Hybrid push: A mechanistic model for initial transcription common to all RNA polymerases.

摘要

The well-studied RNA polymerases fall into two distinct and apparently unrelated classes: the "single subunit" family, represented by bacteriophage T7, mitochondrial, and chloroplast RNA polymerases and the multi-subunit bacterial and eukaryotic enzyme family. All are characterized by a relatively unstable initial phase, exploited as a point of regulation at some promoters, and by a large structural rearrangement associated with the transition to stable elongation. Our previous studies with T7 RNA polymerase have established a specific model for the coupling of hybrid growth to promoter release and to abortive cycling.;We now extend these studies to the more complex and structurally unrelated multi-subunit system to show that the fundamental energetics of abortive cycling is common in all RNA polymerases. In my current research, I have studied the role of the RNA-DNA hybrid, sigma factor, and transcription bubble collapse in abortive cycling and backtracking in initially transcribing complexes of E. coli RNA polymerase. To understand the role of the collapse of the transcription bubble in abortive initiation, I have engineered nicks, gaps and mismatch sequences in DNA templates. As in our earlier studies in the T7 system, my new results show that collapse of the downstream end of the transcription bubble decreases the stability of a halted complex. This increases overall turnover, but not necessarily the probability of abortive dissociation during read-through transcription. The results demonstrate clearly that DNA "scrunching" is not the primary driving force of abortive instability.;In E. coli RNA polymerase, the σ3.2 linker initially occupies the RNA exit channel, presenting an impediment to the nascent RNA-DNA hybrid, and displacement of the σ 3.2 linker is thought to weaken promoter contacts, allowing promoter escape. Exploiting mutations in the σ3.2 linker, GreB as a probe of backtracking, and novel promoter constructs that allow initiation without sigma, we present evidence to support a model, similar to one recently proposed for the single subunit enzymes, in which the pushing of the hybrid against the σ3.2 linker destabilizes the hybrid (note the analog of the σ3.2 linker in eukaryotes is the TFIIB B-finger). Thus abortive cycling is a necessary consequence of the common evolutionary need to couple timed promoter release to the energetics of phosphoryl transfer.