Loss of the Malignant Phenotype of Human Neuroblastoma Cells by a Catalytically Inactive Dominant-Negative hTERT Mutant

Human neuroblastic tumors are embryonal tumors, which derive from cells that migrate from the neural crest tissue and give rise to the sympathetic nervous system (1). These tumors are heterogeneous and display a wide spectrum of differentiation stages from benign ganglioneuroma and well-differentiated tumors to undifferentiated malignant neuroblastoma. Neuroblastoma, the most frequent solid tumor of early childhood, is diagnosed as a disseminated disease (stage 4) in about 60% of the cases. Unlike other malignant diseases, stage 4 in neuroblastoma elicits distinct clinical patterns based on disease distribution and patient's age. A particular subset of metastatic neuroblastoma, called stage 4S, is defined by a localized dissemination of the primary tumor to liver, skin or bone marrow, in infants under 1 year. These 4S neuroblastomas are characterized by their ability to regress spontaneously or differentiate in benign ganglioneuroma, and show an excellent outcome. In contrast, stages 4 in children older than 1 year, with dissemination of tumors to lymph node, bone, bone marrow, liver or other organs, have a dismal outcome despite the use of high dose chemotherapy, showing a need of new therapeutic approaches in these patients (1).

The MYCN oncogene plays a pivotal role in the progression of neuroblastoma to advanced malignancy, and its amplification is correlated with poor clinical outcome (2). At a cytogenetic level, MYCN amplicons are present in nuclei as either extrachromosomal double minutes or intrachromosomal structures known as homogeneous staining regions (HSR). Expulsion of amplified MYCN from either double minutes or HSR has been observed in both neuroblastoma cell lines and tumor samples (3, 4). Caspase-8 plays a crucial role as a mediator of death receptor-triggered apoptosis. It also contributes to various nonapoptotic cellular functions. Indeed, capase-8 deficiency can participate to tumor development (5). It is frequently inactivated, especially in MYCN-amplified neuroblastoma (6). This inactivation has been responsible for the resistance of neuroblastoma cells to death receptor-induced apoptosis (7, 8). Unlike other human cancers, the p53 tumor-suppressor protein encoded by TP53 gene is rarely mutated in neuroblastoma primary tumors (9). However, when mutations occurred, they were found in neuroblastoma with drug resistance acquired over the course of chemotherapy.

In vitro, human malignant neuroblasts can be committed to distinct neural crest lineages: neuroblastic (N-type), substrate-adherent non-neuronal (S-type), or intermediate (I-type) cells that can be distinguished by morphologic and biochemical traits and can undergo transdifferentiation either spontaneously or after chemical induction (10, 11). This conversion process itself is not well understood but appears to be clinically relevant and might be involved in the ability of stage 4S to regress. Therefore the possibility to drive malignant neuroblastoma cells towards a nontumorigenic phenotype remains a real challenge for the development of new therapeutic strategies.

Telomerase is a ribonucleoprotein composed at least of an RNA template (human telomerase RNA, hTR) and a catalytic protein subunit (human telomerase reverse transcriptase, hTERT), responsible for de novo synthesis and elongation of telomeric repeats at chromosomal ends (12). By maintaining telomere length, telomerase controls cell survival. Besides its canonical role, telomerase elicits other functions in several essential cell signaling pathways, including apoptosis (13–15), differentiation (16), DNA damage responses (17), and regulation of gene expression (16, 18). Telomerase activity is absent or poorly detected in most human somatic cells, whereas it is high in germ and tumor cells, including neuroblastoma. Because of its high activity in cancer cells, telomerase is an attractive anticancer therapeutic target (19–22). That aggressive stage 4 neuroblastoma expressed high levels of telomerase activity, whereas favorable tumors had no or little telomerase expression and activity (23, 24), prompted us to investigate the role of this enzyme in the biology of malignant neuroblastoma cells and their response to chemotherapy.

In this study, we characterized the phenotypic and molecular changes linked to telomerase activity modulation in the stage 4 MYCN-amplified neuroblastoma IGR-N-91 cell line (25), used as a canonical model. Established in vitro, IGR-N-91 malignant neuroblasts displayed stable HSR of MYCN, loss of chromosome 1p (26), a mutated TP53 by insertion of exons 7, 8, 9 in tandem (27) and loss of caspase-8 expression (28). In nude mice, these cells are highly tumorigenic and able to metastasize (25). We found that telomerase inhibition due to the stable expression of a dominant-negative variant of hTERT induced, in the original IGR-N-91 cell population, a switch from N to S phenotype with increased ability to undergo drug-induced apoptosis and loss of tumorigenic properties in vivo.

The human MYCN-amplified IGR-N-91 cell line was established from an infiltrated bone marrow collected from a stage 4-neuroblastoma belonging to an 8-year-old boy after unsuccessful adriamycin-vincristine chemotherapy (25). The cells were authenticated by checking for MYCN amplification by real-time quantitative PCR and for the presence of the mutant TP53 carrying exons 7 to 9 duplication (27). Cells were cultured in DMEM/F12 supplemented with 10% FBS, penicillin (50 IU/mL), streptomycin (50 μg/mL), and l-glutamine (2 mmol/L, PAA Laboratories), and incubated at 37°C at 5% CO₂ atmosphere.

Cisplatin was purchased from Sigma, staurosporine from Roche Diagnostics and TRAIL from Enzo Life Sciences.

Cells were transduced with retroviral vectors, which drive expression of both transgenes and GFP from a single bicistronic message as previously described (29, 30).

Total cellular RNA was collected from samples using TRIzol reagent (Invitrogen). RT reaction was carried out using Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnostics) according to the manufacturer's instructions with oligo(dT)₈ primers. The cDNA samples were amplified in a 25 μL reaction mixture containing 50 μmol/L each of the 4 deoxynucleotide triphosphates, 0.5 U Taq DNA polymerase (Promega), and 0.4 μmol/L specific primers. Primer sequences are shown in Supplementary Table S1. The cDNA samples were amplified for 30 cycles (94°C for 30 s, 56°C for 30 s and 72°C for 45 s). For the internal control GAPDH, the annealing temperature was 62°C. The PCR products were fractionated on a 2% agarose gel and visualized under UV light.

The expression of ectopic WT-hTERT and DN-hTERT transcripts was quantified by real-time reverse transcriptase (RT)-PCR using the LightCycler technology and the Light Cycler FastStart DNA MasterPLUS SYBR Green Kit (Roche Diagnostics) according to the manufacturer's instructions. Primer sequences are shown in Supplementary Table S1. The primers allowed the detection of both endogenous hTERT and exogenous WT- or DN-hTERT transcripts. WT-hTERT and DN-hTERT levels were normalized to the expression of GAPDH and to the expression of endogenous hTERT in empty vector control cells.

Genomic DNA (20 ng) was combined with (i) the Taqman MYCN Copy Number Assay (4400292, Applied Biosystems), (ii) the Taqman RNase P Copy Number Reference Assay (4403326, Applied Biosystems), and (iii) the FastStart Universal Probe Master (ROX, Roche Diagnostics). Reactions were run on a 96-well Real-Time StepOnePlus PCR System (Applied Biosystems) and amplified at 50°C for 2 minutes, 95°C for 10 minutes and 40 cycles of 95°C for 15 s and 60°C for 60 s. The number of copies of the target sequence in each sample was determined by relative quantification using the comparative C_T (ΔΔC_T) method.

Telomerase activity was measured using the telomerase PCR-ELISA kit (Roche Diagnostics) according to the manufacturer's instructions and expressed as a percentage of activity detected in empty vector transduced cells. Telomere lengths were determined as described previously (31), using a chemoluminescent assay (Roche Diagnostics).

For cell viability measurement, IGR-N-91 cells were seeded and cultivated for 24 hours, before drug addition. After incubation with the drugs, WST-1 labeling mixture (Roche Diagnostics) was added and the cells incubated for 3 hours. The absorbance of the samples was measured at 550 nm using a microtiter plate reader (Dynex). For bidimensional clonogenic assays, cells were plated in tissue culture plates. Colonies were then fixed and stained using May-Grünwald-Giemsa staining after 2 weeks of incubation at 37°C at 5% CO₂ atmosphere.

Apoptosis was assessed by the TUNEL technology using In Situ Cell Death Detection Kit, TMR red (Roche Diagnostics), according to the manufacturer's instructions. After labelling, cells were mounted in Vectashield mounting medium containing 4-,6-diaminidine-2-phenylindole (DAPI Vector Laboratories) to counterstain nuclei. TMR red labeled apoptotic cells were detected by fluorescent microscopy (Leica DMRD, equipped with ×63 objective).

Cells, fixed and permeabilized as previously described (32), were blocked with 2% bovine serum albumin (BSA) in PBS overnight at 4°C. Then, they were incubated with anti-hTERT primary antibody (Epitomics) overnight, washed twice in PBS and incubated with Alexa Fluor 594-conjugated anti-rabbit secondary antibody (Molecular Probes, Invitrogen) for 1 hour at room temperature. Subsequently, cells were washed 3 times in PBS and mounted in Vectashield mounting medium with DAPI to counter stain nuclei. Cells were observed by conventional fluorescence microscopy.

Total cellular proteins were extracted with SDS lysis buffer (SDS 8%, Tris-HCl 250 mmol/L pH 6.8, 2.5 mmol/L NaF, 0.2 mmol/L sodium orthovanadate, protease inhibitor cocktail from Sigma), separated by SDS-PAGE (10% to 15%) and transferred onto a PVDF membrane. The membranes were saturated with 5% non–fat dry milk powder and 0.5% Tween-20 in TBS 0.5% for 1 hour and probed overnight with the primary antibody. Mouse monoclonal, N-Myc (NCM II 100, Santa Cruz), caspase-8 (551242, BD Pharmingen), CD44 (MEM-263, Abcam), caspase 3 (AM-20, Calbiochem), p53 (554293, BD Pharmingen), NSE (Ab-1, Labvision), and rabbit monoclonal hTERT (Epitomics) were used as primary antibodies. After washing, blots were then incubated with the appropriate horseradish peroxidase-conjugated secondary antibody for 1 hour and then developed by enhanced chemiluminescence (Perkin Elmer Life Sciences). Rabbit polyclonal HRP-conjugated anti-GAPDH (Ab9385, Abcam) was used as a control for equal loading.

DNAs were purified as previously described (31), digested with restriction enzyme (BstNI, HaeII), submitted to electrophoresis through a 1.1% agarose gel and transferred to a Hybond-N nylon membrane (GE Healthcare Europe GmbHBranch). The membrane was dried and baked at 80°C for 30 minutes and the DNA cross-linked onto the membrane by UV radiation. Prehybridization was carried out for 3 hours at 60°C in 7% SDS, 1 mmol/L EDTA (pH 8), 263 mmol/L Na₂HPO₄ and 1% BSA. Hybridization with the ³²P-labeled single-strand M13 DNA probe was carried out at 60°C overnight. The membrane was then washed in trisodium citrate, 30 mmol/L, NaCl, 300 mmol/L, pH 7.0, 0.1% SDS twice at room temperature and once at 60°C, and exposed to X-ray film.

All animal experiments were carried in compliance with European directive 86 of 609, French laws and regulations and were approved by the local Ethics Committee (CEEA IRCIV/IGR n°26, registered with the French Ministry of Research). The IGR-N-91 tumor cells (5 × 10⁶) derived from empty vector, WT-hTERT or DN-hTERT at days 135 and 180 postinfection, in a volume of 80 μL of growth medium were mixed with the same volume of BD Matrigel Matrix (BD Biosciences) and subcutaneously injected into each flank of 6-week-old female nude mice. Experiments were conducted in 3 groups of 10 mice. Each tumor, considered as ellipsoid, was measured in 2 dimensions with a caliper for volume assessment according to the (length × width²)/2 formula. In accordance with animal legal requirements, mice were euthanized once tumors reached 20 mm in diameter.

To study the effects of long-term expression of the catalytically inactive dominant-negative (DN)-hTERT, we introduced by retroviral infection the wild-type (WT)-hTERT, the DN-hTERT or the control empty (EV) vector expressing GFP alone into the human IGR-N-91 neuroblastoma cells. The expression of the transgenes was assessed by real-time RT-PCR, telomerase activity assay, Western blot, and immunofluorescence analyses and compared with the GFP-control cells (Fig. 1A–D). Ectopic DN-hTERT and WT-hTERT transcripts were highly expressed in the transduced cells (Fig. 1A). Western blot (Fig. 1B) and immunofluorescence (Fig. 1D) analyses revealed a high expression of these transgenes and showed that both DN-hTERT and WT-hTERT are mainly nuclear. To investigate whether DN-hTERT exerted a dominant-negative effect on endogenous telomerase enzyme, TRAP assay was carried out to assess telomerase activity of cell extracts. DN-hTERT cells showed a dramatically reduced endogenous telomerase activity (more than 80%), whereas WT-hTERT cells exhibited almost a 4-fold increase level of telomerase activity compared with the control vector transduced cells (Fig. 1D). Although in DN-hTERT cells, telomerase activity was successfully downregulated in early cell passages, it gradually recovered in late passages up to 60% of the control cells by day 160 (Fig. 1E, left). This recovery resulted from the progressive decline in DN-hTERT mRNA levels (Fig. 1E, right). Despite the partial loss of DN-hTERT expression, no change in GFP expression was observed during the long-term culture. Genomic PCR showed the specific and progressive loss of the DN-hTERT sequence while the GFP sequence of the transgene was still present (Supplementary Fig. S1). Of note DN-hTERT sequence was still observed 118 days postinfection in accordance with the detection of DN-hTERT transcripts 115 days postinfection. In contrast, no loss of the transgene was observed in WT-hTERT cells during the period of cellular expansion (Supplementary Fig. S2).

We next investigated whether telomerase activity inhibition influenced telomere length in the IGR-N-91 cells. Despite the telomerase activity inhibition, telomere length was unchanged (Fig. 1F). As expected, WT-hTERT cells exhibited longer telomeres as compared with control cells. As decreased telomerase activity is generally associated with cell growth arrest and loss of viability, the effect of the DN-hTERT transgene on cell proliferation rate was evaluated by regularly counting cell during more than 150 days. The growth kinetics of the DN-hTERT cells did not differ from the WT-hTERT cells or the control vector cells even in the early passages when telomerase activity inhibition was high (Fig. 1G). This lack of effects on growth proliferation and viability was confirmed by bidimensional clonogenic assays (Fig. 1H). Altogether, these data indicate that telomerase inhibition via the expression of DN-hTERT leads neither to telomere shortening nor to loss of in vitro proliferative capacity.

Importantly, DN-hTERT expression in the IGR-N-91 N-type cells induced major changes in cell morphology (Fig. 2A), by developing substrate-adherent or S phenotype with enlarged and flattened nuclei and cytoplasm resembling epithelial morphology. This morphology differed from the neurone-like phenotype of ATRA-treated cells, which presented a small, rounded, loosely adherent cell body, numerous neurite-like processes, and after longer periods of exposure cellular clustering. Note that neither the morphology of WT-hTERT cells nor empty vector expressing cells differed from the N-type parental cells. DNA fingerprint analysis carried out all along the experiment (Fig. 2B) pointed out a common origin for the 3 transduced sublines and excluded a cross-contaminated subline putatively selected throughout in vitro passages of the cell culture. Of note plating the parental IGR-N-91 cells at a clonal density always give rise only to N-type cells and never generate S-type cell colonies. Interestingly, S-type cell features were observed as soon as 39 days after DN-hTERT transduction and remained stable during all the 160 days of cellular expansion culture.

N-type cells express high levels of neuron specific enolase (NSE), an early neuronal progenitor marker but not CD44, in contrast to S-type cells expressing high levels of CD44 but not NSE. Analysis of these 2 markers showed that DN-hTERT cells expressed a high level of CD44 protein and exhibited an important reduction of NSE protein, whereas WT-hTERT cells displayed an increase in NSE and a decrease in CD44 proteins compared with the parental and empty vector transduced cells (Fig. 2C). These results indicated that the level of telomerase activity can actively modulate neuroblastoma cell phenotype.

IGR-N-91 cells showed MYCN-amplification with approximately 60 copies/haploid genome and subsequently elicited a high N-MYC protein level (Fig. 2C, left). Transduction of WT-hTERT did not modify N-MYC protein level (Fig. 2C, right). In contrast, DN-hTERT cells showed a progressive loss of N-MYC protein, which became undetectable by 170 days after transduction. Table 1 shows a progressive and dramatic loss of MYCN copy number in DN-hTERT cells from 70 to 0.5. In contrast, an increase of MYCN copy number was observed in WT-hTERT cells from 70 to 82.

Previous studies have established increased sensitivity to chemotherapeutic drugs after telomerase inhibition in a variety of tumor cell lines, whereas ectopic hTERT expression resulting in an increased telomerase activity is able to protect cells from death through its anti-apoptotic function (14, 33–36). Therefore, we examined cell survival rate in transduced IGR-N-91 cells when treated with drugs such as cisplatin, a DNA damaging agent, staurosporine, a protein kinase inhibitor, or TRAIL, a death receptor ligand. The WST-1 proliferation assay, carried out at both 24 and 48 hours of treatment, showed that DN-hTERT cells exhibited a greater sensitivity to all tested drugs (Fig. 3A). Despite a partial late telomerase activity recovery, the inhibition of proliferation was most striking at day 169 after transduction than at day 35 (Fig. 1E). As expected, a protective effect of WT-hTERT was noticeable after long-term transgene expression and this effect was more pronounced in TRAIL-treated cells.

To determine if the decrease in cell viability caused by DN-hTERT was due to an increase in apoptosis, floating and adherent cells were collected for apoptosis analysis by TUNEL assay (Fig. 3B). Neither GFP control nor WT-hTERT cells revealed significant apoptotic nuclei. In contrast, drug treatments produced significantly high levels of apoptosis in the DN-hTERT cells and sensitization to apoptosis was higher at day 135 after transduction, thus confirming and extending the results obtained with the WST-1 proliferation assays.

To investigate the mechanisms by which hTERT-DN expression sensitizes cells to drug-induced apoptosis, we analyzed the expression pattern of p53 and caspase-8, 2 key functional proteins involved in apoptosis, DNA damage pathways and/or differentiation.

IGR-N-91 cells expressed both the wild-type TP53 and the mutant form carrying exons 7 to 9 duplication (27). Transcript analysis by RT-PCR confirmed the insertion of exons 7, 8, 9 in tandem; amplicons spanning from exon 1 to 11 showed a 1,432 bp and a 1,750 bp band, corresponding to the wild-type and mutant isoform, respectively as confirmed by sequence analysis (Fig. 4A). The mutant p53 amplicon was predominantly detected in the control, WT-hTERT, and DN-hTERT cells up to 21 days posttransduction (Fig. 4A). Interestingly, by day 80 posttransduction, the amount of the p53 mutated form markedly diminished in DN-hTERT cells, whereas the amplicon with a size corresponding to the wild-type transcript increased (Fig. 4A). Finally, the mutant amplicon disappeared over time. When transduction is prolonged to 115 and 160 days, only the wild-type form was detected.

Western blot analysis confirmed that the transduction of DN-hTERT selected for a loss of mutated p53 at the benefit of a wild-type p53 content (Fig. 4B). PCR experiments conducted on DNA samples from DN-hTERT cells at different postinfection times confirmed that this loss of exon 7 to 9 duplication occurred at the genomic level. The amplification of the 1,350 bp DNA fragment spanning the tandem duplication was progressively lost (Supplementary Fig. S3).

Caspase-8, a key mediator of death receptor-induced apoptosis, is frequently inactivated by epigenetic silencing in neuroblastoma cell lines and MYCN amplified tumors (6). This silencing has been shown to be involved in the resistance of neuroblastoma cells to TRAIL-mediated apoptosis (7, 8) and to enhance the metastasis process. Figure 4 shows that in WT-hTERT cells a significant decrease of caspase-8 mRNA was detected compared with empty vector-transduced cells. This decline in expression was marked at the protein level (Fig. 4B). In contrast, in DN-hTERT cells, an increase in caspase-8 mRNA levels is detected as soon as 80 days after transduction (Fig. 4A) and associated with a large increase in protein levels (Fig. 4B). Note that in all transduced cells, the levels of caspase-3 remained similar over time. Importantly, the specific increase of caspase-8 expression by DN-hTERT has been also shown in another stage-4 neuroblastoma cell line (SK-N-BE2) suggesting that this modulation of caspase-8 expression by hTERT is not cell line specific (Supplementary Fig. S4). These results indicate a direct link between hTERT and caspase-8 signaling.

Complete loss of MYCN as well as the acquisition of a differentiated S-type phenotype led us to investigate tumorigenic properties of DN-hTERT cells. Experimental tumors generated in female nude mice after 135 days of culture posttransduction showed a significant increase in tumor growth in animal xenografted with WT-hTERT cells compared with empty vector transduced cells. Animals xenografted with DN-hTERT cells showed palpable nodules up to day 15, and the nodules started slowly growing (Fig. 5). This results in a tumor growth rate dramatically decreased compared with animals inoculated with the empty vector transduced cells. After 180 days postinfection, DN-hTERT cells lose their ability to generate tumor, as no tumor growth could be observed within the 28 days of the experiment. These results showed that the tumorigenic properties of DN-hTERT neuroblasts were lost in vivo.

That undifferentiated aggressive stage 4-neuroblastoma expressed high levels of telomerase activity, whereas favorable differentiated or regressing tumors had no or little telomerase expression and activity, prompted us to investigate the role of this enzyme in the biology of malignant neuroblastoma cells and their response to chemotherapy.

DN-hTERT expression induced major morphologic changes that switched the transduced cell population from a neuronal (N) to a substrate-adherent (S) phenotype, whereas WT-hTERT cells conserved the N-type morphology of the parental cells. These morphologic changes were associated with a modified expression of specific lineage markers (CD44, NSE) and an increase of caspase-8 protein level in DN-hTERT cells, compared with control cells, whereas WT-hTERT expression had the reverse effect. The resulting S-type cell population showed major genomic changes with a dramatic loss of MYCN copy number leading to the disappearance of N-MYC protein and a progressive restoration of wild-type p53. Therefore, both morphologic and molecular evidence show that telomerase impacts malignant neuroblastoma cell fate.

That DN-hTERT and WT-hTERT transductions evoked opposite effects on CD44, NSE, caspase-8 expression, and on MYCN copy number favors a model in which DN-hTERT drives N- to S-type cell population whereas modulating the expression of key genes involved in neuroblastoma biology. This hypothesis is in agreement with previous reports providing functional evidence that hTERT indirectly or directly modulates gene expression (18, 37). To support this cellular conversion, the growth curves showed no modifications in the proliferation rate and all attempts to observe the emergence of S-type cell colonies after plating of the parental IGR-N-91 cell population at a clonal density, failed. Nevertheless, if the DN-hTERT has induced the selection of a subset of S-type cells resident in the N-type IGR-N-91 cell population, that will suggest that DN-hTERT would have selectively induced cell death in N-type cells and/or provided S-type cells a selective growth advantage, whereas WT-hTERT would have pulled the population to a more N-type differentiated phenotype. This supports the idea that DN-hTERT and WT-hTERT exerts cell-specific dependent functions. Of note even though the S-type phenotype was clearly observed by day 39 after transduction, it is possible that the cell population continued to evolve during long-term culture as shown by MYCN copy number changes, the p53 genomic structure and the elimination of DN-hTERT sequence of the transgene without alteration of the GFP sequence as already reported (38, 39). Elimination of MYCN copies was previously reported in neuroblastoma, promoting a tumoral reversion phenotype (3, 4). The restoration of the wild-type sequence of TP53 could result from the deletion of 1 copy of the duplication in DN-hTERT cells as tandem duplications of exons are known to be relatively instable (40). The reversal of a p53 gene mutation restoring a wild-type p53 function has already been described in a TPA-resistant K562 cell subline established from a K562 cell line that expressed a truncated p53 due to the loss of 1 allele and an insertion mutation in exon 5 of the other allele (41). Of major interest is that the molecular alterations observed in DN-hTERT cells are maintained despite the restoration of telomerase activity after long-term culture and could account for the progressive changes in sensitivity to drug-induced apoptosis, as well as the development of an attenuated malignant phenotype. Therefore, these results dramatically differ from those already reported (38, 39) showing that increased sensitivity to drug in DN-hTERT expressing cells requires telomere shortening and that the recovery of telomerase activity is accompanied by the reversion of telomerase inhibition effects (i.e., re-induction of cell proliferation, telomere lengthening and resistance to apoptosis). Therefore, our results highlighted the functional relevance of noncanonical functions of hTERT in neuroblastoma, independent of its role on telomere maintenance. Keeping with this notion, the DN-hTERT IGR-N-91 cell line provide an efficient tool to gain further insights into several important questions on both the noncanonical functions of telomerase and neuroblastoma biology.

Because S phenotype in neuroblastoma cell lines is considered as a phenotype of low malignancy, this in vitro switch from N-type to an S-type could represent a process of cell reversion leading to in vivo tumor regression. In neuroblastoma tumors, MYCN amplification is a marker of dismal outcome and represents the cornerstone of risk-based classification of this disease. The absence of MYCN amplification and the high expression of CD44 are potent markers of a better prognosis (1). Moreover, the loss of caspase-8 expression in neuroblastoma contributes to an aggressive disease resistant to anticancer therapy and a metastatic phenotype (42, 43) while spontaneous regressing tumors (including stage 4S neuroblastoma) frequently express caspase-8. Therefore, we showed here, for the first time, a direct link between hTERT, CD44 and caspase-8, and the copy content of MYCN that can influence the selection process of tumors with a more or less aggressive behavior. This finding is strongly supported by in vivo xenograft experiments in nude mice showing that DN-hTERT cells lose their tumorigenic properties whereas WT-hTERT cells accelerate tumor growth as compared with mock cells.

In conclusion, our major finding is the discovery of a new link between telomerase biology and malignant neuroblast cell fate. Indeed, we showed that through the manipulation of telomerase activity, it can be possible to regulate key genes, leading to the loss of the malignant behavior of neuroblasts, whereas sensitizing them to anticancer drugs results in more effective treatments. This finding has therefore important implications in the development of novel strategies for neuroblastoma therapeutic management.

No potential conflicts of interest were disclosed.

Conception and design: M. Samy, E. Ségal-Bendirdjian

Development of methodology: M. Samy, S. Douc-Rasy, J. Bénard, E. Ségal-Bendirdjian

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M. Samy, C.-H. Gattolliat, S. Douc-Rasy, J. Bénard, E. Ségal-Bendirdjian

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Samy, C.-H. Gattolliat, S. Bombard, S. Douc-Rasy, J. Bénard, E. Ségal-Bendirdjian

Writing, review, and/or revision of the manuscript: M. Samy, S. Bombard, S. Douc-Rasy, J. Bénard, E. Ségal-Bendirdjian

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): M. Samy, F. Pendino, J. Hillion, E. Nguyen, E. Ségal-Bendirdjian

Study supervision: E. Ségal-Bendirdjian

We are grateful to Dr. R. Weinberg (Massachusetts Institute of Technology) for provision of the pBABE-puro-WT-hTERT and pBABE-puro-DN-hTERT constructs). We thank Dr. G. Kellermann (INSERM UMR-S 1007), Dr. P. Forgez (INSERM UMR-S 938), and Dr. N. Insdorf for their critical reading of the manuscript. We thank Pr J. R. Lillehaug (University of Bergen, Norway) for providing access to the cell transduction facilities.

This work was supported by grants from the Institut National de la Santé et de la Recherche Médicale (INSERM), the Association pour la Recherche contre le Cancer (E. Ségal-Bendirdjian), the Fondation de France (E. Ségal-Bendirdjian), the Ligue Nationale Contre le Cancer (Comité Ile de France, M. Samy; Comité Montbéliard, J. Bénard) and the Fédération Enfants et Santé & Société Française de Lutte contre les Cancers et les Leucémies de l'Enfant et de l'Adolescent, France (J. Bénard).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.