Wilfried Roth

Enhanced Notch signaling contributes to uncontrolled cell growth and cell death resistance in cancer. Here, we demonstrate that in colorectal carcinoma cells the Notch1-dependent activation of cell cycle and proliferation is mediated by repression of the cyclin-dependent kinase inhibitor (CDKI) p27. The half-life of p27 significantly increased after siRNA‑mediated knockdown of Notch1. Notch1 depletion altered the transcription of SKP2, KPC1 and KPC2, which are E3-ubiquitin ligase subunits targeting p27 for proteasomal degradation in the nucleus and the cytoplasm, respectively. As a consequence, the levels of p27 in both cellular fractions were elevated upon Notch1 knockdown. Importantly, the downregulation of Notch1 significantly sensitized colorectal cancer cells to chemotherapy and ionizing radiation. Our findings support an important role of p27 in Notch1-dependent oncogenic signaling and suggest that Notch1 is a promising target for an experimental therapy of colorectal carcinoma.

Treatment with the Bcl-2/Bcl-XL inhibitor ABT-737 is a promising novel strategy to therapeutically induce apoptotic cell death in malignant tumors such as glioblastomas. Although many studies have demonstrated that ABT-737 acts synergistically with chemotherapeutic drugs, the possibility of a combined treatment with ionizing radiation (IR) and ABT-737 has not yet been thoroughly investigated. Similarly, the relationship between p53 function and the pro-apoptotic effects of ABT-737 are still obscure. Here, we demonstrate that IR and ABT-737 synergistically induce apoptosis in glioblastoma cells. The sensitivity to ABT-737-mediated cell death is significantly increased by the IR-dependent accumulation of cells in the G2/M cell cycle phase. Wild type p53 function inhibits the efficacy of a combined IR and ABT-737 treatment via a p21-dependent G1 cell cycle arrest. Moreover, mutant as well as wild type p53 counteract the pro-apoptotic activity of ABT-737 by maintaining the expression levels of the Mcl-1 protein. Thus, p53 regulates the sensitivity to ABT-737 of glioblastoma cells. Our results warrant a further evaluation of a novel combination therapy using IR and ABT-737. The efficacy of such a therapy could be substantially enhanced by Mcl-1-lowering strategies.

Defects in the apoptotic signaling cascades contribute to the poor therapeutic response of malignant gliomas. As glioblastomas are characterized by high expression levels of anti-apoptotic Bcl-2 family proteins, we studied the effects of the novel Bcl-2 inhibitor, ABT-737, on malignant glioma cells. ABT-737 treatment released the pro-apoptotic Bax protein from its binding partner Bcl-2 and potently induced apoptotic cell death in glioblastoma cells in vitro and in vivo. The local administration of ABT-737 prolonged the survival in an intracranial glioma xenograft model. Downregulation of Mcl-1 and overexpression of Bcl-2 sensitized the cells to ABT-737-mediated apoptosis. Moreover, ABT-737 potentiated the cytotoxicity of the chemotherapeutic drugs vincristine and etoposide, and of the death ligand TRAIL. As glioma stem cells may play a crucial role for the tumor progression and the resistance to treatment in glioblastomas, we investigated the effects of ABT-737 on the subpopulation of glioma cells exhibiting stem cell characteristics. Inhibition of proliferation and induction of apoptosis by ABT-737 were less efficient in glioma stem cells than in non-stem cell-like glioma cells. As the resistance of glioma stem cells was associated with high Mcl-1 expression levels, ABT-737 treatment combined with downregulation of Mcl-1 could represent a promising novel approach in glioblastoma treatment.

The HIPPI (HIP-1 protein interactor) protein is a multifunctional protein that is involved in the regulation of apoptosis. The interaction partners of HIPPI include HIP-1 (Huntingtin-interacting protein-1), Apoptin, Homer1c, Rybp/DEDAF, and BAR (bifunctional apoptosis regulator). In search for other binding partners of HIPPI, we performed a yeast two hybrid screen and identified BLOC1S2 (Biogenesis of lysosome-related organelles complex-1 subunit 2) as a novel HIPPI-interacting protein. In co-immunoprecipitation assays, BLOC1S2 specifically associates with HIPPI, but not with HIP-1. To study the expression of BLOC1S2 on the protein level, we generated a mouse monoclonal antibody specific for BLOC1S2 and a multiple tissue array comprising 70 normal and cancer tissue samples of diverse origin. BLOC1S2 protein is widely expressed in normal tissue as well as in malignant tumors with a tendency towards lower expression levels in certain subtypes of tumors. On the subcellular level, BLOC1S2 is expressed in an organellar-like pattern and co-localizes with mitochondria. Over-expression of BLOC1S2 in the presence or absence of HIPPI does not induce apoptosis. However, BLOC1S2 and HIPPI sensitize NCH89 glioblastoma cells to the pro-apoptotic actions of staurosporine and the death ligand TRAIL by enhancing caspase activation, cytochrome c release, and disruption of the mitochondrial membrane potential. Given its interaction with HIPPI and its pro-apoptotic activity, BLOC1S2 might play an important functional role in cancer and neurodegenerative diseases.

PEA-15 (phosphoprotein enriched in astrocytes 15 kDa) is a death effector domain-containing protein, which is involved in the regulation of apoptotic cell death. Since PEA-15 is highly expressed in cells of glial origin, we studied the role of PEA-15 in human malignant brain tumors. Immunohistochemical analysis of PEA-15 expression shows strong immunoreactivity in astrocytomas and glioblastomas. Phosphorylation of PEA-15 at Ser¹¹⁶ is found in vivo in perinecrotic areas in glioblastomas and in vitro after glucose deprivation of glioblastoma cells. Overexpression of PEA-15 induces a marked resistance against glucose deprivation-induced apoptosis, whereas small interfering RNA (siRNA)-mediated downregulation of endogenous PEA-15 results in the sensitization to glucose withdrawal-mediated cell death. This antiapoptotic activity of PEA-15 under low glucose conditions depends on its phosphorylation at Ser¹¹⁶. Moreover, siRNA-mediated knockdown of PEA-15 abolishes the tumorigenicity of U87MG glioblastoma cells in vivo. PEA-15 regulates the level of phosphorylated extracellular-regulated kinase (ERK)1/2 in glioblastoma cells and the PEA-15-dependent protection from glucose deprivation-induced cell death requires ERK1/2 signaling. PEA-15 transcriptionally upregulates the Glucose Transporter 3, which is abrogated by the inhibition of ERK1/2 phosphorylation. Taken together, our findings suggest that Ser¹¹⁶-phosphorylated PEA-15 renders glioma cells resistant to glucose deprivation-mediated cell death as encountered in poor microenvironments, for example in perinecrotic areas of glioblastomas.

Induction of apoptosis by the death ligand TRAIL might be a promising therapeutic approach in cancer therapy. However, since not all tumor cells are sensitive to TRAIL, there is a need for the development of strategies to overcome TRAIL-resistance. The results of the present study show that the anti-diabetic drug troglitazone sensitizes human glioma and neuroblastoma cells to TRAIL-induced apoptosis. This process is accompanied by a substantial increase of active caspase 8 and active caspase 3, but it is independent of troglitazone's effects on the nuclear receptor PPAR-γ. Troglitazone induces a pronounced reduction in protein expression levels of the anti-apoptotic FLICE-inhibitory protein (FLIP) without affecting FLIP mRNA levels. Further, protein and mRNA expression levels of the anti-apoptotic protein Survivin significantly decrease upon treatment with troglitazone. Moreover, sensitization to TRAIL is partly accompanied by an up-regulation of the TRAIL receptor, TRAIL-R2. A combined treatment with troglitazone and TRAIL might be a promising experimental therapy because troglitazone sensitizes tumor cells to TRAIL-induced apoptosis via various mechanisms, thereby minimizing the risk of acquired tumor cell resistance.