Background
Neuroblastoma (NB) is the most frequent solid extracranial tumour in children and is a major cause of death from neoplasia in infancy [
1]. These tumours are clinically and biologically heterogeneous, with cell populations differing in their genetic programs, maturation stage and malignant potential [
2]. Clinically, spontaneous regressions and tumour maturation are frequent in infants or in low stage tumours, whereas older children often present at diagnosis with high stage progressive and metastatic disease and their overall prognosis is poor [
2]. Little improvement in therapeutic options has been made in the last decade, requiring a urgent need for the development of new therapies.
Anti-cancer therapies mediate their cytotoxic effect by predominantly inducing apoptosis in tumour cells. Apoptosis may be induced by triggering the death receptors (extrinsic pathway) or the mitochondria (intrinsic pathway) leading to the activation of effector caspases [
3]. Tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) is a promising candidate for therapy of many forms of cancer as it selectively induces cell death in transformed cells, sparing normal tissues [
4]. TRAIL mediates apoptosis by activation of the death receptor pathway. Its interaction with TRAIL-R1 and -R2 receptors leads to recruitment of adaptor FADD and initiator caspase-8 to the DISC, resulting in caspase-8 activation and initiation of a cell death cascade by direct cleavage of effector caspases [
4,
5]. The process is positively regulated and amplified by caspase-3-mediated activation of caspase-8 [
6,
7], and/or by parallel activation of the mitochondrial pathway via caspase-8-dependent cleavage of Bid [
8], resulting in activation of the apoptosome through Bax and Bak oligomerisation and the release of cytochrome-c and Smac/DIABLO into the cytosol. Conversely, negative regulation is promoted by the caspase-8 antagonist c-FLIP [
9] or by anti-apoptotic Bcl-2 and Bcl-x
L-mediated blockade of mitochondria activation [
10]. In addition, other inhibitors of apoptosis proteins (IAPs), such as cIAP-1/-2 and XIAP [
11] interact with effector caspases, which are neutralized by Smac/DIABLO [
3]. Survivin, an other IAP shown to be over-expressed in most tumours, protects cancer cells from apoptosis by interacting with Smac/DIABLO.
Resistance to TRAIL-induced apoptosis in various tumours was described to be caused by the deregulation of diverse signalling molecules such as down-regulation of TRAIL-receptors, caspase-8, caspase-10 or Bax, or over-expression of c-FLIP, Bcl-2, Bcl-x
L or survivin [
12]. In N-type NB cells, resistance to TRAIL was attributed to the down-regulation of caspase-8 expression by hypermethylation or allelic deletion [
13‐
15], as well as to the down-regulation of cell surface TRAIL-R1/-R2 expression [
16]. Numerous TRAIL resistant tumour cell lines were reported to be sensitised to TRAIL by combined treatments with chemotherapeutic agents, cycloheximide (CHX), IFN-γ or irradiation by diverse cell-type specific mechanisms [
17,
18]. We have previously shown that NB cells could be sensitised to TRAIL by subtoxic doses of chemotherapeutic drugs or CHX by the activation of extrinsic and intrinsic apoptotic pathways and caspases-dependent cleavage of XIAP, Bcl-x
L and RIP [
19]. However as chemotherapeutic drugs are non-specifically and highly toxic toward non-tumoral cells, it may be beneficial to develop alternative and less toxic therapeutic strategies that synergise with TRAIL.
Histone deacetylase inhibitors (HDACIs) are a new class of promising anti-cancer agents which inhibit tumour growth both in vitro and in vivo with very low toxicity toward normal cells [
20]. Recently, several HDAC inhibitors have entered Phase I and Phase II clinical trials and demonstrate encouraging anti-tumour activity in a variety of cancer types [
21]. The anti-tumour effect of HDACIs was proposed to result from accumulation of acetylated histones leading to activation of genes involved in inhibition of tumour cell growth [
20]. Altered activities of histone deacetylases or histone acetyl transferases are indeed involved in different human cancer [
22,
23]. HDACIs mechanism of action appears to involve cell cycle arrest, induction of apoptosis and differentiation both in vitro and in vivo [
21,
22]. The mechanisms of induction of apoptosis by HDACIs are cell type specific and involve the activation of the intrinsic apoptotic pathways. HDACIs can also synergise with TRAIL to induce apoptosis by various mechanisms depending on tumour types, such as increased expression of TRAIL-receptors and TRAIL [
24‐
28], decrease in c-FLIP [
24,
26,
29], Bcl-x
L [
30], XIAP and Bcl-2 expression (or activation) [
31], or increased formation of the DISC [
26,
32].
This is the first report describing the detailed mechanisms of TRAIL sensitization by HDACIs in neuroblastoma cells. The analysis of the mechanisms by which the three HDACIs, sodium butyrate (NaB), Trichostatin A (TSA) and suberoylanilide hydroxamic acid (SAHA) enhance the action of TRAIL, demonstrates that HDACIs enhance caspases activation and restores a positive ratio between pro- and anti-apoptotic proteins in favour of apoptosis. In addition, we demonstrate that HDACIs also potentiate TRAIL action by increasing the amplitude and the kinetics of the apoptotic process. The association of TRAIL and HDACIs, two non-toxic anti-cancer agents, could be of therapeutic benefit for the treatment of children with NB.
Methods
Cell culture and reagents
The caspases-8 positive neuroblastoma cells SH-EP, SH-EP FADD-DN [
33], SH-EP-FLIP [
34] and SK-N-AS were grown in DMEM medium supplemented with 10 % of FCS, 200 μg/ml gentamicin (Essex Chemie AG). Cells were incubated with indicated amount of soluble recombinant TRAIL (a gift from J. Tschopp) and cross-linking mouse anti-Flag Ab M2 (Sigma) with a constant ratio of 1/5 of TRAIL to M2 respectively. Sodium butyrate (Fluka) was dissolved in H
2O and stored at -20°C. SAHA (Biovision) and TSA (Sigma) were dissolved in DMSO and store at -20°C. Cells were pretreated 30 min with caspases inhibitors zVAD-fmk (100 μM, Bachem), zVDVAD-fmk (50 μM, Calbiochem), zIETD-fmk, zDEVD-fmk, and zLEHD-fmk (50 μM, R&D systems) before TRAIL or HDACIs treatments.
Cell viability assays
Cells (1–2.5*105/well in 96-well-plates; 100 μl) were plated 24 h before treatment and incubated with TRAIL and/or HDACIs for 48 h. Assays were performed in quadruplicates. Viability was measured using the MTS/PMS cell proliferation kit from Promega according to manufacturer's instructions. Percentage of cell viability as compared to untreated controls was calculated.
Measurement of apoptosis by detection of sub-diploid population
Cells were harvested by trypsinization, washed twice with ice-cold PBS, resuspended in 1 ml of ice-cold PBS, and fixed with 3 ml of 100% ice-cold ethanol for 1 h at 4°C. For staining with propidium iodide (PI), cells were washed twice in ice-cold PBS and incubated for at least 30 min at room temperature in 0.2 ml of PBS containing 200 μg/ml RNase A and 10 μg/ml propidium iodide. The stained cells were analyzed using a FACScan flow cytometer (Becton Dickinson).
Immunoblotting
Whole cell extracts were prepared as already described [
14]. Protein extracts (30–50 μg) were loaded on 12% SDS-PAGE and transferred on nitrocellulose membranes. Blots were saturated with 5% skim milk, 0.1 % Tween 20 in TBS and revealed using mouse monoclonal antibodies to detect caspase-8 (MBL), caspase-2 (Apotech corporation), caspase-3, caspase-7, XIAP, RIP, cytochrome-c (BD Pharmingen), AIF (Santa Cruz), β-actin (Sigma). Polyclonal rabbit antibodies were used to detect caspase-9 (Cell Signaling), Bid, Bcl-x
L (BD Transduction Laboratories), Bim (Imgenex), survivin (R&D systems). Binding of the first antibody was revealed by incubation with either goat anti-mouse IgG (Jackson ImmunoResearch) or goat anti-rabbit IgG (Nordic Immunological Laboratories). Bound antibodies were detected using the Lumi-light western blotting substrate (Roche) according manufacturer's instructions.
Caspases activities
Caspases-8, -2, -3 and -9 protease activities were measured using the caspases-3, -8, -9 colorimetric protease assay kits from MBL and caspases-2 colorimetric substrate VDVAD-pNA from Alexis. Cytosolic lysates were prepared after TRAIL and/or HDACIs treatments according to manufacturer instructions. One hundred μg of protein extracts were incubated with 200 μM of IETD-pNA, VDVAD-pNA, DEVD-pNA, and LEHD-pNA colorimetric substrates for 3 h at 37°C. Cell lysates were incubated with 10 μM of caspase inhibitor (zIETD-fmk, zVDVAD-fmk, zDEVD-fmk, or zLEHD-fmk) for 30 min before addition of the respective caspase substrate for control experiments. Hydrolysed pNA was detected using a microtiter plate reader at 405 nm. Background absorbance from cell lysates and buffers were subtracted from the absorbance of stimulated and unstimulated samples before calculation of relative caspases activities.
Analysis of mitochondrial transmembrane potential
The drop of mitochondrial membrane potential ΔΨm was measured by staining the cells with the fluorescent dye JC-1 [
35] according to manufacturer's protocol (Calbiochem). Loss of ΔΨm resulting in reduction of red aggregates was measured by flow cytometry using the FL2 channel (550–650 nm) (FACScan, Becton Dickinson). Results are given in percentage of cells with low ΔΨm compared to untreated controls.
Cell surface immunostaining
Cells were washed in FACS buffer (RPMI, 10 % FCS, 2 mM EDTA) and stained with mouse monoclonal antibodies anti-TRAIL-R1, -TRAIl-R2 and -TRAIL (Alexis), followed by goat secondary antibody conjugated with FITC (Caltag laboratory) and analysed by FACScan (Beckton Dinkinson).
Transfection with siRNAs
Survivin siRNA targets nucleotides 235–253 of survivin mRNA 5'-AAGGAGCTGGAAGGCTGGGAGTT-3' and control siRNA is composed of the inverse sequence 5'-AAGAGGGTCGGAAGGTCGAGG-3', as described previously [
36]. siRNAs were a gift from the lab of Uwe Zangemeister-Wittke. 180'000 cells/well were plated in 12 wells (1 ml) and transfected 8 h later with 100 nM or 25 nM of siRNAs with 3 μl or 2 μl respectively of Lipofectamine2000 according to manufacturer's instructions (Invitrogen). Sixteen h after transfection, cells were plated in 96 wells (10'000 cells/well) and induced 24 h later with TRAIL and/or HDACIs for 48 h.
Discussion
The present study demonstrates that simultaneous administration of TRAIL and subtoxic doses of HDACIs strongly potentiates the triggering of apoptotic cascade in NB cells. The detailed analysis of the mechanisms of sensitisation reveals that the increase in cell death is mediated by the enhanced activation of the caspases cascade and the pro-apoptotic protein Bid, concomitant with the down-regulation of the anti-apoptotic proteins RIP, Bcl-xL, XIAP, and survivin in a caspases-dependent manner.
Stimulation of the death receptor pathway through enhancement of death receptor expression is one mechanism of sensitisation to TRAIL by HDACIs. Several recent reports have shown that the expression of TRAIL or TRAIL-receptors was induced by HDAC inhibitors in leukaemia cells, breast and colon cancer cells [
24‐
28], while other studies have reported no change in the expression level of DR4, DR5 and DcR2 in melanoma and lymphoma cells [
31,
37]. We show here that unlike chemotherapeutic drugs, Doxorubicin, Cisplatin, Etoposide or Taxol which up-regulate the cell surface expression of TRAIL-R2 in NB cells [
19], HDACIs did not influence any of TRAIL-R1, TRAIL-R2 and TRAIL cell surface expression in NB cells, indicating that sensitisation to TRAIL-induced apoptosis by HDACIs occurs at least partly by different mechanisms.
Several studies reported a change in the expression level of proteins involved in the extrinsic apoptotic pathway such as FADD [
24] and FLIP [
24,
26,
29]. No change in the steady state level of caspases-8, -2, -3, -7 and -9, Bid, FADD or FLIP was detected after stimulation with subtoxic doses of NaB, SAHA or TSA in NB cells. Nevertheless, as previously reported in leukaemia cells [
31], the death receptor pathway is involved in the synergistic induction of apoptosis by simultaneous treatment with TRAIL and HDACIs in NB cells. Indeed, following over-expression of a FADD-DN protein or c-FLIP
L, NB cells became fully resistant to TRAIL, which could not be rescued by co-treatments with HDACIs.
In accordance with previous reports [
25,
26,
32,
37,
40] we demonstrate that HDACIs sensitised NB cells to TRAIL by enhancing the cleavage-mediated activation of the caspases cascade and Bid, while no cleavage was observed with subtoxic doses of HDACIs alone. The enhanced apoptosis induction was caspases-dependent as caspases inhibitors completely abolished NB cell death. Interestingly, TRAIL and HDACIs co-treatments also increased the amplitude of caspases activation. Indeed, a higher level of caspases activities was reached with combined treatment compared to TRAIL alone. This suggests that HDACIs sensitise NB cells to TRAIL-induced apoptosis by enhancing the extent of caspases activation and thereby increasing the magnitude of the apoptotic process.
In addition, the simultaneous addition of TRAIL and HDACIs synergistically affected the mitochondrial pathway as evidenced by the enhanced drop of ΔΨm, the increased caspases-9 activation and cytochrome c and AIF release into the cytosol. This suggests that HDACIs may also act at the level of the mitochondria to sensitise NB cells to TRAIL, as previously reported with other tumour cell lines [
24,
31,
40].
The modulation of the intracellular ratio between pro- and anti-apoptotic proteins could be an other mechanism of HDACIs potentiation to TRAIL-induced apoptosis, which could occur by the increased activity of pro-apoptotic proteins and/or by the down-regulation of anti-apoptotic proteins. The BH3-only protein Bim
EL was reported to be increased by subtoxic doses of TSA and depsipeptide in CCL and Jurkat cells [
32]. Here we show that in NB cells Bim
EL protein level was reduced by treatment with TRAIL and further decreased by co-administration of HDACIs. This was due to caspases-dependent cleavage as the down-regulation of Bim
EL was protected by zVAD. Interestingly, the activation of Bim
EL by caspases-3-dependent cleavage was described to induce a positive feedback amplification of the apoptotic signal by enhancing the affinity of Bim
EL to Bcl-2 [
41]. Therefore, in NB cells Bim
EL may be activated by caspases-mediated cleavage following combined treatment and such Bim
EL activation may participate to TRAIL potentiation by HDACIs through the amplification of the apoptotic signal. In addition, we observed the reduction of the anti-apoptotic protein Bcl-x
L following combined treatments. Hence, the increase of the Bim
EL /Bcl-x
L ratio could enhance mitochondrial permeability leading to the release of pro-apoptotic factors.
The caspases-dependent down-regulation of anti-apoptotic proteins such as XIAP and Bcl-2 [
31] and the role of the down-regulation of Bcl-x
L [
30] by co-treatment with TRAIL and HDACIs were previously described. In an other report it was shown that the reduction of c-FLIP, Bcl-x
L, Bcl-2, and XIAP expression induced by subtoxic doses of the HDACI LAQ824 was independent on proteasome or caspases activity [
26]. In contrast, we demonstrate here that the steady state level of RIP, Bcl-x
L, XIAP and survivin in NB cells was reduced by caspases-dependent cleavages mediated by co-treatments with TRAIL and HDACIs. Interestingly, both the level and the timing of down-regulation of anti-apoptotic proteins were increased by combined treatments compared to TRAIL alone. The concomitant caspases-dependent down-regulation of the anti-apoptotic proteins RIP, Bcl-x
L, XIAP, and survivin, and the activation of the pro-apoptotic proteins Bid and probably Bim
EL increased the ratio between pro- and anti-apoptotic proteins and therefore lowered the threshold of apoptotic signal and contributed to the sensitising effect of HDACIs to TRAIL-induced apoptosis.
We have previously shown that RIP, Bcl-x
L, and XIAP were down-regulated by caspases-dependent cleavages upon co-treatment with TRAIL and chemotherapeutic drugs, while the steady state level of survivin was not affected, in contrast to co-treatment with TRAIL and HDACIs [
19]. This indicates that HDACIs and chemotherapeutic drugs contribute through different ways to TRAIL-induced apoptosis. The reduction of survivin expression mediated by siRNAs results in an increased the sensitivity threshold of NB cells to HDACIs and/or TRAIL. This suggests that the down-regulation of survivin induced by higher doses of HDACIs and TRAIL plays a role in the sensitising effect of HDACIs to TRAIL.
Conclusion
The potentiation by HDACIs to TRAIL-induced apoptosis may be caused by stimulation of the apoptotic cascade through increased activation of both the extrinsic and the intrinsic apoptotic pathways induced by TRAIL and HDACIs, respectively. The apoptotic signal is further enhanced by the caspases-dependent activation of pro-apoptotic proteins such as Bid and Bimel and inactivation of anti-apoptotic proteins such as XIAP, Bcl-xL, RIP, and survivin. It results a change in the equilibrium of pro- to anti-apoptotic molecules that lowers the cell death threshold and strongly favours apoptosis.
These findings may have important implications for the use of TRAIL in cancer therapeutic using recombinant soluble TRAIL. As HDACIs strongly enhance the apoptotic action of TRAIL even at low concentrations, HDACIs may be used in combination with TRAIL to reduce the doses of TRAIL required for inhibition of tumour growth. Recently, several HDAC inhibitors have entered Phase I and Phase II clinical trials and are demonstrating encouraging anti-tumour activity in a variety of cancer types (21). The combination of TRAIL and HDACIs may therefore be an interesting and soft inoffensive new anti-tumour strategy particularly relevant in the treatment of children with highly malignant neuroblastoma.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
AMM performed all major experimental work including FACS analyses, participated in the design and in the coordination of the study and drafted the manuscript. KBB participated in all cell culture experiments and performed the immunoblots, caspases activity assays. KA participated in the immunoblotting experiment and in cells stimulation with drugs. MF helped for the siRNA transfections. RM participated in cell treatments with HDACIs. JMJ and NG were involved in the overall design of the study and helped to draft the manuscript.