Trks and NB Biology In NB tumors, our studies have shown that Trks not only serve as prognostic markers but also impact the biology of these tumors. Neuroblastoma (NB) tumors expressing high levels of BDNF and TrkB are associated with poor 5-year survival outcomes. Our previous studies indicated that BDNF blocked the cytotoxic effects of vincristine, etoposide, adriamycin and cisplatin on NB cells. This chemoresistance was mediated by the TrkB tyrosinie kinase and PI-3kinase. Recent studies have further mapped the signaling molecules downstream of PI3-kinase that mediate BDNF/TrkB rescue of NB cells from chemotherapy. Using genetic approaches we found that activation of AKT can substitute for BDNF and rescue NB cells from death induced by cisplatin and etoposide while a dominant negative AKT can block BDNF's ability to rescue cells. We have used a novel phosphatidylinositol ether lipid analogue, PIA, which blocks AKT kinase activity and shown that it can re-sensitize BDNF treated cells to chemotherapy. While Paclitaxel is an effective drug in many cancers, its activity in Neuroblastoma has been poor. We have identified that the expression of major effector of paclitaxel induced apoptosis, BIM, is inhibited by BDNF/TrkB through activation of MAPK and could be reversed by inhibiting MAPK. This raises the possibility that inhibitors of MAPK would increase the efficacy of paclitaxol in NB. Such studies are currently underway. Since angiogenesis and the expression of VEGF is increased in tumors from patients with a poor prognosis, we reasoned that this signal transduction path may be invovled in the regulation of VEGF. Our studies showed that BDNF activation of TrkB through the PI3-Kinase/AKT/mTOR pathway stimulated HIF1a leading to an increase in VEGF transcription and production of VEGF by NB cells. We identified an number of small molecule inhibitors of these signaling path intermediaries that can block induction of HIF-1a. Moreover, we found that topotecan can inhibit HIF-1a leading to a decrease in VEGF production, thus identifying a potential mechanism by which this drug inhibits NB tumor growth. Neuroblastoma Differentiation The CASZ1 GENE was identified as a novel gene that maps to Chromosome 1p36 the area encompassing the putative neuroblastoma tumor suppressor gene. hCas maps to chromosome 1p36 by FISH, and there is LOH and/or complete deletion of hCas in NB cell lines (8/8) and primary NB tumors (3/3). During normal human fetal development hCas expression peaks at a time comparable to its expression in drosophila, and at a time when migrating neuroblasts form the sympathetic ganglia and the medullary cells of the adrenal glands. In the retinoic acid (RA) induced NB differentiation model, hCas mRNA transiently increases and peaks after 24 hours of RA treatment. The increase in hCas mRNA is inversely correlated with a decrease in expression of the N-myc oncogene. In 8/9 NB cell lines there is an inverse correlation between hCas and N-myc expression (R=-0.894). Decrease in hCas by anti-sense hCas does not change N-myc levels. However, N-myc transfection and overexpression in NB cells consistently decreases hCas expression. Transient transfection of N-myc decreases hCas promoter activity while transfection of a dominant negative N-myc stimulates hCas promoter activity. Thus N-myc negatively regulates hCas. We propose a model in which one allele of hCas is deleted on 1p and the other allele is transcriptionaly silenced by the N-myc oncogene as a novel potential mechanism by which two important genetic events in NB, N-myc amplification and 1p LOH, may contribute to tumorigenesis. hCas involvement in both drosophila and human neural development and its expression in NB differentiation raise the possibility that hCas may be a putative NB tumor suppressor gene on 1p.
