GADD153 and 12-lipoxygenase mediate fenretinide-induced apoptosis of neuroblastoma.

The synthetic retinoid fenretinide induces apoptosis of neuroblastoma cells and in vitro acts synergistically with chemotherapeutic drugs used to treat neuroblastoma. The mechanisms of fenretinide-induced cell death of neuroblastoma cells are complex, involving cellular signaling pathways as yet incompletely defined but, in part, involving the generation of reactive oxygen species (ROS). In an attempt to characterize the mechanism of action of fenretinide, cDNA array filters were screened to identify apoptotic genes regulated in response to treatment of SH-SY5Y cells with fenretinide. Expression of the stress-induced transcription factor, GADD153, was up-regulated at both the protein and mRNA levels in response to fenretinide. Overexpression of GADD153 increased apoptosis in the presence and absence of fenretinide, whereas reduced expression of GADD153 by expression of antisense DNA abrogated the response to fenretinide. Although fenretinide is a partial retinoic acid receptor (RAR)-beta/gamma agonist, RARbeta/gamma antagonists did not block the induction of GADD153 by fenretinide; conversely, the induction of GADD153 was blocked by antioxidants. Enzyme inhibitors were used to identify pathways mediating the ROS-dependent effects of fenretinide: inhibitors of phospholipase A(2) and lypoxygenases (LOX), and specific inhibitors of 12-LOX, but not 5-LOX or 15-LOX, inhibited the induction of ROS, apoptosis, and GADD153 in response to fenretinide. The inhibition of ROS and apoptosis was reversed by the addition of the 12-LOX products, 12 (S)-hydroperoxyeicosatetraenoic acid (12-HpETE) and 12 (S)-hydroxyeicosatetraenoic acid (12-HETE). Fenretinide did not increase free arachidonic acid levels, but increased LOX activity without a detectable increase in 12-LOX protein. These results suggest that fenretinide induces apoptosis via RAR-dependent and -independent pathways in which the RAR-independent pathway is characterized by a fenretinide-dependent increase in 12-LOX activity, leading to the induction of GADD153. The targeting of 12-LOX and/or GADD153 in neuroblastoma cells may thus present a novel pathway for the development of drugs inducing apoptosis of neuroblastoma with improved tumor specificity.

Induction of GADD153 and Bak: novel molecular targets of fenretinide-induced apoptosis of neuroblastoma.

Unlike 13-cis retinoic acid, the synthetic retinoid fenretinide induces apoptosis of neuroblastoma cells and in vitro acts synergistically with the chemotherapeutic drugs, cisplatin, etoposide and carboplatin. The stress-induced transcription factor GADD153 and the Bcl2-related protein Bak are upregulated in response to fenretinide. Although fenretinide is a partial retinoic acid receptor (RAR)-beta/gamma agonist, RARbeta/gamma antagonists do not block the induction of GADD153 or Bak by fenretinide. Conversely, the induction of GADD153 and Bak is blocked by antioxidants. Neither GADD153 or Bak were induced by chemotherapeutic agents but over expression of GADD153 results in increased sensitivity to fenretinide-induced apoptosis. Therefore, fenretinide induces apoptosis via RAR-dependent and -independent pathways in which the RAR-independent pathway is characterised by the reactive oxygen species-dependent induction of GADD153 and Bak. The targeting of GADD153 and Bak in neuroblastoma cells may be novel pathways for the development of drugs inducing apoptosis of neuroblastoma with improved tumour specificity.

Growth and DNA damage-inducible transcription factor 153 mediates apoptosis in response to fenretinide but not synergy between fenretinide and chemotherapeutic drugs in neuroblastoma.

Fenretinide induces apoptosis of neuroblastoma cells in vitro and interacts synergistically with the chemotherapeutic drugs cisplatin and etoposide. The stress-inducible transcription factor known as growth and DNA damage (GADD)-inducible transcription factor 153 is induced in response to fenretinide and in other cell types modulates apoptosis via pro- and antiapoptotic members of the BCL2 family. Because BCL2-family proteins are important in apoptosis induced by chemotherapeutic drugs, GADD153 may be a key mediator of synergy between fenretinide and chemotherapeutic drugs. To investigate this, GADD153 cDNA in sense and antisense orientations was stably transfected into SH-SY5Y neuroblastoma cells using a tetracycline-inducible vector. Increased expression of GADD153 raised the background level of apoptosis and increased apoptosis induced by fenretinide or the chemotherapeutic drugs cisplatin and etoposide. However, there was no increase in synergy between fenretinide and chemotherapeutic drugs. Conversely, expression of antisense-GADD153 virtually abolished the induction of apoptosis in response to fenretinide but overall had no significant effect on apoptosis induced by chemotherapeutic drugs. The effect of antisense-GADD153 on synergy between chemotherapeutic drugs and fenretinide varied with the drug used: there was no effect on synergy between fenretinide and cisplatin, but the combination of fenretinide with etoposide became antagonistic. These results suggest that mechanisms mediating synergy between fenretinide and chemotherapeutic drugs lie upstream of GADD153.

Mechanisms of free-radical induction in relation to fenretinide-induced apoptosis of neuroblastoma.

The mechanisms of fenretinide-induced cell death of neuroblastoma cells are complex, involving signaling pathways mediated by free radicals or reactive oxygen species (ROS). The aim of this study was to identify mechanisms generating ROS and apoptosis of neuroblastoma cells in response to fenretinide. Fenretinide-induced ROS or apoptosis of SH-SY5Y or HTLA 230 neuroblastoma cells were not blocked by Nitro l-argenine methyl ester (l-NAME), an inhibitor of nitric oxide synthase. Flavoprotein-dependent superoxide-producing enzymes such as NADPH oxidase were also not involved in fenretinide-induced apoptosis or ROS generation. Similarly, ketoconazole, a cytochrome P450 inhibitor, and inhibitors of cyclooxygenase (COX) were also ineffective. In contrast, inhibition of phospholipase A(2) or lipoxygenases (LOX) blocked the induction of ROS and apoptosis in response to fenretinide. Using specific inhibitors of LOX, blocking 12-LOX but not 5- or 15-LOX inhibited both fenretinide-induced ROS and apoptosis. The effects of eicosatriynoic acid, a specific 12-LOX inhibitor, were reversed by the addition of the 12-LOX products, 12 (S)-hydroperoxyeicosatetraenoic acid and 12 (S)-hydroxyeicosatetraenoic acid. The targeting of 12-LOX in neuroblastoma cells may thus be a novel pathway for the development of drugs inducing apoptosis of neuroblastoma with improved tumor specificity.

Gangliosides link the acidic sphingomyelinase-mediated induction of ceramide to 12-lipoxygenase-dependent apoptosis of neuroblastoma in response to fenretinide.

BACKGROUND: The lipid second messenger ceramide, which is generated by acidic and neutral sphingomyelinases or ceramide synthases, is a common intermediate of many apoptotic pathways. Metabolism of ceramide involves several enzymes, including glucosylceramide synthase and GD3 synthase, and results in the formation of gangliosides (GM3, GD3, and GT3), which in turn promote the generation of reactive oxygen species (ROS) and apoptosis. Fenretinide, a retinoic acid derivative, is thought to induce apoptosis via increases in ceramide levels, but the link between ceramide and subsequent apoptosis in neuroblastoma cells is unclear. METHODS: SH-SY5Y and HTLA230 neuroblastoma cells were treated with fenretinide in the presence or absence of inhibitors of enzymes important in ceramide metabolism (fumonisin B1, inhibitor of ceramide synthase; desipramine, inhibitor of acidic and neutral sphingomyelinases; and PDMP, inhibitor of glucosylceramide). Small interfering RNAs were used to specifically block acidic sphingomyelinase or GD3 synthase activities. Apoptosis, ROS, and GD3 expression were measured by flow cytometry. RESULTS: In neuroblastoma cells, ROS generation and apoptosis were associated with fenretinide-induced increased levels of ceramide, glucosylceramide synthase activity, GD3 synthase activity, and GD3. Fenretinide also induced increased levels of GD2, a ganglioside derived from GD3. Inhibition of acidic sphingomyelinase but not of neutral sphingomyelinase or ceramide synthase, blocked fenretinide-induced increases in ceramide, ROS, and apoptosis. Exogenous GD3 induced ROS and apoptosis in SH-SY5Y cells but not in SH-SY5Y cells treated with baicalein, a specific 12-lipoxygenase inhibitor. Exogenous GD2 did not induce apoptosis. CONCLUSIONS: A novel pathway of fenretinide-induced apoptosis is mediated by acidic sphingomyelinase, glucosylceramide synthase, and GD3 synthase, which may represent targets for future drug development. GD3 may be a key signaling intermediate leading to apoptosis via the activation of 12-lipoxygenase.

Quantification of MYCN, DDX1, and NAG gene copy number in neuroblastoma using a real-time quantitative PCR assay.

Amplification of the proto-oncogene MYCN is a strong adverse prognostic factor in neuroblastoma patients in all tumor stages. The status of the MYCN gene has become an important factor in clinical decision making and therapy stratification. Consequently, fast and accurate assessment of MYCN gene copy number is of the utmost importance and the use of two independent methods to determine MYCN status is recommended. For these reasons we have developed and evaluated a real-time quantitative PCR (Q-PCR) assay as an alternative for time-consuming Southern blot analysis (SB), and as a second independent technique in parallel with fluorescence in situ hybridization (FISH) analysis. Advantages of Q-PCR are a large dynamic range of quantification, no requirement for post-PCR sample handling and the need for very small amounts of starting material. The accuracy of the assay was illustrated by measurement of MYCN single gene copy changes in DNA samples of two patients with 2p deletion and duplication, respectively. Two different detection chemistries i.e., a sequence specific TaqMan probe and a generic DNA binding dye SYBR Green I were evaluated and shown to yield similar results. Also, two different calculation methods for copy number determination were used i.e., the kinetic method and the comparative C(T) method, and shown to be equivalent. In total, 175 neuroblastoma samples with known MYCN status, as determined by FISH and/or SB, were examined. Q-PCR data were highly concordant with FISH and SB data. In addition to MYCN copy number evaluation, DDX1 and NAG gene copy numbers were determined using a similar Q-PCR strategy. Survival analysis pointed out that DDX1 and/or NAG amplification has no additional adverse effect on prognosis.