Proteomic analysis reveals a novel mechanism induced by the leukemic oncogene Tel/PDGFRβ in stem cells: activation of the interferon response pathways

Objective proteomic analysis offers opportunities for hypothesis generation on molecular events associated with pathogenesis in stem cells. Relative quantification mass spectrometry was employed to identify pathways affected by Tel/PDGFRβ, an oncogene associated with myeloproliferative neoplasia (MPN). Its effects on over 1800 proteins were quantified with high confidence. Of those up-regulated by Tel/PDGFRβ several were involved in the interferon gamma (IFNγ) response. To validate these observations we employed embryonic and myeloid stem cells models which revealed Tel/PDGFRβ-induced STAT1 up-regulation and activation was responsible for modulating the interferon response. A STAT1 target highly up-regulated was ICSBP, a transcriptional regulator of myeloid and eosinophilic differentiation. ICSBP interacts with CBP/p300 and Ets transcription factors, to promote transcription of additional genes, including the Egr family, key regulators of myelopoiesis. These interferon responses were recapitulated using IFNγ stimulation of stem cells. Thus Tel/PDGFRβ induces aberrant IFN signaling and downstream targets, which may ultimately impact the hematopoietic transcriptional factor network to bias myelomonocytic differentiation in this MPN.

Changes in signal transduction downstream from the granulocyte-macrophage colony-stimulating factor receptor during differentiation of primary hemopoietic cells

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a multifunctional cytokine, having different effects on primitive hemopoietic cells and terminally differentiated end-cells of the myeloid lineage. Human primitive hemopoietic cells (CD34+) were obtained from the peripheral blood after mobilization and induced to proliferate and then differentiate with a combination of cytokines in vitro. Cells at different time points were then used to analyze the expression of the GM-CSF receptor and GM-CSF mediated activation of the JAK 2-STAT 5 and MAP kinase pathways. Scatchard analysis as measured by radioligand binding revealed that freshly purified CD34+ cells expressed 36+/-1 high affinity receptors per cell (mean +/- SE, n = 3) and the level of expression was not significantly different after 3 days in culture, but rose five- to tenfold by day 8. The day 0 CD34+ cells were hyporesponsive to GM-CSF, but by 3 days in culture the cells were still morphologically immature but were actively proliferating and exhibited maximal GM-CSF induced JAK 2-STAT 5 and MAP kinase activation at the optimal time point. Further culture of the CD34+ cells resulted in myeloid differentiation associated with prolongation of MAP kinase activation but not JAK 2-STAT 5 activation. These data indicate that the JAK 2-STAT 5 and MAP kinase pathways are independently regulated and that changes in these signaling pathways occur with differentiation.

Differentiation-linked changes in granulocyte-macrophage colony-stimulating factor receptor mediated signalling in the HL-60 promyelocytic cell line

Granulocyte-macrophage colony-stimulating factor (GM-CSF) induces the proliferation and maturation of immature myeloid progenitor cells and primes mature cell function in phagocytes. To investigate whether the biochemical events following the binding of GM-CSF to its receptor are differentiation dependent we analysed GM-CSF mediated activation of the JAK 2-STAT 5 and MAP kinase pathways in undifferentiated HL-60 cells and HL-60 cells induced to differentiate with dimethyl sulphoxide (DMSO) or retinoic acid (RA). GM-CSF stimulated MAP kinase activation in both the undifferentiated and differentiated HL-60 cells. Activation of MAP kinase (expressed as a proportion of total cellular MAP kinase) was maximal at 5 min and of similar magnitude in both cell types. There was, however, a marked difference in the later kinetics of activation, with the response being transient in the undifferentiated cells and disappearing within 15 min, whereas it was prolonged and persisted for at least 60 min in the differentiated cells. GM-CSF mediated activation of STAT 5 was markedly increased (15-20-fold) after differentiation of HL-60 cells but the kinetics of activation did not change. The increase in STAT 5 activation was not due to a change in total cellular STAT 5 expression but correlated with increased JAK-2 protein levels. These data show that in the HL-60 cell model, differentiation modulates the activation of signalling molecules downstream of the GM-CSF receptor.

Granulocyte-macrophage colony stimulating factor receptor alpha and beta chain complexes can form both high and intermediate affinity functional receptors

Scatchard analysis of primary human haemopoietic cells using iodinated GM-CSF suggests that there are low, intermediate and high affinity classes of the GM-CSF receptor. To investigate the molecular basis of this, we generated a clone of transfected NIH3T3 cells that constitutively expressed the human granulocyte-macrophage colony stimulating factor receptor (GM-CSF R) beta chain and inducibly expressed the human GM-CSF R alpha chain. In the cells fully induced to express the alpha chain the overall level of expression of the alpha and beta chains at the cell surface was comparable with that found in primary haemopoietic cells and cell lines. When cells were partially induced to express the alpha chain, the alpha:beta ratio determined by antibody binding was approximately 1:1 and Scatchard analysis revealed a single class of intermediate affinity receptors (Kd = 614+/-88 pM). In cells with fully induced alpha chain expression, the alpha:beta ratio was approximately 3:1 and there was a switch to a dual high and low affinity receptor with K(d)s of 67+/-32 pM and 1.7+/-0.56 nM respectively. The change from intermediate to high affinity was not associated with changes in alphabeta stoichiometry as detected by cross-linking with radiolabelled GM-CSF and gel electrophoresis. Both the high and intermediate affinity receptors were able to activate the STAT 5 and the MAP kinase pathways, although there was a difference in the ligand dose-response curves which was compatible with the different affinities of the receptors. It is proposed that the switch from an intermediate to high affinity receptor was due to the availability of free alpha chains presenting ligand to the alphabeta chain complexes at the surface of the cell membrane.

Tissue specificity of X-chromosome inactivation patterns

The analysis of X-chromosome inactivation patterns has been used in a number of clinical situations such as the identification of carrier status in X-linked genetic disorders and the establishment of the monoclonal origin of tumors. Interpretation of the result obtained requires comparison with the constitutive pattern for the individual, and for hematopoietic malignancies, skin biopsies or cultured fibroblasts have often been used as the control tissue because normal cells of the same lineage as the malignancy are not generally available. However, this assumes that patterns in the different tissues are constitutionally the same. We have therefore compared X-chromosome inactivation patterns from peripheral blood (granulocytes, E- cells, and T cells), skin, and muscle from 20 hematologically normal females, and colonic mucosa from 9 individuals. In 11 patients (55%), the results obtained were similar for all tissues of an individual, but in 9 patients, significant differences were observed between tissues. The most consistent feature was a skewing in peripheral blood (> 75% expression of one allele) but not skin and/or muscle. These studies suggest that skin cannot be used as a control tissue for the interpretation of X-chromosome inactivation patterns in hematopoietic cells.

Frequency of clonal remission in acute myeloid leukaemia

Analysis of X-chromosome inactivation patterns in females has been used to assess clonality of various tumours and for prenatal diagnosis of X-linked disorders. Studies with these methods in acute myeloid leukaemia suggest that a significant proportion of cases have clonal remissions (ie, persistence of the malignant clone), which may represent return to a preleukaemic state. We therefore analysed X-chromosome inactivation patterns with differential methylation patterns of heterozygotes for three DNA probes, HPRT, PGK, and M27 beta, in leukaemic patients and normal controls. As expected, blast cells from 67 of 68 analysable samples (99%) were monoclonal or had a skewed X-inactivation pattern. A skewed pattern in remission was also found in 26 of 77 patients (34%), proportion only slightly greater than control (16/75, 21%). In 7 of 10 patients with a skewed pattern in myeloid cells there was similar skewing in the T cells, which is compatible with the concept of a constitutively skewed X-chromosome inactivation pattern of haemopoietic cells in these patients. Our study illustrates the difficulty of interpreting clonality in individual tumour samples and emphasises the importance of comparisons with non-malignant tissue of the same cell type from that individual and from normal control populations.

Assessment of X-chromosome inactivation patterns using the hypervariable probe M27 beta in normal hemopoietic cells and acute myeloid leukemic blasts

The value of the hypervariable X-linked probe M27 beta for use in the analysis of X-chromosome inactivation patterns in normal blood and bone marrow cells and in the assessment of clonality in acute myeloid leukemia (AML) blast cells has been determined. By electrophoresing samples for 30 h, heterozygosity of the M27 beta locus was demonstrable in 324/415 females (78%) and this value could be increased to 93% by electrophoresing for 50 h. Determination of the X-chromosome inactivation patterns in blood and bone marrow samples from hematologically normal females was possible in approximately 90% of heterozygous individuals. The X-chromosome inactivation ratios obtained with M27 beta were comparable with phosphoglycerate kinase or hypoxanthine phosphoribosyl transferase in 46 individuals heterozygous for one of these genes in addition to M27 beta. The results obtained were closely correlated (r = 0.89). In 21 of 35 (60%) AML blasts, however, there was hypermethylation of the M27 beta locus and clonal analysis was not possible. Hypermethylation was not related to FAB type.

X-chromosome inactivation patterns using HPRT and PGK polymorphisms in haematologically normal and post-chemotherapy females

Clonal analysis using X-chromosome inactivation patterns of the hypoxanthine phosphoribosyl transferase (HPRT) and phosphoglycerate kinase (PGK) genes has been used in the assessment of haematological malignancies. In many analyses normal haemopoietic tissue or constitutive DNA from each individual has not been readily available and it has been assumed that when the ratio of the percentage of the two alleles remaining after differential methylation enzyme digestion is greater than 3/1 then an oligoclonal/monoclonal population of cells is present. We report here in a study of 65 haematologically normal females informative for HPRT or PGK that 15 (23%) had a skewed Lyonization pattern with clonal analysis ratios greater than 3/1. No correlation was observed between age and the X-chromosome inactivation pattern to suggest the detection of pre-malignant clonal states. The incidence was similar in 23 females who had received conventional combination chemotherapy, with six women (26%) having a ratio greater than 3/1. In the majority of individuals therefore, conventional chemotherapy does not appear to alter the patient's X-chromosome inactivation pattern. These results indicate that caution must be exercised in interpreting the results of clonal analysis of any given individual.