¹Leukaemia Research Fund Cellular Development Unit, UMIST, Sackville Street, Manchester, M60 1QD, UK

²Department of Biomolecular Sciences, UMIST, Sackville Street, Manchester, M60 1QD, UK

³Cancer Research Campaign, Department of Experimental Haematology, Paterson Institute for Cancer Research, Christie Hospital, NHS Trust, Manchester, M20 9BX, UK

⁴Molecular and Cellular Pharmacology Group, School of Biological Sciences, University of Manchester, Manchester, M13 9PT, UK

Correspondence to: E Spooncer, Cancer Research Campaign, Department of Experimental Haematology, Paterson Institute for Cancer Research, Christie Hospital, NHS Trust, Manchester, M20 9BX, UK

Chronic myeloid leukaemia is a haemopoietic stem cell disorder, the hallmark of which is the expression of the Bcr-Abl Protein Tyrosine Kinase (PTK). We have previously reported that activation of a temperature sensitive Bcr-Abl PTK in the multipotent haemopoietic cell line FDCP-Mix for short periods resulted in subtle changes including, a transient suppression of apoptosis and no inhibition of differentiation. In contrast, activation of the Bcr-Abl PTK for 12 weeks results in cells that display a delay in differentiation at the early granulocyte stage. Flow cytometric analysis also indicates that the expression of cell surface differentiation markers and nuclear morphology are uncoupled. Furthermore, a significant number of the mature neutrophils display abnormal morphological features. Prolonged exposure to Bcr-Abl PTK results in interleukin-3 independent growth and decreased p53 protein levels. FDCP-Mix cells expressing a dominant negative p53 and p53^null FDCP-Mix cells demonstrate that the reduction in p53 is causally related to the delay in development. Returning the cells to the restrictive temperature restores the p53 protein levels, the growth factor dependence and largely relieves the effects on development. We conclude that prolonged Bcr-Abl PTK activity within multipotent cells results in a reduction of p53 that drives a delayed and abnormal differentiation. Oncogene (2000) 19, 5487-5497.

Bcr-Abl; p53; CML; differentiation; haemopoietic

Introduction

Chronic myeloid leukaemia (CML) is a clonal disorder of pluripotent haemopoietic cells characterized by the presence of the Philadelphia chromosome (Ph⁺), the result of a reciprocal translocation between chromosome 9 and 22 (for a review see Gotoh and Broxmeyer, 1997). The 9:22 translocation fuses the Bcr gene with the Abl gene resulting in the expression of a chimeric protein, Bcr-Abl, that has an essential role in the molecular pathology of CML (Daley et al., 1990; Griffiths et al., 1992; Heisterkamp et al., 1990; Kelliher et al., 1990). The Bcr-Abl protein contains several functional domains including, most importantly, a constitutively activated form of the c-abl protein tyrosine kinase (PTK) (Konopka et al., 1984; Lugo et al., 1990).

CML is a progressive disease with an initial chronic phase in which there is a marked expansion in the late myeloid cell population (reviewed by Clarkson and Strife, 1993 and Verfaille, 1998). CML progenitor cells possess only a subtle defect in maturation, whilst retaining their requirements for bone marrow stromal cells or cytokines to survive and proliferate (Gishizky and Witte, 1992). In the course of the disease, this stage is followed by an accelerated phase that can include myelodysplastic features, hyposegmented neutrophils and basophilia. The blast crisis is the most severe manifestation of the disease in which differentiation is apparently blocked.

Although Bcr-Abl expression plays an initiating role in the pathogenesis of CML, the transformation to blast crisis appears to arise as a consequence of additional events (Ahuja et al., 1991; Foti et al., 1991) that are as yet poorly defined. Alterations in p53 expression are, however, implicated in disease progression. Whilst structural alterations of p53 are very rare in the chronic phase of CML, they are detected in 25-30% of CML blast crisis cases (Ahuja et al., 1989; Mashal et al., 1990; Prokocimer and Rotter, 1994). Although the majority of studies concentrate on p53 mutations, several reports have implicated a mere reduction in wild type (wt) p53 gene dosage as a factor of key importance for the development of the accelerated phase (Nakai and Misawa, 1995; Nakai et al., 1992; Rovira et al., 1995). Nakai et al. (1995) reported that the loss of a wtp53 allele preceded any p53 mutation and that the hemizygous wtp53 expression conferred a growth advantage on the cells. Studies of Bcr-Abl transfected p53^null murine bone marrow and p53^null/Bcr-Abl transgenic mice also implicated p53 in the disease progression (Honda et al., 2000; Skorski et al., 1996). p53^null cells transfected with Bcr-Abl displayed growth factor independence and were morphologically less differentiated than their p53 expressing counterparts. In addition, immunodeficient mice injected with these cells developed a highly aggressive, poorly differentiated acute myeloid leukaemia. Importantly, both Bcr-Abl and the loss of p53 were required for the development of the leukaemic like phenotype. A further reason to assess whether prolonged Bcr-Abl PTK expression influences p53 levels is the apparent ability of p53 to induce differentiation in K562 cells implicating p53 in the abnormal differentiation features of acute phase CML (Feinstein et al., 1992; Kremenetskaya et al., 1997).

Numerous attempts have been made to model CML using IL-3 dependent cell lines (Carlesso et al., 1994; Daley and Baltimore, 1988; Kabarowski et al., 1994; Klucher et al., 1998; Laneuville et al., 1992). Although these systems show the oncogenic potential of Bcr-Abl, they do not address the complex pathogenesis of CML. They have limited value in defining the events that lead to blast crisis, since they display a phenotype similar to that of the blast crisis stage of CML. This illustrates the absolute necessity to express Bcr-Abl in a biological relevant context, that is a haemopoietic progenitor cell capable of myeloid differentiation. Thus, to produce a biologically relevant and experimentally convenient model of CML to define molecular mechanisms underlying the progression of primitive Ph⁺ cells to blast crisis, we expressed a conditional mutant of Bcr-Abl in the primitive multipotent haemopoietic cell line, FDCP-Mix (Pierce et al., 1998a).

We have previously reported that expression of Bcr-Abl PTK activity in the FDCP-Mix cells for up to 10 days did not alter their ability to differentiate or their absolute requirement for growth factors to survive and proliferate (Pierce et al., 1998a). However, in the absence of added growth factor the cells displayed increased survival. In limiting concentrations of growth factors the cells exhibited enhanced survival and proliferation. The tsBcr-Abl transfected FDCP-Mix cells were, thus, similar in many respects to Ph⁺ progenitor cells during the initial chronic phase. They retained their ability to differentiate into mature myeloid cells and also retained their absolute requirement for growth factors to survive and proliferate.

Here we report the effects of prolonged exposure to Bcr-Abl PTK on haemopoietic cell differentiation and identify p53 modulation as a possible key molecular event associated with Bcr-Abl mediated progression in CML.

Results

Prolonged exposure to Bcr-Abl PTK activity results in the acquisition of Bcr-Abl dependent growth factor independence

When FDCP-Mix tsBcr-Abl cells were maintained at the permissive temperature indefinitely under conditions which did not positively select for clonal outgrowth of factor independent clones (i.e. high concentrations of IL-3) they were observed to become factor independent (Figure 1). After 9-15 weeks culture the tsBcr-Abl FDCP-Mix cells cultured at the permissive temperature (32°C) but not those maintained at the restrictive temperature (39°C) could proliferate in the absence of IL-3 (Figure 1a). The average time to onset of overt factor independent proliferation at 32°C in the presence of IL-3 was 12 weeks (12±3 weeks, mean±s.d., n=7). This effect was dependent on the presence of an active BCR-ABL PTK because, as previously shown, the parental FDCP-Mix and FDCP-Mix neo cells do not give rise to factor independent cells (Just et al., 1991). More importantly, when the factor independent tsBcr-Abl cells were transferred back to the restrictive temperature, growth factor dependence was restored (Figure 1b).

When cells were plated at 39°C in semi-solid media without IL-3 after prolonged culture at 32°C no factor independent colonies grew, indicating that there is not a subset of cells which are factor-independent, regardless of BCR-ABL activity. All the experiments presented have been repeated with multiple clones of FDCP-Mix cells expressing tsBcr-Abl.

Recent reports indicating that Bcr-Abl mediated growth factor independence arises due to autocrine production of growth factors (Jiang et al., 1999). However, we were unable to detect any IL-3, SCF or G-CSF production in the tsBcr-Abl FDCP-Mix cells using RT-PCR (data not shown). It therefore appears that the observed growth factor independence does not arise due to the production of these cytokines.

Differentiation is delayed and cell morphology abnormal as a consequence of prolonged exposure to Bcr-Abl PTK activity

CML is a progressive disease in which normal myeloid differentiation is lost in accelerated phase and blast crisis. To see if this results from long-term exposure to Bcr-Abl PTK activity (as we have observed for cytokine independence) we tested the ability of tsBcr-Abl FDCP-Mix cells to undergo myeloid cell differentiation.

TsBcr-Abl FDCP-Mix cells cultured for 12 weeks at the restrictive temperature in high concentrations of IL-3 maintain a primitive cell morphology, as do the parental FDCP-Mix cells (Figure 2a,b,c). When the cells were switched to conditions that induced granulocyte differentiation, at either the permissive or restrictive temperature, they underwent granulocyte development with the clonal extinction of primitive colony forming cells (Figures 2e,f and 3d,e). Thus, without prior exposure to PTK activity the cells respond to induction of myeloid cell differentiation in the absence or presence of active Bcr-Abl PTK with a similar pattern of differentiation to the parental FDCP-Mix cells (Figures 2d and 3a,b). As in previous experiments, we note a slight maturation delay as a consequence of growing the cells at 32°C (Figure 3a,b) and a further tendency towards a maturation delay in the presence of the Bcr-Abl PTK (Figure 3a,e). However, this is not statistically significant (see footnote to Figure 3 and Pierce et al., 1998a).

The tsBcr-Abl FDCP-Mix cells cultured for 12 weeks at 32°C in IL-3 also maintained a primitive cell morphology (Figure 2h,i). When tsBcr-Abl FDCP-Mix cultured for 12 weeks at 32°C were switched to the restrictive temperature and cultured in conditions which induce myeloid development, normal cell maturation was observed with the formation of neutrophils and monocytes (Figures 2n,o and 3f). However, when induced to differentiate at the permissive temperature there was evidence of both reduced maturation and abnormal mature cell morphology in about 20% of the cells (Figures 2k,l and 3c). Under these conditions the cultures contained a large number of immature cells as well as mature cells, indicative of a developmental block or delay. The morphology of many of the cells was also atypical. There were a significant number of cells with large, multilobed nuclei (Figure 2k,l) and an increase in the number of large multinucleate cells (Figure 2k,l). These cell types were scored as abnormal in the differential morphology. Following 7 days in culture under myeloid differentiation conditions the cells were unable to form colonies in soft gels in the presence of IL-3 (data not shown). The discordant development was, therefore, not associated with persistence of clonogenic cells under differentiation conditions. Furthermore, control cells showed no such acquisition of a developmental abnormality and were unaffected by prolonged exposure to culture at 32°C (Figures 2g,j,m and 3a). The data show very clearly that there is a consistent significant difference between panel (c) (maintained at 32°C and differentiated at 32°C) and all other groups for the delayed development and abnormal morphology (see footnote to Figure 3). The other groups are not statistically different from one another. Despite their abnormal appearance, the neutrophils present were capable of phagocytosis (see Figure 2k). Undifferentiated FDCP-Mix cells were unable to phagocytose FITC conjugated latex beads. However, cells induced to undergo differentiation were capable of phagocytosis, whether from cells cultured long-term at 39°C or 32°C and induced to differentiate at 39°C or 32°C (data not shown).

The unusual development we observed was dependent on the long-term exposure to Bcr-Abl PTK and the continued presence of Bcr-Abl PTK during differentiation. This is demonstrated by the reversion of the differentiation to normal following long-term maintenance at the permissive temperature and differentiation at the restrictive temperature (Figures 2n,o and 3f).

To confirm the abnormal phenotype and/or delayed development we examined the expression of two cell surface markers (Gr-1 and Mac-1) associated with maturing and mature neutrophils and macrophages. Parental and tsBcr-Abl expressing cells were negative for Gr-1 and Mac 1 expression under culture conditions that promote self-renewal. As expected, when induced to differentiate parental FDCP-Mix cells stained positively with GR-1 and Mac 1 antibodies. Mac 1 expression was positive though variable, in all the cells. Gr-1 expression was seen in subsets of the FDCP-Mix cells. Short-term exposure to Bcr-Abl PTK decreased the number of cells expressing both Gr-1 and Mac 1 (compare Figure 4a,e,f). Growing tsBcr-Abl FDCP-Mix at the permissive temperature for 12 weeks and then allowing them to differentiate at the restrictive temperature showed that the long-term exposure to Bcr-Abl had affected the ability of the cells to express Gr-1 and Mac 1 (compare Figure 4b,d,f). This is probably due to the persistence of changes in the tsBcr-Abl FDCP-Mix cells after the switch back from permissive to restrictive temperature.

Most notable was the fact that Mac 1 expression was completely abrogated by growing and differentiating tsBcr-Abl FDCP-Mix cells at the permissive temperature (compare Figure 4a,c). Gr-1 expression was increased in these cells compared to non-differentiated cells, but the expression was much lower than seen when control FDCP-Mix were induced to differentiate (a). Furthermore all the tsBcr-Abl cells exposed to Bcr-Abl chronically (c) expressed low levels of Gr-1, whereas in control samples (Figure 4a,b,f) two populations were present, those not expressing Gr-1 at all and those with relatively high levels of expression. It should be pointed out here that Figure 2k,l shows quite clearly that cells from Figure 4c contain morphologically recognizable neutrophils and macrophages. Thus, mature cells which normally express Gr-1 and Mac-1 are not doing so as a direct consequence of Bcr-Abl PTK exposure. Surface phenotype and morphological status have been uncoupled by the Bcr-Abl oncogene.

The onset of discordant development correlates with p53 modulation

Alterations in p53 expression are implicated in CML disease progression. We therefore investigated the effects of prolonged exposure to Bcr-Abl PTK on p53. We observed a reduction in wtp53 levels during the emergence of the growth factor independence and discordant development in the FDCP-Mix tsBcr-Abl cells. The level of expression and the conformation of p53 was investigated using immunoprecipitation with a range of monoclonal antibodies. Four anti-p53 antibodies were used: Ab240, 248, 421 and Ab1620. In immunoprecipitation Ab240 recognizes only mutant p53, Ab248 and 421 are pan-specific, whilst Ab1620 recognize epitopes present in wtp53 alone. The parental and tsBcr-Abl cells at both 32°C (18 h) and 39°C were positive for the p53 epitopes 248, 421 and 1620 (Figure 5a). Failure to detect the 240 epitope suggests that the FDCP-Mix cells contain wtp53 in the presence of Bcr-Abl PTK activity. We also failed to detect any mutant p53 following prolonged exposure to Bcr-Abl PTK (Figure 5b). However, there was a marked reduction in the amount of p53 detected with antibodies Ab248, 421 and Ab1620 (Figure 5b). Western blotting confirmed this decrease in p53 expression following prolonged exposure to Bcr-Abl PTK (Figure 5c). Interestingly, when the cells with decreased levels of p53 were transferred back to the restrictive temperature, the p53 increased to levels observed in cells not exposed to Bcr-Abl PTK activity (Figure 5d). Thus, as with the growth factor independence and altered differentiation, the p53 down regulation arising from prolonged exposure to Bcr-Abl PTK activity was reversible i.e. it was still reliant on Bcr-Abl PTK activity. It is important to emphasize that we never observed a complete loss of p53 protein expression.

RNA analysis was undertaken to investigate whether or not the modulation of p53 protein occurred at the transcriptional or post-transcriptional level. This clearly demonstrated that there is no alteration in the levels of p53 RNA expression following prolonged exposure to Bcr-Abl PTK activity (data not shown) suggesting that the lower p53 levels occurs as a result of a decrease in p53 protein stability or translation. Similar observations with p53 protein stability have been made in AML where overexpression of Mdm-2 leading to increased degradation of p53 has been suggested to be the mechanism by which the cells escape p53 regulation (Seliger et al., 1996). However, we saw no evidence of increased Mdm-2 expression in our system (data not shown).

Loss or inactivation of p53 results in delayed development in the presence of Bcr-Abl PTK activity

We have shown that there is a correlation between the loss of p53 and the acquisition of factor independence, abnormal and delayed development in multipotent cells. Can any of the effects be directly attributed to the loss of p53? We assessed this possibility in two ways, by using a p53^null FDCP-Mix cell line retrovirally transduced to express tsBcr-Abl and by expression of a dominant negative (DD) p53 in tsBcr-Abl FDCP-Mix cells. When the tsBcr-Abl PTK was activated in p53^null FDCP-Mix by switching the cells from the non-permissive to the permissive temperature, there was no evidence of growth factor independence (Figure 6). The shift to factor independence in the tsBcr-Abl FDCP-Mix grown at 32°C for more than 12 weeks cannot, therefore, be solely due to the loss of p53 protein. Similarly, the expression of DDp53 did not affect the cytokine requirements of tsBcr-Abl FDCP-Mix cells when they were switched from the restrictive to the permissive temperature (data not shown). These two approaches show that p53 alone has no effect on cytokine requirement for proliferation in Bcr-Abl expressing multipotent cells.

We next considered the role of p53 in myeloid cell development. When the tsBcr-Abl, p53^null FDCP-Mix and control cells were cultured in differentiation conditions at the permissive and restrictive temperatures for Bcr-Abl PTK activity we observed a profound delay in development of the tsBcr-Abl p53^null FDCP-Mix only when grown at the permissive temperature (Figure 7a). After 7 days the cultures contained blast cells and primitive myeloid progenitor cells at about three times the level seen in the control cultures. After 10 days there were a few early myeloid progenitors present but the tsBcr-Abl p53^null FDCP-Mix cultures did not persist beyond 14 days. These data imply that the loss of p53 (Figure 5) is causally related to the delayed development seen in tsBcr-Abl FDCP-Mix cells.

This was further confirmed in the tsBcr-Abl FDCP-Mix DDp53 cells. When these cells were induced to undergo myeloid differentiation at both the restrictive and permissive temperature we observed a delay in development at the permissive temperature for Bcr-Abl PTK activity (Figure 7b). The inhibition of p53 action by the DDp53, therefore, recapitulates the delayed development observed as a consequence of long-term exposure to Bcr-Abl activity which is accompanied by a reduction of p53. The differentiation delay resulting from DDp53 is greater than that observed in the LT tsBcr-Abl FDCP-Mix cells. This may be due to the fact we observe a reduction, but not complete loss of p53 in these cells (Figure 5).

We conclude that p53 loss is involved with the mechanisms that result in the delayed differentiation in tsBcr-Abl FDCP-Mix cells exposed to Bcr-Abl PTK activity for >12 weeks. We did not observe morphologically abnormal cells (seen in the LT tsBcr-Abl cells differentiated at the permissive temperature) in either the tsBcr-Abl, p53^null FDCP-Mix or tsBcr-Abl FDCP-Mix DDp53 above 2% of the total cell population.

Discussion

To identify the long-term consequences of Bcr-Abl PTK expression in primitive haemopoietic cells we have used a temperature sensitive kinase mutant of Bcr-Abl expressed in the multipotent haemopoietic FDCP-Mix cells. These cells represent a model system with appropriate biological characteristics to allow convenient, meaningful investigations into the molecular mechanisms of CML.

CML is a progressive disease and, although colony forming cells in chronic phase are factor dependent, this is less clearly the case in blast crisis where the dependency on exogenous stimuli is decreased (MaguerSatta et al., 1998; Metcalf et al., 1974). In keeping with this, following an average of 12 weeks exposure to Bcr-Abl PTK activity, growth factor independent tsBcr-Abl FDCP-Mix cells became evident. This phenomenon has been previously reported by Kabarowski et al. (1994) in the lymphoid cell line Ba/F3. The growth factor independence was still dependent upon Bcr-Abl activity. When the tsBcr-Abl FDCP-Mix cells were placed at the restrictive temperature growth factor dependence was restored. Our observations support the suggestion that, in the progression of the growth factor dependence to independence, the FDCP-Mix Bcr-Abl cells may be following similar molecular mechanisms that occur in the progression of CML.

In view of the nature of CML, far more important was the fact that the cells maintained at the permissive temperature for many weeks lost the ability to undergo differentiation at the rate normally observed in FDCP-Mix cells. Furthermore, there was not only developmental retardation but also acquisition of atypical neutrophil morphology in the cells that did differentiate. Hyposegmentation was one atypical feature observed that has previously been reported as being associated with CML accelerated phase in some patients (Williams, 1995).

The Bcr-Abl induced aberrant development was also accompanied by abnormalities in Gr-1 and Mac-1 expression during maturation. Both short-term and chronic exposure to Bcr-Abl activity inhibits Gr-1 expression (Gr-1 is a protein of unknown function expressed during granulocytic differentiation (Fleming et al., 1993)).

Chronic exposure to Bcr-Abl PTK also totally abrogates the ability of differentiating FDCP-Mix cells to express Mac 1, an effect which is partially reversible by allowing the tsBcr-Abl FDCP-Mix cells to differentiate at the permissive temperature. When FDCP-Mix cells are only exposed to Bcr-Abl PTK activity during differentiation there is also inhibition of Mac 1 expression. Mac 1 (known as CD11b, _M2 integrin and complement receptor 3) is expressed on macrophages and granulocytes, and mediates adherence via a receptor function for ICAM-1 and fibrinogen (Leenen et al., 1994; Lub et al., 1996). Pedersen (1982) has noted that CML neutrophils display subnormal adhesiveness and delayed emigration to extravascular sites. It is possible that Bcr-Abl mediated changes in the expression and activation of integrins accounts for such observations, our data with FDCP-Mix cells supports this.

More importantly, the morphological data with the expression profiles for Mac 1/Gr-1 indicate that the nuclear morphological maturation of FDCP-Mix cells and expression of cell surface maturation markers is discordant. This resonates with the theories of Strife and Clarkson (1998) who stated in their seminal review paper 'altered tyrosine kinase activity of the P210^Bcr-Abl protein, or some other biochemical alteration associated with the t(9;22) translocation, causes a subtle garbling in the transmission of regulatory messages through the transduction pathways, with the result that the leukemic stem cells' normal sequence of maturation is distorted so that the cytoplasmic and nuclear maturation proceeds asynchronously'. The data we present supports such a theory by showing, in a model system, that Bcr-Abl directly affects development and in some cells leads to discordance in the expression of surface markers compared to nuclear morphology.

The fact that this aberrant differentiation depends on the continued presence of tsBcr-Abl PTK activity is important: it indicates that persistent activation of Bcr-Abl is essential to drive significant changes in the ability of these cells to develop normally. This, plus the observation that Ph⁺ cells in blast crisis require less cytokine stimulation, as do tsBcr-Abl FDCP-Mix cells at 32°C, shows that we have, in part, mimicked the progression of CML into accelerated phase.

p53 has a key role in sensing DNA damage and cell cycle regulation (for review see Levine, 1997) whilst also being involved in the regulation of differentiation (Amariglio et al., 1997; Sabapathy et al., 1997; Shounan et al., 1996; Soddu et al., 1996). Alterations in p53 expression are heavily implicated in CML disease progression (see Introduction). This is not only due to the appearance of structural alterations in blast crisis (Ahuja et al., 1989; Mashal et al., 1990; Prokocimer and Rotter, 1994) but also observations on the effects of reduced wtp53 levels and the ability of p53 to induce differentiation in the CML blast crisis cell line K562.

During the emergence of the growth factor independence and discordant development in the FDCP-Mix tsBcr-Abl cells we observed a reduction in wtp53 levels. These data support a role for reduced p53 levels in disease progression. The altered regulation of p53 expression was not due to mutation, as it was reversible, and the Bcr-Abl PTK did not alter the transcription rate of p53.

The experiments with the dominant negative p53 and p53^null tsBcr-Abl FDCP-Mix cells confirm that the reduction in p53 alone is not the cause of changes in the cytokine requirements or the abnormal development in the LT tsBcr/Abl FDCP-Mix cells. However, reduction of p53 does combine with Bcr-Abl activity to retard neutrophil/macrophage development.

These findings confirm the study by Skorski et al. (1996). p53^null cells transfected with Bcr-Abl were morphologically less differentiated than their p53 expressing counterparts. Here, too, both Bcr-Abl and the loss of p53 were required for the development of the leukaemic like phenotype. Our observations further those of Skorski in that they not only implicate p53 in disease progression but also imply a role for Bcr-Abl in the control of p53 protein stability. There is a precedent for this, Bcr-Abl has previously been shown to be involved in the regulation of protein turnover for the Abi proteins, a family of Abl interacting proteins (Dai et al., 1998). These observations are of particular interest since JNK has been implicated in the regulation of p53 protein stability (Fuchs et al., 1998a,b) and is known to be activated by Bcr-Abl and in CML patient samples (Burgess et al., 1998; Raitano et al., 1995).

The decrease in p53 may not only lead to altered developmental rates but also predispose the cells to further mutations. Inactivation of p53 increases the rate of homologous recombination (Mekeel et al., 1997; Sturzbecher et al., 1996), an important event in chromosomal translocations, providing a link between p53 and increased genetic instability in the late stages of tumorigenesis.

Whilst only 25-30% of patients in blast crisis are reported to have alterations in p53 expression, the effects we observed in this model, i.e. a reduction in p53 protein levels are much harder to quantify and would not have been detected in the vast majority of studies to date which have concentrated on the appearance of p53 mutations. Since a simple reduction in p53 gene dosage is now being recognized as a possible factor in tumorigenesis of many other tissues (Venkatachalam et al., 1998) we propose that the reduction in p53 expression as well as (or preceding) mutations to p53 may be an important contributor to the progression of CML from the chronic to blast crisis.

Given the fact that long-term exposure to Bcr-Abl PTK activity induces a loss of p53, a delay in development and dysplastic granulocyte production, we believe that, in part, we are recapitulating the accelerated phase of CML. The modelling of CML in the FDCP-Mix multipotent cell line offers the opportunity to identify the initial molecular events leading to the phenotypic, pathology-related changes seen in multipotent haemopoietic cells expressing an activated Bcr-Abl PTK.

We are currently attempting to assess p53 protein levels in progenitor cells from patients with CML at presentation and in accelerated phase and to compare these with normal progenitor cell populations. In this way the significance of our results to CML can be established.

Materials and methods

Maintenance and differentiation of FDCP-Mix Cells

FDCP-Mix A4 cells were maintained in Fischers medium supplemented with pre-selected batches of horse serum (20% v/v) and IL-3. The cells were subcultured twice a week to a cell concentration of 6-8´10⁴ cells/ml and maintained at 32°C or 39°C in 5% CO₂ in air. In these culture conditions the FDCP-Mix maintain a primitive blast cell phenotype. p53^null multipotent FDCP-Mix cells were generated from the bone marrow of p53^null mice using long-term marrow cultures as previously described (Spooncer et al., 1986). Cells were induced to undergo myeloid differentiation by culture in Iscoves Modified Dulbeccos Medium supplemented with foetal calf serum (20% v/v), recombinant (r) murine GM-CSF (50 U/ml), r-human G-CSF (3000 U/ml) and r-murine IL-3 (0.1 ng/ml) as previously described (Pierce et al., 1998b).

Retroviral transfection

FDCP-Mix A4 and p53^null FDCP-Mix cells were transfected with the retroviral vector pM5-neo or pM5-neo carrying a p210 tsBcr-Abl cDNA as described by Pierce et al. (1998a). The p210 tsBcr-Abl cDNA encodes for a temperature sensitive kinase mutant of p210 Bcr-Abl. Expression is unaffected by temperature but its tyrosine kinase activity is present at 32°C (permissive) and inactive at 39°C (restrictive temperature). TsBcr-Abl FDCP-Mix A4 cells were transfected with the retroviral vector pM5-hygro or pM5-hygro carrying a truncated p53 which acts as a dominant negative inhibitor (DD) of p53 function. The DDp53 cDNA was constructed by site directed mutagenesis as described by Shaulian et al. (1992). Expression is unaffected by temperature. Experiments were performed using at least two different clones of the tsBcr-Abl FDCP-Mix cells.

Western blotting and immunoprecipitation

Western blotting and immunoprecipitation were carried out as previously described (Spooncer et al., 1994). The expression of p53 was analysed by Western blotting using a polyclonal antibody to p53 (a gift from T Hupp, Dundee). To investigate p53 conformation the antibodies Ab240, 421, 1620 (Oncogene Science) and Ab248 (a gift from T Hupp, Dundee) were used for immunoprecipitation.

Measurement of DNA synthesis

DNA synthesis was measured using a [³H]thymidine incorporation assay as previously described (Pierce et al., 1998b).

Northern blotting

RNA extracts were prepared with TRIzol reagent (Gibco BRL, Paisley, UK). Northern blotting was undertaken with radiolabelled probes by standard techniques (Sambrook et al., 1989).

Flow cytometry

Flow cytometry analysis was performed using a Becton Dickinson FacsVantage instrument. Cells expressing Gr-1 and Mac 1 were identified as previously described in Evans et al. (1999).

Acknowledgements

We thank Sue Slack for her excellent technical assistance. This work was supported by the Leukaemia Research Fund and the Cancer Research Campaign.

Ahuja H, Bar-Eli M, Advani SH, Benchimol S and Cline MJ. (1989). Proc. Natl. Acad. Sci. USA 86, 6783-6787. MEDLINE

Ahuja H, Bar-Eli M, Arlin Z, Advani S, Allen SL, Goldman J, Snyder D, Foti A and Cline M. (1991). J. Clin. Invest. 87, 2042-2047. MEDLINE

Amariglio F, Tchang F, Prioleau MN, Soussi T, Cibert C and Mechali M. (1997). Oncogene 15, 2191-2199. MEDLINE

Burgess GS, Williamson EA, Cripe LD, Litz-Jackson S, Bhatt JA, Stanley K, Stewart MJ, Kraft AS, Nakshatri H and Boswell HS. (1998). Blood 92, 2450-2460. MEDLINE

Carlesso N, Griffin JD and Druker BJ. (1994). Oncogene 9, 149-156. MEDLINE

Clarkson B and Strife A. (1993). Leukemia 7, 1683-1721. MEDLINE

Dai ZH, Quackenbush RC, Courtney KD, Grove M, Cortez D, Reuther GM and Pendergast AM. (1998). Genes Dev. 12, 1415-1424. MEDLINE

Daley GQ and Baltimore D. (1988). Proc. Natl. Acad. Sci. USA 85, 9312-9316. MEDLINE

Daley GQ, Van ER and Baltimore D. (1990). Science 247, 824-830. MEDLINE

Evans CA, Pierce A, Winter SA, Spooncer E, Heyworth CM and Whetton AD. (1999). Blood 94, 1504-1514. MEDLINE

Feinstein E, Gale RP, Reed J and Canaani E. (1992). Oncogene 7, 1853-1857. MEDLINE

Fleming TJ, Fleming ML and Malek TR. (1993). J. Immunol. 151, 2399-2408. MEDLINE

Foti A, Ahuja HG, Allen SL, Koduru P, Schuster MW, Schulman P, Bar-Eli M and Cline MJ. (1991). Blood 77, 2441-2444. MEDLINE

Fuchs SY, Adler V, Buschmann T, Yin Z, Wu X, Jones SN and Ronai Z. (1998a). Genes Dev. 12, 2658-2663. MEDLINE

Fuchs SY, Adler V, Pincus MR and Ronai Z. (1998b). Proc. Natl. Acad. Sci. 95, 10541-10546.

Gishizky ML and Witte ON. (1992). Science 256, 836-839. MEDLINE

Gotoh A and Broxmeyer HE. (1997). Curr. Opin. Hematol. 4, 3-11. MEDLINE

Griffiths SD, Healy LE, Ford AM, Bennett CA, Voncken JW, Heisterkamp N, Groffen J and Greaves MF. (1992). Oncogene 7, 1391-1399. MEDLINE

Heisterkamp N, Jenster G, ten Hoeve J, Zovich D, Pattengale PK and Groffen J. (1990). Nature 344, 251-253. MEDLINE

Honda H, Ushijima T, Wakazono K, Oda H, Tanaka Y, Aizawa S, Ishikawa T, Yazaki Y and Hirai H. (2000). Blood 95, 1144-1150. MEDLINE

Jiang X, Lopez A, Holyoake T, Eaves A and Eaves C. (1999). PNAS 96, 12804-12809. MEDLINE

Just U, Stocking C, Spooncer E, Dexter TM and Ostertag W. (1991). Cell 64, 1163-1173. MEDLINE

Kabarowski JH, Allen PB and Wiedemann LM. (1994). EMBO J. 13, 5887-5895. MEDLINE

Kelliher MA, McLaughlin J, Witte ON and Rosenberg N. (1990). Proc. Natl. Acad. Sci. USA 87, 6649-6653. MEDLINE

Klucher KM, Lopez DV and Daley GQ. (1998). Blood 91, 3927-3934. MEDLINE

Konopka JB, Watanabe SM and Witte ON. (1984). Cell 37, 1035-1042. MEDLINE

Kremenetskaya OS, Logacheva NP, Baryshnikov AY, Chumakov PM and Kopnin BP. (1997). Oncol. Res. 9, 155-166. MEDLINE

Laneuville P, Sun G, Timm M and Vekemans M. (1992). Blood 80, 1788-1797. MEDLINE

Leenen PJM, Debrujin MFTR, Voerman JSA, Campbell PA and Vanewijk W. (1994). J. Immunol. Meth. 174, 5-19.

Levine AJ. (1997). Cell 88, 323-331. MEDLINE

Lub M, van Kooyk Y and Figdor CG. (1996). J. Leukoc. Biol. 59, 648-655. MEDLINE

Lugo TG, Pendergast AM, Muller AJ and Witte ON. (1990). Science 247, 1079-1082. MEDLINE

MaguerSatta V, Burl S, Liu L, Damen J, Chahine H, Krystal G, Eaves A and Eaves C. (1998). Oncogene 16, 237-248. MEDLINE

Mashal R, Shtalrid M, Talpaz M, Kantarjian H, Smith L, Beran M, Cork A, Trujillo J, Gutterman J and Deisseroth A. (1990). Blood 75, 180-189. MEDLINE

Mekeel KL, Tang W, Kachnic LA, Luo CM, De Frank JS and Powell SN. (1997). Oncogene 14, 1847-1857. MEDLINE

Metcalf D, Moore MAS, Sheridan JW and Spitzer G. (1974). Blood 43, 847-859. MEDLINE

Nakai H and Misawa S. (1995). Leuk. Lymphoma 19, 213-221. MEDLINE

Nakai H, Misawa S, Horiike S, Maekawa T, Kashima K and Ishizaki K. (1995). Br. J. Haematol. 90, 147-155. MEDLINE

Nakai H, Misawa S, Toguchida J, Yandell DW and Ishizaki K. (1992). Cancer Res. 52, 6588-6593. MEDLINE

Pedersen B. (1982). Br. J. Haematol. 51, 339-344. MEDLINE

Pierce A, Owen-Lynch PJ, Spooncer E, Dexter TM and Whetton AD. (1998a). Oncogene 17, 667-672. MEDLINE

Pierce A, Whetton AD, Owen-Lynch PJ, Tavernier E, Spooncer E, Dexter TM and Heyworth CM. (1998b). J. Cell. Sci. 111, 815-823.

Prokocimer M and Rotter V. (1994). Blood 84, 2391-2411. MEDLINE

Raitano AB, Halpern JR, Hambuch TM and Sawyers CL. (1995). Proc. Natl. Acad. Sci. 92, 11746-11750. MEDLINE

Rovira A, Urbano-Ispizua A, Cervantes F, Rozman M, Vives-Corrons JL, Montserrat E and Rozman C. (1995). Ann. Hematol. 70, 129-133. Article MEDLINE

Sabapathy K, Klemm M, Jaenisch R and Wagner EF. (1997). EMBO J. 16, 6217-6229. Article MEDLINE

Sambrook J, Fritsch EF and Maniatis T. Molecular Cloning; a laboratory manual, Second edn. Cold Spring Harbor, New York.,

Seliger B, Papadileris S, Vogel D, Hess G, Brendel C, Storkel S, Ortel J, Kolbe K, Huber C, Huhn D and Neubauer A. (1996). Eur. J. Haematol. 57, 230-240. MEDLINE

Shaulian E, Zauberman A, Ginsberg D and Oren M. (1992). Mol. Cell. Biol. 12, 5581-5592. MEDLINE

Shounan Y, Dolinkov A, MacKenzie KL, Miller M, Chan YY and Symonds G. (1996). Leukemia 10, 1619-1628. MEDLINE

Skorski T, Nieborowska-Skorska M, Wlodarski P, Perrotti D, Martinez R, Wasik MA and Calabretta B. (1996). Proc. Natl. Acad. Sci. USA 93, 13137-13142. MEDLINE

Soddu S, Blandino G, Scardigli R, Coen S, Marchetti A, Rizzo MG, Bossi G, Cimino L, Crescenzi M and Sacchi A. (1996). J. Cell. Biol. 134, 193-204. MEDLINE

Spooncer E, Fairbairn L, Cowling GJ, Dexter TM, Whetton AD and Owen Lynch PJ. (1994). Leukemia 8, 620-630. MEDLINE

Spooncer E, Heyworth CM, Dunn A and Dexter TM. (1986). Differentiation 31, 111-118. MEDLINE

Strife A and Clarkson B. (1988). Sem. Hematol. 25, 1-19.

Sturzbecher HW, Donzelmann B, Henning W, Knippschild U and Buchhop S. (1996). EMBO J. 15, 1992-2002. MEDLINE

Venkatachalam S, Shi Y, Jones SN, Vogel H, Bradley A, Pinkel D and Donehower LA. (1998). EMBO J. 17, 4657-4667. Article MEDLINE

Verfaille CM. (1998). Hematology-Oncology Clinics of North America 12, 1.

Williams J. (1995). Williams Hematology. McCraw-Hill Inc: London.,

Figure 1 Prolonged exposure to active p210 BCR/ABL PTK results in the acquisition of growth factor independent proliferation which is reversible. (a) FDCP-Mix and tsBcr-Abl clones 2 and 16 maintained long-term (LT) in 10 ng/ml murine interleukin-3 (mIL3) at 32°C or 39°C were monitored once a week for growth factor independent proliferation by [³H]thymidine incorporation in the absence of added IL-3. Cells were washed and seeded at 2´10⁵ cells/ml in Fischer's medium for 18 h prior to addition of [³H]thymidine. Data are expressed as a percentage of thymidine incorporation observed at a maximally stimulating concentration of IL-3 (10 ng/ml IL-3 with 10% horse serum (HS)) and is a representative experiment of seven. FDCP-mix tsBcr-Abl clone 2 LT 39°C, assay at 39°C, FDCP-mix tsBcr-Abl clone 2 LT 32°C, assay at 32°C. [left to right hatched box] FDCP-mix tsBcr-Abl clone 16 LT 39°C, assay at 39°C. [right to left hatched box] FDCP-mix tsBcr-Abl clone 16 LT 32°C, assay at 32°C. (b) Growth factor independent proliferation and its reversible nature. The proliferation rate of FDCP-Mix and tsBcr-Abl clones 2 and 16 maintained long-term at 32°C or 39°C was assessed by [³H]thymidine incorporation at both 32°C and 39°C in the absence of added IL-3. Data are expressed as a percentage of thymidine incorporation observed in maximally stimulating conditions (10 ng/ml of IL-3 with 10% HS) at the relevant assay temperature and is a representative experiment of three. FDCP-mix control maintained LT at 39°C prior to thymidine uptake. FDCP-mix control maintained LT 32°C prior to thymidine uptake. [left to right hatched box] FDCP-mix tsBcr-Abl clone 2 maintained LT at 39°C prior to thymidine uptake. [tight to left hatched box] FDCP-mix tsBcr-Abl clone 2 maintained LT at 32°C prior to thymidine uptake (capable of factor independent growth, see (a). [vertical ruled box] FDCP-mix tsBcr-Abl clone 16 maintained LT at 39°C prior to thymdine uptake. [horizontal ruled box] FDCP-mix tsBcr-Abl clone 16 maintained LT at 32°C prior to thymidine uptake (capable of factor independent growth, see (a)

Figure 2 Micrographs of FDCP-Mix and FDCP-Mix tsBcr-Abl. Morphology of FDCP-Mix (column 1) and tsBcr-Abl clones 2 (column 2) and 16 (column 3). Row 1 (a,b,c); 12 weeks or longer growth in conditions for maintenance of a primitive phenotype at 39°C. Row 2 (d,e,f); maintained with a primitive phenotype at 39°C and differentiated for 10 days at 39°C. Row 3 (g,h,i); maintained with a primitive phenotype at 32°C for 12 weeks or longer. Row 4 (j,k,l); maintained with a primitive phenotype at 32°C and differentiated for 10 days at 32°C, Row 5 (m,n,o); maintained with a primitive phenotype at 32°C and differentiated for 10 days at 39°C. Magnification ´2000

Figure 3 Prolonged exposure to active p210 Bcr-Abl PTK results in discordant development. Parental FDCP-Mix cells and tsBcr-Abl clones 2 and 16 maintained at 32°C or 39°C for extended periods were plated in culture conditions which promoted granulocytic development and incubated at 32°C and 39°C. Viable cell counts and levels of primitive clonogenic cells were monitored over a 10 day period. Cytospins were prepared, stained with May-Grunwald-Giemsa and differential morphology scored for greater than 100 cells per slide. The differential categories shown are blast cells and early (immature) granulocytes, mature neutrophil granulocytes, monocytes and cells of abnormal morphology as categorized in the main text. The data shown are from four separate experiments for (a) and (b) and (c-f) are compiled from six differentiation experiments using both tsBcr-Abl FDCP-Mix cells clone 2 and clone 16 (both clones showed the abnormal and delayed development). Statistical analysis is shown in the table below:

Figure 4 The effect of long and short-term exposure to Bcr-Abl PTK activity on expression of Mac 1 and Gr-1. TsBcr-Abl FDCP-Mix and FDCP-Mix control cells were cultured in differentiation inducing conditions (see Figures 2 and 3) for 7 days followed by flow cytometric analysis of Gr-1 and Mac 1 expression using flow cytometry. Isotype control antibody (dotted), Gr-1 (dashed) and Mac 1 (solid) expression are shown on the same diagram. Results shown are from one of four experiments in which the same results were recorded

Figure 5 Changes in p53 protein level occur on prolonged exposure to Bcr-Abl PTK activity. The level and conformation of p53 in the parental FDCP-Mix and tsBcr-Abl cells was assessed by immunoprecipitation and Western blotting. (a) Immunoprecipitation of p53 with a range of antibodies from tsBcr-Abl FDCP-Mix cells at 32°C and 39°C. (b) Immunoprecipitation of p53 from tsBcr-Abl FDCP-Mix clone 16 following prolonged (>12 weeks) incubation at 32°C and 39°C. (c) Western blot analysis of total p53 levels in tsBcr-Abl FDCP-Mix clones 2 and 16 following prolonged (>12 weeks) incubation at 32°C and 39°C. (d) Immunoprecipitation of p53 from tsBcr-Abl FDCP-Mix clone 16 at 32°C following prolonged (>12 weeks) incubation at 32°C and after an 18 h switch to 39°C. No effect was seen in the appropriate neo transfected control cells (not shown for conciseness). Equal protein loading was assured by protein assays and visualization of filters prior to processing. Film exposure times were varied as appropriate for each experiment

Figure 6 Thymidine incorporation of p53^null FDCP-Mix. The proliferation of tsBcr-Abl p53^null FDCP-Mix cells at 32°C and 39°C in response to a range of IL-3 concentrations was measured by a 4 h [³H]thymidine incorporation following overnight incubation at each temperature. All cells were maintained at 39°C prior to the experiment. Results shown are from one experiment that is representative of three. Wild type FDCP-Mix tsBcr/Abl cells exhibit the same profile of thymidine incorporation as the p53^null isolates (data not shown for clarity of the figure)

Figure 7 Myeloid differentiation of tsBcr-Abl p53^null FDCP-Mix and DDp53 tsBcr-Abl FDCP-Mix. The tsBcr-Abl p53^null FDCP-Mix (a) and DDp53 tsBcr-Abl FDCP-Mix (b) cells were subject to myeloid differentiation as described for Figure 3. The histogram legend is as shown in Figure 3. Results shown are the mean±s.e.m. of three experiments and were subject to a two-tailed paired t-test. In (a) the percentage blasts at 7 days were compared between control and tsBcr-Abl at both temperatures and tsBcr-Abl at 32°C and 39°C. In (b) the percentage blasts at 7 days were compared between control and DDp53 at both temperatures and DDp53 at 32°C and 39°C. **Indicates a value for P of less than 0.01

Table 1 Students t-test analysis of differential morphologies at day 10 of differentiation. Bold font indicates the group in which Bcr-Abl PTK is active during both prior maintenance and differentiaton of the cells