First published online January 25, 2007
Stem Cells
Vol. 25 No.
5
May 2007, pp.
1119
-1125
doi:10.1634/stemcells.2006-0637; www.StemCells.com
© 2007 AlphaMed Press
Severe Hypoxia Defines Heterogeneity and Selects Highly Immature Progenitors Within Clonal Erythroleukemia Cells
Serena Giuntolia,
Elisabetta Rovidaa,
Antonella Gozzinib,
Valentina Barbettia,
Maria Grazia Cipolleschia,
Massimo Olivottoa,
Persio Dello Sbarbaa
aDipartimento di Patologia e Oncologia Sperimentali della Università degli Studi di Firenze, Firenze, Italy;
bDivisione di Ematologia, AUO Careggi, Firenze, Italy
Key Words. Leukemia stem cells • Severe hypoxia • Culture-repopulating ability • 5-Fluorouracil resistance
Correspondence: Persio Dello Sbarba, M.D., Ph.D., Dipartimento di Patologia e Oncologia Sperimentali, Viale G.B. Morgagni, 50, 50134 Firenze, Italy; Telephone: 39-055-4282325; Fax: 39-055-4282333; e-mail: persio{at}unifi.it
Received October 6, 2006;
accepted for publication January 12, 2007.
First published online in STEM CELLS EXPRESS January 25, 2007.
 |
ABSTRACT
|
|---|
We showed that resistance to severe hypoxia defines hierarchical levels within normal hematopoietic populations and that hypoxia modulates the balance between generation of progenitors and maintenance of hematopoietic stem cells (HSC) in favor of the latter. This study deals with the effects of hypoxia (0.1% oxygen) in vitro on Friend's murine erythroleukemia (MEL) cells, addressing the question of whether a clonal leukemia cell population comprise functionally different cell subsets characterized by different hypoxia resistance. To identify leukemia stem cells (LSC), we used the Culture Repopulating Ability (CRA) assay we developed to quantify in vitro stem cells capable of short-term reconstitution (STR). Hypoxia strongly inhibited the overall growth of MEL cell population, which, despite its clonality, comprised progenitors characterized by markedly different hypoxia-resistance. These included hypoxia-sensitive colony-forming cells and hypoxia-resistant STR-type LSC, capable of repopulating secondary liquid cultures of CRA assays, confirming what was previously shown for normal hematopoiesis. STR-type LSC were found capable not only of surviving in hypoxia but also of being mostly in cycle, in contrast with the fact that almost all hypoxia-surviving cells were growth-arrested and with what we previously found for HSC. However, quiescent LSC were also detected, capable of delayed culture repopulation with the same efficiency as STR-like LSC. The fact that even quiescent LSC, believed to sustain minimal residual disease in vivo, were found within the MEL cells indicates that all main components of leukemia cell populations may be present within clonal cell lines, which are therefore suitable to study the sensitivity of individual components to treatments.
Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION
|
|---|
We showed previously that resistance to severe hypoxia defines hierarchical levels within normal hematopoietic populations; in vitro colony-forming cells (CFC) are highly hypoxia-sensitive, as opposed to their progenitors, which are hypoxia-resistant [1–5]. Indeed, severe hypoxia enhances maintenance in vitro of normal stem cells capable of short-term hematopoietic reconstitution (STR-HSC). This finding was later confirmed for human long-term repopulating (LTR) stem cells reconstituting NOD/SCID mice [6, 7]. We also showed that hypoxia protects stem cells from stimuli that induce lineage commitment to differentiation [5]. Such a protection may be necessary also for the maintenance of leukemia stem cells (LSC), as treatment of leukemic populations with differentiating agents is not unlikely to result in the suppression of leukemic growth [8]. Based on the above, we hypothesized LSC to be metabolically adapted to survive in a severely hypoxic environment. This feature would be crucial for the long-term persistence of neoplasia, as well as its progression to a more aggressive phenotype, hypoxia being a strong inducer of genomic instability [9]. Although these concepts have been developed for solid tumors, oxygen levels also may be particularly low around leukemia cells. Indeed, bone marrow, which is physiologically characterized by a high level of cell crowding and poor vascularization [1], can undergo a further reduction of oxygenation in patients with leukemia [10] as a result of the fast growth of leukemia cells and anemia, which commonly develops as a consequence of leukemic growth.
The effects of hypoxia on leukemia cell populations so far have been scarcely investigated. We previously found that hypoxia enhances the maintenance of pre-CFC potential in cultures of CD34+ cells from patients with chronic myeloid leukemia (CML) and, within CML cell lines, selects imatinib-resistant highly immature progenitors, where the BCR/Abl protein is suppressed [11, 12]. In the study reported here, targeted to deepen the characterization of the effects of severe hypoxia on leukemia progenitor and stem cells, we used the murine Friend's erythroleukemia (MEL) cell line to address the question of whether cell subsets characterized by different sensitivity to hypoxia are detectable within a clonal leukemia cell line and whether these differences define the level within the progenitor/stem cell hierarchy. To identify LSC, we used the Culture Repopulating Ability (CRA) assay, developed in our laboratory [2] and extensively applied [3, 5, 7] to quantify in vitro normal STR-HSC and later adapted to study their leukemic counterpart [11, 12].
We found that the MEL cell line is functionally and metabolically heterogeneous in that it comprises hypoxia-sensitive CFC as well as hypoxia-resistant culture repopulating cells (CRC), which correspond to STR-type LSC. This confirmed for a leukemia cell population what we showed previously for normal hematopoiesis: that a severely hypoxic environment maintains stem cells but not clonogenic progenitors. LSC were found to be capable of not only surviving but also cycling in severe hypoxia. The relevance of these findings to LSC maintenance in vivo is discussed.
|
MATERIALS AND METHODS
|
|---|
Cells and Culture Conditions
MEL cells were cultured in RPMI 1640 medium supplemented with 50 units/ml penicillin, 50 µg/ml streptomycin, and 10% fetal bovine serum (all from EuroClone, Paignton, U.K., http://www.euroclone.net/), and incubated at 37°C in a water-saturated atmosphere containing 5% CO2 and 95% air. Experiments were performed with cells from maintenance cultures subcultured in fresh medium 24 h before plating (time zero) at 3 x 105/ml. Incubation in normoxia (21% O2) was carried out in a conventional cell culture incubator in a 5% CO2, 95% air water-saturated atmosphere. Incubation in severe hypoxia (0.1% O2) was carried out in a Ruskinn Concept 400 anaerobic incubator, flushed with a preformed gas mixture (0.1% O2, 5% CO2, 95% N2) and water-saturated. This incubator allows easy entry and exit of materials and sample manipulations without compromising the hypoxic environment. 5-Fluorouracil (5FU) (Sigma) was added to cultures at 100 µg/ml 2 days before cell recovery and count or transfer.
Clonal Assays
Clonogenic semisolid cultures were established in MethoCult medium (Stem Cell Technologies, Vancouver, BC, Canada, http://www.stemcell.com). The number of cells plated was chosen, after preliminary tests, based on the expected colony-formation efficiency of the cell population assayed. The number of colonies formed was scored at day 7 of incubation. Colony-formation efficiency was calculated by dividing the number of colonies developed by the number of cells plated. The number of CFC in liquid cultures incubated in either normoxia or hypoxia was measured by transferring cells to semisolid cultures incubated in any case in normoxia. The total number of CFC in liquid cultures was calculated by multiplying the colony-formation efficiency values by the number of viable cells in liquid culture.
The CRA Assay
This assay estimates the culture-repopulating power of hematopoietic cells subjected to a selection treatment (e.g., hypoxia) in liquid culture 1 (LC1) by means of their transfer to nonselective conditions (e.g., normoxia) in liquid culture 2 (LC2) and following a further incubation therein. Each incubation has a standard 6–7-day duration. Before being replated into LC2, cells selected in LC1 were centrifuged, resuspended in fresh medium, and counted. CRA has been shown to reflect stem cells endowed with marrow-repopulating ability in vivo, therefore representing a simple and economic method to detect in vitro STR-HSC [2]. The maintenance of culture-repopulating cells (CRC) after selection in LC1 is quantified by the ratio of peak (day 6) cell numbers in LC2 to peak cell numbers in nonselective LC1 established with equal numbers of cells. It is worth noting that, in the latter cultures, repopulation values are usually higher because nonselective conditions spare selection-sensitive progenitors (for instance, CFC). In some experiments, extended CRA was measured after a prolonged (to day 18) LC2 incubation.
Measures of Cell Viability, Apoptosis, and Cell Cycle Phases
The number of viable cells was counted in a hemocytometer by trypan blue exclusion. To quantify apoptosis, 5 x 105 cells were washed with ice-cold phosphate-buffered saline (PBS) and resuspended in 100 µl of binding buffer (HEPES-buffered saline solution with 2.5 mM CaCl2). fluorescein isothiocyanate (FITC)-labeled annexin-V (Roche Diagnostics) and propidium iodide (PI; 10 ng/ml) were added. Cells were then incubated for 15 minutes in the dark at room temperature (RT). After the addition of 400 µl of binding buffer and agitation, flow cytometry was performed using a FACSscan (BD Biosciences, San Jose, CA, http://www.bdbiosciences.com). Annexin-V+/PI– cells were defined as "early apoptotic," and annexin-V+/PI+ cells as "late apoptotic." To determine cell cycle phase distribution after incubation in normoxia or hypoxia, 5 x 105 cells were washed with ice-cold PBS, fixed in 75% ethanol in PBS (30 minutes, on ice), treated with RNAase (2.5 µg/ml in PBS) at 37°C for 30 minutes and resuspended in 500 µl of PI solution in PBS (10 ng/ml), before flow cytometry. To resolve G0 from G1, 5 x 105 cells incubated for 7 days in normoxia or hypoxia were labeled with a FITC-conjugated 
-human-Ki67 mouse monoclonal antibody also capable of reacting with murine cells (BD Pharmingen Europe, Milan, Italy, http://www.bdbiosciences.com).
Protein Separation and Detection
Cells (5 x 106) were washed twice with ice-cold PBS containing 100 µM Na3VO4 and solubilized by incubating for 10 minutes at 95°C in Laemmli buffer (62.5 mM Tris/HCl, pH 6.8, 10% glycerol, 0.005% bromphenol blue, and 2% SDS). Lysates were clarified by centrifugation (20,000g, 10 minutes, RT) and protein concentration in supernatants was determined by the BCA method. Aliquots (30 µg/sample) were boiled for 10 minutes in the presence of 100 mM 2-mercaptoethanol, subjected to SDS-polyacrylamide gel electrophoresis (PAGE) in 15% polyacrylamide minigels, and then transferred onto polyvinylidene difluoride membranes (Hybond-P; GE Healthcare, Chalfont St. Giles, Buckinghamshire, U.K., http://www.gehealthcare.com) by electroblotting. Membranes were blocked in PBS containing 0.1% Tween-20 (TPBS) and 5% bovine serum albumin (3 h, RT), and then in a 1:500–1:1000 dilution of -cyclin-A (Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), -cyclin-D1 (Santa Cruz Biotechnology Inc.), -histone-H4 (Upstate Biotechnology, Lake Placid, NY, http://www.upstate.com) rabbit antibodies. The horseradish peroxidase-conjugated anti-IgG secondary antibody was from Sigma. Antibody-coated protein bands were visualized by ECL chemiluminescence detection (GE Healthcare). Stripping was performed by incubating membranes in stripping buffer (62.5 mM Tris/HCl, pH 6.7, 2% SDS, and 100 mM 2-mercaptoethanol; 3 x 10 minutes; 50°C) and then extensively washing with PBS containing 0.1% Tween-20.
|
RESULTS
|
|---|
Figure 1A shows the kinetics of the total number of viable cells in MEL cultures incubated for 7 days in normoxia (21% O2) or severe hypoxia (0.1% O2). In normoxia, cell number increased approximately eightfold over the first 3 days of incubation, to decrease thereafter as an effect of culture crowding. In hypoxia, this increase was suppressed and cell number at day 3 was markedly lower than at time zero. The CFC content of normoxic or hypoxic liquid cultures was measured by transferring cells, at different times, from these cultures to clonal assays in semisolid cultures incubated (in any case) in normoxia (Fig. 1B). The number of CFC markedly increased in the course of incubation in normoxia, whereas in hypoxia this increase was suppressed, and CFC number was reduced at the end of incubation to values well below those of time zero. Data from Figure 1A and 1B are computed in Figure 1C to calculate the percentage maintenance of viable cells and CFC in hypoxia versus normoxia at day 3 (i.e., at the peak of culture expansion in normoxia). This maintenance was approximately 2% and 4%, respectively, indicating that the bulk cell population as well as the CFC subset are highly sensitive to hypoxia.
The time-course of apoptosis in MEL cell cultures is shown in Figure 2A. Approximately 25% of cells were in apoptosis (early plus late apoptosis) at time zero of cultures. In normoxia, the percentage of apoptotic cells decreased in the course of early culture expansion, to reach higher (40%–50%), stable levels thereafter. In hypoxia, the early reduction of apoptosis observed in normoxia was suppressed, and the total percentage of apoptotic cells increased progressively to reach 70%–80% after day 3 of incubation. The effects of hypoxia on cell cycling were then determined. The flow-cytometric analysis of cell cycle showed that hypoxia did not significantly alter phase distribution with respect to normoxia within the first 3 days of incubation (Fig. 2B). A prolonged incubation (day 7) in hypoxia determined an accumulation of almost all cells in G0/G1. When G0 was resolved from G1 by means of -Ki67 labeling, in hypoxia, most of the viable cells were found in G0, whereas the percentage of G0 cells was much lower in normoxia. These results indicated that viable cells surviving 7 days of incubation in hypoxia (Fig. 2A) were growth-arrested (Fig. 2B). This conclusion was confirmed by determining the expression of cyclins (Fig. 2C). Cyclin-A (characteristic of the S-phase) was found reduced in hypoxia at day 3 to levels comparable with those of normoxia at day 6. On the other hand, cyclin-D1 (produced in early G1) was suppressed before day 6 of incubation in hypoxia, whereas cyclin-D1 remained at relatively high levels throughout incubation in normaxia (day 10). Thus, in hypoxia, cells started to reduce their entering into S (cyclin-A) as early as day 3, to undergo cycle arrest (cyclin-D1) before day 6. Taken together, the results of Figure 2 indicate that hypoxia markedly reduced survival and cycling of MEL cells in culture.
View larger version (38K):
[in this window]
[in a new window]
|
Figure 2. Effects of hypoxia on MEL cell cycling and apoptosis. (A): Proapoptotic effects of hypoxia. The percentages of annexin-V-positive/propidium iodide-negative (early apoptosis, gray) and annexin-V-positive/propidium iodide-positive (late apoptosis, black) were measured at the indicated times in cultures incubated in normoxia or hypoxia. Data represent means ± SEM of three independent experiments. (B): Effects of hypoxia on cell cycle phase distribution. Flow cytometry of cells incubated in hypoxia or normoxia for the indicated times was performed after staining with propidium iodide. The percentages of cells in the G2/M (dark gray), S (light gray), or G0/G1 (white) phases of the mitotic cycle are reported (days 0–3). Flow cytometry of cells incubated in hypoxia or normoxia for 7 days was also performed after staining with -Ki67 fluorescein isothiocyanate-conjugated antibodies. The percentages of cells in the G0 (dotted white) or G1 (dashed white) phases are reported. Data represent means ± SEM of three independent experiments. (C): Effects of hypoxia on the expression of cyclins. Cells were lysed in Laemmli buffer, and proteins were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotting with the indicated antibodies, on the same membrane before and after stripping. Anti-H4 immunoblotting was used to check equalization of protein amount across loaded samples. Data are from one representative experiment out of two. Abbreviation: WB, Western blot.
|
|
Cells selected in hypoxia, in a new set of experiments, were subjected to the CRA assay [2] to measure their STR-type stem cell potential (Fig. 3 Cells recovered from hypoxic LC1 at day 7, when the few surviving cells (Figs. 1A, 2A) were growth-arrested (Fig. 2B), were replated into LC2, and incubated in normoxia to determine therein the kinetics of viable cell number (Fig. 3A). LC2 was repopulated with the standard kinetics of CRA assay [2]. Thus, the MEL cell population contained hypoxia-resistant CRC. The percentage maintenance of CRC in hypoxic versus normoxic LC1 was then calculated (Fig. 3B) as described in Materials and Methods. At the end of selection in hypoxia (day 7 of LC1), 73.8% of the time-zero CRA potential was maintained. This result contrasts sharply with the maintenance values in hypoxia obtained for the bulk of population and the CFC subset (Fig. 1C). Thus, in contrast with the latter components of the MEL cell population, CRC were substantially hypoxia-resistant. The expansion potential of hypoxia-selected CRC was then determined by sequentially subculturing (four cycles) LC2 cells at the peaks of expansion (of each cycle). The cumulative increase of viable cell number within these subcultures (points of the graph) closely matched that of cultures established with equal numbers of cells always incubated in normoxia (line of the graph), confirming the hypoxia resistance of CRC (Fig. 3C). Taken together, the results so far presented showed that: (a) hypoxia inhibits MEL cell growth and survival; (b) the MEL cell population is heterogeneous in its response to hypoxia, and progenitor subsets exist that are hypoxia-sensitive (CFC) or hypoxia-resistant (CRC); and (c) hypoxia-resistant CRC represent the progenitors ultimately responsible for regeneration and population maintenance within MEL cell cultures.
View larger version (21K):
[in this window]
[in a new window]
|
Figure 3. Culture Repopulating Ability (CRA) assay of hypoxia-selected MEL cells. (A): Time course of the number of viable cells repopulating LC2. Hypoxia-selected cells recovered from day-7 hypoxic LC1 were centrifuged, resuspended in fresh medium, and replated at 3 x 104/ml in LC2 to be incubated in normoxia. LC2 repopulation was then assessed by determining the kinetics of the number of viable cells. (B): Maintenance of culture-repopulating cells (CRC) in hypoxia as detected by CRA assay. To calculate the percentage maintenance of CRC-type leukemia stem cells (column five) after selection in hypoxic LC1, the peak (day 6) value of (A) is computed in column 4. Maintenance is quantified by the ratio of this value to the corresponding value (peak number of viable cells) in nonselective LC1 established with equal numbers of cells (columns 1 and 3) and incubated in normoxia (column 2). (C): Cumulative, long-term cell output from hypoxia-selected CRC. LC2 established with cells selected in hypoxic LC1 as indicated in (A) were serially subcultured weekly (i.e., at peak times) and incubated always in normoxia (points). Nonselective LC1 (as in B, column 2) were serially subcultured, at peak times, under identical conditions and always incubated in normoxia (line). Note that the line is therefore not to be considered to interpolate the experimental points. All data of Figure 3 represent means ± SEM of three independent experiments. Abbreviations: LC1, liquid culture 1; LC2, liquid culture 2.
|
|
The above results prompted us to determine the effects of 5FU, a cycle-specific cytotoxic drug, on MEL cell subsets exhibiting different levels of sensitivity to hypoxia, on CRC in particular. 5FU was administered to LC1 at day 5, and the numbers of viable cells (Fig. 4A) or CFC (Fig. 4B) were counted 2 days later (day 7). The treatment with 5FU in normoxia reduced these numbers almost to zero (Fig. 4, A and B, left). In hypoxia, 5FU was much less effective, in keeping with the fact that most viable cells were growth-arrested (Fig. 2B). In particular, two thirds of cells and one fifth of CFC surviving hypoxia seemed 5FU-resistant (Fig. 4A, 4B, right). However, it is noteworthy that, because hypoxia-resistant cells and CFC represented only 1.9% and 4.1% of the total, respectively (Fig. 1C), their 5FU-resistant fractions were extremely small minorities (ranging between 0.8 and 1.2%). The exact nature of these fractions remains to be determined, although semiquantitative RT-PCR for the β-globin gene indicated that relatively more mature cells were selected by hypoxia (data not shown).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 4. Effects of 5-fluorouracil (5FU) on hypoxia-selected MEL cell subsets. (A): Effects of 5FU on the number of viable cells in hypoxia. 5FU was administered (+) or not (–) at day 5 to LC1 incubated in normoxia (left) or hypoxia (right), and the numbers of viable cells were counted 2 days later (day 7). Right, the difference was not statistically significant. Data represent means ± SEM of three independent experiments. (B): Effects of 5FU on the number of CFC in hypoxia. 5FU was administered (+) or not (–) at day 5 to LC1 incubated in normoxia (left) or hypoxia (right); cells were recovered 2 days later (day 7) and replated into semisolid cultures incubated in all cases in normoxia to score colonies 7 days after replating. Data represent means ± SEM of three independent experiments. (C): Effects of 5FU on hypoxia-selected culture-repopulating cells (CRC) detected by Culture Repopulating Ability (CRA) assay. 5FU was administered (+) or not (–) to LC1 in hypoxia 2 days before cell recovery (day 7), and transfer to LC2 was established with identical numbers of viable cells. LC2 were incubated in normoxia until cells nonpretreated with 5FU reached the peak of expansion (day 6, as shown inFig. 3 A). Data represent means ± SEM of three independent experiments. (D): Effects of 5FU on hypoxia-selected CRC detected by extendedCRA assay. Cells pretreated with 5FU (black) from the experiments described in (C) were incubated in normoxia until cell culture reached the peak of expansion. Data are from one representative experiment out of two. Abbreviations: 5FU, 5-fluorouracil; LC1, liquid culture 1; LC2, liquid culture 2.
|
|
Figure 4C shows the effects of 5FU on hypoxia-selected CRC. Hypoxic LC1 were treated with 5FU from day 5 to day 7, when cells were transferred to LC2 to be incubated in normoxia. The effects of 5FU were then assessed by counting the total number of viable cells at peak of LC2. Only 3.3% of hypoxia-resistant CRC survived 5FU (+5FU/–5FU; Fig. 4C), corresponding to 2.4% of the total content of CRC of MEL cell cultures, as calculated according to column five of Fig. 3B. Although this percentage was 3–4 times higher than those of viable cells and CFC, the main result of Figure 4C is that CRC were capable of cycling in severe hypoxia, and most of them were actually cycling. This is in striking contrast with the fact that the vast majority of cells surviving hypoxia (Fig. 2A) were growth-arrested (Fig. 2B). It is noteworthy that cycling CRC, easily revealed by a functional assay such as CRA, were instead undetected in the experiments shown in Figure 2B, indicating that flow cytometric analyses of cell cycle are of value for cell bulk but not for a very small cell population, such as that of CRC.
The rather unexpected findings shown in Figure 4C led to a greater effort to detect a 5FU-resistant, quiescent CRC subset of the MEL cell population, possibly corresponding to cells responsible for long-term persistence of leukemia in vivo. The experiments of Figures 3A and 4C had been carried out within the timeframe of standard CRA assay shown to quantify STR-type stem cells [2]. We thus established extendedCRA assays in which LC2 repopulation was measured after a prolonged incubation. In Figure 4D, hypoxia-selected, 5FU-treated cells corresponding to the black column in Figure 4C were incubated in LC2 3 times longer, until day 18. The number of viable cells in culture underwent a 12-day lag phase followed by a rapid increase, peaking at day 17 or 18. It is noteworthy that the peak value, as well as the time difference between the beginning of culture expansion and the peak, were very much in keeping with those of the standard CRA assay (Fig. 3A). Thus, the MEL cell population includes a subset of hypoxia- and 5FU-resistant CRC capable of sustaining delayed culture repopulation as efficiently as 5FU-sensitive CRC recruitable to rapid repopulation.
|
DISCUSSION
|
|---|
This study was based on the adaptation to leukemia cells [11, 12] of the CRA assay developed in our laboratory [2] and successfully applied [3, 5, 7, 13] for the quantification in vitro of the stem cell potential of normal hematopoietic cell populations. The CRA assay, in combination with incubation under severe hypoxia as a selection procedure, showed that the MEL cell population includes highly hypoxia-sensitive CFC as well as their hypoxia-resistant, CRC-type progenitors. Thus, functionally different MEL cell subsets were identified, and a hierarchy within the MEL cell population was defined on the basis of previous studies on normal hematopoietic cells [1]. Taking into account other less immature cell subsets detectable within the population, such as nonclonogenic progenitors and postmitotic precursors committed to maturation, a high-level of heterogeneity of MEL cells emerged. It is noteworthy that such a hierarchical heterogeneity, far from being limited to primary leukemia cell populations, which presumably progress in vivo to a polyclonal state, was found instead within a stabilized, clonal leukemia cell line, and is therefore to be considered a general property of leukemia cell populations. This implies that questions about functional differences between leukemia stem and progenitor cell compartments, as well as aspects of neoplastic progression, can be addressed by dissecting clonal leukemia populations such as cell lines.
The model of cancer as a heterogeneous tissue containing a core subset of neoplastic stem cells has emerged over the past few years for solid tumors as well as leukemias [14, 15], leading to the concept of LSC [16]. However, cell characters underlying heterogeneity of neoplastic progenitors are far from well defined [17]. In this study, the CRA assay allowed us to detect STR-type LSC and show that they are highly hypoxia-resistant. Thus, the second and most important result of our study is the definition of an important metabolic character, adaptation to hypoxia, as a peculiarity of LSC with respect to the bulk of leukemia cell population, but not to their normal counterpart, the HSC. Indeed, resistance to hypoxia is believed, on one hand, to be relevant to the maintenance of cancer stem cells in the poorly vascularized core of solid tumors [18, 19] and, on the other hand, to be a feature of HSC, in keeping with the physiologically high-level of cell crowding and poor vascularization of bone marrow [1]. It is worth pointing out that adaptation to hypoxia is a property of stem cells that is expressed independently of their actual location in areas where oxygen tension is markedly reduced and, therefore, does not require LSC to be confined in those areas. That LSC may remain hypoxia-adapted independently of their actual tissue location is supported by the fact that normal circulating HSC exhibit the same resistance to hypoxia as bone marrow HSC [3, 7].
Hypoxia-adapted STR-type LSC, far from being "frozen" in a quiescent state, resulted mostly in cycle, as they were highly sensitive to 5FU. However, the level of 5FU-sensitivity of LSC was similar to that measured for normal "activated" STR-HSC by the CRAcell assay [2]. In the same study, a relatively 5FU-resistant and therefore relatively quiescent STR-HSC subset was detected by the CRACFC assay [2], which is carried out within the standard CRA time frame (peak of LC2 repopulation at day 6–7) suitable to quantify STR-HSC. On the other hand, here and in another recent study dealing with CML cells [12], it was impossible to resolve different subsets of STR-type LSC within leukemia cell populations by comparing CRAcell (Fig. 3A, 3B) to CRACFC (data not shown). Thus, hypoxia-resistant STR-type LSC seem to be a more homogeneous, mostly cycling cell subset, whereas STR-HSC are less homogeneous and relatively more quiescent [2, 5, 13]. This conclusion that stem cell hierarchy within the MEL cell population is different from normal is in keeping with current views on the issue [20]. The higher probability of STR-type LSC to cycle, compared with HSC, may represent the basis for the developmental advantage of leukemic over normal hematopoiesis. Such a feature, which must be confirmed by further studies, may turn out to be of some therapeutic interest, in view of the possibility of targeting STR-type LSC with a good therapeutic index with respect to STR-HSC. It is worth pointing out here that, as hypoxia maintains stem cell compartments, the function of cycling in hypoxia of LSC, as well as HSC, is most probably to sustain self-renewal, whereas generation of more mature progenitors is restrained by hypoxia itself, as we suggested previously [5]. Furthermore, these results seem to definitively confirm, under nearly anoxic conditions (0.1% O2), the phenomenon of cell proliferation in hypoxia (1% O2) that we observed previously [4, 5]. Metabolic and signaling questions underlying this phenomenon are worth being addressed in depth, especially considering that current knowledge on regulation of cell cycling derives from cells tuned on aerobic metabolism, a biochemical setting completely different from that which favors stem cell maintenance.
A subset of hypoxia- and 5FU-resistant LSC was revealed by the extendedCRA assay, which was not applied to the study of normal HSC [2]. To definitively establish whether this assay actually detects mid term/LTR-type LSC requires transplantation of the hypoxia- and 5FU-selected cell population into NOD/SCID mice. In any case, it is a fact that the MEL cell population includes a cell subset capable of persisting in quiescence, which is believed to define LSC sustaining minimal residual disease. Taken together, the data collected by the CRA and the extendedCRA assays indicate that the MEL stem cell compartment is composed of two different stem cell subsets, capable of repopulating cultures with comparable efficiency. One subset is composed of "ready-to-go," always cycling, STR-like LSC, which sustains continuous regeneration of cultures when these are incubated under conditions permissive for clonogenesis (normoxia), are capable of surviving hypoxia, and can be rescued to clonogenesis upon transfer back to normoxia (Fig. 3A). The other subset is composed of quiescent LSC placed at a longer time-distance from recruitment even when transferred to normoxia but able, once rescued to clonogenesis, to repopulate cultures with a kinetics similar to that of the always cycling STR-like LSC (compare Figs. 4D and 3A). The presence of a quiescent stem cell subset within a stabilized leukemia cell line may appear as an unexpected finding [20].
|
CONCLUSIONS
|
|---|
This work showed that hypoxia-resistance defines leukemic stem cells, much like normal stem cells, and that different subsets of hypoxia-resistant LSC exist. STR-type LSC were found to be capable not only of surviving in hypoxia but also being mostly in cycle, in contrast with the fact that almost all hypoxia-surviving cells were growth-arrested and with what previously found for HSC. Quiescent LSC were also detected that were capable of delayed culture repopulation with the same efficiency as STR-like LSC. Thus, all main components of leukemia cell populations, including different subsets of LSC as well as hypoxia-sensitive progenitors, were found within a clonal cell line. This study points to cell lines such as the MEL cells as suitable models to study neoplastic progression as well as the sensitivity to treatment of individual components of the progenitor and stem cell compartments. Furthermore, it clearly emerges from the data reported here that the hypoxia-resistant LSC represent a main target of therapeutic protocols aimed at the definitive eradication of disease and that high therapeutic index protocols might be based on treatments selectively active within a severely hypoxic environment.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
|
|---|
The authors indicate no potential conflicts of interest.
|
ACKNOWLEDGMENTS
|
|---|
We thank Dr. Roberto Caporale, AOU Careggi, General Laboratory, Dept. 4, Florence, for his cooperation in collecting and interpreting immunophenotypical data. This work was supported by Istituto Superiore di Sanità (National Program "Stem Cells"), Ente Cassa di Risparmio di Firenze and Fondazione Cassa di Risparmio di Volterra. S.G. was the recipient of a fellowship from Associazione Italiana Leucemie.
|
REFERENCES
|
|---|
- Cipolleschi MG, Dello Sbarba P, Olivotto M. The role of hypoxia in the maintenance of hematopoietic stem cells. Blood 1993;82:2031–2037.[Abstract/Free Full Text]
- Cipolleschi MG, Rovida E, Ivanovi
Z et al. The expansion of murine bone marrow cells preincubated in hypoxia as an in vitro indicator of their marrow-repopulating ability. Leukemia 2000;14:735–739.[CrossRef][Medline]
- Ivanovi Z, Dello Sbarba P, Trimoreau F et al. Primitive human HPCs are better maintained and expanded in vitro at 1 percent than 20 percent oxygen. Transfusion 2000;40:1482–1488.[CrossRef][Medline]
- Ivanovi Z, Bartolozzi B, Bernabei PA et al. Incubation of murine bone marrow cells in hypoxia ensures the maintenance of marrow-repopulating ability together with the expansion of committed progenitors. Br J Haematol 2000;108:424–429.[CrossRef][Medline]
- Ivanovi Z, Belloc F, Faucher JL et al. Hypoxia maintains and interleukin-3 reduces the pre-colony-forming cell potential of dividing CD34(+) murine bone marrow cells. Exp Hematol 2002;30:67–73.[CrossRef][Medline]
- Danet GH, Pan Y, Luongo JL et al. Expansion of human SCID-repopulating cells under hypoxic conditions. J Clin Invest 2003;112:126–135.[CrossRef][Medline]
- Ivanovic Z, Hermitte F, Brunet de la Grange P et al. Simultaneous maintenance of human cord blood SCID-repopulating cells and expansion of committed progenitors at low O2 concentration (3%). STEM CELLS 2004;22:716–724.[Abstract/Free Full Text]
- Lotem J, Sachs L. Epigenetics wins over genetics: induction of differentiation in tumor cells. Semin Cancer Biol 2002;12:339–346.[CrossRef][Medline]
- Reynolds TY, Rockwell S, Glazer PM. Genetic instability induced by the tumor microenvironment. Cancer Res 1996;56:5754–5757.[Abstract/Free Full Text]
- Jensen PO, Mortensen BT, Hodgkiss RJ et al. Increased cellular hypoxia and reduced proliferation of both normal and leukaemic cells during progression of acute myeloid leukemia in rats. Cell Prolif 2000;33:381–395.[CrossRef][Medline]
- Desplat V, Faucher JL, Mahon FX et al. Hypoxia modifies proliferation and differentiation of CD34(+) CML cells. STEM CELLS 2002;20:347–354.[Abstract/Free Full Text]
- Giuntoli S, Rovida E, Barbetti V et al. Hypoxia suppresses BCR/Abl and selects Imatinib-insensitive progenitors within clonal CML populations. Leukemia 2006;20:1291–1293.[CrossRef][Medline]
- Hermitte F, Brunet de la Grange P, Belloc F et al. Very low O2concentration (0.1%) favors G0 return of dividing CD34+ cells. STEM CELLS 2006;24:65–73.[Abstract/Free Full Text]
- Reya T, Morrison SJ, Clarke MF et al. Stem cells, cancer, and cancer stem cells. Nature 2001;414:105–111.[CrossRef][Medline]
- Al-Hajj M, Clarke MF. Self-renewal and solid tumor stem cells. Oncogene 2004;23:7274–7282.[CrossRef][Medline]
- Warner JK, Wang JC, Hope KJ et al. Concepts of human leukemic development. Oncogene 2004;23:7164–7177.[CrossRef][Medline]
- Jordan CT, Guzman ML. Mechanisms controlling pathogenesis and survival of leukemic stem cells. Oncogene 2004;23:7178–7187.[CrossRef][Medline]
- Harris AL. Hypoxia—a key regulatory factor in tumour growth. Nat Rev Cancer 2002;2:38–47.[CrossRef][Medline]
- Blagosklonny MV. Antiangiogenic therapy and tumor progression. Cancer Cell 2004;5:13–17.[CrossRef][Medline]
- Blagosklonny MV. Target for cancer therapy: proliferating cells or stem cells. Leukemia 2006;20:385–391.[CrossRef][Medline]
Top
Abstract
Introduction
) or hypoxia (
) were counted at the indicated times. Data represent means ± SEM of three independent experiments. (B): Time course of the number of CFC. Time-zero cells or cells recovered at the indicated times from cultures incubated in normoxia (gray bar) or hypoxia (black bar) described in (A) were centrifuged, counted, and replated in semisolid medium, to be incubated for 7 days in any case in normoxia. The numbers of colonies developed are shown as an index of CFC. Data represent means ± SEM of three independent experiments. (C): Percentage maintenance in hypoxia of viable cells and CFC. Data from (A and B) are computed to calculate the percentage maintenance of viable cells and CFC in hypoxia versus normoxia at day 3 (i.e., at the peak of culture expansion in normoxia). Abbreviation: CFC, colony-forming cell.