et al., 1997), it was logical to hypothesize that this pathway may play a role in p53 conformational modulation. In order to test the involvement of MAPK activity in Tpo-mediated p53 conformational change, we positively confirmed that MAPK was activated in M07e cells following Tpo stimulation and that this activation could be suppressed with the MAPK pathway inhibitor PD98059.

PD98059, with an EC₅₀ of 5 M, is a selective inhibitor of MAPK Kinase (MEK-1), shown previously to markedly suppress MAPK activity in a variety of cell systems (Dudley et al., 1995; Pang et al., 1995). We incubated factor starved M07e cells with either control media or media containing 10 M PD98059 for 1 h, followed by a 15 min incubation in either control or Tpo-containing media. We assessed MAPK activation by first immunoprecipitating the dual (Thr 202/Tyr 204) phosphorylated species of MAPK from cell lysates. Phosphorylation of MAPK on Thr 202 and Tyr 204 has been shown to be the critical event conferring enzymatic activity to MAPK (Payne et al., 1991; Sturgill et al., 1988). The dual phosphorylated MAPK immunoprecipitates were washed with lysis buffer and kinase buffer and then incubated with the well-characterized MAPK target Elk-1 with non-limiting amounts of ATP. Only activated MAPK will be able to phosphorylate Elk-1 (at serine 383) in this kinase reaction assay. The reaction was terminated by the addition of 3´SDS sample buffer and separated by SDS - PAGE. A Western blot was then performed using an antibody specific for phospho-Elk-1. The Western blot in Figure 2a shows that Tpo can dramatically induce the phosphorylation of Elk-1 and hence activation of MAPK. It also shows that PD98059 can suppress this Tpo-mediated activation of MAPK. Figure 2b shows that equal amounts of Elk-1 were included in the kinase reactions and gel loading after the membrane shown in Figure 2a was stripped and re-probed with a second, non-phospho-specific (pan) -Elk-1 antibody. Half of the whole cell lysate from which MAPK was immunoprecipitated was separated by SDS - PAGE and probed with a pan-MAPK antibody to demonstrate that the lysates contained comparable amounts of MAPK (Figure 2c).

MAPK pathway plays a potent role in mediating Tpo-induced conformational changes in p53 and Tpo-enhanced viability of M07e cells

Factor starved M07e cells were incubated at a density of 1´10⁶ viable cells/ml for 1 h in the presence or absence of 10 M PD98059. Following this, the cells were incubated for 24 h in serum-free media containing or lacking Tpo and analysed by flow cytometry for the buildup of Pab 240⁺ p53. Figure 3a shows that only Tpo-stimulated cells possess Pab 240⁺ p53 when compared with time 0 or control media cultures. PD98059 suppressed the ability of Tpo to induce a conformational shift in p53 by up to 80% (P=0.021). This suggests that the MAPK pathway can play a potent role in p53 conformational change. It was shown previously that Tpo-induced conformational change in p53 coincides with nuclear exclusion and retarded DNA binding ability of p53 (Ritchie et al., 1997). Therefore, it can also be suggested that the MAPK pathway can mediate Tpo-enhanced survival by suppressing the pro-apoptotic phenotype of p53. To more directly test the role of MAPK in the Tpo-enhanced survival of M07e cells, we measured apoptosis induction in M07e cells that were incubated for 1 h in the presence or absence of 10 M PD98059 and then incubated in control media or media containing Tpo. Figure 3b reveals that Tpo was less able to maintain the viability of M07e cells pretreated with PD98059 when compared to cells pretreated with control media (P=0.033). This supports the argument that the MAPK pathway has a critical role in Tpo viability signaling and may confer that viability by functionally inactivating p53 through conformational modulation. We measured apoptosis with the Merocyanin 540 dye (MC540) which fluoresces in the red, or FL-2, channel (Reid et al., 1996).

To provide further evidence that the MAPK pathway is involved in the modulation of p53 conformation as well as viability maintenance, a retroviral vector (pBabe) containing a constitutively active Raf1 (pBp-Raf1-CX, Diaz et al., 1997) or the empty vector were either packaged and used to transduce M07e cells or used directly to transfect M07e cells. Transduction of M07e cells with the pBp-Raf1-CX construct resulted in a significant upregulation (P=<0.01) of p53 in the promoter conformation in the absence of Tpo stimulation when compared with cells transduced with virus containing the empty vector (Figure 3c). In addition, M07e cells transduced with the pBp-Raf1-CX and grown in the absence of Tpo for 24 h had a significantly lower apoptotic population when compared with the control vector cultures as shown in Figure 3d (P=0.015). The M07e cells that were transiently tranfected with pBp-Raf1-CX also contained a significant level (P=<0.01) of p53 in the promoter conformation in the absence of Tpo stimulation when compared with M07e cells transfected with the empty vector (Figure 3e). M07e cells transfected with pBp-Raf1-CX displayed a lower apoptotic rate in the absence of Tpo stimulation when compared with those transfected with the empty vector as shown in Figure 3f (P=0.038). A MAP kinase assay was performed to measure levels of MAPK activity in both the transduced and transfected M07e cells. Figure 4 demonstrates that M07e cells transduced with the pBp-Raf1-CX construct possessed a high level of activated MAPK in the absence of Tpo stimulation. M07e cells transfected with the pB-Raf1-CX construct also possessed elevated levels of MAPK activity, albeit lower than the transduced cells or cells stimulated with Tpo. We confirmed equal amounts of MAPK protein in the cell lysates (data not shown).

Using MC540, which detects phosphatidylserine buildup on the outer leaflet of the plasma membrane (a physiologically relevant event in the early stages of apoptosis), we were able to determine the percentage of cells undergoing apoptosis at the given time points by flow cytometry (Reid et al., 1996). As shown in the left side of Figure 5a, M07e cells factor starved for 36 h that are almost 50% non-viable as measured by trypan blue exclusion (Ritchie et al., 1996), contain an MC540 dim and an MC540 bright population (Figure 5b). GM-CSF-stimulated cells that were in log-phase growth at 36 h, were predominantly MC540 dim. By sorting the MC540 bright cells from the MC540 dim cells in the factor-starved cultures through fluorescence activated cell sorting (FACS), it can be seen that MC540 dim cells possess `normal' forward (cell size, FSC) vs side (cell density, SSC) scatter profile (Figure 5c) similar to factor-stimulated cultures (Figure 5a) and a healthy-looking, blastic morphology (Figure 5d). In comparison, MC540 bright cells represent a smaller, denser population of cells by forward vs. side scatter profile and have typical apoptotic morphology. Furthermore, sorting on MC540 dim vs bright cells previously revealed that the MC540 bright cells possessed a subdiploid content of DNA that was fragmented (a hallmark of apoptosis) using the Hoechst 33342 dye and conventional electrophoresis (Reid et al., 1996). That study also revealed that the MC540 dim cells were diploid and did not contain fragmented DNA. In our analyses, a marker was drawn in between the MC540 bright and dim populations (as shown in Figure 5b as M1) and the percentage of cells residing within that MC540 bright population was determined.

It was possible that PD98059 was inducing apoptosis in a p53-independent manner resulting in an indirect retardation of Tpo-mediated viability enhancement. To resolve this issue, the protein levels of p53 were negated in M07e cells with antisense oligonucleotides, that have been successfully used in previous studies to markedly suppress p53 protein expression (Bi et al., 1993; Ritchie et al., 1997). As shown in Figure 6a, M07e cells treated with the antisense oligonucleotides were markedly resistant to PD98059-induced apoptosis when compared to the scrambled sense oligos (P=0.008). This is convincing evidence that the PD98059-induced death in the presence of survival factor stimulation is predominantly a p53-dependent process in M07e cells. Furthermore, it strengthens the argument that p53 is a downstream target of MAPK-mediated suppression of apoptosis. Figure 6b shows that the p53-specific antisense oligonucleotides were effective in suppressing p53 protein levels during these assays when detected with conformation-independent p53 antibody DO-1. Figure 6c shows that the cell lysates contained comparable amounts of protein as measured by glyceraldehyde-3 phosphate dehydrogenase protein levels.

JAK/STAT signaling plays an insignificant role in mediating Tpo-induced conformational changes in p53 and Tpo-enhanced viability of M07e cells

It has been shown repeatedly that JAK2, and to a lesser degree Tyk2, are the two members of the JAK family activated by Tpo stimulation (Drachman and Kaushansky, 1997; Gurney et al., 1995; Morita et al., 1996). We wanted to determine if the JAK2 tyrosine kinase is activated in our subline of M07e cells following Tpo stimulation. It has been previously demonstrated that JAK2, but not Tyk2, is activated by Tpo in M07e cells, but the subline of M07e cells used in that study respond proliferatively to this cytokine (Tortolani et al., 1995), while our M07e subline does not proliferate well in response to Tpo (Ritchie et al., 1996). It was therefore possible that Tpo was not activating JAK2 in our subline. 1´10⁶ factor starved and viable M07e cells were incubated for 30 min in either control media or media containing Tpo. Following this, JAK2 was immunoprecipitated from the cell lysates. After transfer to PVDF membranes, JAK2 was probed with anti-phosphotyrosine antibodies to measure the degree of JAK2 phosphorylation, which is in turn an indirect measure of activation (Stahl et al., 1994). Figure 7a shows that JAK2 is markedly tyrosine-phosphorylated in M07e cells following Tpo-stimulation. Equal JAK2 protein content in the immunoprecipitates was demonstrated with -JAK2 antibody. To substantiate the use of Tyrphostin AG490, which blocks JAK2 kinase activity with relative selectivity (Meydan et al., 1996), we measured the level of phosphorylated STAT5, a direct downstream target of JAK2, by immunoprecipitating STAT5 from lysates of Tpo-stimulated M07e cells preincubated with, or without, 0.3 M of the AG490 compound and subsequently determined the tyrosine phosphorylation status of STAT5 with a 1 : 1 mixture of -phosphotyrosine clones 4G10 and PY99. It has been previously demonstrated that tyrosine phosphorylation of STAT5 in M07e cells coincided with its ability to form DNA bound complexes (Bacon et al., 1995). Figure 7b shows that AG490 treated cells have a marked decrease in phosphorylated STAT5 protein levels, suggestive of suppressed activation of JAK2. Equal STAT5 protein content in the immunoprecipitates was demonstrated with -STAT5 antibody.

We wanted to test the possibility that JAK signaling in M07e cells may potentially play a role in p53 conformational change and viability in M07e cells. With regard to p53 conformational change, AG490 had very little effect as compared to control Tpo-treated cells (P=0.300) when these cells were analysed by flow cytometry (Figure 8a). In addition, AG490 did not have a notable effect on Tpo-enhanced viability in M07e cells (Figure 8b) when stained with the MC540 dye. This set of data indicates that JAK signaling may not play a significant role in Tpo-induced conformational changes in p53, nor in mediating apoptosis suppression directly.

One report provided evidence that, in contrast to STAT5, STAT3 phosphorylation may not completely parallel JAK2 phosphorylation (Drachman and Kaushansky, 1997). The box 1 and 2 domains, which mediate JAK2, STAT5, and STAT3 phosphorylation, were shown to be critical for proliferation. Two tyrosines in the 10 C-terminal residues of c-Mpl can mediate STAT3 phosphorylation as well but play a less vital role in mitogenesis. One of these two latter tyrosines is also critical for Shc phosphorylation, which is believed to couple c-Mpl with the Ras/Raf-1/MAPK pathway. In addition, STAT3 can be activated at Ser 727 by MAPK (Chung et al., 1997), which may account for STAT3 being activated through the C-terminal portion of c-Mpl. It has also been demonstrated that STAT3 may play a role in suppressing apoptosis (Fukada et al., 1996).

We therefore wanted to test the individual role of STAT3 in p53 conformational changes and viability. Using an antisense oligonucleotide that was directed against the mRNA of human STAT3 using the same codon targeting strategy as the p53 antisense oligonucleotides, we were able to diminish the protein levels of STAT3 by over 50%, when compared with scrambled sense oligonucleotide treatment, and roughly 65% when compared to media control. 1´10⁶ factor starved and viable M07e cells were incubated for 48 h in serum free media alone or containing 50 ng/ml Tpo each with or without 10 M of STAT3-specific oligonucleotide sequences. At 48 h, the cells were collected, lysed and the protein content between samples adjusted equally. Western blotting of PVDF membranes (probed with an -STAT3 antibody) is shown in Figure 9. Figure 10a and b show that neither the antisense or scrambled sense oligonucleotides have a significant effect on Tpo-induced conformational changes in p53 (P=0.507 for AS, P=0.251 for SS) or on Tpo-enhanced survival (P=0.776 for AS, P=0.695 for SS) of M07e cells when analysed by flow cytometry at 24 and 48 h, respectively. To allow sufficient time for STAT3 protein expression diminishment, we supplemented the media during an initial 24 h incubation period with Tpo to maintain cellular integrity. The STAT3 antisense oligonucleotides were effective biologically, as they markedly suppressed Tpo and GM-CSF induced synergistic proliferation as measured by tritiated thymidine incorporation when compared to the scrambled sense oligos (data not shown). These data suggests that, although the two C-terminal tyrosine residues of c-Mpl may confer the anti-apoptotic effects of Tpo, and are potentially involved in STAT3 activation, STAT3 alone does not appear to play a substantial role in Tpo-enhanced suppression of p53-mediated growth factor withdrawal-induced apoptosis. It also suggests that complete suppression of STAT3 protein levels is not required to abrogate synergistic proliferation induction. The fact that STAT3 does appear to be involved in at least synergistic proliferation in our subline of M07e cells further strengthens the argument that Tpo induces proliferation and differentiation/survival through two separate mechanisms.

Phosphatidylinositol-3-phosphate Kinase (PI-3K) plays an insignificant role in Tpo-induced conformational changes in p53

Some reports have demonstrated the ability of Tpo to activate PI-3K, which has been shown to suppress apoptosis in some cell systems (Párrizas et al., 1997; Sattler et al., 1997; Zauli et al., 1997). We therefore tested the involvement of PI-3K in p53 conformational change and Tpo-enhanced viability in M07e cells using wortmannin, a well-characterized inhibitor of PI-3K activity that we have previously shown to be biologically effective at a significant level in this cell line (Gotoh et al., 1997; Takahira et al., 1997). Factor starved M07e cells were pretreated with either control media or media containing 0.5 mM wortmannin for 1 h. Figure 11a shows that wortmannin did not have a significant effect on Tpo-induced conformational change in p53 (P=0.664) when used alone. Nor did wortmannin augment or enhance PD98059's effect on p53 conformational changes (data not shown). However, wortmannin did weakly but significantly block the ability of Tpo to suppress apoptosis (P=0.019) as shown in Figure 11b.

The data presented herein demonstrates that MAPK has the principal ability to mediate Tpo-induced conformational changes in p53. It also demonstrates that MAPK is the major apoptosis-suppressing subsystem within the cell following Tpo stimulation, with PI-3K playing a minor and potentially contributory role in Tpo's anti-death activities.

Discussion

Cells of the hematopoietic system die by apoptosis when deprived of growth or viability factors (Williams et al., 1990). It has been convincingly shown that this death process can be moderated by p53 (Blandino et al., 1995; Yonish-Rouach et al., 1991; Zhu et al., 1994). The over-expression of wild-type p53 in p53-deficient leukemic cells induces cell death that is suppressible by hematopoietic cytokines such as SLF, IL-3, IL-6, and erythropoietin (Epo), while the introduction of dominant negative mutant p53 retards the apoptotic process following growth factor withdrawal (Abrahamson et al., 1995; Blandino et al., 1995; Yonish-Rouach et al., 1991). It follows that hematopoietic cytokines must somehow counteract p53 activity by either functionally disabling p53, by suppressing downstream targets of p53, or suppressing p53 protein expression.

Conformational modulation appears to be one way through which p53 is functional inactivated during cytokine stimulation, for when p53 adopts the promoter conformation it is sequestered from the nucleus and, more importantly, is greatly weakened in its ability to bind DNA (Hainaut and Milner, 1993a; Ritchie et al., 1997; Zerrahn et al., 1992). These events can be translated into an inability of p53 to upregulate the expression of pro-apoptotic bax and downregulate anti-apoptotic bcl-2. Although a considerable amount of work is being done in an effort to understand the molecular mechanics of this conformational change, details concerning the signaling pathways employed during cytokine signaling which interface with the p53-modifying machinery have yet to be reported. Of clinical interest, AML samples most commonly contain wild-type p53, but in the promoter conformation as a function of growth stimulation (Zhang et al., 1992). In vivo, it is conceivable that cytokines may contribute to leukemogenesis by functionally inactivating p53, via MAPK pathway-mediated conformational modulation, in a hematopoietic cell that possesses an activated proto-oncogene such as dysregulated c-Myc or dysregulated p21^Ras capable of co-transformation with `mutant' p53. Identification of signal transduction cascades and molecular machinery that lead to this altered state of p53 could narrow efforts towards finding chemo-therapeutic targets which would restore p53 activity and enhance chemo- and radio-sensitivity in these cells.

We report here that a signal transduction pathway ubiquitously activated during hematopoietic cytokine stimulation, the MAPK pathway, is largely responsible for effecting the conformational modulation in p53. This effect was relatively free of influence from JAK/STAT and PI-3K activities. The fact that JAK2 as well as STAT5 appear to be activated normally following Tpo stimulation in our subline of M07e cells lessens the possibility that a dysfunctional JAK/STAT pathway exists. This suggests that the MAPK pathway may be at least partially involved in uncoupling the proliferative and survival-inducing effects of Tpo. However, the MAPK pathway appears to also be involved in megakaryocytic differentiation (Dorsch et al., 1997; Melemed et al., 1997). We could not test this in M07e cells since they do not appear to be capable of differentiating.

Hypothetically, differentiation and p53-suppressive signals transmitted through the same pathway would prove beneficial to a developing megakaryocyte. The endomitotic process that ensues during megakaryocytopoiesis would present itself as genetic amplification to p53, one of the events which p53 specifically prevents. Therefore, using a common signaling pathway would ensure that if the signal to become polyploid were initiated in a megakaryocyte, a signal would also be simultaneously generated which suppresses cellular sentinels that detect polyploidization and prohibit DNA synthesis, such as p53.

The next logical question is; where does the MAPK pathway actually interface with the conformational machinery? Is it necessarily a phosphorylation event? Some evidence suggests that the upregulation of p53 in the promoter conformation is independent of changes in phosphorylation (Milne et al., 1994). However, murine p53 is a substrate for MAPK activity in vitro (Picksley et al., 1992). We have not tested for changes in p53 phosphorylation status during cytokine stimulation in M07e cells. Genetic analysis of the MAPK cascade will certainly prove useful in elucidating what pathway components are involved in this process. In addition, any relationship the MAPK pathway may have with intracellular redox subsystems that potentially regulate the redox state of p53, namely glutathione and thioredoxin, would warrant investigation. Our observation that PI-3K is somewhat involved in suppressing apoptosis, but not in the conformational modulation of p53, opens up the possibilities that conformational modulation of p53 is not the only means of suppressing p53-mediated apoptosis through Tpo signaling, that Tpo may suppress apoptosis in a p53-dependent and p53-independent manner and that both of these last two scenarios are in effect. For example, Akt, which can be activated by cytokine signaling through PI-3K, can induce the phosphorylation and inactivation of Bad, a pro-apoptotic member of the Bcl-2 family that can functionally substitute for Bax (Franke et al., 1995; Yang et al., 1995). Although the ability of p53 to transactivate bad has yet to be conclusively demonstrated, p53 is a potent modulator of other Bcl-2 family member protein levels and may also contribute to raising Bad protein levels (Miyashita and Reed, 1995). The role MAPK plays with regard to apoptosis suppression certainly appears to be dominant to the role PI-3K plays in this cytokine/cell system. PI-3K may therefore supplement the anti-apoptotic activities of MAPK while serving another more vital function such as adhesion or proliferation induction (Dorsch et al., 1997; Takahira et al., 1997; Zauli et al., 1997).

Materials and methods

Cytokines, cells and reagents

Recombinant human (rhu) GM-CSF was a kind gift of the Immunex Corp. and rhuTpo was purchased from R&D Systems. Concentrations used in these assays were 100 U/ml for GM - CSF and 50 ng/ml for Tpo. The human growth factor-dependent subline, M07e, was obtained from Genetics Institute (Boston, MA). The M07e subline, and its parental line M07, have been biologically characterized previously (Avanzi et al., 1988; Hendrie et al., 1991; Miyazawa et al., 1991). M07e cells were maintained in RPMI 1640 (Bio Whittaker) with 20% heat inactivated fetal bovine serum (FBS) with 100 U/ml GM - CSF. Factor starvation was accomplished by washing day 3 or day 4 cells twice with RPMI 1640 and incubating them for 18 h in RPMI 1640 with 0.1% bovine serum albumin (BSA) to synchronize the cells in G₀/G₁ of the cell cycle and remove residual GM - CSF and serum-induced signals. Serum-free media consisted of RPMI 1640 with 2% dilipidated, deionized and dialyzed BSA, 10 mg/ml insulin, 200 mg/ml saturated human transferrin, 50 mM beta-mercaptoethanol, 40 mg/ml LDL, all from Sigma Chemical Co. All other reagents were purchased from Sigma except the following: Merocyanin (MC 540) was purchased from Molecular Probes. The MEK-1 inhibitor PD98059 and the MAPK assay kit were purchased from New England Biolabs and JAK2 inhibitor AG490 was purchased from Calbiochem.

Flow cytometric analysis

For flow cytometric detection of intracellular protein levels, 1´10⁶ cells were collected at the indicated time points per condition and washed in PBS with 0.1% BSA. The samples were then simultaneously permeabilized and fixed in 500 l of a 4% paraformaldehyde, 0.1% saponin, 10 mM HEPES solution for 15 min at 4°C. Cells were then washed twice with PBS with 0.1% BSA and incubated with 1 g/ml of the primary antibody or appropriate isotype control for 20 min at 4°C. After washing twice, the samples were incubated with the appropriate secondary FITC- or Tri Color (TC)-conjugated antibody for 20 min. Stained cells were then washed and resuspended in 400 l of a 4% paraformaldehyde 10 mM HEPES solution for flow cytometric analysis. Ten thousand events were collected per sample. For staining M07e cells with MC540, 20 (L of working stock (10 g/ml) is added to 1 - 10´10⁶ cells in 500 l of PBS. This staining solution/cell suspension is incubated at room temperature for 15 min, and washed once with PBS. MC540 is analysed in the FL2 channel on a logarithmic scale. The amplitude gain of the FL2 channel was adjusted so that the MC540 dim peak had a mean channel fluorescence of ~10. A marker was set between the MC540 dim and MC540 bright populations and the cells that resided within the marker were considered positive (Reid et al., 1996). For the sorting experiment, MC540 dim from MC540 bright cells were separated on a Becton Dickinson FACS sorter based upon FL2 intensity. 1´10⁶ cells were collected from each of these two populations into PBS. Morphological examination of these cells was carried out by gently centrifuging (500 g for 4 min) M07e cells onto microscope slides in a cytospin apparatus (Shanndon). Slides were stained with a Hema-tek machine using Wright-Giemsa. Pictures were taken with a scope-integrated camera at 1000´ magnification under oil immersion.

Antibodies, Western blotting and immunoprecipitations

Antibodies used in these studies include; -p53 (Pab 240), an IgG₁ murine monoclonal that recognizes the promoter conformation of p53, and pan -MAPK were purchased from Oncogene Science. -dual phosphorylated-MAPK, a murine monoclonal and -phospho-Elk-1, a rabbit IgG polyclonal, were both purchased from New England Biolabs. -p53 (DO-1), an IgG murine monoclonal which recognizes all forms of p53, -JAK2, an IgG rabbit polyclonal, -STAT3, an IgG rabbit polyclonal, -STAT5, an IgG rabbit polyclonal, -Elk-1, an IgG rabbit polyclonal, and -phosphotyrosine (PY99), a murine IgG monoclonal, were all purchased from Santa Cruz Biotechnology, Inc. -glyceraldehyde-3 phosphate dehydrogenase (GAPDH) was purchased from Biodesign International. Anti-phosphotyrosine (anti-p-Tyr) 4G10 murine monoclonal was purchased from Upstate Biotechnology. All secondary, HRP-labeled antibodies were purchased from Amersham Life Science. All secondary, fluorochrome-conjugated antibodies as well as flow cytometric isotype controls were purchased from Caltag. Cells were lysed in ice cold 50 mM Tris, 150 mM NaCl, 5 mM EDTA, 1% NP-40, with common inhibitor cocktail added. Lysates analysed by Western blot were adjusted for total protein content and equal protein amounts were loaded per well. Samples were separated by SDS - PAGE and transferred to Immobilon-P membranes (Millipore), blocked with 2% BSA, probed and developed with ECL Western blot detection kit (Amersham Life Science). Immunoprecipitations were performed as previously described (Tauchi et al., 1994). Briefly, cell lysates were incubated with appropriate antibody for 4 h at 4°C on a rotary drum. Immunoprecipitates were collected by protein A/G agarose beads (Santa Cruz Biotechnology) and were gently washed five times before being resuspended in SDS - PAGE loading buffer.

Antisense oligonucleotides

The 18-mer phosphorothiolated p53- and STAT3-specific oligonucleotides were synthesized by an automated process (Biotechnology Facility, Indiana University). The sequences of the p53-specific oligonucleotides were; 5'-CGG CTC CTC CAT GGC AGT-3' (antisense); 5'-CAA GGA GTA GGC ACC GGT-3' (scrambled sense). The target region of p53 mRNA, the affinity for the target region, and the translation inhibition efficiency have been described previously (Bi et al., 1993). The oligonucleotides were resuspended at a final concentration of 5 M in serum-free assay medium. STAT3-specific oligonucleotides were; 5'-CCA TTG GGC CAT CCT GTT-3' (antisense); 5'-CAA GGA GTA GGC ACC GGT-3' (scrambled sense). The STAT3 antisense oligonucleotides target the -1 and -2 and first four codons of the human STAT3 mRNA. Cells were incubated with oligonucleotides during a 24 h factor starvation or Tpo-stimulation period and up to 72 h post-stimulation, with addition of fresh oligos every 24 h.

MAPK activity assay

The MAPK activity assay kit was purchased from New England Biolabs. 2´10⁶ cells per well of a six well plate were collected by centrifugation and washed with ice cold PBS. Cells were lysed on ice for 20 min and the supernatant collected by microcentrifugation for 10 min at 18 000 g at 4°C. To 200 l of lysate was added 2 l of -phospho-MAPK and this was incubated overnight on a rotary drum at 4°C. Fifty l of 50% protein A/G-agarose was added to this mixture and incubated on a rotary drum for 3 h at 4°C. The agarose pellet was collected, washed twice with lysis buffer and twice with kinase buffer (both supplied by the New England Biolabs). The agarose pellet was resuspended in 50 l kinase buffer with 1 l of 10 mM ATP and 1 l (2 g/L) of Elk-1 fusion protein. These kinase cocktails were incubated at 33°C for 30 min and terminated with 25 l of SDS - PAGE sample buffer. Twenty l of this was separated on a 10% SDS - PAGE. The membrane was probed with -phospho-Elk-1. This assay detects Elk-1 that has been phosphorylated at Ser 383, the site of phosphorylation by MAPK.

Genomic DNA isolation and sequencing

Genomic DNA was isolated with the PureGene kit (Gentra Systems, Minneapolis, MN, USA) and the sequence of exons 4 - 9 of p53 was determined in a bi-directional and overlapping fashion by Cybergene Corporation (Sweden) using proprietary primers. The determined genomic sequence of M07e cell p53 was compared to that reported by Ed Harlow and colleagues (Harlow et al., 1985).

Transfections, transductions and plasmids

Dr Mark Marshall generously provided the plasmid pBp-Raf1-CX (Diaz et al., 1997) based on the retroviral vector pBabepuro (Morgenstern and Land, 1990). pBp-Raf1-CX contains the Ki Ras 17 amino acid membrane localization signal. Isoprenylation of this sequence leads to plasma membrane association and constitutive activation of Raf1 kinase. In pBp-Raf1-CX, the activated Raf1-CX is transcriptionally regulated by the Molony Murine Leukemia Virus LTR and puromycin resistance gene is transcriptionally regulated by the SV40 early promoter. M07e cells (1´10⁶ cells/0.5 ml DMEM) were transfected with the 10 g of the expression vectors by electroporation (0.22 v, 960 F) in 4 mm cuvettes (Midwest Scientific) with Gene Pulser Transfection Apparatus (BioRad). The pBabepuro and pBp-Raf1-CX retroviral vectors were shuttle packaged through BOSC23 ecotropic packaging cells (Pear et al., 1993) into PA317 amphotropic packaging cells (Miller and Buttimore, 1986). M07e cells were transduced with supernatants collected from PA317 cells. Transfected and transduced cells were cultures and stained for p53 confirmation and apoptosis as previously described.

Statistical analysis

Samples were compared for statistically significant differences using the Student's t-test to derive a P value. All error bars represent standard deviation (s.d.).

Acknowledgements

These studies were supported by Public Health Service grants RO1 HL 56416, RO1 HL 54037, RO1 DK 53674 and by a project in PO1 HL 53586 from the National Institutes of Health to HEB. AR is supported by NIH training program T32 DK 07519 to HEB. The authors wish to thank Rebecca Miller, Linda Chung, and Cynthia Booth for their excellent secretarial and administrative support.

Abrahamson JL, Lee JM and Bernstein A. (1995). Mol. Cell. Biol. 15, 6953-6960. MEDLINE

Alexander WS, Maurer AB, Novak U and Harrison-Smith M. (1996). EMBO J 15, 6531-6540. MEDLINE

Avanzi GC, Lista P, Giovinazzo B, Miniero R, Saglio G, Benetton G, Coda R, Cattoretti G and Pegoraro L. (1988). Br. J. Haematol. 69, 359-366. MEDLINE

Bacon CM, Tortolani PJ, Shimosaka A, Rees RC, Longo DL and O'Shea JJ. (1995). FEBS Lett. 370, 63-68. MEDLINE

Bi S, Lanza F and Goldman JM. (1993). Leukemia 7, 1840-1845. MEDLINE

Blandino G, Scardigli R, Rizzo MG, Crescenzi M, Soddu S and Sacchi A. (1995). Oncogene 10, 731-737. MEDLINE

Brizzi MF, Dentelli P, Lanfrancone L, Rosso A, Pelicci PG and Pegoraro L. (1996). Oncogene 13, 2067-2076. MEDLINE

Broudy VC, Lin NL and Kaushansky K. (1995). Blood 85, 1719-1726. MEDLINE

Carver-Moore K, Broxmeyer HE, Luoh SM, Cooper S, Peng J, Burstein SA, Moore MW and de Sauvage FJ. (1996). Blood 88, 803-808. MEDLINE

Chung J, Ucida E, Grammer TC and Blenir J. (1997). Mol. Cell. Biol. 17, 6508-6516. MEDLINE

Cohen-Solal K, Debili N, Vainchenker W and Wendling F. (1997). Thromb. Haemost. 78, 37-41. MEDLINE

Diaz B, Barnard D, Filson A, MacDonald S, King A and Marshall M. (1997). Mol. Cell. Biol. 17, 4509-4516. MEDLINE

Dorsch M, Fan PD, Danial NN, Rothman PB and Goff S. (1997). J. Exp. Med. 186, 1947-1955. MEDLINE

Drachman JG and Kaushansky K. (1997). Proc. Natl. Acad. Sci. USA 94, 2350-2355. Article MEDLINE

Dudley DT, Pang L, Decker SJ, Bridges AJ and Saltiel AR. (1995). Proc. Natl. Acad. Sci. USA 92, 7686-7689. MEDLINE

Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, Morrison DK, Kaplan DR and Tsichlis PN. (1995). Cell 81, 727-736. MEDLINE

Fukada T, Hibi M, Yamanaka Y, Takahashi-Tezuka M, Fujitani Y, Yamaguchi T, Nakajima K and Hirano T. (1996). Immunity 5, 449-460. MEDLINE

Gotoh A, Ritchie A, Takahira H and Broxmeyer HE. (1997). Ann. Hematol. 75, 207-213. MEDLINE

Gurney AL, Wong SC, Henzel WJ and de Sauvage FJ. (1995). Proc. Natl. Acad. Sci. USA 92, 5292-5296. MEDLINE

Hainaut P and Milner J. (1993a). Cancer Res. 53, 4469-4473. MEDLINE

Hainaut P and Milner J. (1993b). Cancer Res. 53, 1739-1742. MEDLINE

Harlow E, Williamson NA, Ralston R, Helfman DM and Adams TE. (1985). Mol. Cell. Biol. 5, 1601-1610. MEDLINE

Hendrie PC, Miyazawa K, Yang YC, Langefeld CD and Broxmeyer HE. (1991). Exp. Hematol. 19, 1031-1037. MEDLINE

Kaushansky K, Broudy VC, Lin N, Jorgensen MJ, McCarty J, Fox N, Zucker-Franklin D and Lofton-Day C. (1995). Proc. Natl. Acad. Sci. USA 92, 3234-3238. MEDLINE

Kinoshita T, Shirouzu M, Kamiya A, Hashimoto K, Yokoyama S and Miyajima A. (1997). Oncogene 15, 619-627. MEDLINE

Melemed AS, Ryder JW and Vik TA. (1997). Blood 90, 3462-3470. MEDLINE

Meydan N, Grunberger T, Dadi H, Shahar M, Arpaia E, Lapidot Z, Leeder JS, Freedman M, Cohen A, Gazit A, Levitzki A, Roifman and Cm. (1996). Nature 379, 645-648. MEDLINE

Miller AD and Buttimore C. (1986). Mol. Cell. Biol. 6, 2895-2902. MEDLINE

Milne DM, Campbell DG, Caudwell FB and Meek DW. (1994). J. Biol. Chem. 269, 9253-9260. MEDLINE

Milner J. (1991). Proc. R Soc. Lond. B Biol. Sci. 245, 139-145. MEDLINE

Milner J and Watson JV. (1990). Oncogene 5, 1683-1690. MEDLINE

Miyashita T and Reed JC. (1995). Cell 80, 293-299. MEDLINE

Miyazawa K, Hendrie PC, Mantel C, Wood K, Ashman LK and Broxmeyer HE. (1991). Exp. Hematol. 19, 1110-1123. MEDLINE

Morgenstern JP and Land H. (1990). Nuc. Acids Res. 18, 3587-3596.

Morita H, Tahara T, Matsumoto A, Kato T, Miyazaki H and Ohashi H. (1996). FEBS Lett. 395, 228-234. MEDLINE

Murakami M, Narazaki M, Hibi M, Yawata H, Yasukawa K, Hamaguchi M, Taga T and Kishimoto T. (1991). Proc. Natl. Acad. Sci. USA 88, 11349-11353. MEDLINE

Nagata Y and Todokoro K. (1995). FEBS Lett. 377, 497-501. MEDLINE

Pang L, Sawada T, Decker SJ and Saltiel AR. (1995). J. Biol. Chem. 270, 13585-13588. MEDLINE

Párrizas M, Saltiel AR and LeRoith D. (1997). J. Biol. Chem. 272, 154-161. MEDLINE

Payne DM, Rossomando AJ, Martino P, Erickson AK, Her JH, Shabanowitz J, Hunt DF, Weber MJ and Sturgill TW. (1991). EMBO. J. 10, 885-892. MEDLINE

Pear WS, Nolan GP, Scott ML and Baltimore D. (1993). Proc. Natl. Acad. Sci. USA 90, 8392-8396. MEDLINE

Picksley SM, Meek DW and Lane DP. (1992). Oncogene 7, 1649-1651. MEDLINE

Porteu F, Rouyez MC, Cocault L, Benit L, Charon M, Picard F, Gisselbrecht S, Souyri M and Dusanter-Fourt I. (1996). Mol. Cell. Biol. 16, 2473-2482. MEDLINE

Rainwater R, Parks D, Anderson ME, Tegtmeyer P and Mann K. (1995). Mol. Cell. Biol. 15, 3892-3903. MEDLINE

Ramsfjell V, Borge OJ, Cui L and Jacobsen SE. (1997). J. Immunol. 158, 5169-5177. MEDLINE

Reid S, Cross R and Snow EC. (1996). J. Immunol. Methods 192,, 43-54.

Ritchie A, Gotoh A, Gaddy J, Braun SE and Broxmeyer HE. (1997). Blood 90, 4394-4402. MEDLINE

Ritchie A, Vadhan-Raj S and Broxmeyer HE. (1996). Stem Cells 14, 330-336. MEDLINE

Rouyez MC, Boucheron C, Gisselbrecht S, Dusanter-Fourt I and Porteu F. (1997). Mol. Cell. Biol. 17, 4991-5000. MEDLINE

Sasaki K, Odai H, Hanazono Y, Ueno H, Ogawa S, Langdon WY, Tanaka T, Miyagawa K, Mitani K, Yazaki Y et al. (1995). Biochem. Biophys. Res. Commun. 216, 338-347. Article MEDLINE

Sattler M, Salgia R, Durstin MA, Prasad KV and Griffin JD. (1997). J. Cell. Physiol. 171, 28-33. Article MEDLINE

Sitnicka E, Lin N, Priestley GV, Fox N, Broudy VC, Wolf NS and Kaushansky K. (1996). Blood 87, 4998-5005. MEDLINE

Skoda RC, Seldin DC, Chiang MK, Peichel CL, Vogt TF and Leder P. (1993). EMBO. J. 12, 2645-2653. MEDLINE

Stahl N, Boulton TG, Farruggella T, Ip NY, Davis S, Witthuhn BA, Quelle FW, Silvennoinen O, Barbieri G, Pellegrini S et al. (1994). Science 263, 92-95. MEDLINE

Sturgill TW, Ray LB, Erikson E and Maller JL. (1988). Nature 334, 715-718. MEDLINE

Takahira H, Gotoh A, Ritchie A and Broxmeyer HE. (1997). Blood 89, 1574-1584. MEDLINE

Tauchi T, Feng GS, Marshall MS, Shen R, Mantel C, Pawson T and Broxmeyer HE. (1994). J. Biol. Chem. 269, 25206-25211. MEDLINE

Tortolani PJ, Johnston JA, Bacon CM, McVicar DW, Shimosaka A, Linnekin D, Longo DL and JJ OS. (1995). Blood 85, 3444-3451. MEDLINE

Vigon I, Mornon JP, Cocault L, Mitjavila MT, Tambourin P, Gisselbrecht S and Souyri M. (1992). Proc. Natl. Acad. Sci. USA 89, 5640-5644. MEDLINE

Welham MJ, Duronio V, Sanghera JS, Pelech SL and Schrader JW. (1992). J. Immunol. 149, 1683-1693. MEDLINE

Williams GT, Smith CA, Spooncer E, Dexter TM and Taylor DR. (1990). Nature 343, 76-79. MEDLINE

Xia Z, Dickens M, Raingeaud J, Davis RJ and Greenberg ME. (1995). Science 270, 1326-1331. MEDLINE

Yamada M, Komatsu N, Okada K, Kato T, Miyazaki H and Miura Y. (1995). Biochem. Biophys. Res. Commun. 217, 230-237. MEDLINE

Yang E, Zha J, Jockel J, Boise LH, Thompson CB and Korsmeyer SJ. (1995). Cell 80, 285-291. MEDLINE

Yonish-Rouach E, Resnitzky D, Lotem J, Sachs L, Kimchi A and Oren M. (1991). Nature 352, 345-347. MEDLINE

Zauli G, Bassini A, Vitale M, Gibellini D, Celeghini C, Caramelli E, Pierpaoli S, Guidotti L and Capitani S. (1997). Blood 89, 883-895. MEDLINE

Zerrahn J, Deppert W, Weidemann D, Patschinsky T, Richards F and Milner J. (1992). Oncogene 7, 1371-1381. MEDLINE

Zhang W and Deisseroth AB. (1994). Leuk. Lymphoma 14, 251-255. MEDLINE

Zhang W, Hu G, Estey E, Hester J and Deisseroth A. (1992). Oncogene 7, 1645-1647. MEDLINE

Zhu YM, Bradbury DA and Russell NH. (1994). Br. J. Cancer 69, 468-472. MEDLINE

Figure 1  Time course of Tpo-induced upregulation of p53 in the promoter conformation. (a) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with 50 ng/ml Tpo for up to 24 h and then analysed for the level of promoter conformation p53 (Pab 240) or all p53 (DO-1) by flow cytometry as described in Materials and methods. Data is expressed graphically as a histogram and numerically as a percent change in the mean channel fluorescence (MCF) from the background isotype (IgG1a) control. (b) The ratio of the MCF of Pab 240 stained samples to the MCF of DO-1 stained samples at the corresponding time point was determined and expressed as a percentage. This includes data from two separate experiments. Error bars represent one standard deviation (±s.d.)

Figure 2 Tpo induces the activation of p44/p42 MAPK in M07e cells, which is suppressed with the MEK-1 inhibitor PD98059. Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 10 M PD98059 (PD) for 1 h at 37°C. Following this, these two cultures were incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 15 min. The cells were collected, lysed and activated MAPK was immunoprecipitated from the lysates with dual-phosphospecific (Thr 202/Tyr 204) antibody. The immunoprecipitates were then incubated with Elk-1 fusion protein and excess cold ATP for 30 min at 33°C. The reaction was terminated with 3x SDS sample buffer and separated by SDS - PAGE. After transfer to PVDF membranes, Elk-1 phosphorylated at Ser 383 (MAPK site) was detected with phosphospecific Elk-1 antibody. (a) IgG H.C. is the immunoprecipitating immunoglobulin heavy chain. This blot represents one of four separate experiments that yielded similar results. (b) The membrane in (a) was stripped and re-probed with a non-phosphospecific Elk-1 antibody to demonstrate equal amounts of Elk-1 in the kinase reactions. (c) M07e cell lysates (without immunoprecipitation with -MAPK) were probed with a pan-p44/p42 MAPK antibody to demonstrate comparable amounts of MAPK in the cell lysate preparations. This blot represents one of four separate experiments that yielded similar results

Figure 3 MAPK pathway plays a potent role in mediating Tpo-induced conformational changes in p53 and Tpo-enhanced viability of M07e cells. (a) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 10 M PD98059 (PD) for 1 h at 37°C. Following this, these two cultures were incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 24 h. The cells were then collected and analysed for the level of promoter conformation p53 by flow cytometry using Pab 240 as described in Materials and methods. Data is expressed as a percent change in the mean channel fluorescence (MCF) from the background isotype (IgG_1a) control. This data represents an average of four separate experiments. (b). M07e cells were treated as in (a) except they were instead analysed for the extracellularization of phosphatidylserine using the MC540 dye as described in Materials and methods. Data is expressed as the percentage of cells that stained brightly for MC540. This data represents an average of three separate experiments. (c) 1´10⁶ M07e cells/ml were transduced with either pBabe vector control or pBp-Raf1-CX construct as described in Materials and methods. The cells were either prepared for intracellular detection of p53 in the promoter conformation or (d) stained with MC540 to measure apoptosis. (e) 2´10⁶ M07e cells/ml were transiently transfected with either pBabe vector control or pBp-Raf1-CX construct as described in Materials and methods. The cells were either prepared for intracellular detection of p53 in the promoter conformation or (f) stained with MC 540 to measure apoptosis

Figure 4 M07e cells transduced or tranfected with pBp-Raf1-CX contain elevated levels of MAPK activity in the absence of Tpo stimulation. Either 1´10⁶ M07e cells/ml were transduced or 2´10⁶ M07e cells/ml were transiently transfected with either pBabe vector control or pBp-Raf1-CX construct as described in Materials and methods. Cell lysates were prepared and a MAPK assay was performed as described in Materials and methods and in the legend to Figure 2

Figure 5  MC540 dye used to detect apoptosis. 1´10⁶ M07e cells were either factor starved for 36 h or factor stimulated for 36 h with 100 U/ml GM-CSF. The cells were collected and stained with 400 ng/ml MC540 as described in Materials and methods. (a) Forward (FSC, cell size) vs. Side (SSC, cell density) scatter profile of factor starved and factor stimulated cells. (b) MC540 staining of factor starved and factor stimulated cells. Factor starved cells possess MC540 dim (left peak, R1) and MC540 bright (right peak, R2) populations. Apoptotic cells reside within the MC540 bright population. (c) Factor starved cells were sorted on a FACS sorting machine based upon MC540 (FL2) intensity, and the FSC vs SSC profiles of those two populations analysed. (d) Cellular morphology of factor starved cells sorted on a FACS sorting machine based upon MC540 (FL2) intensity, and stained with Wright - Giemsa. All pictures were taken at 1000´ magnification under oil immersion

Figure 6 Negation of p53 protein levels with anti-sense oligonucleotide reverses the negative effects of PD98059 on Tpo-enhanced cellular viability. (a) 1´10⁶ M07e cells/ml were incubated with 5 M of either antisense or scrambled sense p53-specific oligonucleotides for 36 h in serum-free media. The cells were then incubated for 1 h at 37°C with or without 10 M PD98059 (PD). Following this, the cells were incubated for 48 h in the absence (M) or presence of 50 ng/ml Tpo (T) and then analysed for apoptosis with the MC540 dye as described in Materials and methods. This data represents an average of three separate experiments. (b) Western blot analysis of whole cell lysates taken at 48 h to confirm that p53 antisense oligonucleotides were sufficiently suppressing p53 protein expression. This experiment was performed three times yielding similar results. (c) Equal loading of cell lysates was confirmed with anti-GAPDH by Western blot analysis

Figure 7 JAK2 and STAT5 are tyrosine phosphorylated in our subline of M07e cells that do not respond proliferatively to Tpo. JAK2 inhibitor AG490 markedly suppresses STAT5 tyrosine phosphorylation (a) 1´10⁶ factor starved M07e cells/ml were incubated for 30 min in control serum-free media or media containing 50 ng/ml Tpo (T). The cells were then lysed and JAK2 immunoprecipitated. Following transfer to PVDF membranes, the level of tyrosine phosphorylated JAK2 was detected with a 1 : 1 mixture of -phosphotyrosine (P-Tyr) clones 4G10 and PY99. Media (M) and Tpo (T) samples are compared to factor starved, time 0 lysates (0). The bottom panel shows that the immunoprecipitates all contained comparable amounts of JAK2. This experiment was performed three times yielding similar results. (b) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 0.3 M AG490 (AG) for 1 h at 37°C. Following this, these two cultures were incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 15 min. The cells were collected, lysed, and STAT5 immunoprecipitated. Following transfer to PVDF membranes, the level of tyrosine phosphorylated STAT5 was detected with a 1 : 1 mixture of -phosphotyrosine (P-Tyr) clones 4G10 and PY99. Media (M) and Tpo (T) samples are compared to factor starved, time 0 lysates (0). The bottom panel shows that the immunoprecipitates all contained comparable amounts of STAT5. This experiment was performed three times yielding similar results

Figure 8 JAK2 inhibitor AG490 has an insignificant effect on Tpo-induced conformational changes in p53 and Tpo-enhanced viability of M07e cells. (a) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 0.3 M AG490 (AG; a JAK2 inhibitor) for 1 h at 37°C. Following this, these two cultures were incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 24 h. The cells were then collected and analysed for the level of promoter conformation p53 by flow cytometry using Pab 240 as described in Materials and methods. Data is expressed as a percent change in the mean channel fluorescence (MCF) from the background isotype (IgG_1a) control. This data represents an average of three separate experiments. (b) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 0.3 M AG490 (AG) for 1 h at 37°C. Following this, these two cultures were incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 48 h. The cells were then collected and analysed for the extracellularization of phosphatidylserine using the MC540 dye as described in Materials and methods. Data is expressed as the percentage of cells that stained brightly for MC540. This data represents an average of three separate experiments

Figure 9 STAT3-specific antisense oligonucleotides suppress STAT3 protein expression dose dependently. 1´10⁶ factor starved M07e cells/ml were incubated for 48 h in incremental amounts of STAT3 antisense or scrambled sense oligonucleotides. The cells were then collected, lysed and the whole cell level of STAT3 detected by Western blot analysis as described in Materials and methods. This experiment was performed three times yielding similar results

Figure 10 STAT3-specific antisense oligonucleotides have an insignificant effect on Tpo-induced conformational changes in p53 and Tpo-enhanced viability of M07e cells. (a) 1´10⁶ cells/ml were incubated in serum-free media with or without 10 M STAT3 antisense (AS) or scrambled sense (SS) oligonucleotides plus 50 ng/ml Tpo (to maintain cellular viability) for 24 h at 37°C. These cultures were then incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 24 h. The cells were collected and analysed for the level of promoter conformation p53 by flow cytometry using Pab 240 as described in Materials and methods. Data is expressed as a percent change in the mean channel fluorescence (MCF) from the background isotype (IgG_1a) control. This data represents an average of three separate experiments. (b) 1´10⁶ cells/ml were incubated in serum-free media with or without 10 M STAT3 antisense (AS) or scrambled sense (SS) oligonucleotides plus 50 ng/ml Tpo (to maintain cellular viability) for 24 h at 37°C. These cultures were then incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 48 h. The cells were collected and analysed for the extracellularization of phosphatidylserine using the MC540 dye as described in Materials and methods. Data is expressed as the percentage of cells that stained brightly for MC540. This data represents an average of three separate experiments

Figure 11 Wortmannin has an insignificant effect on Tpo-induced conformational changes in p53 but a weak but significant effect on Tpo-enhanced viability of M07e cells. (a) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 0.5 mM wortmannin (W) for 1 h at 37°C. These two cultures were then incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 24 h. The cells were then collected and analysed for the level of promoter conformation p53 by flow cytometry using Pab 240 as described in Materials and methods. Data is expressed as a percent change in the mean channel fluorescence (MCF) from the background isotype (IgG_1a) control. This data represents an average of three separate experiments. (b) Factor starved M07e cells were incubated at 1´10⁶ cells/ml in serum-free media with or without 0.5 mM wortmannin (W) for 1 h at 37°C. These cultures were then incubated in the absence (M) or presence of 50 ng/ml Tpo (T) for 48 h. The cells were collected and analysed for the extracellularization of phosphatidylserine using the MC540 dye as described in Materials and methods. Data is expressed as the percentage of cells that stained brightly for MC540. This data represents an average of three separate experiments