RICERCA ITALIANA     

Medilloblastoma: molecular pathways of neoplastic development and progression to identify novel therapeutic approaches

Università degli Studi di Roma "La Sapienza"

Abstract

Several genetic and epigenetic changes are described in MB, which affect signaling pathways involved in specific developmental steps and may be responsible for tumorigenesis. A major deregulated pathway in MB is the Sonic hedgehog (SHH) pathway. The SHH pathway promotes the proliferation of cerebellar granule progenitors. The WNT pathway is also involved and mutations of WNT proteins have been reported in MB. ). In the intercalation of WNT and EGFR networks activates several pathways that are thought to contribute to radiation resistance and enhanced proliferation. Less is known about the role of Notch. Insights for MB tumorigenesis arise also from animal models. The gene expression profiles of murine tumors from mutations in different pathways shared a common gene expression patterns presumed to be tumor specific. Complex interconnections between the different pathways are emerging and suggest that dysregulation of different pathways may cooperate in the pathogenesis of MB. A important result of these studies is that such pathways may represent potential therapeutic targets for novel treatments for MB. There is evidence that the molecular mechanisms promoting tumour growth and those regulating tumor angiogenesis are correlated. A variety of angiogenic factors have been implicated, individually and together, in the neo-angiogenesis of MB. The VEGF has an autocrine activity on the neoplastic cells and is not only a relevant factor for tumour cell proliferation, but it is also a target for other molecules. b-catenin is one of those and its relationship with VEGF seems to be of particular interest, considering the possible role b-catenin and the WNT transcription pathway in the MB pathogenesis. Recent evidence suggests that MBs might originate from cancer stem cells which may represent the specific target for new therapeutic compounds. Among these, retinoic acid and other naturally occurring retinoids act as morphogenetic agents during embryonic development, and regulate the growth and differentiation of a wide variety of cell types including stem cells. The present research program, composed of four Research Units (RU), will focus on different biologic aspects of the tumorigenesis of MB in the effort to open new therapeutic roads. The following tasks will be pursued: Task1: To evaluate the main molecular pathways of SHH and NOTCH involved in MB tumorigenesis; Task 2.To investigate the WNT signaling pathway and the signal transduction pathway turned on by EGFR in response to ionizing radiation in MB cell lines. Task 3 To investigate the autocrine role of VEGF and its relationship with b-catenin in MB cell lines. Task 4 Evaluation of the antitumor activity of retinoid related molecules (RRMs) and all trans-retinoid (ATRA) in MB preclinical models. <<<

Principal Investigator

Felice Giangaspero Università degli Studi di ROMA "La Sapienza"

Research Objectives

The present research program is composed of four Research Units (RU) whose single projects will focus on different biologic aspects involved in the tumorigenesis of medulloblastoma (MB) in the effort to open new therapeutic roads to improve further the outcome for children affected by this tumor. Un important goal of the whole program is to create a network which will help the investigators to collect a relative high number of new cases of MB and to shear research tools as cell lines, murine models and, when obtained, cancer stem cells.
The principal objectives are the following:
•To evaluate the molecular pathways of SHH and NOTCH involved in MB tumorigenesis. Complex interconnections between the different pathways involved in MB tumorigenesis are emerging and all togheter suggest that dysregulation of different pathways may cooperate in the pathogenesis of MB.
•To evaluate the role of WNT pathway in ionizing radiation response and the molecular cross-talk between beta-catenin, p53 and signal transduction pathways turned on by EGFR in human MB cell lines.
•To understand the possible relationship between b-catenin (generally the WNT pathway) and the outocrine effect of VEGF in MB cell line. Assuming that the success of an oncologic therapeutic approach will depend upon a multiple and concurrent attacks to all those mechanisms promoting and sustaining tumour cell growth, it can hypothesized that those agents regulating the neoplastic neovascolarization are expected to have a potential important role in the treatment of MB.
•To evaluate the antitumor activity of new therapeutic compounds as Retinoid related molecules (RRMs) and all trans retinoid acid (ATRA) in MB preclinical models including primary cell cultures, cancer stem cells and xenografts and to investigate the underlying molecular pathways involved in RRMs and ATRA treatment response. <<<

Timescale

24 months

National and international background

Medulloblastoma (MB) is the most common malignant brain tumor of childhood. Current treatment includes maximal surgical resection, whole neuroaxis irradiation, and chemotherapy. Cure rate is approximately 40% in children with metastatic/unresectable MB, and 65% in patients with average risk disease. In addition, treatment-related morbidity can be devastating in children, who may suffer severe long-term physical and cognitive deficits (1). Although the survival has improved considerably within the past two decades, however, patients with similar neoplasms and receiving identical therapies, can have widely disparate clinical outcome (1). It would be advantageous to tailor specific therapies to individual lesions so that patients are not over or under treated. Clinical, histological, and molecular factors have been proposed for such MB stratification (2) . Moreover the basic mechanisms of tumor progression which can explain the variable biologic behavior of such neoplasm is only beginning to emerge.
Most MB are thought to originates from transformed granule cell precursor (GCP) of the cerebellum (although some may arise from neuronal precursors cells in the subventricular zone). Several genetic and epigenetic changes are described in MB, which affect signaling pathways involved in specific developmental steps and may be responsible for tumorigenesis (3,4,5,6). A major deregulated pathway in MB is represented by the Sonic hedgehog (SHH) signaling pathway. Germline mutations in components of the SHH (Ptc and the Gli-repressor SUFU) activate ligand-independent signals and are responsible for the inherited Gorlin’s syndrome characterized by increased susceptibility to the development of MB (6,7,8). Somatic mutations of Ptc, SUFU and Smo are observed in 25% sporadic MB while epigenetic activation of Gli-1 gene expression also occurs (1,6,9). The SHH pathway is a major regulator of GCP since SHH, an agonist ligand secreted by Purkinje cells, promotes the proliferation of granule progenitors of the external granular layer (EGL) when binds to the receptor Ptch, thus suggesting that uncontrolled activation of this mitogenic pathway sustains tumorigenesis (3,6).
The WNT pathway, as the SHH signalling, is also involved in a number of developmental processes and tumorigenesis and mutations in components of this system have been demonstrated in approximately 15% of sporadic MB predicting aberrant pathway activation (10). In the absence of WNT signal, beta-catenin is subjected to degradation because of its association with Axin, GSK3-beta and the tumor suppressor APC. On activation of WNT signal, GSK3-beta is inhibited and accumulated beta-catenin translocated into the nucleus where it acts as a transcriptional factor of a number of target genes involved in cell cycle progression, apoptosis and differentiation (7, 11-13) such as cyclin D1, c-myc and ARF (14, 15). In particular ARF encodes a potent tumor suppressor that antagonizes MDM2, a negative regulator of p53, but can also induce a p53-independent apoptosis (16). Moreover, in the absence of a functional p53 pathway beta-catenin is rendered free to exert its transcriptional effects. On the other hand beta-catenin induces the expression of WISP-1 gene whose product can block p53-mediated apoptosis relying on the activation of AKT kinase. AKT is a well-established anti-apoptotic protein: it is activated by PI3K and engages direct interaction with MDM2. The amount of available beta-catenin for target gene activation can also be increased by shifting protein from the cadherin-bound pool to the cytoplasmic pool (17,18). EGFR and ErbB2 binding to their cognate ligand are able to phosphorylate beta-catenin at tyrosine residue, which causes dissociation of beta-catenin from the cadherin-catenin complex. The beta-catenin thus liberated can be recycled to the cytoplasmic pool stimulating the WNT signalling (15, 16, 22). In the intercalation of WNT and EGFR networks GSK3-beta is both a central negative regulator of beta-catenin and a favorite AKT target: phosphorylation of GSK3-beta by AKT inactivates the former. EGFR (ErbB1 o HER-1) is a member of the ErbB family of receptor tyrosine kinases that also includes the closely related ErbB2, ErbB3 and ErbB4 receptors. Signal transduction through the EGFR, activates several pathways that are thought to contribute to radiation resistance, including decreased apoptosis, cell cycle alterations, and enhanced proliferation. Cyclin D1, a key regulator of the cell cycle, is EGFR-activated through Ras-Raf-MEK-ERK protein kinase cascade as well as the c-myc, another oncogenic protein. Agents that inhibit EGFR signalling may block this mechanism and enhance the cell killing effectiveness of radiation therapy or serve as radiation sensitizers. Gefitinib (ZD 1839 “Iressa”) is a selective inhibitor of EGFR blocking the signal transduction pathways involved in proliferation and survival, offering a therapeutic target to increase the cytotoxic effect of radiotherapy (19, 20, 21) Radiation therapy plays a major role in the treatment of MB (22). Response to radiation varies by individual tumor and it is often limited by tumor radiation resistance and the rapid proliferation of cells after radiation resulting in tumor repopulation. Several factors are believed to contribute to the radioresistance. The p53 tumor-suppressor plays a critical role in the prevention of human cancer playing a key role in regulating apoptosis especially in the setting of radiation exposure. It has been established that exist a crosstalk between p53 and two pivotal regulatory proteins: beta-catenin and the AKT/PKB downstream protein of EGFR pathway (23). It has been, recently, observed that nuclear accumulation of beta-catenin in MB is a marker of favorable outcome; this suggests that Wnt pathway activation in MB may be peculiar, affecting the balance between cell proliferation and cell death to produce tumor cells that are prone to apoptotic stimuli or particularly radiosensitive (11).
High expression of Notch receptor and of its target gene have been also demonstrated in MB (24,25) suggesting an etiologic role for the Notch cascade in this tumor. However, less is known about the role of Notch, that is required to complete neurogenesis and maintains granule progenitor proliferation by activating, together with SHH, the differentiation inhibitor Hes-1 (26).
Insights for MB tumorigenesis arise also from animal models (27,28). Deletion of one Ptc1 allele in mice results in the development of MB underlying the critical tumorigenic role of sonic hedgehog (SHH) (27). Targeted deletion of genes controlling cell cycle checkpoints, apoptosis and DNA repair (p53, Rb, DNA Ligase IV and PARP-1) (28,29) or DNA damage (ionizing radiation in a susceptible phase of neonatal progenitors, but not during adulthood) in Ptc1+/- mice (30), also enhance MB development, by promoting the accumulation of genetic defects in other cellular pathways. All of the MB models are claimed to be histologically similar to their human counterpart. However no information are, so far, available on the morphological spectrum and the presence of anaplastic progression in such murine model. Lee et al (31) comparing the gene expression profiles from a number of mouse MB models demonstrated that all MB exhibited gene expression profiles that were more similar to developing cerebellum than to adult cerebellum, supporting the prevailing hypothesis that MB arise from cerebellar granule cell precursors in the external germinal layer during development. Surprisingly, the gene expression profiles of tumors arising from mutations in different genetic pathways shared a number of common gene expression patterns that were distinct from normal developing and adult cerebellum and were presumed to be tumor specific.
Complex interconnections between the different pathways involved in MB tumorigeneis such as Shh/Gli and the Notch pathway or the Shh/Gli and Wnt pathway (9-12) are emerging and all together suggest that deregulation of different pathways may cooperate in the pathogenesis of MB. A very important result of these studies is that such pathways may represent potential therapeutic targets for development of novel treatments for MB. For example, antagonists of the Shh pathway, including cyclopamine, has been proved to block proliferation of MB cells in vitro and in vivo (32,33).
There is evidence that the molecular mechanisms promoting tumour growth and those regulating tumor angiogenesis are correlated (34). Medulloblastomas often shows marked neovascularization (35) and actually the quantitative assessment of angiogensis has been also proposed as a possible criteria to be incorporated in the anaplasia grading system for MB (36). A variety of angiogenic factors have been implicated, individually and together, in the neo-angiogenesis of MB highlighting once again the complexity of the biological processes regulating tumour neovascularization (37). Thus, investigation into one of this area could bring insight into the other one. The VEGF is one of the most powerful mitogenes for the endothelial cells. Different isoforms of the VEGF exist, generated by mechanism of alternative splicing of the same gene. The VEGF other than having this paracrine activity on the endothelial cells, it has also an autocrine activity on the neoplastic cells, considering that also tumour cells express VEGF receptors (38). VEGF is not only a relevant factor for tumour cell proliferation, but it is also a target for other molecules, including some of those involved in maintaining cell structure integrity and in regulating gene transcription. b-catenin is one of those and its relationship with VEGF seems to be of particular interest, considering the possible role b-catenin and the WNT transcription pathway in the MB pathogenesis (39,40,41,43)). It has been shown that b-catenin influences VEGF expression in colon cancer cells. More precisely, it has been shown that within the VEGF-A promoter there are seven TCF (T-cell factor) binding domains (42). The final result of the genetic alteration of the b-catenin gene and/or of other ones coding for molecules forming the WNT transcription pathway, is the different intracellular distribution of the molecule which thus becomes segregated with the nucleus, where, interacting with specific transcriptional factors (LEF-TCF), influences the transcription of numerous genes regulating cell proliferation (c-myc; cyclin D, c-jun, etc..) (14). Assuming that the success of an oncologic therapeutic approach will depend upon a multiple and concurrent attacks to all those mechanisms promoting and sustaining tumour cell growth, it can hypothesized that those agents regulating the neoplastic neovascolarization are expected to have a potential important role in the treatment of MB.
Accumulating evidence suggests that MBs might originate from a small fraction of tumor cells, termed cancer stem cells (CSCs), which have undergone neoplastic transformation, and possess the ability to proliferate and self-renew and uniquely maintain the tumor growth. Interestingly, the genes that are often mutated or overexpressed in MB, such Shh, Wnt, and Notch, have been demonstrated to be involved in regulation of self-renewal in normal stem cells from many tissues (43-46). Cancer stem cells may represent the specific target for new therapeutic compounds. Among these, retinoic acid (RA) and other naturally occurring retinoids act as morphogenetic agents during embryonic development, and regulate the growth and differentiation of a wide variety of cell types including stem cells. The biological effects of retinoids in target cells are mediated by binding through families of nuclear hormone receptors (RAR alpha, beta and gamma, and RXR alpha, beta and gamma), which show spatially and temporally restricted distribution patterns during normal embryogenesis. The abundant expression of RARbeta in the developing central nervous system suggests that RARbeta may act as the major transducer of the retinoid signal during normal neurogenesis (47). RA administration in mouse embryos impairs rostral CNS demarcation (the area from which cerebellum arises) and down-regulates the expression of the transcription factor OTX2 (49). Low concentrations of all trans-retinoic acid (ATRA) and other RAR agonists induce significant growth-inhibitory effects in both primary cultures and established MB cell lines through induction of apoptosis (49,50). ATRA does not appear to be effective in MB cell lines without OTX2 expression. The antiproliferative effects of ATRA has been linked to downregulation of OTX2 expression (51,52). Retinoid-based therapies are currently used in haematological and solid tumors (53, 54). In order to overcome the limitations of ATRA in the clinical setting, such as acquired resistance and side-toxic effects, many synthetic retinoids have been generated. CD437, which is the prototype of this novel class of atypical retinoids, was originally designed as RAR gamma selective agonist and is endowed with potent antitumor activity and strong apoptogenic potential in different tumor types. Interestingly, a novel RRM analogue, ST1926, with either a better pharmacological profile and a stronger antitumour activity than CD437, appears to maintain RAR gamma agonistic activity but loses receptor selectivity (55,56). In different tumor models, CD437 and ST1926 were shown to induce DNA damage, cell cycle arrest in G1 or G2 phases and apoptosis (56,57). Because of their strong apoptogenic potential in different tumor types, RRMs might represent an interesting class of agents in MB, which appears already susceptible to ATRA-mediated programmed cell death. <<<