|Home | About | Journals | Submit | Contact Us | Français|
Acute promyelocytic leukemia (APL) results from a blockade of granulocyte differentiation at the promyelocytic stage. All-trans retinoic acid (ATRA) induces clinical remission in APL patients by enhancing the rapid differentiation of APL cells and the clearance of PML-RARα, APL's hallmark oncoprotein. In the present study, we demonstrated that both autophagy and Beclin 1, an autophagic protein, are upregulated during the course of ATRA-induced neutrophil/granulocyte differentiation of an APL-derived cell line named NB4 cells. This induction of autophagy is associated with downregulation of Bcl-2 and inhibition of mTOR activity. Small interfering RNA-mediated knockdown of BECN1 expression enhances apoptosis triggered by ATRA in NB4 cells but does not affect the differentiation process. These results provide evidence that the upregulation of Beclin 1 by ATRA constitutes an anti-apoptotic signal for maintaining the viability of mature APL cells, but has no crucial effect on the granulocytic differentiation. This finding may help to elucidate the mechanisms involved in ATRA resistance of APL patients, and in the ATRA syndrome caused by an accumulation of mature APL cells.
Acute promyelocytic leukemia (APL) is characterized by an arrest of the terminal differentiation of myeloid cells into granulocytes.1 The genetic hallmark of this disease is a reciprocal chromosomal translocation that joins the RAR (retinoic acid receptor) α gene to the PML (promyelocytic leukemia) gene.2 The resulting chimeric protein, PML-RARα, is thought to be responsible for the arrest of granulopoiesis by directly inhibiting the transcription of the retinoic acid target gene, a mechanism that could account for its leukemogenic properties.3–5 Administration of pharmacological doses of All-trans retinoic acid (ATRA) to APL patients produces a clinical remission of the disease by inducing the maturation of promyelocytes and the degradation of the PML-RARα protein, leading to the eradication of the leukemic stem cells.5–8 However, some of these patients relapse due to the development of ATRA resistance by leukemic cells. Moreover, one of the main complications of ATRA treatment of APL patients is the ATRA syndrome, which is caused by the accumulation of mature granulocytes and can lead to a fatal outcome in APL patients.9 Being able to identify the molecular pathways that can enhance the effectiveness of ATRA-induced both differentiation of APL cells and death of mature cells will therefore be helpful in designing new strategies for improving the treatment of patients affected by APL.
Macroautophagy (referred to as autophagy hereafter) is a lysosomal process that induces the degradation of cytoplasmic constituents such as proteins and organelles, thus leading to cell renovation. In response to environmental and hormonal changes, autophagy is induced to provide cells with the nutrients and energy necessary for cell survival and cellular remodeling.10–12 Paradoxically, under certain circumstances, autophagy contributes to programmed cell death.13 During the autophagy process, a cell sequesters cytoplasmic material within double-membrane vesicles known as autophagosomes, and delivers them to lysosomes for degradation by lytic enzymes. At the molecular level, the Atg proteins (the products of autophagy-related genes) orchestrate the sequential steps of generating autophagosomes.14 An early event in the initiation of autophagy is the binding of Beclin 1 (a mammalian ortholog of yeast Atg6) to the class III PtdIns 3-kinase (hVps34) which promotes the recruitment of other Atg proteins to the phagophore membrane.15,16 Beclin 1 possesses a BH3 only-domain which is necessary for its binding to Bcl-2 and Bcl-XL. This binding represses Beclin 1-dependent autophagy in several experimental settings.17–19
Recent studies have established the role of autophagy during mammalian development and cell differentiation, processes that require an extensive cellular remodeling by a rapid change in intracellular contents, such as organelles and proteins.20,21 For example, differentiation of T-lymphocytes and erythrocytes requires a selective removal of mitochondria by autophagy.22–24 Along this line, mice lacking the essential autophagy-related gene Atg7 in the hematopoietic system develop severe anemia and lymphopenia as a result of defective removal of mitochondria.23 Similarly, Atg7 and Atg5 are involved in normal adipocyte differentiation suggesting an essential function of autophagy in adipogenesis.25,26 Autophagy is also an important event for megakaryocytic maturation of the chronic myelogenous leukemia,27 and differentiation of neuroblastoma and glioma stem progenitor cells, as well.28,29 Moreover, autophagy is required for the pre-implantation development of mouse embryos,30 and its deficiency in murine models lacking the key autophagy-regulatory genes (e.g., Ambra1, Atg5 and Atg7) induces the accumulation of abnormal proteins during the development of the nervous system20 suggesting its essential role in developmental processes.
However, the function of autophagy-regulatory pathways in ATRA-induced neutrophil/granulocyte differentiation of APL cells is still not well documented. In the present study, we investigated the autophagic responses during the course of ATRA-induced maturation of an APL-derived cell line, NB4 cells. In particular, we explored the involvement of Beclin 1 in regulating NB4 cell death and differentiation, two processes that occur during ATRA treatment of APL cells.
To find out whether autophagy regulates the differentiation of APL cells, autophagic activity was evaluated during the course of ATRA-induced maturation of NB4 cells into neutrophils/granulocytes. We first investigated the amount of autophagosomes present in cells by an ultrastructural analysis based on electron microscopy studies. As shown by the electron micrographs in Figure 1A, the treatment of NB4 cells with ATRA promoted the accumulation of vacuoles containing degraded materials. Autophagy was also assessed by evaluating the amount of LC3 (microtubule-associated protein 1 light chain 3) protein conjugated with phosphatidyl-ethanolamine (LC3-II), a well-known marker of autophagosomes. As shown in Figure 1Ba, ATRA triggered an increase in the level of LC3-II protein expression revealing once again the accumulation of autophagosomes during the differentiation of promyelocytes to neutrophils/granulocytes. This accumulation was enhanced in the presence of lysosomal protease inhibitors (Fig. 1Bb), supporting the notion that ATRA promotes the activation of autophagy flux rather than inhibiting either autophagosome maturation or the lysosomal degradation of the autophagosomal cargo. To gain further insight into the mechanisms involved in the induction of autophagy, we examined the expression and activation of two autophagy-regulatory pathways involving Beclin 1 and mTOR, respectively. As shown in Figure 1C, ATRA treatment of NB4 cells caused an upregulation of Beclin 1 and a significant downregulation of the Bcl-2 protein, which inhibits autophagy by its interaction with Beclin 1.17 ATRA also caused a decrease in the level of phosphorylation of the p70S6 kinase, a substrate of mTOR, suggesting that it has an inhibitory effect on mTOR activity (Fig. 1C). To find out whether the ATRA-induced upregulation of autophagy and Beclin 1 observed in the acute promyelocytic leukemia cells (NB4) extends to other leukemia cell lines, we investigated the effect of ATRA on the autophagic responses of the HL60 cell line, an in vitro model of myeloid leukemia. As shown in Figure 1D, ATRA treatment of HL60 cells resulted in the accumulation of both LC3-II and Beclin 1 proteins, supporting the notion that the upregulation of autophagy and Beclin 1 by ATRA may not be a response restricted to promyelocytic leukemia cells.
Because Beclin 1 plays a pivotal function in the regulation of cell death processes,16,31–33 we further investigated whether this protein plays a role in the viability of APL cells differentiated by ATRA. For this study, we downregulated Beclin 1 expression by using a specific siRNA that targets BECN1, and then evaluated autophagy and apoptosis during ATRA treatment of NB4 cells. As shown in Figure 2A, siRNA against BECN1 reduced both the basal level of Beclin 1 and its ATRA-upregulated level. The silencing of BECN1 inhibited the basal autophagy but did not affect the induction of autophagy triggered by ATRA, as revealed by immunoblot analysis of the LC3-II amount (Fig. 2A). To verify that Beclin 1 is able to regulate inducible autophagy in NB4 cells, the effect of BECN1 siRNA on autophagy was examined following incubation of NB4 cells in nutrient-free medium (NF), a condition known to stimulate canonical autophagy. Under starvation conditions, autophagy was induced in NB4 cells, which was significantly reduced following the silencing of BECN1 expression by a specific siRNA (Fig. 2A). Taken together, these results suggest that Beclin 1 is required for starvation-induced autophagy in NB4 cells, whereas it has no crucial role in the stimulation of autophagy induced by ATRA. We also evaluated apoptosis following the silencing of BECN1 expression in NB4 cells treated with ATRA. As shown in Figure 2B, ATRA treatment of NB4 cells promoted a time-dependent accumulation of apoptotic cells (characterized by the accumulation of condensed and fragmented nuclei), which was enhanced following the knockdown of BECN1 expression by a specific siRNA. Apoptosis was also assessed by measuring the loss of mitochondrial transmembrane potential (Fig. 2C). Our results showed that the ATRA-mediated loss of mitochondrial transmembrane potential was significantly enhanced when the upregulation of Beclin 1 was prevented by a specific BECN1 siRNA, which again suggests that Beclin 1 has an anti-apoptotic effect under this condition. Furthermore, the ATRA-induced cleavage of caspase-3, which is another feature of apoptosis, was promoted when Beclin 1 upregulation was reduced by a specific BECN1 siRNA (Fig. 2D). Taken together, these results demonstrate that the upregulation of Beclin 1 during ATRA treatment is a prosurvival event that increases the viability of mature APL cells.
Given the parallels between Beclin 1 upregulation and ATRA-induced differentiation of APL cells, we next explored whether these two responses are linked. To this end, we downregulated Beclin 1 expression using a siRNA that targets BECN1, and then evaluated the neutrophil/granulocyte differentiation process. As expected, ATRA treatment resulted in the differentiation of NB4 cells that was revealed by the increased expression of the cell surface marker CD11c (Fig. 3A), the presence of cells with lobed nuclei (Fig. 3B), and the accumulation of superoxide anions (Fig. 3C), which are characteristic features of neutrophil/granulocyte differentiation. Inhibition of Beclin 1 upregulation using a specific siRNA against BECN1 did not prevent ATRA-induced CD11c expression (Fig. 3A), nuclei morphological changes (Fig. 3B) or accumulation of superoxide anions (Fig. 3C). Taken together, these results suggest that ATRA-induced Beclin 1 upregulation is not crucial for neutrophil/granulocyte maturation of APL cells.
Autophagy is a critical modulator of cancer cell fate in virtue of its ability to regulate cell survival, differentiation and proliferation.21,34 In the present study, we investigated the link between autophagy, cell death and differentiation in the context of ATRA-induced granulocytic differentiation of the acute promyelocytic leukemia (APL)-derived cell line. We provide, here, the first evidence showing that Beclin 1 is upregulated by ATRA, and that this event constitutes an anti-apoptotic signal that prolongs the life span of mature APL cells. These data are in accordance with several previous reports, demonstrating that Beclin 1 displays a cytoprotective effect against numerous stressful situations.15,16 Notably, Wang et al. have shown that Beclin 1 plays an anti-apoptotic role during the course of vitamin D3-induced differentiation of the HL60 cell line,31 an in vitro model of myeloid leukemia cells. This supports the possibility that the prosurvival function of Beclin 1 in mature cells is not restricted to the APL cell model. Along this line, we showed that autophagy and Beclin 1 were also upregulated by ATRA in the myeloid leukemia HL60 cell line. Such upregulations were also observed during the megakaryocytic differentiation of the chronic myelogenous leukemia K562 cell line, supporting the idea that autophagy and differentiation are interconnected processes.27
Interestingly, Wang et al. revealed that Beclin 1 exerts an anti-apoptotic effect by impeding the binding of the pro-apoptotic protein Bad to Bcl-XL during the differentiation of HL60 cells induced by vitamin D3. Similarly, Vázquez and Colombo demonstrated that the anti-apoptotic effect that Coxiella burnetti (a gram negative bacterium) exerts in infected host cells is prevented by Beclin 1 depletion and by the expression of a Beclin 1 mutant with defective Bcl-2 binding, suggesting that the interaction of Beclin 1 and Bcl-2 may modulate apoptosis.33 These results support the idea, that under some circumstances, the BH3-only domain of Beclin 1 may competitively disrupt the binding of pro-apoptotic proteins (e.g., Bad) to Bcl-2 or Bcl-XL, thus avoiding the induction of apoptosis. Paradoxically, the results of Ciechomska et al. showed that the binding of Beclin 1 to Bcl-2 does not modify the Bcl-2-mediated protection against apoptotic stimuli that initiates endoplasmic reticulum or mitochondria death signaling pathways.35 Thus, Beclin 1-regulated Bcl-2 anti-apoptotic function occurs only in certain circumstances. Our findings open the question as to whether or not the accumulation of Beclin 1 during ATRA-induced neutrophil/granulocyte differentiation of APL cells exerts an anti-apoptotic function by displacing the BH3-only pro-apoptotic proteins from their binding to Bcl-2 and Bcl-XL. Further studies are also required to elucidate the molecular mechanisms underlying Beclin 1 upregulation by ATRA. Interestingly, recent reports have demonstrated that autophagy contributes to the activation of neutrophils36 and the induction of a neutrophil-specific cell death program called Neutrophil extracellular traps (NETs) cell death.37 In addition, autophagy is involved in a novel form of programmed necrosis that occurs in neutrophils exposed to the inflammatory cytokine, GM-CSF.38 Thus, the biological significance of Beclin 1 functions in normal neutrophil life span under both physiological and inflammatory conditions deserves to be investigated. In fact, to the best of our knowledge there is no evidence in the literature showing that neutrophil function is abnormal in Atg-deficient mice.
We found that inhibiting the ATRA-mediated Beclin 1 upregulation by a specific siRNA directed against BECN1 did not impair ATRA-induced autophagy or the differentiation of APL cells, whereas it did affect the viability of mature APL cells. These findings suggest that Beclin 1 had no crucial role in the induction of autophagy and differentiation in ATRA-treated APL cells. However, we cannot definitely rule out the role of Beclin 1 in these processes since extinction of Beclin 1 by siRNA was not complete in our experiments, and this may be insufficient to prevent autophagy and differentiation induced by ATRA. Conversely, our results (data not shown) and those recently published by two other groups,39,40 indicate that rapamycin, an inhibitor of mTOR that activates autophagy, enhanced the differentiating-promoting effect of ATRA. Along these lines, recent studies have demonstrated that PML-RAR α, a hallmark oncoprotein of APL, is degraded by autophagy via a mechanism that involves ULK 1 (Unc-51-like kinase 1, an Atg1 homolog), a downstream target of mTOR. Moreover, a recent paper that was published while our paper was under review, has demonstrated the implication of p62/SQSTM1, a cargo receptor for the degradation of ubiquitinated substrates by autophagy, in the clearance of PML-RARα.41 Whether the autophagic clearance of the PML-RARα oncoprotein contributes to the anti-leukemogenic properties of ATRA in APL patients or not is a matter for investigation in further studies. The clinical significance of the prosurvival role of Beclin 1 in the outcome of APL patients treated with ATRA, in particular those who develop ATRA syndrome, deserves also further investigation.
ATRA (R2625), Hoechst 33258 (14530) and E64d (516485) were purchased from Sigma. Rapamycin (553211) and pepstatin A methyl ester (516485) were obtained from Calbiochem. Dihydroethidium (D-11347) and tetramethylrhodamine methylester (TMRM) (T-668) were purchased from Molecular Probes. Antibodies against the following proteins were used: phospho-p70S6 kinase (9204) and p70S6 kinase (9202), caspase-3 (9665) from Cell Signaling Technology, Inc., LC3 (L7543) and β-actin (118K4846), from Sigma, CD11c (333145) and Beclin 1 (612113) from BD Bioscience. All cell culture media were purchased from Gibco-Invitrogen. Nonspecific control small interfering RNA (siRNA) or a specific siRNA that targets BECN1 were synthesized by Eurogentec, and have been previously validated and described in reference 42.
The acute promyelocytic leukemia-derived cell line, NB4, and the myeloid leukemia cell line, HL60, were grown at 37°C in 5% CO2 in an RPMI medium supplemented with 2 mM L-glutamine and 10% decomplemented fetal calf serum. For transfection of siRNAs experiments, NB4 cells (5 × 106 cells in 0.5 ml of Opti-MEM medium) were mixed either with 0.5 nmoles of a nonspecific control siRNA or a specific siRNA that targets BECN1 and electroporated with 260 V and 1,050 mF. They were then diluted at a density of 1 × 105 cells/ml and treated as described in the figure legends.
Cells were fixed for 1 h at 4°C in 1.6% glutaraldehyde in a 0.1 M Sörensen phosphate buffer (pH 7.3), washed and fixed again in aqueous 2% osmium tetroxide, dehydrated in ethanol, embedded in Epon, and processed for electron microscopy using a FEI Tecnai Spirit transmission electron microscope at 80 kV in ultrathin sections stained with uranyl acetate and lead citrate.
Cell extracts were prepared in 10 mM Tris, pH 7.4, 1% SDS, 1 mM sodium vanadate, treated with benzonase nuclease for 5 min at room temperature, and boiled for 3 min. Fractions (30–50 µg) of cellular extract proteins were subjected to SDS-PAGE using a Tris/glycine buffer system based on the method of Laemmli. After electrophoresis, proteins were transferred to a Hybond-C super nitrocellulose transfer membrane (Amersham Biosciences, RPN203G). Protein loading was assessed by Ponceau red staining of membranes. Blots were then incubated with primary antibodies using the manufacturer's protocol followed by the appropriate horseradish peroxidase-conjugated secondary antibody. Immunostained proteins were visualized on X-ray film using the enhanced chemiluminescence (ECL) detection system.
Neutrophil/granulocyte differentiation in the NB4 cells was determined by (i) morphological assessment of cells stained with May-Grunwald Giemsa, (ii) analysis of changes in the expression of the CD11c cell surface marker and (iii) quantification of anion superoxide production by using dihydroethidium (HE) dye as previously described in references 43 and 44.
Apoptosis was determined by (i) the quantification of apoptotic nuclei (i.e., condensed and fragmented apoptotic nuclei) following Hoechst 33258 staining; (ii) measuring the mitochondrial transmembrane potential (ΔΨm) by using TMRM dye. Briefly, cells were collected and loaded with 200 nM TMRM for 30 min at 37°C in Krebs Ringer Buffered Saline supplemented with 20 µM verapamil. Cells were then analyzed using a Facscalibur (Beckman Coulter) flow cytometer; (iii) the determination of caspase-3 activity by protein gel blotting analyses of the appearance of cleaved forms of caspase-3.
All experiments were repeated at least three times. Results are representative of three independent experiments and are expressed as means ± standard deviation. Student's distribution probability density function was used for calculation of p values. p < 0.05 was considered statistically significant. The software used was GraphPad PRISM.
This work was supported by funds from the Institut National de la Santé et de la Recherche Médicale (INSERM), the University of Bordeaux and IFR 66, as well as by a grant from Ligue contre le Cancer comité de la Gironde and by a regional administration grant from Région Aquitain. A.T. is supported by a Ph.D. fellowship from the Conseil Régional Aquitaine/INSERM. We thank Gaëlle Labrunie for her technical support and Pippa McKelvie-Sebileau for help with the English editing.
No potential conflicts of interest were disclosed.