RNA polymerase II subunit Rpb9 is important for transcriptional fidelity in vivo

Abstract

The fidelity of yeast RNA polymerase II (Pol II) was assessed in vivo with an assay in which errors in transcription of can1-100, a nonsense allele of CAN1, result in enhanced sensitivity to the toxic arginine analog canavanine. The Pol II accessory factor TFIIS has been proposed to play a role in transcript editing by stimulating the intrinsic nuclease activity of the RNA polymerase. However, deletion of DST1, the gene encoding the yeast homolog of TFIIS, had only a small effect on transcriptional fidelity, as determined by this assay. In contrast, strains containing a deletion of RPB9, which encodes a small core subunit of Pol II, were found to engage in error-prone transcription. rpb9Delta strains also had increased steady-state levels of can1-100 mRNA, consistent with transcriptional errors that decrease the normal sensitivity of the can1-100 transcript to nonsense-mediated decay, a pathway that degrades mRNAs with premature stop codons. Sequences of cDNAs from rpb9Delta strains confirmed a significantly increased occurrence of transcriptional substitutions and insertions. These results suggest that Rpb9 plays an important role in maintaining transcriptional fidelity, whereas TFIIS may serve a different primary purpose.

Fig. 1.

Design of assay for transcriptional accuracy in vivo. Rare transcriptional errors by an error-prone Pol II correct the premature stop codon and/or stabilize the can1–100 mRNA to NMD, allowing increased production of Can1 protein, which is detected as enhanced sensitivity to canavanine.

Fig. 2.

Contributions of TFIIS, Rpb9, and the NMD pathway to canavanine sensitivity. (a) Mid-log phase cultures of the various yeast strains were serially diluted to equal concentrations and spotted onto synthetic complete medium lacking arginine and containing the indicated concentrations of canavanine. Growth was evaluated after 2 days. (b) Same as in a for the indicated strains, except that growth was evaluated after 4 days.

Fig. 3.

Northern blot of CAN1 transcripts. Total RNA was isolated from the indicated strains and processed as described in Materials and Methods.

Fig. 4.

Effect of absence of the NMD pathway on growth of yeast lacking TFIIS or Rpb9. Mid-log phase cultures of the indicated yeast strains were serially diluted and spotted onto synthetic complete medium. Growth was evaluated after 2 days.