An integrated model for termination of RNA polymerase III transcription

Abstract

RNA polymerase III (RNAPIII) synthesizes essential and abundant non-coding RNAs such as tRNAs. Controlling RNAPIII span of activity by accurate and efficient termination is a challenging necessity to ensure robust gene expression and to prevent conflicts with other DNA-associated machineries. The mechanism of RNAPIII termination is believed to be simpler than that of other eukaryotic RNA polymerases, solely relying on the recognition of a T-tract in the non-template strand. Here we combine high-resolution genome-wide analyses and in vitro transcription termination assays to revisit the mechanism of RNAPIII transcription termination in budding yeast. We show that T-tracts are necessary but not always sufficient for termination and that secondary structures of the nascent RNAs are important auxiliary cis-acting elements. Moreover, we show that the helicase Sen1 plays a key role in a fail-safe termination pathway. Our results provide a comprehensive model illustrating how multiple mechanisms cooperate to ensure efficient RNAPIII transcription termination.

Architecture of the yeast Pol III pre-termination complex and pausing mechanism on poly-dT termination signals

“…Metagene analysis of Pol III distribution around tRNA genes (Figure 4B) indicates that a considerable fraction of wild type Pol IIIs reads through the primary terminator (i.e. the first T-tract after the 3' end of the gene), in agreement with previous studies (Turowski et al, 2016;Xie et al, 2021). Importantly, we observed a significant increase in Pol III signal downstream of the primary terminator in the three mutants, indicating global transcription termination defects (Figure 4B and individual examples in Figure 4C).…”

“…Instead, read-through (RT) products could be observed that corresponded to transcripts that were released once Pol III reached the end of the dsDNA. While our promoter-dependent transcription assay does not allow to distinguish between Pol III pausing and Pol III termination, a similarly prepared batch of purified Pol III could recently been shown to efficiently terminate on a 5-dT termination sequence on immobilized DNA templates (Xie et al, 2021). Together, these results confirmed that highly-purified Pol III is capable to pause, and presumably terminate, on its own in vitro on a poly-dT termination signal in a promoter-dependent transcription setup.…”

“…Yeast genome was obtained from Saccharomyces Genome Database (S. cerevisiae genome version R64-2-1). tRNA coordinate files (from 5' end to the end of the primary terminator) were obtained from (Xie et al, 2021). For each tRNA gene, the primary terminator was defined as the 1st T-tract after the 3' end of the mature tRNA.…”

“…A more recent study, in which a Sen1 mutant that failed to interact with Pol III in S. cerevisiae was investigated, also revealed that disrupting the Sen1-Pol III interaction causes increased termination RT of Pol III (Xie et al, 2021). However, the later study concluded that S. cerevisiae Sen1 more likely functions as a “fail-safe” termination factor that acts on secondary poly-dT termination sites residing downstream of the first termination signal (Xie et al, 2021). In that scenario, Sen1 would trigger the release of Pol III that failed to terminate on its primary termination site and got stuck on downstream poly-dT signals.…”

Sen1 is a key regulator of transcription-driven conflicts

“…The interactions of Sen1 with the replisome and RNAPIII are mutually exclusive, which implicates the existence of two distinct Sen1-containing complexes presumably with different functional roles. Interestingly, however, both interactions are impaired by the sen1-3 mutation, which likely alters a shared region of interaction (Xie et al, 2021). These findings indicate that the sen1-3 mutant allows untangling the function of Sen1 in NNS-dependent termination from its functions at the replisome and RNAPIII transcription.…”