PROJECT SUMMARY/ABSTRACT Condensate models of transcription in which dynamic, membraneless bodies in the nucleus concentrate large numbers of transcriptional components have recently been described, providing a new framework to solve mysteries in transcription. Early studies of RNA in complex coacervates revealed that RNA can have positive and negative effects on proteinaceous condensates and may operate to control gene expression. Preliminary data confirms the prediction that RNAs at low levels, like those present during early transcription initiation, will stimulate condensates with the coactivator Mediator, and at higher levels, dissolve them. Transcriptional dysregulation is a hallmark of cancer, but how oncogenic transcription maintains high levels of expression remains unresolved. It is possible that the RNA-mediated behavior of condensates is altered in cancerous cells, and perhaps key oncogenes like MYC have a function in this behavior that has not been described before. Thus, this proposal will test the hypothesis that excess levels of oncogenic proteins compensate for higher RNA production by making transcriptional condensates more resistant to dissolution. The drivers of oncogenesis and metastasis may have a physicochemical effect on transcriptional dysregulation that was not anticipated before the advent of the condensate model of transcription. To test this hypothesis, the proposal includes aims that will characterize oncogenic condensates and evaluate the ability of oncogenic proteins to stabilize them in cancer cells. As part of this proposal, relevant training will include strengthening and learning new skills in super-resolution microscopy, cellular engineering, protein biochemistry, and cancer biology. The results from this proposal will provide direct evidence that biophysical phenomenon, mediated by the product of transcription, regulate gene expression in normal and malignant states. They will provide potential new routes to develop therapies against hard-to-target transcription factors and transcriptional dependencies in cancer.