Lay Summary

LAY ABSTRACT:

Millions of cells are formed every day in the developing brain of children. Medulloblastoma, a pediatric tumor, occurs when the proliferation of cells in the cerebellum (a lower part of the brain) becomes uncontrolled. The Notch pathway is a key mechanism that governs cell proliferation in many biological contexts. Aberrant up-regulation of Notch signals is associated with medulloblastoma. Re-gaining control of Notch could help cure medulloblastoma patients. Our goal is to better understand the molecular mechanisms that control Notch signals in brain cells and, thus, to define novel therapeutic targets for the benefit of medulloblastoma patients. We recently identified the L3MBTL3 gene as a new modulator of Notch signals. Importantly, previous studies have shown that the L3MBTL3 genes is deleted in medulloblastoma patients. We hypothesize that the L3MBTL3 deletions observed in medulloblastoma patients result in the aberrant regulation of Notch signals, thus supporting tumorigenesis. In our project, we propose to test this hypothesis by studying the extent to which inhibiting L3MBTL3 modulate medullobastoma tumor progression in mouse model of medulloblastoma. Our study could offer critical mechanistic insights on the role of the L3MBTL3 in medullobastoma that could be harnessed in the future for the therapeutic benefit of medullobastoma patients. For example, if our hypothesis that L3MBTL3 suppresses medullobastoma tumorigenesis is validated, strategies aiming at increasing the actvity of L3MBTL3 in medullobastoma cells could provide the means for manipulating Notch signaling for therapeutic benefit in medullobastoma patients.

Executive Summary

TECHNICAL ABSTRACT:

Notch signaling modulates medulloblastoma tumorigenesis. Using our proteomic platform, we recently discovered an interaction between RBPJ, a key co-factor of Notch for the mediation of Notch signals, and L3MBTL3, a methyllysine reader for which homozygous deletions are observed in medulloblastoma. In a recent publication, we demonstrated that L3MBTL3 is at the heart of a molecular mechanism linking the KDM1A demethylase to Notch signal modulation. We hypothesize that malfunction of this molecular mechanism may contribute to the previously suggested tumor suppressor role of L3MBTL3 in medulloblastoma. We propose to further characterize the role of L3MBTL3 in this disease context. We will use our novel “floxed” L3mbtl3 conditional KO mouse to investigate the extent to which the cerebellum-specific KO of L3mbtl3 modulates tumorigenesis in a genetically engineered mouse model of medulloblastoma. This functional analysis will be complemented by molecular and structural studies aiming at understanding the role of L3MBTL3 as a repressor of Notch signaling in medulloblastoma cells. Our study could offer critical mechanistic insights on the role of the L3MBTL3 in medullobastoma that could be harnessed in the future for the therapeutic benefit of medullobastoma patients. For example, if our hypothesis is validated, i.e., L3MBTL3 suppresses MB tumorigenesis, the availability of the structure of the RBPJ-L3MBTL3 complex could provide the means for manipulating Notch signaling for therapeutic benefit in medulloblastoma.

JEAN-FRANCOIS RUAL (PI)  -  BIOGRAPHY:

I was trained at the Harvard Medical School in the laboratories of Dr. M. Vidal and Dr. S. Artavanis-Tsakonas, both internationally renowned for their seminal contributions in the field of interactome mapping (MV) and Notch signalingpathway(SAT). My training in proteomics and my combined expertise in both yeast two-hybrid and protein complex purification technologies are unique, as attested by my publication record in the field of “interactome” mapping. As a project leader at the Center for Cancer Systems Biology and as a contributor to many productive collaborative efforts, I have learned the value of a team-oriented approach to research. My organizational skills and my willingness to take responsibility have already been tested as I was given the opportunity to lead projects that were voluminous and complex. For example, the CCSB human interactome project, involved coordination among 30 scientists with different fields of expertise (Rual et al., Nature, 2005). More recently, the international effort I have led for the characterization of L3MBTL3 as a repressor of Notch signaling required the coordination of a multi-institutional collaboration between investigatorswith a wide range of complementary expertise ranging from molecular, cellular, structural and computational biology to biochemistry, proteomics as well as Drosophilaand C. elegans genetics (Xu et al., EMBO Journal, 2017).

I was appointed Assistant Professor of Pathology at University of Michigan on July 2011. The research interests of the Rual lab lie in the field of cancer systems biology and the use of proteomic approaches to study cellular networks.  We focus on the systematic analysis of protein interactions in biological systems and their relationship to human disease, with particular interest in the Hedgehog and the Notchpathways and their role in cancer. Our “interactome” mapping efforts have led to the discovery of PIN1 as a novel modulator of Hedgehog signaling (Xu et al., Neoplasia, 2017) as well as L3MBTL3 as a novel modulator of Notch signaling (Xu et al., EMBO Journal, 2017). In collaboration with Dr. Camelo-Piragua, we recently demonstrated in a mouse model of medulloblastoma that the loss of Pin1 impairs tumor development (Xu et al., Neoplasia, 2017). Using a similar approach, the goal of the proposed grant application is to investigate the molecular interplay between L3MBTL3 and RBPJ for the regulation of Notch signals in medulloblastoma cells and the biomedical relevance of L3MBTL3 to medulloblastoma tumorigenesis.

BACKGROUND AND SIGNIFICANCE

Note: figures and references are shown in the "Research proposal" section.

Medulloblastoma (MB) is the most common malignant brain tumor of childhood^1,2. MB is located in the cerebellum and presents embryonal histological features with densely packed small round cells and hyperchromatic nuclei. Genomics applied to MB defined four MB subgroups, each characterized by a distinct molecular/genetic signature, patient demographics and clinical profile (WNT, Sonic Hedgehog or SHH, Group#3 and #4)^3,4. More recently, integrative genomic analyses further underscored the highly heterogeneous and complex nature of MB with a large spectrum of molecularly distinct consensus subgroups and subtypes within them^5-7. Current therapeutic approaches to MB are based on surgery, radiation and non-targeted chemotherapy, and are indistinguishably applied to all MB subgroups. These therapies have led to significant improvements, with a 70% survival rate¹, but these results are achieved at a high cost to quality of life, e.g., cognitive or hormonal deficiencies^8,9. Alternative therapeutic approaches are needed and molecular stratification of MB patients has yet to be routinely implemented in the clinic. The use of SHH signaling inhibitors for the treatment of SHH-MB patients is an area of active interest¹⁰and clinical trials are currently under way, e.g., NCT01878617. However, studies in both humans and mice have previously shown that resistance to such treatment is quickly acquired, leading to relapse^11,12. A multi-therapy approach to MB could help circumvent these limitations by reducing the emergence of cells resistant to single-target inhibitors. The Notch signaling pathway, a key contributor to cellular stemness in both normal and disease contexts, including SHH-MB, represents a candidate target in a multi-therapy approach.

The Notch signaling pathway defines a conserved cell signaling mechanism governing juxtacrine signals between adjacent cells that is essentialto biological development in metazoa^13,14. The biological action of Notch is highly pleiotropicand aberrant Notch signaling leads to a broad spectrum of developmental disorders¹⁵and many cancers¹⁶.The developmental outcome of Notch signals depends on cellular context and can influence cell fate in a remarkable number of different ways, e.g., differentiation, proliferation or apoptosis^13,14. Thus, various context-specific mechanisms, many of which likely remain to be uncovered, allow the Notch building block to be “re-used” in different flavors at various junctures within the developmental framework. Identifying these context-specific modulators of Notch signals is not only essential to understanding the plasticity of Notch as a regulator of cell fate during morphogenesis, it could also provide novel clues to manipulating Notch for therapeutic benefit in human disease.

Notch, a therapeutic target in MB. Notch has emerged as an essential contributor to brain tumorigenesis^17-19, including cerebellar MB (SHH^20-26and Group 3⁵), consistent with its key roles in neural development²⁷as well as its broad contribution to oncogenesis across tissues¹⁶. In SHH-induced mouse models of MB, Notch2 and Notch target genes, e.g. HES5, showed high level of expression^21,26. Inhibition of Notch signal by soluble Delta ligand or inhibitors of the γ-secretase protease complex, which is essential for proteolytic cleavage of Notch and subsequent signaling,impairs the growth of MB-derived cells²¹. Remarkably, the Gfap promoter-driven expressionof an oncogenic form of the Notch receptor in mouse cerebellar granular neuron progenitor cells (CGNPs), i.e. the “cell of origin” in SHH-MB²⁸,results in CGNPhyper-proliferation and MBformation²⁵.Similarly, the aberrant regulation of modulators of Notch signals, e.g. Numb, Neuralized,Jag2, Itch and Ncor1,is associatedwith MB tumorigenesis^29-35. In normal physiological context, Notch2 promotes CGNP proliferation whereas Notch1 expression is associated with postmitotic cellular differentiation^24,36. A similar trend is observed in MB, i.e. Notch2 is overexpressed in primary tumors, and promotes tumor proliferation, whereas Notch1 is largely undetectable and suppresses tumor cell growth²⁴. Finally, a recent, comprehensive pathway analysis of recurrent genetic events observed in across MB tumors revealed significant overrepresentation of genes involved in the Notch pathways in Group 3⁵. Altogether,notwithstanding the ongoing debate^37,38,these observations underscore the clinical relevance of Notch signaling as a therapeutic target in MB.

Current Notch-directed therapies based on the use of inhibitors of the γ-secretase protease complex, which is essential for proteolytic cleavage of Notch and subsequent signaling,are being tested in clinical trials for MB and other brain tumors^39,40. However, the use of these “general” inhibitors of Notch signaling leads to substantial side effects associated with the ubiquitous knock down of a pathway as pleiotropic as Notch^41-43, thus reducing their therapeutic index and their potential as curative agents. Modulation of Notch signals is context-dependent and involves a complex network of proteins^14,44.The use of a context-specific, Notch-directed therapeutic approach should provide the desired healing benefit with minimum side-effects to treat Notch-related brain tumors. Our objective is to discover cerebellum-specific Notch pathway proteins and associated protein interactions that are essential in MB to maintain tumorigenic potential, thus defining Notch-directed, context-specific therapeutic targets.

HYPOTHESIS AND SPECIFIC AIMS

Using our proteomic platform for systematic protein interaction mapping^45,46, we recently discovered an interaction between RBPJ, a key transcription factor for the regulation of Notch signaling, and L3MBTL3, a methyllysine reader⁴⁷. L3MBTL3 co-localizes withRBPJ on chromatin and recruits the histone demethylase KDM1A at the enhancers of Notch/RBPJ target genes to repress their expression⁴⁷. Importantly, deletions of the L3MBTL3 locus are observed in MB patients^5,48,49 and expression of L3MBTL3 in MB-derived cells inhibits cell growth⁴⁸. We hypothesizethat the RBPJ-L3MBTL3 interaction is at the heart of a molecular mechanism linking the histone demethylase KDM1A to Notch signal repression and that malfunction of this molecular mechanism contributes to L3MBTL3’s putative tumor suppressor role in MB (Fig. 1). Hence, L3MBTL3 KO may promote tumorigenesis through aberrant “de-repression” of Notch/RBPJ target genes. We propose to test this hypothesis by characterizing both in vitro and in vivo the role of the RBPJ-L3MBTL3 interaction in cell- and animal-based models of MB.

SA#1: Cell-based. We will test if the tumor suppressor role of L3MBTL3 in MB-derived cells is dependent on its ability to modulate Notch signals. This hypothesis will be tested, for example, by using an RBPJ-interaction defective allele of L3MBTL3⁴⁷. MB cell growth inhibition will also be tested in presence (over-expression) or absence (RNAi) of KDM1A, the histone demethylase that we have identified as being recruited by L3MBTL3 to RBPJ complexes.

SA#2: In vivo characterization. The relevance of L3MBTL3 to MB remains to be demonstrated in vivo. To test if L3mbtl3 KO can initiate MB tumorigenesis “on its own”, we will use our novel conditional “floxed” L3mbtl3^fl/fl KO mouse in combination with a Math1-Cre+mouse to specifically drive the KO of L3mbtl3in cerebellar granular neuron progenitor cells. Moreover, we will crossbreed the [Math1-Cre+; L3mbtl3^fl/fl] mouse with a Ptch1^+/- mouse⁵⁰ to investigate the extent to which loss of L3mbtl3 may further promote tumorigenesis in this Ptch1^+/- mouse model of low incidence SHH-MB⁵⁰.

Impact: The completion of these aims will provide critical mechanistic insights into how L3MBTL3 collaborates with RBPJ to repress Notch target genes in MB cells, and into the role of L3MBTL3 in MB tumorigenesis. These results could be harnessed in the future for the therapeutic benefit of MB patients. For example, if our hypothesis is validated, i.e., L3MBTL3 suppresses MB tumorigenesis, the availability of the structure of the RBPJ-L3MBTL3 complex could provide the means for manipulating Notch signaling for therapeutic benefit in MB.

Note: figures and references are shown in the "Research proposal" section.

Description of Research Proposal

Budget

Collaborations and Conflicts of Interest