Deciphering transcriptional dynamics in vivo by counting nascent RNA molecules

Abstract

Transcription of genes is the focus of most forms of regulation of gene expression. Even though careful biochemical experimentation has revealed the molecular mechanisms of transcription initiation for a number of different promoters in vitro, the dynamics of this process in cells is still poorly understood. One approach has been to measure the transcriptional output (fluorescently labeled messenger RNAs or proteins) from single cells in a genetically identical population, which could then be compared to predictions from models that incorporate different molecular mechanisms of transcription initiation. However, this approach suffers from the problem, that processes downstream from transcription can affect the measured output and therefore mask the signature of stochastic transcription initiation on the cell-to-cell variability of the transcriptional outputs. Here we show theoretically that measurements of the cell-to-cell variability in the number of nascent RNAs provide a more direct test of the mechanism of transcription initiation. We derive exact expressions for the first two moments of the distribution of nascent RNA molecules and apply our theory to published data for a collection of constitutively expressed yeast genes. We find that the measured nascent RNA distributions are inconsistent with transcription initiation proceeding via one rate-limiting step, which has been generally inferred from measurements of cytoplasmic messenger RNA. Instead, we propose a two-step mechanism of initiation, which is consistent with the available data. These findings for the yeast promoters highlight the utility of our theory for deciphering transcriptional dynamics in vivo from experiments that count nascent RNA molecules in single cells.