Brain tumors are the leading cause of cancer-related death in children. Poor understanding of the interactions between tumor cells, especially cancer stem cells (CSCs), and normal brain tissues have played a role. Due to the difficulties of obtaining tumor tissues, there is a limited availability of biologically accurate and matched in vitro (tumor organoid) and in vivo (PDOX) model systems. Normal brain tissue is more difficult to obtain, making it nearly impossible to develop normal brain organoids. To overcome these barriers, we propose to utilize our existing (n=127) and future established patient derived orthotopic xenograft (PDOX) models as the source supply of tumor cells to develop matching tumor organoids. These PDOX models replicate the biology of original patient tumors and represent a wide spectrum of the newly discovered molecular subtypes of malignant brain tumors. Using 24 PDOX models, we have established 13 (54.1%) tumor organoids. Recognizing the rarity of normal brain tissues, we propose to use autopsied tissues. Indeed, we have cryopreserved viable cells of normal cerebrum, cerebellum, brain stem, and sub-ventricular zone cells from children of different ages and showed that the autopsied cerebral cells can proliferate in vitro to form neurospheres. We, therefore, hypothesize that 1) autopsied normal brain tissues and PDOX models can provide critically needed and biologically accurate cells for the development of normal and tumor organoid, and 2) the growth of these organoid are dependent on the growth factor milieu of their location and modulated by the dynamic interactions between different types of cells. Radiation therapy is one of the most important treatment regimens for pediatric brain tumors. Since mutation of histone modification genes are frequent in pediatric brain tumors, our 3rd hypothesis is that such epigenetic changes affect responses and resistance toward radiation, and successful targeting in organoid and PDOX tumors will lead to new combination therapies. We propose three Specific Aims.
Aim 1 : Develop normal brain organoids from autopsied normal cerebrum, cerebellum, brain stem and subventricle zone from all age groups and establish tumoral organoids from PDOX models and identify their location- and developmental-stage dependent niche factor combinations, respectively.
Aim 2 : Understand the dynamic interactions between tumor (CSCs and non-stem tumor) cells and normal brain cells. We will use co-culture assays to identify the niche factor combinations that dictate cellular diversity and survival of CSC and normal brain cells, and perform global multi-omics analysis and single cell RNAseq to elucidate the molecular mechanisms, followed by in vivo validation of our findings in PDOX models.
Aim 3 : Determine the role of histone methylation in regulating CSC survival and self-renewal of malignant brain tumors when perturbed by radiation. We will use normal and tumoral organoids to determine how pre-existing histone methylation status affect their survival and to identify key radiation-resistant methylation(s) by combining radiation with an informer set of small-molecule probes that selectively target distinct nodes in tumor epigenetic regulatory circuity in organoids and in PDOX models.
Our study will provide a new mechanistic understanding of brain tumor cell biology during their interaction with normal brain cells and identify a new set of genetic and epigenetic drivers that can be targeted in our large panel of organoids and PDOX mouse models of pediatric malignant brain tumors.
