Current Research and Scholarly Interests
My laboratory studies childhood brain tumors with a particular focus on medulloblastoma and pediatric high grade glioma. We utilize computational/genomic and cell biological approaches to 1) understand the molecular and cellular basis of these diseases, 2) identify and target the genetic, epigenetic and metabolic vulnerabilities in these tumors and 3) inform the current and next generation of clinical trials focused on bringing the most promising and novel therapies to patients with these diseases.
Functional annotation of the medulloblastoma genome
We are currently combining genome-wide RNAi with chemical biology and chemical genomic screens to systematically define the functional consequence of targeting genes and biological pathways enriched in medulloblastoma. Using this approach, we have uncovered genetic, epigenetic and metabolic vulnerabilities the clinically aggressive MYC-amplified/Group3 tumors, which we are developing into novel therapies.
Genome-wide mapping of cell-identity genes in medulloblastoma
We are characterizing the epigenetic states of various childhood brain tumors and how modulation of these epigenetic states alters their behavior. We have identified bromodomain and extra-terminal containing proteins (BET) as critical effectors of gene expression in two of the most lethal forms of medulloblastoma (MYC-amplified/'Group3' tumors and p53-mutated SHH tumors) and have established BET inhibitors as promising therapies for treating patients with these tumors. BET proteins, such as BRD4, facilitate gene expression through their recognition of acetylated lysines, such as at the K27 residue on the histone tail, and subsequent positive interactions with the pTEFb and Pol II transcriptional machinery. BET proteins have also been identified as critical components of large transcriptional co-activator complexes referred to as "super-enhancers" that activate key cell identity genes in development and oncogenesis. We hypothesize the genes regulated by super-enhancers represent high-value therapeutic target and are now mapping and functionally annotating super-enhancer regions in various subtypes of medulloblastoma with a focus on MYC-amplified/Group3 and SHH subforms of this disease.
Epigenomic dysregulation in pediatric glioblastoma and DIPG
We have a keen interest in understanding how mutations in Histone H3 proteins, that commonly occur in pediatric high grade gliomas, influence oncogenesis. The mutations in these histones substitute a methionine at the lysine-27 residues of either H3.1 or H3.3 proteins ("K27M" mutant histone) and we and others have shown these K27M mutant histones are potent inhibitors of the PRC2 complex, resulting in a striking global hypomethylation of Histone K27 residues across the genome and in turn, global changes in DNA methylation and gene expression patterns in glioblastoma and DIPG. One of our major goals is to identify vulnerabilities that arise on account of K27M mutation in these cancers through "synthetic lethal" screens and then to exploit these vulnerabilities for therapeutic gain.
Contribution of DDX3 and RNA helicases in cancer
Using whole-exome and genome sequencing, we have identified a number of previously unreported mutations in medulloblastoma. We are functionally characterizing several of these using in vitro and in vivo assays and determining whether such events carry clinical significance in terms of prognosis, treatment response or potential for targeted therapy. Through this effort, we have identified an RNA helicase, DDX3X, as one of the most commonly mutated genes in medulloblastoma, occurring predominantly in either WNT pathway or SHH pathway activated tumors. We are now characterizing how mutations in DDX3 alter its global interactions with proteins and RNAs using IP-MS and CLIPseq, respectively, and identifying how these shifts facilitate cance