Deacetylase inhibitors modulate proliferation and self-renewal properties of leukemic stem and progenitor cells

Abstract

Acute myeloid leukemia (AML) is a highly malignant disease that is not curable in the majority of patients. Numerous non-random genetic abnormalities are known, among which several translocations such as PLZF/RARα or AML1/ETO are known to aberrantly recruit histone deacetylases. Deacetylase inhibitors (DACi) are promising drugs leading to growth inhibition, cell cycle arrest, premature senescence and apoptosis in malignant cells. It is believed that DACi may have clinical efficacy by eradicating the most primitive population of leukemic stem and progenitor cells, possibly by interfering with self-renewal.

The aim of the study was to investigate the effects of DACi on leukemic stem and progenitor cells using murine transduction-transplantation models of hematopoietic cells harboring the leukemia-associated fusion proteins (LAFP) PLZF/RARα or a truncated AML1/ETO protein (AML1/ETO exon 9). We show that the self-renewal and short-term repopulation capacity of AML1/ETO- or PLZF/RARα-expressing Sca1⁺/lin^- stem and progenitor cells are profoundly inhibited by clinically applicable concentrations of the DACi dacinostat and vorinostat. To further investigate the mechanisms underlying these effects, we examined the impact of DACi on the transcription factor c-MYC and the Polycomb group protein BMI1, which are induced by LAFP and involved in leukemic transformation. In AML1/ETO or PLZF/RARα-positive 32D cells, DACi-mediated antiproliferative effects were associated with downregulation of BMI1 and c-MYC protein levels. Similar effects were demonstrated in primary samples of cytogenetically defined high-risk AML patients. In conclusion, DACi may be effective as maintenance therapy by negatively interfering with signaling pathways that control survival and proliferation of leukemic stem and progenitor cells.

Introduction

Acute myeloid leukemia (AML) is an aggressive malignant disorder affecting mostly older patients. In spite of intensive treatment, the majority of AML patients relapse and die of their disease. As residual leukemic stem and progenitor cells (LSC) are a potential reservoir for AML relapse, their response has to be taken into account when evaluating the efficacy of novel therapies. LSC are functionally defined by their capacity to initiate the disease upon inoculation into irradiated mice due to their extensive proliferation and self-renewal capacity.^1,2 Thus, targeting of transplantable LSC while sparing normal hematopoietic stem cell function is a matter of intensive research.

Deacetylase inhibitors (DACi) are promising drugs that lead to growth inhibition, cell cycle arrest, premature senescence and apoptosis of malignant cells.³ The underlying molecular mechanisms include epigenetic reprogramming of the transcriptome by alteration of the acetylation status of core histones and thus modification of the chromatin structure. DACi additionally induce selective proteasomal degradation of histone decetylase 2 (HDAC2), DNA methyltransferases and other proteins responsible for aberrant gene repression and signaling.^4-6 This effect equally alters the cellular program, but to date, the underlying mechanisms remain elusive and have not yet been studied in the leukemic stem and progenitor compartment.

Balanced translocations such as t(8;21) and t(11;15) and the corresponding leukemia-associated fusion proteins (LAFP) AML1/ETO and PLZF/RARα, respectively, are a hallmark of AML. Ectopic expression of PLZF/RARα or a truncated AML1/ETO protein (AML1/ETO exon 9) in murine hematopoietic stem cells and transplantation into syngeneic mice recapitulates the leukemic phenotype characterized by a differentiation block and an increased self-renewal capacity.^7,8 AML1/ETO as well as PLZF/RARα have been found to induce the transcription factor c-MYC,^9,10 and ectopic expression of c-MYC in murine bone marrow elicited an aggressive myeloid leukemia by conferring self-renewal capacity to committed myeloid progenitor cells.^11,12 The Polycomb group (PcG) protein BMI1 is essential for the leukemic transformation property of PLZF/RARα and a target gene of c-MYC.^13,14 The PcG family of proteins maintains the repressed transcriptional state of their target genes und thus contributes to regulation of stem cell renewal.¹⁵ As part of the Polycomb-repressive complex-1 (PRC1), BMI1 mediates the interaction between PLZF/RARα and PRC1, resulting in repression of retinoic acid responsive genes.

As aberrant recruitment of histone deacetylase activity is a common oncogenic feature of LAFP, we have analyzed the impact of DACi on leukemic AML1/ETO and PLZF/RARα-positive stem and progenitor cells using the potent DACi dacinostat and vorinostat. Both DACi belong to the group of hydroxamic acid derivatives which have shown anti-tumor activity in pre-clinical models and proven its efficacy in patients with hematologic malignancies including AML.^16-18 The aim of the study was to investigate the effects of DACi on leukemic stem and progenitor cells using doses that are effective and allow long-term treatment.

Results

Deacetylase inhibitors suppress proliferation of AML fusion protein-expressing 32D cells

We studied the impact of DACi using two well characterized models of acute leukemia induced by PLZF/RARα or a truncated form of AML1/ETO (AML1/ETO exon 9), respectively.^7,8 Both leukemia-associated fusion proteins (LAFP) were cloned into proviral constructs and retrovirally transduced into 32D cells. Expression of the fusion proteins was confirmed by western blot (Fig. 1A and B). The LAFP-transduced 32D cells were cultured for 7 d with dacinostat (10 and 20 nM) or vorinostat (1 and 2 µM). Selection of the respective DACi doses was based on previously proven inhibition of deacetylase activity in primary AML progenitor cells and clinical relevance.^4,19 DACi treatment efficiently inhibited proliferation of LAFP- as well as mock-transduced 32D cells by the factor 5–21 compared with untreated controls (Fig. 1C). This anti-proliferative effect was dose-dependent, but not statistically significantly different between LAFP- and mock-transduced 32D cells.

Figure 1. Effect of DACi treatment on proliferation of 32D cells expressing AML1/ETO or PLZF/RARα. Expression of the oncofusion genes AML1/ETO exon 9 (A) and PLZF/RARα (B) in 32D cells was demonstrated by western blotting using the indicated antibodies. Lysates of Kasumi and AML1/ETO transduced 293T cells were used as positive controls for AML1/ETO. (C) The 32D cells expressing mock, AML1/ETO or PLZF/RARα were treated with dacinostat (10 nM, 20 nM) or vorinostat (1 µM, 2 µM) for 7 days, and proliferation was determined by trypan blue exclusion of viable cells. Results are given as fold change reduction in cell growth of DACi-treated cells compared with untreated controls. The means of three separate experiments performed in duplicates +/− SEM are given.

Deacetylase inhibitors impair short-term engraftment potential of leukemic stem cells

To determine whether DACi are targeting the leukemic stem and progenitor cell compartment, we utilized a transduction-transplantation mouse model (Fig. 2A). PLZF/RARα and AML1/ETO were retrovirally transduced into Sca1⁺/lin^- hematopoietic stem cells (HSC). FACS analysis of GFP positive cells confirmed constant transduction efficiencies of > 60–70% (Fig. 2B). Addition of dacinostat (20 nM) or vorinostat (2 µM) to a 7-day culture of 2.5 x 10⁴ mock- or LAFP-expressing Sca1⁺/lin^- cells resulted in a six- to 22-fold reduction of cell numbers compared with untreated controls (n = 3; Fig. 2C). Importantly, the antiproliferative effect of the DACi on the bulk of mock- or LAFP-transduced HSC did not vary significantly.

Figure 2. Effect of DACi for regulation of leukemia-initiating potential of AML1/ETO and PLZF/RARα positive stem cells. (A) Experimental strategy for studying the influence of DACi dacinostat and vorinostat on the biology of murine HSC. Sca1⁺/lin^- bone marrow (BM) cells were infected with the indicated retroviruses and maintained for one week in liquid culture supplemented with the indicated growth factors in the presence of DACi. All cells were inoculated into lethally irradiated recipients that were then sacrificed at day 12 after transplantation. (B) GFP reporter gene expression is given by FACS analysis of mock, AML1/ETO and PLZF/RARα infected Sca1⁺/lin^- cells. One out of three representative experiments is shown. (C) Proliferation of mock, AML1/ETO and PLZF/RARα expressing Sca1⁺/lin^- HSC exposed to DACi (20 nM dacinostat, 2 µM vorinostat). The cells were treated with indicated DACi for seven days and proliferation was determined by trypan blue exclusion of viable cells. Reported is the mean of fold change reduction of cell number with SEM compared with untreated controls in vitro (n = 3). (D) Numbers of spleen colonies are given as fold change reduction in CFU-S compared with untreated controls (n = 3). The colony numbers show the mean of three independent CFU-S12 experiments with SEM of mock, AML1/ETO and PLZF/RARα expressing Sca1⁺/lin^- HSC exposed to DACi (dacinostat, vorinostat) for seven days in vitro and transplanted into three mice per group. P-values: p = 0.02 (dacinostat-treated AML1/ETO expressing Sca1⁺/lin^- HSC), p = 0.005 (vorinostat-treated PLZF/RAR Sca1⁺/lin^- HSC) compared with mock-treated cells. (E) The expression of the leukemia-associated fusion proteins (LAFP) AML1/ETO and PLZF/RARα in the spleen colonies was assessed by RT-PCR. β-actin was used as a control. (n = 2) (F) For verification of the RT-PCR analyses, one quantitative real-time PCR analysis of the transgenes AML1/ETO and PLZF/RARα of the spleen colonies was performed. Internal reference gene was GAPDH. Results are represented as 2^-ΔΔCT values.

Engraftment and in vivo proliferation capacity of the DACi-pretreated leukemic stem and progenitor cells were assessed in spleen colony-forming unit (CFU-S) assays²⁰ by transplanting all progeny grown from 2.5 × 10⁴ mock or LAFP-transduced Sca1⁺/lin^- cells, i.e., 15–30 × 10⁴ untreated and 1–6 × 10⁴ DACi-treated cells. Spleen colony formation was not affected by exposure to DACi in mice receiving mock-transduced cells (Fig. 2D). In contrast, DACi-pretreatment of LAFP-transduced Sca1⁺/lin^- cells led to a significant reduction in CFU-S numbers of PLZF/RARα- or AML1/ETO-positive cells in the spleen of transplanted mice as analyzed by RT-PCR (Fig. 2D and E). Quantitative RT-PCR of spleen cells confirmed a 4–7 log reduction of LAFP-positive cells (Fig. 2F) suggesting profound depletion of short-term LSC by the potent DACi dacinostat and vorinostat.

Deacetylase inhibitors exhaust in vitro self-renewal potential of murine AML1/ETO- and PLZF/RARα-positive HSC

To determine if dacinostat and vorinostat inhibit the self-renewal potential of the LAFP-positive HSC, serial replating experiments were performed as shown in Figure 3A. We have previously reported that LAFP such as PML/RARα and PLZF/RARα confer an increased replating efficiency to Sca1⁺/lin^- HSC, which correlates with an aberrant self-renewal potential.²¹ Here, Sca1⁺/lin^- stem cells carrying AML1/ETO or PLZF/RARα had a serial replating capacity far exceeding that of mock infected controls (n = 3; more than six vs. three rounds of plating, Fig. 3B–D), which was significantly impaired by DACi treatment. In the presence of dacinostat or vorinostat, colony formation of AML1/ETO-positive cells was significantly reduced from the third plating on and completely abrogated at the sixth plating. Using PLZF/RARα-positive cells, the same results were obtained for dacinostat, while residual colony formation was observed in the presence of vorinostat. Thus, prolonged treatment with moderate doses of DACi may severely inhibit or even eliminate self-renewing leukemic stem and progenitor cells.

Figure 3. Effect of DACi treatment on the replating efficiency of Sca1⁺/lin^- HSC expressing AML1/ETO and PLZF/RARα. (A) Experimental strategy for studying the influence of the indicated DACi on the biology of murine HSCs. Sca1⁺/lin^- BM cells were infected with the indicated retroviruses and plated in semi-solid medium with the indicated growth factors to determine the serial plating potential in the presence of either dacinostat or vorinostat. Replating efficiency of murine Sca1⁺/lin^- HSC expressing mock (B), AML1/ETO (C) and PLZF/RARα (D) upon exposure to dacinostat (10 nM, 20 nM) or vorinostat (1 µM, 2 µM). Numbers of platings (first to sixth) and CFU numbers are provided. Error bars show means of triplicates with SEM, n = 3.

DACi decrease BMI1- und c-MYC protein expression in LAFP-positive 32D cells

We next aimed to examine the effect of DACi on c-MYC as a potential mechanism underlying LSC targeting in LAFP-positive cells. Due to a limited amount of Sca1⁺/lin^- HSC, AML1/ETO or PLZF/RARα-transduced 32D cells were used for western blot analyses. Treatment with dacinostat or vorinostat for 48 h led to a dose-dependent reduction of c-MYC protein levels (Fig. 4A). The polycomb group protein BMI1 is another key player involved in regulation of the proliferative potential of leukemic stem and progenitor cells.²² Similar to c-MYC, BMI1 was also consistently reduced following DACi treatment. These inhibitory effects were observed in both LAFP- and mock-transduced 32D cells (Fig. 4A). As the cell cycle regulatory gene p21 is a known target of DACi treatment,²³ we analyzed its expression and found it to be increased in LAFP-positive and mock 32D as expected (Fig. 4B).

Figure 4. Effect of DACi on Wnt target genes in Mock, AML1/ETO and PLZF/RARα expressing 32D cells. Infected 32D cells (mock, AML1/ETO and PLZF/RARα) were cultured in the presence or absence of dacinostat (10 and 20 nM) or vorinostat (1 and 2 µM) for 48 h. (A) Nuclear extracts were prepared. Protein expression of BMI1, c-MYC and Ac-H3 was assessed by western blot analysis with the indicated antibodies. (B) Q-RT-PCR analysis of p21 expression. Results are represented as 2^-ΔΔCT values. Internal reference gene was GAPDH. One representative experiment is shown.

Effect of dacinostat and vorinostat on primary patient samples of high-risk AML

Risk classification of AML is based on cytogenetic and molecular genetic aberrations. Accordingly, AML1/ETO which is present in about 12% of AML cases is associated with a good prognosis.²⁴ The same holds true for the rare PLZF/RARα-positive acute promyelocytic leukemia.²⁵ To assess the potential clinical implications of our findings, we collected leukemic samples from AML patients with genetically defined high-risk disease, who often fail standard chemotherapy.^26,27

Figure 5 displays the results of several AML samples representing the most frequent and clinically relevant high-risk features: complex cytogenetic abnormalities, i.e., three or more chromosomal aberrations (FFM03), monosomy of chromosome 7 (FFM12), a monosomal karyotype, i.e., presence of one single monosomy in addition to a complex aberrant karyotype (FFM05), and a normal karyotype with presence of an internal tandem duplication in the FMS-related tyrosine kinase 3 gene (FLT3-ITD; FFM23).²⁴ Here, we show that primary AML samples contain primitive leukemic progenitor cells that are capable of in vitro proliferation for several weeks in presence of the cytokines interleukin-3, stem cell factor, flt3-ligand and thrombopoietin. Within six weeks, an exponential cell growth resulting in a 2–4 log expansion was observed in all three AML samples tested (Fig. 5A).

Figure 5. AML cells after DACi treatment. CD34⁺/CD38^- progenitors isolated from AML patients and enriched by magnetic column separation were cultured in the presence or absence of DACi (dacinostat (20 nM) or vorinostat (2 µM). Proliferation was determined by trypan dye exclusion assay. (A) Exponential cell proliferation of untreated and DACi-treated cells of FFM03, FFM12, FFM23 and FFM05 cells. (B) Regulation of stem-cell specific proteins by DACi. AML cells were cultured in the presence or absence of dacinostat (10 and 20 nM) or vorinostat (1 and 2 µM) for 48 h, and protein lysates were prepared. Altered expression of ac-Histone-3, p21, c-MYC and BMI1 was observed by western blot analysis for FFM05.

In the presence of dacinostat or vorinostat, the AML cultures consistently declined, confirming our results obtained in AML1/ETO or PLZF/RARα-expressing 32D cells. Hyperacetylation of histone H3 and induction of p21 served as a proof of the DAC inhibitory activity in FFM05 cells (Fig. 5B). Importantly, downregulation of c-MYC and BMI1 by the potent DACi dacinostat and vorinostat was also verified in this primary AML sample.

Discussion

In the majority of AML patients, a small population of leukemic stem and progenitor cells persists after intensive chemotherapy and represents a reservoir of cells that may lead to recurrent disease. Targeting these LSC may improve patient outcome by reducing the probability of AML relapse.

In our study, we demonstrate that the DACi dacinostat and vorinostat effectively diminish self-renewal and short-term repopulating capacity of AML1/ETO and PLZF/RARα-positive stem and progenitor cells. The well characterized LAFP AML1/ETO exon 9 (referred to as AML1/ETO) and PLZF/RARα are known to confer LSC properties to normal HSC²⁸ and induce an acute leukemia in C57BL/6N mice,^8,29 rendering these mouse models excellent read-outs for leukemogenesis. We confirmed that Sca1⁺/lin^- HSC carrying AML1/ETO or PLZF/RARα possess an almost indefinite replating capacity³⁰ and give rise to leukemic colonies in the CFU-S assay, thus demonstrating engraftment and in vivo proliferation of the transplanted LSC.

After treatment of the LAFP-transduced HSC with dacinostat or vorinostat, only few PLZF/RARα- or AML1/ETO-expressing spleen cells could be detected by qRT-PCR, suggesting profound depletion of leukemia-initiating cells. Expression of LAFP in the spleens was routinely evaluated to discriminate between leukemic and residual normal cells, which were present in the transplanted population as we did not isolate transduced Sca1⁺/lin^- cells in view of consistently high transduction efficiencies exceeding 60–70%.

These results are in agreement with our previously published data showing considerable loss of normal committed progenitor cells in response to DACi, as measured by cell proliferation, while largely sparing the stem cell fraction giving rise to CFU-S.¹⁹ Here, we provide evidence that leukemic short-term repopulating stem cells are more susceptible to DACi-mediated damage than their normal counterparts.

In methylcellulose culture, our HSC exhausted their proliferative capacity after the third plating, while LAFP-transduced HSC form colonies for more than three passages as previously reported.³⁰ Serial replating and day 12 spleen colony-forming unit assays (CFU-S) are thus both suitable to detect functional, self-renewing LSC.²⁰ We observed a dose-dependent suppression of the self-renewal capacity of AML1/ETO-positive cells by dacinostat and vorinostat. This observation is consistent with a gradual loss of self-renewing LSC over time and is in agreement with our previous report demonstrating a diminishing replating capacity of PLZF/RARα-positive LSC upon dacinostat treatment.³¹

As primary LSC are rare, pathway analysis was performed using transduced IL-3 dependent myeloid 32D cells as a well-accepted model system. Functional impairment of murine 32D but also of human AML cells was associated with downregulation of c-MYC and BMI1. The proliferative potential of leukemic stem and progenitor cells lacking BMI1 is compromised, because they eventually undergo proliferation arrest and show signs of differentiation and apoptosis, leading to failure of the leukemic cells to initiate disease in transplanted syngeneic hosts.²² Dacinostat and vorinostat reduced BMI1 protein levels leading to reversion of the leukemic phenotype. As BMI1 is expressed in all AML cells analyzed so far, it may thus support proliferation and self-renewal of leukemic and progenitor cells independently of PLZF/RARα.^32,33 Accordingly, its depletion may be associated with a more general antileukemic or even antitumor effect of DACi, based on our results in AML1/ETO-positive hematopoietic cells and, by another group, in various breast cancer cell lines.³⁴ Downregulation of c-MYC by DACi has been reported previously^35-38 and may also account for inhibition of LSC capacity. Here, a reduced protein level of c-MYC was recognized as early as 6 h after addition of DACi (data not shown), suggesting that c-MYC repression is a direct effect of DACi treatment. In contrast, the regulation of the c-MYC target gene BMI1¹³ by DACi was found to be indirect and independent from c-MYC in breast cancer.³⁴ Further studies will have to show if BMI1 is a direct target of c-MYC in our leukemic models.

In summary, we present evidence that DACi interfere with signaling pathways controlling survival and proliferation of leukemic stem and progenitor cells. Thus, abrogation or severe reduction of in vitro self-renewal by the hydroxamic acid analogs dacinostat and vorinostat may be associated with beneficial clinical effects, e.g., inducing sensitivity of LSC to chemotherapy.³⁹ Our results have potential impact on the clinical development of DACi. To date, DACi have not been approved for the treatment of AML patients. Although complete remissions have been reported, most patients do not enjoy a satisfactory and sustained antileukemic effect and/or suffer from considerable dose-dependent side effects. According to our data showing a substantial inhibition of LSC, potent DACi may be more relevant in the setting of minimal residual disease requiring prolonged treatment.

Materials and Methods

Cells lines and reagents

The hematopoietic progenitor cells 32D were maintained in RPMI-1640 (Invitrogen) plus 10% fetal calf serum (FCS) (Invitrogen) supplemented with 10 ng/ml of m-IL3 (Peprotech). The ecotropic Phoenix packaging cell line was cultured in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen) containing 10% FCS. Dacinostat was kindly provided by Novartis Pharmaceuticals and vorinostat (suberoylanilide hydroxamic acid, SAHA) by Biozol.

AML samples

Peripheral blood samples were obtained from AML patients at diagnosis or relapse. All patients gave written informed consent. Collection of patient samples was approved by the ethics committee of the Goethe-University of Frankfurt. Baseline morphology, cytogenetics and cell surface antigen analysis were performed as part of the routine clinical evaluation of the patients. Isolation of mononuclear cells and the subsequent immunomagnetic selection of CD34⁺ or CD34⁺CD38^- cells were performed as previously described.⁴⁰

Culture and analysis of leukemic progenitor cells

CD34⁺ cells were maintained in X-Vivo (Lonza) supplemented with 10% FCS (Hyclone), 1% L-Glutamine (Invitrogen), interleukin-3, thrombopoietin (25 ng/mL each), stem cell factor and Flt-3 ligand (50 ng/mL each, Peprotech) for seven days. Either dacinostat or vorinostat were added to the cultured cells. Flow cytometry was performed as previously described.⁴⁰ For the colony assay, cells harvested from suspension culture were plated in methylcellulose (Methocult® GF H4434, CellSystems Biotech) ± dacinostat or vorinostat. Colonies (> 20 cells) were counted after 12 d and replated if possible. Data were given as mean ± SEM and compared by the Student’s t or Mann-Whitney U-test, as appropriate. p-values < 0.05 were considered as significant.

Isolation of Sca1⁺/lin^- murine HSCs

Sca1⁺/lin^- HSCs were isolated from female C57BL/6J mice from 6 to 12 weeks of age (Janvier) killed by cervical dislocation. Bone marrow (BM) was harvested from femora and tibiae by flushing the bones with a syringe and 26-gauge needle. Cells were “lineage depleted” using the Lineage Cell Depletion Kit (Miltenyi Biotech GmbH) according to the manufacturer’s instructions. Sca1⁺ cells were purified by immunomagnetic beads using EasySep column-free system according to the manufacturer’s instruction (StemCell Technologies, Inc.). Purified cells were pre-stimulated prior to further use for two days in medium containing mIL-3 (20 ng/mL), mIL-6 (20 ng/mL) and murine stem cell factor (mSCF: 100 ng/mL; Peprotech).

Retroviral infection

Phoenix cells were transfected with retroviral vectors as described before.⁴¹ Retroviral supernatant was collected at days two and three after transfection. Target cells (32D or Sca1⁺/lin^- murine HSC) were plated onto retronectin-coated (Takara-Shuzo) nontissue culture-treated 24-well plates and exposed to the retroviral supernatant for three hours at 37°C, in the presence of polybrene if 32D cells were used. Cells were centrifuged at 600 g for 45 min. Infection was repeated four times and infection efficiency had to be 70%, as assessed by the detection of green fluorescent protein (GFP)-positive cells by fluorescence-enhanced cell sorting (FACS).

Colony assays, replating efficiency, murine Sca1⁺/lin^- cells

At day two after infection, Sca1⁺/lin^- cells were plated at 5,000 cells/ml in methylcellulose (Methocult® GF M3534, CellSystems Biotech StemCell) and treated with DACi (dacinostat, vorinostat). On day 12 after plating, the colony number was counted. After removing the methylcellulose, 5,000 cells/ml were plated again in fresh methylcellulose treated with DACi. Replating efficiency was determined by serial plating and counting.

Day 12 spleen colony-forming unit assay (CFU-S12), murine Sca1⁺/lin^- cells

After seven days of culture of Mock, AML1/ETO and PLZF/RARα infected Sca1⁺/lin^- cells (treated with DACi: dacinostat, vorinostat), all grown cells, from initially 25,000 Sca1⁺/lin^- cells, isolated from C57B/6 mice (Ly5.2), were injected into lethally irradiated (12 Gy) female C57b/6 mice (Ly5.2), eight to 12 weeks of age. Three mice per group were transplanted mice and euthanized 12 d later. Spleens were either fixed in Tellesnizky’s fixative for five minutes and subsequently transferred to 70% ethanol or used for RNA isolation and subsequent RT-PCR analysis.

Western blotting

Western blotting was done according to widely used protocols with the following antibodies: anti-β-actin, anti-acetyl-Histone-3 (from Cell Signaling Technology), anti-c-MYC, anti-BMI1 (from Santa Cruz Biotechnologies). All antibodies were diluted in 5% low fat dry milk or 5% BSA. Blocking was performed in 5% low fat dry milk; washing was performed in TBS containing 0.1% Tween 20 (TBS-T).

Real-time PCR-TaqMan (qRT-PCR)

Total RNA and first strand DNA were obtained according to standard protocols. TaqMan-PCR was performed in triplicate using the ABI PRISM 7700 (Applied Biosystems). The related “assays-on-demand” for p21 transcripts were used according to the manufacturer's instructions (Assay-ID: p21 - Mm00432448_m1) (Applied Biosystems). Internal reference gene was glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Assay-ID: Mm99999915_g1). The expression data were evaluated using the ΔΔCT method.⁴² Analysis of AML1/ETO and PLZF/RARα in the spleen cells was performed in reference laboratories in Frankfurt (Dr. Heike Pfeifer, Department of Medicine II, Johann Wolfgang Goethe-University) and Munich (PD Dr. Susanne Schnittger) by quantitative real-time PCR (qRT-PCR) using GAPDH as a housekeeping gene, as previously described.

RT-PCR

Total RNA and cDNA were obtained according to standard protocols. For detection of the AML1/ETO and PLZF/RARα fusion transcripts by RT-PCR, the following primer pairs were used with an annealing temperature of 55°C; for AML1/ETO: AML1b_fwd (TTGTCGGTCGAAGTGGAAGA) and ETO_rev (GCGCCATTCAAGGCTGTAG) and for PLZF/RARα: PLZF_fwd (AGGCTGTGGAGCAGCAGCACAGGAAG) and RAR_rev (TCTGGATGCTGCGGCGGAAGAA).

Disclosure of Potential Conflicts of Interest

G.B. has received honoraria and travel grants from Novartis Pharma GmbH.

Authorship

Contribution: A.R., K.S., M.K., S.W., A.V., J.R., C.O. and B.B. performed experiments; G.B., O.H.K., H.S. and M.R. designed the study and analyzed and interpreted the data; A.R., G.B., O.H.K. and M.R. wrote the paper; and all authors reviewed/revised the manuscript and gave their final approval of the version to be published.