¹Departments of Medicine and Hematology, Glasgow Royal Infirmary, Glasgow, UK

²Terry Fox Laboratory, British Columbia Cancer Agency Vancouver, Canada

³University of British Columbia, Vancouver, Canada

Correspondence to: C J Eaves, Terry Fox Laboratory, 601 West 10th Avenue, Vancouver, BC, Canada V5Z 1L3; Fax: (604) 877-0712

Chronic myeloid leukemia (CML) has been studied intensively for many years; yet its treatment remains problematic and its biology remains elusive. In chronic phase, the leukemic clone appears to be maintained by a small number of BCR-ABL-positive hematopoietic stem cells that differentiate normally and amplify slowly. In contrast, as these cells enter the intermediate stages of lineage restriction, their progeny are selectively expanded and generate an enlarged pool of neoplastic progenitors. Recent analyses of purified subsets of primitive CML cells have provided a coherent explanation for this dichotomous behavior of BCR-ABL-positive stem and progenitor cells based on the discovery of an unusual autocrine IL-3/G-CSF mechanism activated in them. This only partially counteracts in vivosignals that maintain normal stem cells in a quiescent state but, when active in CML stem cells, promotes their differentiation in favor of their self-renewal. In more differentiated CML progenitors, the same mechanism has a more potent mitogenic effect which is then extinguished when the cells enter the terminal stages of differentiation. Thus, further expansion of the clone is limited until inevitably additional mutations are acquired that further distort or override the regulatory mechanisms still operative in the chronic phase.

Leukemia (2002) 16, 549-558. DOI: 10.1038/sj/leu/2402444

chronic myeloid leukemia; stem cells; BCR-ABL; autocrine

Introduction

The chronic phase of chronic myeloid leukemia (CML) is characterized by the presence of an abnormally amplified multilineage clone that produces functionally normal mature blood cells of all types except for those of the T lineage.¹ However, disease symptoms usually develop as a result of the markedly elevated output of granulocytes that is the biologic hallmark of this disease. The expanded clone responsible is believed to be initiated by the chance occurrence in a hematopoietic stem cell of a rare mutational event in which the majority of the 3' end of the c-abl gene from chromosome 9 replaces the 3' end of the BCR-1 gene on chromosome 22.^2,3 A characteristically shortened derivative chromosome 22 (known as the Philadelphia chromosome or 'Ph'), thus created by this reciprocal exchange of chromosome ends, has been recognized for four decades as diagnostic of >90% of cases with features associated with CML.^4,5 The BCR-ABL fusion gene harbored on the Ph is transcriptionally active and encodes a 210 kDa oncoprotein (p210^BCR-ABL) that has transforming activity in both primary mouse bone marrow cells and established cell lines.^6,7,8,9,10 P210^BCR-ABL differs from the normal p145^c-ABL in its preferential location in the cytoplasm and its constitutively elevated tyrosine kinase activity. Mutational analyses suggest that both of these properties of p210^BCR-ABL are critical to its transforming activity.^11,12 Recent findings indicate that the formation of the BCR-ABL gene may not, alone, be sufficient to produce disease. This is suggested by the finding that cells containing detectable levels of BCR-ABL transcripts exist in the blood of a high proportion of normal adults.^13,14 In addition, even when a BCR-ABL⁺ clone does expand to the point where a disease entity is recognized (ie to an extent that allows >10¹² cells to be produced daily), this occurs only after a prolonged latent period (on average, 6-7 years, as calculated from the follow-up of survivors of the atomic bomb explosions in Japan at the end of World War II¹⁵). Thus the net growth advantage that expression of BCR-ABL confers on a CML stem cell must be relatively subtle. In light of these considerations, it is perhaps not surprising that the identification of therapies able to effectively target the CML stem cell compartment have remained elusive. In this review, we will summarize some of our recent studies of these cells that provide new insights into their deregulated growth and suggest new avenues for exploration in a clinical setting.

The Ph⁺ BCR-ABL⁺ stem cell: basic concepts

Cell of origin of the disease

Although the granulocytic lineage is the one that is typically the most affected in CML, an accompanying hypercellular marrow containing an increased number of erythroblasts and megakaryocytes was recognized more than 50 years ago as a typical feature of CML. Indeed, this latter observation formed the original basis of the concept that CML arises in a pluripotent stem cell,¹⁶ later confirmed by a variety of elegant studies using cytogenetic, FISH and X-linked markers of clonal expansion.^{4,17,18,19,20,21,22} These showed that erythroid cells, megakaryocytes, and B and T cell precursors, are all commonly produced within the neoplastic clone. More recently, Gunsilius et al²³ reported that the BCR-ABL gene could be found in endothelial cells produced from a CML patient's cells. If confirmed, these data might place the level of the BCR-ABL mutation in an even earlier cell type, ie one with endothelial as well as hematopoietic differentiation potential.

Morphology and phenotype of the neoplastic cells

CML results in the production of granulocytes, red cells and platelets that, although derived from Ph⁺/BCR-ABL⁺ precursors, are morphologically indistinguishable from their normal counterparts. A definitive diagnosis of CML therefore relies on the cytogenetic or molecular demonstration in blood or marrow of either the Ph chromosome or a BCR-ABL gene rearrangement. Similarly, since the cells in the Ph⁺/BCR-ABL⁺ clone undergo normal differentiation, analysis of changes in cell markers that accompany this process in normal cells do not allow individual Ph⁺ and Ph^- elements to be distinguished from one another at any stage of differentiation. This also applies to functionally defined progenitors, including those classified as various types of colony-forming cells (CFCs), as well as even more primitive cells detected by their ability to generate progeny CFCs after 5-8 weeks in co-cultures containing stromal cells and referred to as long-term culture-initiating cells (LTC-ICs).^24,25 However, some differences are seen at the progenitor population level due to the fact that a much higher proportion of primitive Ph⁺ cells are actively cycling relative to their normal (Ph^-) counterparts. For example, most primitive CML progenitors are characterized by an upregulated expression of HLA-DR^26,27,28 which is known to represent an early response to growth factor activation by their predominantly quiescent, primitive normal counterparts.²⁹

Transplantability

The ability of Ph⁺ CML cells to engraft humans was first discovered by accident following the transfusion of leukopenic patients with freshly leukapheresed blood cells from donors with CML at a time when leukemic stem cells were not known to be present at high levels in these collections.^30,31 More recently, further evidence that CML cells can engraft has been provided by analysis of cells from CML recipients of retrovirally marked autografts at the time of subsequent disease relapse.³² However, none of these studies provided any quantitative indication of the relative numbers of transplantable CML stem cells in a given patient nor whether their engraftment properties are affected by the disease process. Interestingly, in other studies with similar types of autotransplants, prolonged disease-free outcomes have been encountered,^33,34 suggesting that the number of CML 'stem' cells in many patients may actually be quite small and/or that many of these cells are not readily transplantable. Attempts have also been made to produce the disease in xenogeneic hosts by transplantation of immunodeficient mice with CML patients' cells. Thus far, engraftment of irradiated SCID or NOD/SCID mice with Ph⁺/BCR-ABL⁺ cells from chronic phase CML patients has been found to require the transplantation of very large inocula, consistent with experience in patients.^35,36,37 Moreover, in most cases, a selective engraftment of normal (Ph^-) cells has been seen, particularly when the more primitive human cells detectable in these mice are analyzed.^{35,36,37,38,39} Thus, although it is clear that some CML cells can produce leukemic progeny following their intravenous injection, none of the xenotransplant models used, to date, have lived up to initial hopes to consistently transfer the human disease to animal hosts at frequencies or levels adequate for their further study on a routine basis.

More recently, it has been found that the transplantability of normal hematopoietic stem cells can fluctuate according to their cell cycle status.^40,41 This may explain, at least in part, the relatively poor engraftment obtained with transplants of primitive CML cells. Only one-twentieth of the potentially transplantable stem cells in normal human cord blood or fetal liver actually seed to the marrow of NOD/SCID mice,^42,43 and the ability of even this fraction to engraft markedly decreases both as they enter G₁ from G₀ and then again as they transit S/G₂/M.^41,44 The proportion of many primitive types of CML progenitors that are proliferating is known to be much higher than is the case for their normal counterparts.²⁵ It would thus be anticipated that this would also be true for the CML stem cell population. Accordingly, these cells would be expected, on this basis alone, to appear selectively compromised in their ability to engraft NOD/SCID mice. Conversely, the detection of any Ph⁺ progeny in NOD/SCID mice might then indirectly infer the presence, in chronic phase CML patients, of rare G₀/G₁ stem cell populations. As discussed below, we have recently shown that such G₀/G₁ CML cells can be found in all patients examined to date.

Deregulated hematopoietic cell functions in CML

The BCR-ABL fusion protein has a wide range of substrates, and it is likely that many of these contribute to producing the altered behavior of Ph⁺/BCR-ABL⁺ stem cells. Many theories have been formulated to explain the myeloid expansion that is typical of chronic phase CML. These emphasize to varying degrees the abnormalities in control of proliferation, apoptosis, cellular adhesion and the organization of the cytoskeleton that have been noted in various studies of BCR-ABL expressing cells. As noted above, marked numerical perturbations occur in the subpopulations that make up the CML clone as the neoplastic stem cells begin to differentiate. Of interest, the leukemic stem cell compartment appears to remain quite small but entry into the intermediate compartments of progenitors (of all lineages) is accompanied by a rapid expansion of these cells. However, ultimately an increased output of mature cells is largely restricted to the granulocyte lineages. It is therefore likely that the cell context in which BCR-ABL is expressed has a significant impact on how cellular behavior is altered. In this review, we have focused on mechanisms relevant to the deregulated turnover that is characteristic of the most primitive elements in the CML clone and their ability to generate an expanded population of intermediate progenitor types.

Proliferative activity and stem cell turnover

Evidence of a perturbation of the cell cycle regulation of Ph⁺/BCR-ABL⁺ stem cells is inferred from changes that affect neoplastic cells detected by surrogate assays for these and closely related populations of primitive Ph⁺/BCR-ABL⁺ cells. These include assays for LTC-ICs and CFCs defined either by their high proliferative potential (HPP)^25,45 and/or their multipotent differentiation ability (referred to as colony-forming units-granulocyte/erythroid/macrophage/megakaryocyte, or CFU-GEMMs.^25,45,46 However, it has been a challenge to acquire data even on these cell types, both because of their relatively low frequency in the blood and marrow of CML patients and because of their lack of a unique phenotype, which precludes their direct distinction or separation. Also, because of the frequent prevalence of residual normal cells in the most primitive compartments and the wide variation in proportions of different types of normal and leukemic cells in individual CML patients, the progenitors investigated in any given experiment (or their clonal progeny) must be genotyped. This, in turn, limits the number of samples that can be analyzed and, hence the ultimate precision of the data obtained.

Nevertheless, the results of several approaches have provided strong support for the view that the turnover of BCR-ABL⁺ stem cells, like that of their immediate progeny, is deregulated. The first approach involves exposing the cells to an agent that is specifically toxic to S-phase cells (eg high specific activity of ³H-thymidine or hydroxyurea) for 15-30 min. This methodology was first used to demonstrate that the HPP-CFCs in the marrow of normal adults are largely quiescent, as are all of the CFCs normally found in the blood. In contrast, analogous measurements applied to CFCs from CML patients showed that the majority of the leukemic CFCs, regardless of their location (in bone marrow or blood), are at any given moment in cycle.⁴⁵ Later, a more prolonged ³H-thymidine suicide procedure was used to show that in normal adults, most (>90%) of the LTC-ICs are quiescent, whereas in CML, most (>90%) of their leukemic counterparts are proliferating.^47,48

A second approach to assessing the cell cycle status of primitive hematopoietic cells has exploited FACS technology for isolating viable cells in G₀, G₁ and S/G₂/M based on differences in their DNA and RNA content as visualized by co-staining with Hoechst 33342 and Pyronin Y (Figure 1). Such experiments have confirmed that there is an increased proportion of primitive leukemic cells (CFCs and LTC-ICs) in chronic phase CML patients that are cycling (in S/G₂/M).⁴⁸ Unexpectedly, these analyses also provided the first direct evidence that there is, in addition, a consistently detectable fraction of G₀ CML cells within these same primitive leukemic cell populations. Furthermore, using a high resolution procedure for tracking cell division, it was possible to isolate a subset of Ph⁺/BCR-ABL⁺ CFCs that remained viable and functionally unaltered, but did not enter mitosis for 5 days in the presence of a growth factor cocktail that is potently mitogenic for primitive normal cells (Figure 2). These latter studies thus demonstrated the depth of quiescence that a proportion of primitive CML progenitors reach in vivo that, nevertheless, can be reversed after several days of growth factor stimulation in vitro. Interestingly, single cell RT-PCR analyses showed that the primitive quiescent CML cells (isolated by either strategy) expressed BCR-ABL transcripts. This latter observation suggests that the BCR-ABL oncoprotein alone may not be sufficient in these cells to counteract all of the inhibitory cues emanating from the environment in vivo. Indeed, such a possibility was raised a decade ago from studies indicating that exposure of primitive cycling CML CFCs to non-toxic concentrations of TGF- in vitro can transiently arrest their entry into S-phase.⁴⁹

We also found that relatively low numbers (10⁵) of primitive CD34⁺ G₀ CML cells were able to engraft irradiated NOD/SCID mice and produce detectable numbers of leukemic progeny 6-8 weeks later.⁴⁸ This indicates that the quiescence of these transplantable leukemic cells is also reversible in vivo, thus pointing to the likely importance of these cells in sustaining and spreading the disease in patients. The consistently demonstrable presence of this previously unrecognized compartment of primitive, transiently quiescent CML cells in chronic phase patients provides a possible explanation for both the historic failure of intensive (but non-myeloablative) chemotherapy regimens to eradicate chronic phase disease,^50,51 and the occasional observation of late relapse many years post transplant.^52,53 On the other hand, these clinical findings could also have other explanations and neither permits any direct quantitative inferences to be made about the frequency of primitive neoplastic cells that are quiescent in an individual patient. The availability of methods that allow the reproducible isolation of these cells and their further characterization should now facilitate investigation of the role of these primitive quiescent leukemic cells in disease progression and response to therapy (see below).

A third approach to evaluating normal and leukemic stem cell dynamics in CML patients has involved a comparison of telomere length measurements performed on their granulocytes (which are typically >95% leukemic) and on those from age-matched normal individuals. Telomere length in normal somatic cells decreases with each cell division and changes in telomere length can thus serve as an indicator of their mitotic history, with shortening to a critical level acting as a checkpoint to signal senescence.⁵⁴ The main factor that elongates telomeres is the enzyme telomerase (TERT), whose expression is normally restricted to primitive cell types, particularly cells of the germ line.^55,56 However, low levels of TERT have also been documented in later types of primitive hematopoietic cells⁵⁷ and in certain lymphocytes.⁵⁸ In tissue culture, when human cell lines are immortalized, telomerase expression is frequently seen to be upregulated and high levels of telomerase are characteristic of most human cancers.⁵⁷ Conversely, forced overexpression of telomerase can extend the number of cell divisions possible and contribute to full malignant transformation, suggesting a causal relationship between telomere shortening, senescence and the circumvention of this response by malignant cells.^59,60,61,62 Interestingly, telomere length comparisons between normal and CML cells have only recently become informative. This is largely due to the very wide variation in telomere length seen between normal individuals, even when those of the same age are compared,^63,64 compounded by the imprecise methodology available until recently for making telomere length measurements. Techniques that utilize flow fluorescence in situ hybridization (flow-FISH) or quantitative FISH⁶⁵ have provided a major advance in this regard as has been their application to assessments of either very large numbers of individuals or, even better, of non-clonal (eg T-lineage) cells from the same patient.

Telomere length measurements performed on granulocytes from large numbers of normal individuals of different ages now suggest that, in the adult, all of the stem cells from which they originate will divide, on average, only once every 1-2 years.⁶³ Several studies have now reported that the leukemic cells (which are predominantly mature myeloid cells in chronic phase CML patients) have shorter telomeres than normal granulocytes or bone marrow cells.^66,67,68 In the study that we undertook, telomere length measurements were obtained on peripheral blood leukocyte cells from >150 CML samples and from >300 normal individuals. In some of the CML samples, purified endogenous normal (T) cells were also analyzed. These normal vs CML comparisons revealed an average overall difference in telomere length between the normal and leukemic cells of ~1 kb, indicative of a somewhat faster turnover of the leukemic stem cells (Figure 3). Assuming no change in other kinetic parameters, this difference can be estimated to represent ~10 more self-renewal divisions than would be executed by normal stem cells between the time the clone was initiated and the time of diagnosis (ie ~1-2 extra divisions per year over an estimated average 6 to 7 year latent period).

An even greater loss of telomere length is seen when blast phase CML cells are analyzed by quantitative methods.⁶⁹ In fact, our finding of a significant correlation between reduced telomere length and the proportion of patients entering blast phase disease within 2 years of the time when telomere length measurements were performed,⁶⁷ provides strong evidence that telomere shortening may be a rate-limiting step contributing to the onset of blast crisis. More recently, a significant correlation between the degree of telomere shortening at diagnosis and the prognostic index, as measured by the Hasford score, has been demonstrated (MW Drummond, T Holyoake, unpublished findings). Furthermore, in these latter studies, the degree of telomere shortening could be correlated with disease progression from early chronic phase, to accelerated phase and finally into blast crisis (Figure 4).

Taken together, these data not only support previous evidence that CML stem cells cycle in vivo more actively than their Ph^- counterparts, they strongly indicate that telomere length may serve as a useful surrogate marker of disease progression. They also support the concept that reduction of telomere length to a critical point may drive the selection of a clone of blast phase cells in which expression of telomerase has been upregulated. It is thus interesting to speculate that strategies to inhibit the telomerase activity in these cells might have therapeutic potential by inducing their senescence.⁷⁰

In vitro response to STI571

STI571 is a small molecule inhibitor, with great specificity for the p210^BCR-ABL tyrosine kinase. In vitro, this agent selectively suppresses the growth of primary CML CFCs and of BCR-ABL⁺ cell lines. In phase I-II clinical trials, initial results with single agent STI571 therapy have appeared far superior to those achievable using any other non-transplantation treatment modality, to date, leading to the reappearance of Ph^-/BCR-ABL^- hematopoiesis in >50% of chronic phase patients within 1 year of continuous treatment.⁷¹ These clinical results, although preliminary, have raised much interest in the possibility that STI571 will prove to be an important advance in the long-term treatment of patients with CML. However, given previous knowledge of the slow but eventual ability of a small Ph⁺/BCR-ABL⁺ stem cell compartment to re-initiate disease even after several years of apparent normal hematopoiesis, it could be particularly important to understand the relative sensitivity of the quiescent CML progenitor cells to STI571.

As a first step towards this goal, we have recently begun to compare the ability of STI571 to initiate an apoptotic response in cycling vs quiescent CML cells. Initial results show that a 6-day exposure in serum-free culture to 10 M STI571 is, as expected,^10,72 sufficient to kill >90% of the CD34⁺lin^- CML population, the majority of which are proliferating cells. In contrast, when the small quiescent Ph⁺/BCR-ABL⁺ subset was selectively isolated (by their resistance to cell cycle entry for 4 days in vitro), these cells were found to be fully resistant to the same STI571 concentration (Figure 5).⁷³ We therefore anticipate that STI571, as a therapeutic single agent, will be unlikely to achieve long-term cures unless administered over very long periods of time, and that the use of additional agents that target the primitive quiescent leukemic cells more effectively will prove necessary to fully eradicate the disease. The recent reports of WT-1 as a relevant antigen in CML⁷⁴ and the proposed role of WT-1 to induce cellular quiescence in primitive precursors⁷⁵ suggest that immunotherapy against this antigen may be an interesting candidate.⁷⁶

Another mechanism proposed for the resistance of some CML cells to STI571 in vivo is through binding to plasma alpha-1-acid glycoprotein (AGP). However, we have also now shown that, although plasma levels of this glycoprotein are significantly higher in CML patients than in normal controls, purified AGP from CML patients' plasma does not block the effect of STI571 on either Ph⁺ cell lines or primary CD34⁺lin^- cells from chronic phase patients.⁷⁷ These results cast doubt on the likelihood that AGP will prove a useful target for enhancing the anti-leukemic activity of STI571 in vivo.

BCR-ABL-induced growth factor independence

As discussed previously, the increased turnover of primitive CML cells has been recognized for many years. Nevertheless, consensus has not been reached as to the precise molecular mechanism responsible. In part, this may be due to the relative paucity of data from studies of primitive CML cells from patients with chronic phase disease in spite of an abundance of information from studies of BCR-ABL-transduced cell lines or various Ph⁺/BCR-ABL⁺ cell lines established from samples obtained from CML patients in blast crisis, or from mature CML cells that are actually incapable of further division. Interestingly, one of the earliest findings from studies of BCR-ABL-transduced hematopoietic cell lines was the activation of an autocrine mechanism which could confer partial or complete growth factor autonomy in cells that were previously factor-dependent.^78,79 However, until recently, investigators have struggled to extrapolate these findings to primary CML cells, at least for patients in the chronic phase of the disease. Moreover, leukemic CFCs were well known to exhibit a normal dose response to exogenous growth factor stimulation when assayed under standard in vitro conditions. Many groups have therefore concluded that the growth requirements of CML progenitors are unaltered and a role for autocrine mechanisms has been largely and generally discounted as irrelevant to the pathogenesis of naturally arising CML. For the same reasons, the acquired factor independence of many of the BCR-ABL⁺ cell lines available has focused attention on the anti-apoptotic activity of the BCR-ABL oncoprotein. Although the results of experiments to examine this potential role of p210^BCR-ABL have been conflicting, at least five studies comparing the effect of growth factor withdrawal on normal vs leukemic CFC survival have shown the response of the leukemic cells to be superior.^{80,81,82,83,84} In the three studies from our own group, the number of CFCs was found not only to be maintained upon factor withdrawal but also to have increased several-fold within 3 weeks of culture in serum-free medium. Importantly, this growth factor-independent proliferation was shown to be a property exclusive to the primitive Ph⁺/BCR-ABL⁺ cells and is lost when they reach the stage where they lose expression of CD34. Conversely, this capacity for autonomous growth is most pronounced in the most primitive subsets of Ph⁺/BCR-ABL⁺ CD34⁺ cells, as identified by their lack of expression of CD45RA, CD71^-, or CD38^- (Figure 6). The unexpected but convincing picture of primitive CML cell factor independence that emerged from these experiments prompted us to undertake a renewed search for evidence of an autocrine mechanism. The combined results of subsequent RT-PCR analyses, high sensitivity ELISAs, intracellular flow cytometry and bioactivity assays provided definitive evidence of the consistently activated production of IL-3 and G-CSF (but not several other growth factors) in CD34⁺ cells isolated directly from chronic phase patients. Moreover, the activation of IL-3 and G-CSF in these cells was shown to mirror precisely their differentiation stage-specific growth factor independence (Figure 7). Antibody neutralization experiments confirmed the dependence of the primitive leukemic cell proliferation on this autocrine loop although intracellular binding of these two growth factors to receptors on the endoplasmic reticulum likely contributes to the general inability of exogenous anti-IL-3 and anti-G-CSF antibodies to completely abrogate the autonomous growth behavior.

The constitutive production of IL-3 and G-CSF by primitive CML cells fits remarkably with previous Ph⁺ cell line work, with the clinical picture of CML and with the known stage-specific responses of normal cells to high levels of these cytokines. IL-3, as a single factor, has been shown to support the survival of primitive normal bone marrow cells and to activate proliferation of quiescent LTC-ICs and CFCs.⁸⁵ However, in the absence of adequate co-stimulation by FL and SF, excess IL-3 delivers a potent differentiation-inducing stimulus to CD34⁺CD38^- cells including the LTC-ICs. Transplantation of murine bone marrow cells, transgenically engineered or retrovirally transduced to produce IL-3 induces a myeloproliferative syndrome that shares many similarities with human CML.^86,87,88 Likewise, administration of pharmacological doses of IL-3 enhances the production of basophils and mast cells,^89,90 and G-CSF strongly promotes terminal granulopoiesis in vivo,^90,91 both of which are features of CML. Enhanced proliferation of the Ph⁺ clone due to autocrine growth factor production has thus emerged as a potentially important mechanism underlying the 'apparent' growth factor independence of primitive CML cells and the accelerated expansion of the clone as cells transit the intermediate stages of differentiation.

These key observations have also allowed previous, apparently discrepant, findings regarding the factor dependence of CML CFCs to be explained. CFC assays rely on the ability of CD34⁺ cells to initiate cell division as single cells and then produce colonies of terminally differentiating CD34^- cells by executing their final amplifying divisions. Thus, initiation of colony formation by CML progenitors plated in the absence of growth factors would have to occur in the absence of any paracrine stimulation and the majority of the CML cells subsequently generated in each colony would be relatively late types of factor-dependent cells. Thus, in standard semi-solid assays of purified progenitors, only a few abortive colony starts would be expected in the absence of added growth factors, as was originally observed by Strife et al.⁸⁰

Signaling changes induced by BCR-ABL

BCR-ABL-expressing cells have also been found to show many similarities to those induced by IL-3 stimulation⁹² or forced IL-3 overexpression.^86,87,88 In particular, the increased phosphorylation of STAT5, previously thought to be an immediate function of the BCR-ABL oncoprotein, has now been shown in primary CML CD34⁺ cells to occur largely as a consequence of BCR-ABL-induced activation of IL-3 auto-stimulation, since STAT5 activation in these cells could be blocked by neutralization of IL-3 activity.⁸³ We have also shown that the primitive, quiescent Ph⁺/BCR-ABL⁺ cells, that are present in most CML patients, are silent with respect to detectable levels of IL-3 and G-CSF mRNA, although BCR-ABL transcripts are present and both IL-3 and G-CSF transcripts reappear when the cells spontaneously re-enter the cell cycle in vitro.²⁸ The fact that the primitive, quiescent CML cells remain viable in the absence of exogenous IL-3 or G-CSF makes two strong inferences. The first is that BCR-ABL alone may provide a sufficient anti-apoptotic signal to keep these cells alive under conditions of factor deprivation that their normal counterparts may not withstand. The second is that the autocrine IL-3 (and G-CSF) mechanism may be required for their factor-independent proliferation, as indicated by our antibody neutralization studies.

Subsequent studies have demonstrated that the production of IL-3 is consistently activated in murine bone marrow cells transduced with a BCR-ABL retrovirus.^93,94 We have confirmed these findings and have further shown that this occurs in very primitive cells.⁹⁵ Furthermore, activation of expression of both IL-3 and G-CSF in this model occurs a few days after transduction but following the onset of BCR-ABL expression (unpublished findings), again consistent with a causal linkage of factor-independent growth and the autocrine IL-3/G-CSF mechanism secondary to BCR-ABL expression. At present, very little else is known about the mechanism by which the BCR-ABL oncoprotein elicits this autocrine activity, except that its SH2 domain appears to be important for this to occur in vitro.⁹⁶

Recently, it has been shown that BCR-ABL-transduced murine myeloid cells that are unable to produce IL-3 can still display leukemic activity.⁹⁷ Although initially this may seem to contravene the conclusions reached above, in fact this is not surprising at all. There are a number of signaling pathways which BCR-ABL is known to activate that may be candidates for promoting primitive CML cell viability independently (as well as downstream) of IL-3 or G-CSF, eg the ras, PI3K/Akt and Jak/Stat pathways. In addition, we have found BCR-ABL-activated GM-CSF as well as G-CSF to exert potent paracrine effects in BCR-ABL-transduced normal and IL-3^-/- mouse bone marrow cells.⁹⁵ Furthermore, in the murine models employed, the levels of BCR-ABL transcription and translation are likely to be greater than in primary chronic phase CML cells due to the use in transduction models of a powerful promoter in the retroviral LTR. In addition, in murine models, leukemogenic activity is measured by the generation of a rapidly fatal, polyclonal, 'CML-like' syndrome within a few weeks after the transplantation of millions of transduced cells into myeloablated recipients. This contrasts markedly with the long latent period of many years for the clonal disorder that produces CML in people. The occurrence, in this model, of acute B cell leukemias in addition to 'CML-like' disease, may also suggest a number of different target cells, some of which may not be as early in the stem cell hierarchy as the pluripotent cell mutated in primary disease. Our finding of a discrepancy between the presence of BCR-ABL transcripts and IL-3/G-CSF transcripts in the quiescent stem cell fraction²⁸ suggests that IL-3 and G-CSF activation may depend on a threshold level of BCR-ABL transcripts within the cells. For example, in the presence of low levels of BCR-ABL transcripts (and p210^bcr-abl), IL-3 and G-CSF are not activated and the cells remain viable but do not divide. As BCR-ABL transcripts rise above a certain threshold, transcription of the IL-3 and G-CSF genes is activated and this is accompanied by the entry of these cells into cycle. Under conditions where BCR-ABL transcripts are more elevated (eg in retrovirally transduced or actively proliferating CD34⁺ CML cells), transcription of both IL-3 and G-CSF (and even GM-CSF) is never shut off and a state of more sustained proliferation may be achieved. Thus the autocrine IL-3/G-CSF mechanism that is characteristic of primitive chronic phase CML cells may be key to their slow but inexorable expansion that results in a disease with a course quite distinct from the human Ph⁺ acute leukemias. Future investigation of both human and murine models of CML using various transduction and transplant strategies should allow further examination of these possibilities as well as providing more refined models for evaluating new therapeutics.

Acknowledgements

TLH is supported by a Leukemia Research Fund (LRF) Senior Lectureship. MWD is supported by a LRF Clinical Training Fellowship. Much of the work of the authors reported was obtained with support from the Sylvia Aitken Trust, the Scottish National Blood Transfusion service, the National Cancer Institute of Canada (with funds from the Terry Fox Run and the Canadian Cancer Society), as well as the Leukemia Research Fund of Canada.

1 Fialkow PJ, Jacobson RJ, Papayannopoulou T. Chronic myelocytic leukemia: clonal origin in a stem cell common to the granulocyte, erythrocyte, platelet and monocyte/macrophage. Am J Med 1977; 63: 125-130. MEDLINE

2 Groffen J, Stephenson JR, Heisterkamp N, De Klein A, Bartram CR, Grosveld G. Philadelphia chromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22. Cell 1984; 36: 93-99. MEDLINE

3 Shtivelman E, Lifshitz B, Gale RP, Canaani E. Fused transcript of abl and bcr genes in chronic myelogenous leukaemia. Nature 1985; 315: 550-554. MEDLINE

4 Raskind WH, Fialkow PJ. The use of cell markers in the study of human hematopoietic neoplasia. Adv Cancer Res 1987; 49: 127-167. MEDLINE

5 Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature 1973; 243: 290-293. MEDLINE

6 Daley GQ, Van Etten RA, Baltimore D. Induction of chronic myelogenous leukemia in mice by the P210^bcr/abl gene of the Philadelphia chromosome. Science 1990; 247: 824-830. MEDLINE

7 Heisterkamp N, Jenster G, ten Hoeve J, Zovich D, Pattengale PK, Groffen J. Acute leukemia in bcr/abl transgenic mice. Nature 1990; 344: 251-253. MEDLINE

8 Lugo TG, Pendergast AM, Muller AJ, Witte ON. Tyrosine kinase activity and transformation potency of bcr-abl oncogene products. Science 1990; 247: 1079-1082. MEDLINE

9 Evans CA, Owen-Lynch J, Whetton AD, Dive C. Activation of the Abelson tyrosine kinase activity is associated with suppression of apoptosis in hemopoietic cells. Cancer Res 1993; 53: 1735-1738. MEDLINE

10 Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S, Zimmermann J, Lydon NB. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 1996; 2: 561-566. MEDLINE

11 Konopka JB, Witte ON. Detection of c-abl tyrosine kinase activity in vitro permits direct comparison of normal and altered abl gene products. Mol Cell Biol 1985; 5: 3116-3123. MEDLINE

12 Van Etten RA, Jackson P, Baltimore D. The mouse type IV c-abl gene product is a nuclear protein, and activation of transforming ability is associated with cytoplasmic localization. Cell 1989; 58: 669-678. MEDLINE

13 Biernaux C, Loos M, Sels A, Huez G, Stryckmans P. Detection of major bcr-abl gene expression at a very low level in blood cells of some healthy individuals. Blood 1995; 86: 3118-3122. MEDLINE

14 Bose S, Deininger M, Gora-Tybor J, Goldman JM, Melo JV. The presence of typical and atypical BCR-ABL fusion genes in leukocytes of normal individuals: biologic significance and implications for the assessment of minimal residual disease. Blood 1998; 92: 3362-3367. MEDLINE

15 Ichimaru M, Ishimaru T, Mikami M, Yamada Y, Ohkita T. Incidence of leukemia in a fixed cohort of atomic bomb survivors and controls, Hiroshima and Nagasaki October 1950-December 1978. Technical Report RERF TR 13-81 Radiation Effects Research Foundation: Hiroshima, 1981.

16 Dameshek W. Some speculations on the myeloproliferative syndromes. Blood 1951; 6: 372-375.

17 Whang J, Frei E III, Tjio JH, Carbone PP, Brecher G. The distribution of the Philadelphia chromosome in patients with chronic myelogenous leukemia. Blood 1963; 22: 664-673.

18 Rastrick JM, Fitzgerald PH, Gunz FW. Direct evidence for presence of Ph¹ chromosome in erythroid cells. Br Med J 1968; 1: 96-98. MEDLINE

19 Tough IM, Jacobs PA, Court Brown WM, Baikie AG, Williamson ERD. Cytogenetic studies on bone-marrow in chronic myeloid leukaemia. Lancet 1963; 1: 844-846.

20 Martin PJ, Najfeld V, Hansen JA, Penfold GK, Jacobson RJ, Fialkow PJ. Involvement of the B-lymphoid system in chronic myelogenous leukaemia. Nature 1980; 287: 49-50. MEDLINE

21 Bernheim A, Berger R, Preud'Homme JL, Labaume S, Bussel A, Barot-Ciorbaru R. Philadelphia chromosome positive blood B lymphocytes in chronic myelocytic leukemia. Leuk Res 1981; 5: 331-339. MEDLINE

22 Takahashi N, Miura I, Saitoh K, Miura AB. Lineage involvement of stem cells bearing the Philadelphia chromosome in chronic myeloid leukemia in the chronic phase as shown by a combination of fluorescence-activated cell sorting and fluorescence in situ hybridization. Blood 1998; 92: 4758-4763. MEDLINE

23 Gunsilius E, Duba H-C, Petzer AL, Kahler CM, Grunewald K, Stockhammer G, Gabl C, Dimhofer S, Clausen J, Gastl G. Evidence from a leukaemia model for maintenance of vascular endothelium by bone-marrow-derived endothelial cells. Lancet 2000; 355: 1688-1691. MEDLINE

24 Udomsakdi C, Eaves CJ, Swolin B, Reid DS, Barnett MJ, Eaves AC. Rapid decline of chronic myeloid leukemic cells in long-term culture due to a defect at the leukemic stem cell level. Proc Natl Acad Sci USA 1992; 89: 6192-6196. MEDLINE

25 Eaves C, Cashman J, Eaves A. Defective regulation of leukemic hematopoiesis in chronic myeloid leukemia. Leuk Res 1998; 22: 1085-1096. MEDLINE

26 Udomsakdi C, Eaves CJ, Lansdorp PM, Eaves AC. Phenotypic heterogeneity of primitive leukemic hematopoietic cells in patients with chronic myeloid leukemia. Blood 1992; 80: 2522-2530. MEDLINE

27 Verfaillie CM, Miller WJ, Boylan K, McGlave PB. Selection of benign primitive hematopoietic progenitors in chronic myelogenous leukemia on the basis of HLA-DR antigen expression. Blood 1992; 79: 1003-1010. MEDLINE

28 Holyoake TL, Jiang X, Jorgensen HG, Graham S, Alcorn MJ, Laird C, Eaves AC, Eaves CJ. Primitive quiescent leukemic cells from patients with chronic myeloid leukemia spontaneously initiate factor-independent growth in vitro in association with up-regulation of expression of interleukin-3. Blood 2001; 97: 720-728. Article MEDLINE

29 Srour EF, Brandt JE, Leemhuis T, Ballas CB, Hoffman R. Relationship between cytokine-dependent cell cycle progression and MHC class II antigen expression by human CD34+HLA-DR - bone marrow cells. J Immunol 1992; 148: 815-820. MEDLINE

30 Levin RH, Whang J, Tjio JH, Carbone PP, Frei E III, Freireich EJ. Persistent mitosis of transfused homologous leukocytes in children receiving antileukemic therapy. Science 1963; 142: 1305-1311.

31 Graw RGJ, Buckner CD, Whang-Peng J, Leventhal BG, Kruger G, Berard C, Henderson ES. Complication of bone-marrow transplantation. Graft-versus-host disease resulting from chronic-myelogenous-leukaemia leucocyte transfusions. Lancet 1970; 2: 338-341. MEDLINE

32 Deisseroth AB, Zu Z, Claxton D, Hanania EG, Fu S, Ellerson D, Goldberg L, Thomas M, Janicek K, Anderson WF, Hester J, Korbling M, Durett A, Moen R, Berenson R, Heimfeld S, Hamer J, Calvert L, Tibbits P, Talpaz M, Kantarjian H, Champlin R, Reading C. Genetic marking shows that Ph⁺ cells present in autologous transplants of chronic myelogenous leukemia (CML) contribute to relapse after autologous bone marrow in CML. Blood 1994; 83: 3068-3076. MEDLINE

33 McGlave PB, De Fabritiis P, Deisseroth A, Goldman J, Barnett M, Reiffers J, Simonsson B, Carella A, Aeppli D. Autologous transplants for chronic myelogenous leukaemia: results from eight transplant groups. Lancet 1994; 343: 1486-1488. MEDLINE

34 Carella AM, Frassoni F, Melo J, Sawyers C, Eaves C, Eaves A, Apperely J, Tura S, Hehlmann R, Lerma E, Reiffers J, Goldman J. New insights in biology and current therapeutic options for patients with chronic myeloid leukemia. Haematologica 1997; 82: 478-495. MEDLINE

35 Sirard C, Lapidot T, Vormoor J, Cashman JD, Doedens M, Murdoch B, Jamal N, Messner H, Addey L, Minden M, Laraya P, Keating A, Eaves A, Lansdorp PM, Eaves CJ, Dick JE. Normal and leukemic SCID-repopulating cells (SRC) co-exist in the bone marrow and peripheral blood from CML patients in chronic phase while leukemic SRC are detected in blast crisis. Blood 1996; 87: 1539-1548. MEDLINE

36 Wang JCY, Lapidot T, Cashman JD, Doedens M, Addy L, Sutherland DR, Nayar R, Laraya P, Minden M, Keating A, Eaves AC, Eaves CJ, Dick JE. High level engraftment of NOD/SCID mice by primitive normal and leukemic hematopoietic cells from patients with chronic myeloid leukemia in chronic phase. Blood 1998; 91: 2406-2414. MEDLINE

37 Lewis ID, McDiarmid LA, Samels LM, Bik To L, Hughes TP. Establishment of a reproducible model of chronic-phase chronic myeloid leukemia in NOD/SCID mice using blood-derived mononuclear or CD34⁺ cells. Blood 1998; 91: 630-640. MEDLINE

38 Dazzi F, Capelli D, Hasserjian R, Cotter F, Corbo M, Poletti A, Chinswangwatanakul W, Goldman JM, Gordon MY. The kinetics and extent of engraftment of chronic myelogenous leukemia cells in non-obese diabetic/severe combined immunodeficiency mice reflect the phase of the donor's disease: an in vivo model of chronic myelogenous leukemia biology. Blood 1998; 92: 1390-1396. MEDLINE

39 Verstegen MMA, Cornelissen JJ, Terpstra W, Wagemaker G, Wognum AW. Multilineage outgrowth of both malignant and normal hemopoietic progenitor cells from individual chronic myeloid leukemia patients in immunodeficient mice. Leukemia 1999; 13: 618-628. MEDLINE

40 Habibian HK, Peters SO, Hsieh CC, Wuu J, Vergilis K, Grimaldi CI, Reilly J, Carlson JE, Frimberger AE, Stewart FM, Quesenberry PJ. The fluctuating phenotype of the lympho-hematopoietic stem cell with cell cycle transit. J Exp Med 1998; 188: 393-398. MEDLINE

41 Glimm H, Oh I, Eaves C. Human hematopoietic stem cells stimulated to proliferate in vitro lose engraftment potential during their S/G₂/M transit and do not reenter G₀. Blood 2000; 96: 4185-4193. MEDLINE

42 Cashman JD, Eaves CJ. High marrow seeding efficiency of human lymphomyeloid repopulating cells in irradiated NOD/SCID mice. Blood 2000; 96: 3979-3981. MEDLINE

43 van Hennik PB, de Koning AE, Ploemacher RE. Seeding efficiency of primitive human hematopoietic cells in nonobese diabetic/severe combined immune deficiency mice: implications for stem cell frequency assessment. Blood 1999; 94: 3055-3061. MEDLINE

44 Gothot A, Van der Loo JCM, Clapp W, Srour EF. Cell cycle-related changes in repopulating capacity of human mobilized peripheral blood CD34⁺ cells in non-obese diabetic/severe combined immune-deficient mice. Blood 1998; 92: 2641-2649. MEDLINE

45 Eaves CJ, Eaves AC. Cell culture studies in CML. In: Hinton K (ed.) Baillière's Clinical Haematology, 1st edn Bailliere Tindall/WB, Saunders: London, 1987, pp 931-961.

46 Fauser AA, Messner HA. Proliferative state of human pluripotent hemopoietic progenitors (CFU-GEMM) in normal individuals and under regenerative conditions after bone marrow transplantation. Blood 1979; 54: 1197-1200. MEDLINE

47 Ponchio L, Cashman J, Zoumbos N, Eaves CJ, Eaves A. Primitive CML cells show a deregulation of their cycling status both in vivo and in longterm cultures which is not normalized in the presence of interferon-. Blood 1995; 86: 493a.

48 Holyoake T, Jiang X, Eaves C, Eaves A. Isolation of a highly quiescent subpopulation of primitive leukemic cells in chronic myeloid leukemia. Blood 1999; 94: 2056-2064. MEDLINE

49 Cashman JD, Eaves AC, Eaves CJ. Granulocyte-macrophage colony-stimulating factor modulation of the inhibitory effect of transforming growth factor- on normal and leukemic human hematopoietic progenitor cells. Leukemia 1992; 6: 886-892. MEDLINE

50 Goto T, Nishikori M, Arlin Z, Gee T, Kempin S, Burchenal J, Strife A, Wisniewski D, Lambek C, Little C, Jhanwar S, Chaganti R, Clarkson B. Growth characteristics of leukemic and normal hematopoietic cells in Ph¹+ chronic myelogenous leukemia and effects of intensive treatment. Blood 1982; 59: 793-808. MEDLINE

51 Kantarjian HM, Vellekoop L, McCredie KB, Keating MJ, Hester J, Smith T, Barlogie B, Trujillo J, Freireich EJ. Intensive combination chemotherapy (ROAP 10) and splenectomy in the management of chronic myelogenous leukemia. J Clin Oncol 1985; 3: 192-200. MEDLINE

52 Gratwohl A, Hermans J, Niederwieser D, Frassoni F, Arcese W, Gahrton G, Bandini G, Carreras E, Vernant JP, Bosi A, de Witte T, Fibbe WE, Zwaan F, Michallet M, Ruutu T, Devergie A, Iriondo A, Apperley J, Reiffers J, Speck B, Goldman J. Bone marrow transplantation for chronic myeloid leukemia: long-term results. Bone Marrow Transplant 1993; 12: 509-516. MEDLINE

53 Yong ASM, Goldman JM. Relapse of chronic myeloid leukaemia 14 years after allogeneic bone marrow transplantation. Bone Marrow Transplant 1999; 23: 827-828. MEDLINE

54 Blackburn EH. Structure and function of telomeres. Nature 1991; 350: 569-572. MEDLINE

55 Blasco MA, Lee H-W, Hande MP, Samper E, Lansdorp PM, DePinho RA, Greider CW. Telomere shortening and tumor formation by mouse cells lacking telomerase RNA. Cell 1997; 91: 25-34. Article MEDLINE

56 Lee H-W, Blasco MA, Gottlieb GJ, Horner JW III, Greider CW, DePinho RA. Essential role of mouse telomerase in highly proliferative organs. Nature 1998; 392: 569-574. Article MEDLINE

57 Shay JW, Bacchetti S. A survey of telomerase activity in human cancer. Eur J Cancer 1997; 33: 787-791. Article MEDLINE

58 Weng N-P, Hathcock KS, Hodes RJ. Regulation of telomere length and telomerase in T and B cells: a mechanism for maintaining replicative potential. Immunity 1998; 9: 151-157. MEDLINE

59 Vaziri H, Benchimol S. Reconstitution of telomerase activity in normal human cells leads to elongation of telomeres and extended replicative life span. Curr Biol 1998; 8: 279-282. MEDLINE

60 Morales CP, Holt SE, Ouellette M, Kaur KJ, Yan Y, Wilson KS, White MA, Wright WE, Shay JW. Absence of cancer-associated changes in human fibroblasts immortalized with telomerase. Nat Genet 1999; 21: 115-118. Article MEDLINE

61 Jiang XR, Jimenez G, Chang E, Frolkis M, Kusler B, Sage M, Beeche M, Bodnar AG, Wahl GM, Tlsty TD, Chiu CP. Telomerase expression in human somatic cells does not induce changes associated with a transformed phenotype. Nat Genet 1999; 21: 111-114. Article MEDLINE

62 Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu C-P, Morin GB, Harley CB, Shay JW, Lichtsteiner S, Wright WE. Extension of life-span by introduction of telomerase into normal human cells. Science 1998; 279: 349-353. Article MEDLINE

63 Rufer N, Brummendorf TH, Kolvraa S, Bischoff C, Christensen K, Wadsworth L, Schultzer M, Lansdorp PM. Telomere fluorescence measurements in granulocytes and T lymphocyte subsets point to a high turnover of hematopoietic stem cells and memory T cells in early childhood. J Exp Med 1999; 190: 157-167. MEDLINE

64 Slagboom PE, Droog S, Boomsma DI. Genetic determination of telomere size in humans: a twin study of three age groups. Am J Hum Genet 1994; 55: 876-882. MEDLINE

65 Rufer N, Dragowska W, Thornbury G, Roosnek E, Lansdorp PM. Telomere length dynamics in human lymphocyte subpopulations measured by flow cytometry. Nat Biotechnol 1998; 16: 743-747. MEDLINE

66 Engelhardt M, Mackenzie K, Drullinsky P, Silver RT, Moore MAS. Telomerase activity and telomere length in acute and chronic leukemia, pre and post-ex vivo culture. Cancer Res 2000; 60: 610-617. MEDLINE

67 Brummendorf TH, Holyoake TL, Rufer N, Barnett MJ, Schulzer M, Eaves CJ, Eaves AC, Lansdorp PM. Prognostic implications of differences in telomere length between normal and malignant cells from patients with chronic myeloid leukemia measured by flow cytometry. Blood 2000; 95: 1883-1890. MEDLINE

68 Boultwood J, Fidler C, Shepherd P, Watkins F, Snowball J, Haynes S, Kusec R, Gaiger A, Littlewood TJ, Peniket AJ, Wainscoat JS. Telomere length shortening is associated with disease evolution in chronic myelogenous leukemia. Am J Hematol 1999; 61: 5-9. Article MEDLINE

69 Boultwood J, Peniket A, Watkins F, Shepherd P, McGale P, Richards S, Fidler C, Littlewood TJ, Wainscoat JS. Telomere length shortening in chronic myelogenous leukemia is associated with reduced time to accelerated phase. Blood 2000; 96: 358-361. MEDLINE

70 Read M, Harrison RJ, Romagnoli B, Tanious FA, Gowan SH, Reszka AP, Wilson WD, Kelland LR, Neidle S. Structure-based design of selective and potent G quadruplex-mediated telomerase inhibitors. Proc Natl Acad Sci USA 2001; 98: 4844-4849. MEDLINE

71 Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, Ford JM, Lydon NB, Kantarjian H, Capdeville R, Ohno-Jones S, Sawyers CL. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344: 1031-1037. MEDLINE

72 Deininger MWN, Goldman JM, Lydon N, Melo JV. The tyrosine kinase inhibitor CGP57148B selectively inhibits the growth of BCR-ABL-positive cells. Blood 1997; 90: 3691-3698. MEDLINE

73 Graham SM, Jorgensen HG, Allan E, Pearson C, Alcorn MJ, Richmond L, Holyoake TL. Primitive quiescent, Philadelphia positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood 2002; 99: 319-325. Article MEDLINE

74 Gao L, Bellantuono I, Elsässer A, Marley SB, Gordon MY, Goldman JM, Stauss HJ. Selective elimination of leukemic CD34⁺ progenitor cells by cytotoxic T lymphocytes specific for WT1. Blood 2000; 95: 2198-2203. MEDLINE

75 Ellisen LW, Carlesso N, Cheng T, Scadden DT, Haber DA. The Wilms tumor suppressor WT1 directs stage-specific quiescence and differentiation of human hematopoietic progenitor cells. EMBO J 2001; 20: 1897-1909. MEDLINE

76 Campbell JD, Cook G, Holyoake TL. Evolution of bone marrow transplantation - the original immunotherapy. Trends Immunol 2001; 22: 88-92. MEDLINE

77 Jorgensen HG, Elliott MA, Allan EK, Carr DE, Holyoake TL, Smith KD. Alpha-1 acid glycoprotein expressed in the plasma with chronic myeloid leukemia patients does not mediate significant in vitro resistance to STI571 (Glivec). Blood 2002; 99: 713-715. Article MEDLINE

78 Hariharan IK, Adams JM, Cory S. BCR-ABL oncogene renders myeloid cell line factor independent: potential autocrine mechanism in chronic myeloid leukemia. Oncogene Res 1988; 3: 387-399. MEDLINE

79 Sirard C, Laneuville P, Dick J. Expression of bcr-abl abrogates factor-dependent growth of human hematopoietic M07E cells by an autocrine mechanism. Blood 1994; 83: 1575-1585. MEDLINE

80 Strife A, Lambek C, Wisniewski D, Wachter M, Gulati SC, Clarkson BD. Discordant maturation as the primary biological defect in chronic myelogenous leukemia. Cancer Res 1988; 48: 1035-1041. MEDLINE

81 Bedi A, Zehnbauer BA, Barber J, Sharkis S, Jones R. Inhibition of apoptosis by BCR-ABL in chronic myeloid leukemia. Blood 1994; 83: 2038-2044. MEDLINE

82 Maguer-Satta V, Burl S, Liu L, Damen J, Chahine H, Krystal G, Eaves A, Eaves C. BCR-ABL accelerates C2-ceramide-induced apoptosis. Oncogene 1998; 16: 237-248. MEDLINE

83 Jiang X, Lopez A, Holyoake T, Eaves A, Eaves C. Autocrine production and action of IL-3 and granulocyte colony-stimulating factor in chronic myeloid leukemia. Proc Natl Acad Sci USA 1999; 96: 12804-12809. MEDLINE

84 Jiang X, Fujisaki T, Nicolini F, Berger M, Holyoake T, Eisterer W, Eaves C, Eaves A. Autonomous multi-lineage differentiation in vitro of primitive CD34⁺ cells from patients with chronic myeloid leukemia. Leukemia 2000; 14: 1112-1121. MEDLINE

85 Ponchio L, Eaves C. Steel factor and Flk-2/Flt-3 ligand alone trigger quiescent human LTC-IC into S-phase more effectively than primitive quiescent clonogenic progenitor cells. J Hematother 1995; 4: 217.

86 Chang JM, Metcalf D, Lang RA, Gonda TJ, Johnson GR. Nonneoplastic hematopoietic myeloproliferative syndrome induced by dysregulated multi-CSF (IL-3) expression. Blood 1989; 73: 1487-1497. MEDLINE

87 Wong PMC, Chung S, Dunbar CE, Bodine DM, Ruscetti S, Nienhuis AW. Retrovirus-mediated transfer and expression of the interleukin-3 gene in mouse hematopoietic cells result in a myeloproliferative disorder. Mol Cell Biol 1989; 9: 798-808. MEDLINE

88 Just U, Katsuno M, Stocking C, Spooncer E, Dexter M. Targeted in vivo infection with a retroviral vector carrying the interleukin-3 (multi-CSF) gene leads to immortalization and leukemic transformation of primitive hematopoietic progenitor cells. Growth Factors 1993; 9: 41-55. MEDLINE

89 Saito H, Hatake K, Dvorak AM, Lerferman KM, Donvenberg AD, Arai N, Ishizaka K, Ishizaka T. Selective differentiation and proliferation of hematopoietic cells induced by recombinant human interleukins. Proc Natl Acad Sci USA 1988; 85: 2288-2292. MEDLINE

90 Metcalf D. Control of granulocytes and macrophages: molecular, cellular, and clinical aspects. Science 1991; 254: 529-533. MEDLINE

91 Souza LM, Boone TC, Gabrilove J, Lai PH, Zsebo KM, Murdock DC, Chazin VR, Bruszewski J, Lu H, Chen KK, Barendt J, Platzer E, Moore MAS, Mertelsmann R, Welte K. Recombinant human granulocyte colony-stimulating factor: effects on normal and leukemic myeloid cells. Science 1986; 232: 61-65. MEDLINE

92 Sattler M, Salgia R. Activation of hematopoietic growth factor signal transduction pathways by the human oncogene BCR/ABL. Cytokine Growth Factor Rev 1997; 8: 63-79. MEDLINE

93 Zhang X, Ren R. BCR-ABL efficiently induces a myeloproliferative disease and production of excess interleukin-3 and granulocyte-macrophage colony-stimulating factor in mice: a novel model for chronic myelogenous leukemia. Blood 1998; 92: 3829-3840. MEDLINE

94 Li S, Ilaria RL, Million RP, Daley GQ, Van Etten RA. The p190, p210, and p230 forms of the BCR/ABL oncogene induce a similar chronic myeloid leukemia-like syndrome in mice but have different lymphoid leukemogenic activity. J Exp Med 1999; 189: 1399-1412. MEDLINE

95 Jiang X, Yip C, Eisterer W, Eaves A, Eaves C. Reduced tyrosine phosphorylation of p210^BCR-ABL and lack of STAT5 activation in BCR-ABL-transduced IL-3^-/- primitive hematopoietic cells. Blood 2000; 96: 92a.

96 Anderson SM, Mladenovic J. The BCR-ABL oncogene requires both kinase activity and src-homology 2 domain to induce cytokine secretion. Blood 1996; 87: 238-244. MEDLINE

97 Li S, Gillessen S, Tomasson MH, Dranoff G, Gilliland DG, Van Etten RA. Interleukin 3 and granulocyte-macrophage colony-stimulating factor are not required for induction of chronic myeloid leukemia-like myeloproliferative disease in mice by BCR/ABL. Blood 2001; 97: 1442-1450. Article MEDLINE

Figure 1 FACS strategy for the isolation of cells in different phases of the cell cycle. Lineage depleted, CD34-enriched cells were labeled with propidium iodide (PI), CD34-FITC, Hoechst 33342 and Pyronin Y. An initial gate was set on viable PI^- cells and a second on the CD34⁺ fraction. Thereafter, cells were identified and sorted in G₀/G₁/SG₂M of the cell cycle according to their level of Hoechst and Pyronin Y staining. (Reproduced with permission by the American Society of Hematology, Washington DC, from Ref. 48.)

Figure 2 FACS strategy for the isolation of deeply quiescent CD34⁺ cells. Lineage-depleted, CD34⁺ cells were labeled with CFSE and a homogeneous population of viable CD34⁺, CFSE⁺ cells was isolated by FACS. These cells were then cultured for 3 days in the presence of five stimulatory growth factors (Flt3-ligand, Steel factor, IL-3, IL-6 and G-CSF) selected to drive primitive cells into cell division. At the end of the third day the cells were analyzed in terms of residual CFSE fluorescence and those cells which had not yet undergone cell division were identified in M1. The quiescent cells present in M1 were then isolated by FACS and cultured in standard CFC assays. (Reproduced with permission by the American Society of Hematology, Washington DC, from Ref. 48.)

Figure 3 Telomere length in blood or marrow cells from CML patients (solid symbols) is shorter than in blood cells from age-matched normal individuals (open symbols). Telomere length was determined by flow-FISH. (Reproduced with permission by the American Society of hematology, Washington DC, from Ref. 67.)

Figure 4 Loss of telomere fluorescence in CML (as measured by flow-FISH), in the myeloid compartment of peripheral blood. Fluorescence intensity is expressed as molecules of equivalent soluble fluorochrome units (MESF), with 10^-3 MESF approximating to 0.5 kb of telomeric repeat sequence. Granulocytes (or blasts in blast crisis CML patients) were gated using their known forward and side light scattering characteristics. Total peripheral blood leukocytes were used when the granulocyte population was not clearly defined. The bars represent mean values. Significant loss of telomere signal is seen in both accelerated phase disease (AP, n = 11) and myeloid blast crisis (BC, n = 4) as compared to chronic phase (CP, n = 32) within 1 year of diagnosis (P values as shown). Median ages for all three series were similar (56, 61 and 58 years respectively), thereby excluding age-related telomere shortening as a major cause for the observed differences (unpublished observations). The range of telomere signals at all stages is wide, a phenomenon also observed for normal individuals of the same age (see Figure 3).

Figure 5 Demonstration that quiescent CD34⁺ CML cells are insensitive to STI571. Using the FACS strategy outlined in the legend to Figure 2, purified CD34⁺, CFSE⁺ CML cells were cultured in the presence of the same five growth factor combination as described in Figure 2 and in the presence (right dot plot) or absence (left dot plot) of 10 M STI571 (provided by Novartis, Basel, Switzerland). As shown, although all the dividing cells (identified to the left hand side of the dotted line) were killed by STI571, the quiescent cells (right hand side of dotted line) remained viable (PI-) and retained a high level of CD34 expression. (Reproduced with permission by the American Society of Hematology, Washington DC, from Ref. 73.)

Figure 6 Factor-independent growth of CML cells in culture decreases with their differentiated state. Highly purified CD34⁺CD71^-CD45RA^- (diamonds), CD34⁺CD71⁺CD45RA⁺ (squares), and CD34^- (triangle) CML cells (n = 5) or normal marrow cells (circles, n = 3) were cultured at 10⁵ cells per ml in SFM with (solid symbols) or without (open symbols) the same five growth factors described in Figure 2. Viable cell numbers were determined by hematocytometer counts of trypan blue-excluding cells. Results are expressed as the mean fold-change compared with input values ± s.e.m. (a) CML patients with predominantly Ph⁺ LTC-IC (n = 3). (b) CML patients with predominantly Ph^- LTC-IC (n = 2). (Reproduced with permission by the National Academy of Science of the United States of America, Washington, DC, from Ref. 83.)

Figure 7 Schematic representation of the differential effects of different levels of IL-3 and G-CSF stimulation on primitive normal cells by comparison to the observed alterations in CML patients where an autocrine IL-3/G-CSF loop in the earliest leukemic compartments has been demonstrated. (Reproduced with permission by the National Academy of Science of the United States of America, Washington, DC, from Ref. 83.)