PROJECT SUMMARY Oncogene amplification is a key driver of cancer progression and can either occur within chromosomes or via formation of circular extrachromosomal DNA (ecDNA). ecDNA is detected in a quarter of cancer samples and half of all cancer types, and is associated with poor patient outcomes. Despite the prevalence of ecDNA- mediated oncogene amplification in cancer, we have a limited understanding of how oncogene expression is regulated on ecDNAs and how these molecules are maintained in cancer cells. My goal is to elucidate how oncogenes are amplified and dysregulated on ecDNA in cancer using multiplexed genetic perturbations, epigenomic profiling and novel genetic screening methods in cancer cell line models. In the F99 phase, I will systematically identify unique transcriptional dependencies of ecDNA-harbored oncogenes. ecDNA is linked to high levels of oncogene overexpression and accessible chromatin, suggesting that oncogene expression on ecDNA may be uniquely regulated. In my dissertation work so far, I have used imaging, chromatin conformation and epigenetic perturbation approaches to discover a novel mechanism by which ecDNAs cluster with one another in the interphase nucleus and engage in cooperative, intermolecular oncogene activation. These observations suggest that oncogene expression is regulated differently on extrachromosomal oncogene amplicons compared to chromosomal loci. I hypothesize that differential regulation of gene expression on ecDNA depends on unique transcriptional regulators. I will use a combination of flow cytometry, CRISPR screening, single-cell transcriptomics by Perturb-seq, and bulk epigenomic profiling to identify unique transcriptional regulators of oncogenes amplified on ecDNA in a panel of cancer cell lines and patient-derived neurospheres. In the K00 phase, I will elucidate the mechanism of retention of ecDNA-harbored oncogenes in cancer cells. ecDNA lacks centromeres and is uncoupled from the mitotic spindle during cell division. Therefore, ecDNA segregates randomly between daughter nuclei. Nevertheless, ecDNA is retained, and even selected for, during tumorigenesis, suggesting an uncharacterized mechanism for ecDNA segregation. Surprisingly, live cell imaging during mitosis showed strong colocalization of ecDNA with chromosomes, suggesting that ecDNAs may be able to co-opt chromosomal segregation mechanisms despite lacking centromeres. I hypothesize that specific genetic elements on ecDNA enable hitchhiking onto chromosomes in order to partition into daughter nuclei during cancer cell division. I propose to identify DNA elements and protein mediators that enable retention of ecDNAs using a shotgun episome-based genetic screen, mitotic chromatin conformation capture, proximity proteomics, CRISPR screening and integration of epigenomic datasets. Together, elucidating the mechanisms of extrachromosomal oncogene upregulation and amplicon retention in cancer cells will reveal potential therapeutic oppor...