The leukemic stem cells in patients with chronic myeloid leukemia (CML) are well known to be clinically resistant to conventional chemotherapy and may also be relatively resistant to BCR-ABL-targeted drugs. Here we show that the lesser effect of imatinib mesylate (IM) on the 3-week output of cells produced in vitro from lin À CD34 þ CD38 À CML (stem) cells compared with cultures initiated with the CD38 þ subset of lin À CD34 þ cells is markedly enhanced (410-fold) when conditions of reduced growth factor stimulation are used. Quantitative analysis of genes expressed in these different CML subsets revealed a differentiation-associated decrease in IL-3 and G-CSF transcripts, a much more profound decrease in expression of BCR-ABL than predicted by changes in BCR expression, decreasing expression of ABCB1/MDR and ABCG2 and increasing expression of OCT1. p210 BCR-ABL and kinase activity were also higher in the lin À CD34 þ CD38 À cells and formal evidence that increasing BCR-ABL expression decreases IM sensitivity was obtained from experiments with a cell line model. Nevertheless, within the entire CD34 þ subset of CML cells, BCR-ABL expression was not strongly affected by changes in cell cycle status. Taken together, these results provide the first evidence of multiple mechanisms of innate IM resistance in primitive and quiescent CML cells. Leukemia (2007) 21, 926-935.

Chronic myeloid leukemia (CML) has long served as a paradigm for generating new insights into the cellular origin, pathogenesis and improved approaches to treating many types of human cancer. Early studies of the cellular phenotypes and genotypes represented in leukemic populations obtained from CML patients established the concept of an evolving clonal disorder originating in and initially sustained by a rare, multipotent, self-maintaining hematopoietic stem cell (HSC). More recent investigations continue to support this model, while also revealing new insights into the cellular and molecular mechanisms that explain how knowledge of CML stem cells and their early differentiating progeny can predict the differing and variable features of chronic phase and blast crisis. In particular, these emphasize the need for new agents that effectively and specifically target CML stem cells to produce non-toxic, but curative therapies that do not require lifelong treatments.

In this report we describe a quantitative in vitro assay for the most primitive type of leukemic precursors yet defined in patients with chronic myeloid leukemia (CML). This assay is based on the recently described "long-term culture-initiating cell" (LTC-IC) assay for primitive normal human hematopoietic cells. Such cells, when cocultured with competent fibroblast feeder layers, give rise after a minimum of 5 weeks to multiple single and multilineage clonogenic progenitors detectable in secondary semisolid assay cultures. Similar cultures initiated by seeding a highly enriched source of leukemic cells from patients onto normal feeders showed the clonogenic cell output after 5 weeks to be linearly related to the input innoculum over a wide range down to limiting numbers of input cells, thus allowing absolute frequencies of leukemic LTC-ICs to be determined using standard limiting dilution analysis techniques. Leukemic LTC-IC concentrations in CML marrow were found to be decreased, on average to <10% ofthe normal LTC-IC concentration in normal marrow, but were greatly increased (up to > 10 times) in CML blood. Assessment of the number of clonogenic cells produced per leukemic LTC-IC by comparison to normal blood or marrow LTC-IC values showed this function to be unchanged in leukemic LTC-ICs [i.e., 3.1 ± 0.4 clonogenic cells per CML LTC-IC (mean ± SEM, n = 6) versus 3.7 ± 1.2 (n = 3) and 4.3 ± 0.4 (n = 5), respectively, for normal blood and marrow LTC-ICs].In contrast, leukemic LTC-IC maintenance in LTC proved to be highly defective by comparison to normal LTC-IC of either blood or marrow origin. Thus, when cells from primary LTC were subcultured into secondary LTC-IC assays, leukemic LTC-IC rapidly declined (>30-fold) within the first 10 days of culture, whereas normal LTC-IC numbers remained unchanged during this period. These findings illustrate how self-maintenance and differentiation events in primitive human hematopoietic cells can be differentially modulated by an oncogenic process and provide a framework for further studies of their manipulation, analysis, and therapeutic exploitation. (Ph') (1). The initial cell transformed, and hence the origin of the leukemic clone, is believed to be a totipotent hematopoietic cell with lymphoid as well as myeloid differentiation potential since Ph'-positive cells in these lineages are frequently demonstrable (2). This has suggested that production of the BCR-ABL kinase in a totipotent hematopoietic cell gives it a selective growth advantage. Recent experiments involving retroviral infection of murine bone marrow (BM) cells with BCR-ABL constructs are consistent with this (3, 4), although an underlying molecular mechanism has not been determined. In particular, the biological consequences of BCR-ABL kinase expression in very primitive human hematopoietic cells have been difficult to investigate because methods for their selective isolation have not been available.CML patients with elevated leukocyte (WBC) counts show dramatic increases in the number of Ph'-posi...

To develop a purification strategy for isolating the most primitive hematopoietic stem cells present in normal human marrow we have combined cell separation techniques with an assay for cells that initiate sustained hematopoiesis in vitro in the presence of irradiated human marrow adherent cells. These “feeders” were established by subculturing 2- to 6-week-old primary long-term marrow culture adherent layers at a density of 3 x 10(4) irradiated cells per square centimeter. Test “long-term culture (LTC)-initiating cells” were plated on top of the feeders and the cocultures then maintained as standard long-term marrow cultures with half-media changes and removal of half of the nonadherent cells each week. The total number of myeloid, erythroid, and multilineage clonogenic progenitors present after 5 weeks was used to provide a quantitative assessment of the number of LTC-initiating cells originally added. Using this assay, the density, light scatter, and two cell surface antigen properties of LTC- initiating cells have been defined and compared with cells capable of directly forming colonies in methylcellulose. While the majority of the clonogenic cells were found in the high forward light scatter (FLS) “blast” window, LTC-initiating cells had significantly lower FLS properties and in this respect were more similar to lymphocytes. LTC- initiating cells also expressed less HLA-DR antigen than clonogenic cells. The majority of LTC-initiating cells were found in the top 2% of the CD34 (My10) fluorescence profile, whereas clonogenic cells were found throughout the top 5% of the CD34 fluorescence profile. By combining low FLS, low orthogonal light scatter (OLS), low HLA-DR expression, and high CD34 expression, a population could be obtained that was enriched for LTC-initiating cells approximately 800-fold over unseparated marrow. This population contains only 0.06% of the marrow cells and 2% of the total clonogenic cells, but retains 50% to 60% of the LTC-initiating cells present in the original marrow. The ability to purify these two populations independently shows that the LTC and clonogenic assays identify distinct, although not necessarily nonoverlapping cell types in human marrow. Since clonogenic cells are derived from LTC-initiating cells, the LTC assay clearly detects a more primitive population. The availability of a simple approach that allows the purification of such cells by three orders of magnitude in high yield should be useful for the investigation of early events in hematopoiesis as well as for the definitive isolation of human hematopoietic stem cells with long-term in vivo repopulating potential.

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.

Replating experiments have shown that the selfrenewal of pluripotent hemopoietic stem cells can be studied in vitro by clonal analysis techniques. The number of daughter stem cells detectable in individual primary clones produced in vitro varies markedly from one clone to another. These findings are consistent with a general model of stem cell differentiation in which the choice to self-replicate or not is ultimately determined at the single-cell level by a mechanism involving a random-event component that is intrinsic to the stem cell itself. Hemopoietic stem cells were identified by their ability to generate macroscopic-sized colonies having a visible erythroid component (i.e., gross red color) in standard methylcellulose assays containing medium conditioned by pokeweed mitogen-treated spleen cells and erythropoietin. In assays of replated primary or secondary colonies, inclusion of irradiated marrow-cell feeders was found to be an additional requirement. The mixed erythroid-megakaryocyte-granulocyte nature ofcolonies identified simply as macroscopic and erythroid was confirmed by cytochemical stains for lineage-specific markers. Marked variation in self-renewal was a feature of marrow stem cells both before and after maintenance in flask culture, although the overall self-renewal capacity exhibited by flask-cultured cells was n5-fold higher. Variation in self-renewal was not correlated with primary colony size, which also varied over a wide range (0.2-9 x 105 nucleated cells per colony). Variation in stem cell selfrenewal has been previously associated with hemopoietic stem cell proliferation in vivo. Its persistence in vitro in assays of dilute single-cell suspensions casts doubt on the significance of microenvironmental influences in directing stem cell differentiation.

Numerous factors that can influence the proliferation and differentiation in vitro of cells at various stages of hematopoiesis have been identified, but the mechanisms used by stromal cells to regulate the cycling status of the most primitive human hematopoietic cells are still poorly understood. Previous studies of long-term cultures (LTC) of human marrow have suggested that cytokine-induced variations in stromal cell production of one or more stimulators and inhibitors of hematopoiesis may be important. To identify the specific regulators involved, we performed Northern analyses on RNA extracted from human marrow LTC adherent layers, or stromal cell types derived from or related to those present in the adherent layer. These analyses showed marked increases in interleukin-1 beta (IL-1 beta), IL-6, and granulocyte colony-stimulating factor (G-CSF) mRNA levels within 8 hours after treatments that lead to the activation within 2 days of primitive hematopoietic progenitors in such cultures. Increases in granulocyte-macrophage (GM)-CSF and M-CSF mRNA were also sometimes seen. Bioassays using cell lines responsive to G-CSF, GM-CSF, and IL-6 showed significant elevation in growth factor levels 24 hours after IL- 1 beta stimulation. Neither IL-3 nor IL-4 mRNA was detectable at any time. In contrast, transforming growth factor-beta (TGF-beta) mRNA and nanogram levels of TGF-beta bioactivity in the medium were detected at all times in established LTC, and these levels were not consistently altered by any of the manipulations that stimulated hematopoietic growth factor production and primitive progenitor cycling. We also found that addition of anti-TGF-beta antibody could prolong or reactivate primitive progenitor proliferation when added to previously stimulated or quiescent cultures, respectively. Together, these results indicate a dominant negative regulatory role of endogenously produced TGF-beta in unperturbed LTC, with activation of primitive hematopoietic cells being achieved by mechanisms that stimulate stromal cells to produce G-CSF, GM-CSF, and IL-6. Given the similarities between the LTC system and the marrow microenvironment, it seems likely that the control of human stem cell activation in vivo may involve similar variations in the production of these factors by stromal cells.

Recent studies with long-term mouse marrow cultures have indicated the importance of the adherent layer as a primary reservoir of the most primitive stem cells, from which derivative stem cells and more differentiated progenitors are continuously generated. We have now examined the role of the adherent cell layer in long-term human marrow cultures from this point of view. Prerequisite to such an undertaking was the development of a nontoxic and reproducible method for detaching the adherent layer and making it into a single-cell suspension suitable for characterization by colony assays. Both trypsin and collagenase could be used to obtain suspensions that met these criteria. Lack of toxicity was demonstrated by the preservation of CFU-E, BFU-E, and CFU- C plating efficiency in fresh human marrow cell suspensions exposed to the same enzymatic treatments. Collagenase treatment of long-term marrow culture adherent layers was considered superior because it freed all hemopoietic colony-forming cells but left some of the other cells still adherent. Using this method, we found that CFU-C, BFU-E, and CFU- G/E were consistently detectable in the adherent layer for at least 8 wk, with the majority of the BFU-E and CFU-G/E being located in the adherent layer (70%-75% after 2–3 wk and more than 90% by 7–8 wk). Although corresponding numerical differences in adherent and nonadherent CFU-C populations were not observed, the colonies derived from them showed marked differences in the size they achieved; the adherent layer being the exclusive site of CFU-C, with a very high proliferative capacity. These findings emphasize the importance of assessing the progenitor content of the adherent layer of long-term human marrow cultures and provide an appropriate methodology.