The Kinetics of RNA Flow Across Subcellular Compartments

Abstract

Throughout their lifetimes, eukaryotic RNAs traverse across subcellular compartments. The rates of RNA flow across the cell are impacted by regulatory processes, including splicing, export, and ribosome loading. In turn, these processes impact the stability and fate of the transcripts upon which they act. RNA flow rates ultimately determine the dynamic pool of transcripts available for translation. However, current genome-wide techniques that measure RNA turnover quantify the time between synthesis and degradation but lack spatial information about where transcripts reside in the cell during this time, obscuring the subcellular kinetics of RNA flow. To measure rates of RNA flow genome-wide, I first developed subcellular TimeLapse sequencing, a method that tracks the age of RNAs as they shuttle across subcellular compartments. I used this technique to quantify RNA half-lives in mouse NIH-3T3 and human K562 cells at subcellular resolution by fitting a kinetic model to these data in a Bayesian framework. Using this approach, I measured the rates at which transcripts are released from chromatin, exported out of the nucleus, loaded onto polysomes, and degraded in both the nucleus and cytoplasm. All RNA flow rates displayed substantial variability genome-wide, and transcripts from genes with related functions flowed across subcellular compartments with similar kinetics. I then identified candidate determinants of RNA flow by analyzing the variability in kinetics between different transcripts. The targets of RNA binding proteins and transcription factors experienced distinct RNA flow rates. By verifying these associations, I report that two RNA binding proteins, DDX3X and PABPC4, regulate the nuclear export of their target transcripts. I next measured RNA poly(A) tail lengths with subcellular resolution and show that transcripts residing on chromatin for longer times had extended tails, while the reverse trend was observed for mRNAs in the cytoplasm. Finally, additional genetic and molecular features that underlie RNA flow rates were identified using a machine learning model. Collectively, this work characterizes the life cycle of mammalian mRNAs, revealing the many lives of RNA transcripts and the molecular features underlying their fates.