Kinetic competition between RNA Polymerase II and Sen1-dependent transcription termination

Abstract

The essential helicase-like protein Sen1 mediates termination of RNA Polymerase II (Pol II) transcription at snoRNAs and other noncoding RNAs in yeast. A mutation in the Pol II subunit Rpb1 that increases the elongation rate increases read-through transcription at Sen1-mediated terminators. Termination and growth defects in sen1 mutant cells are partially suppressed by a slowly transcribing Pol II mutant and are exacerbated by a faster-transcribing Pol II mutant. Deletion of the nuclear exosome subunit Rrp6 allows visualization of noncoding RNA intermediates that are terminated but not yet processed. Sen1 mutants or faster-transcribing Pol II increase the average lengths of preprocessed snoRNA, CUT, and SUT transcripts, while slowed Pol II transcription produces shorter transcripts. These connections between transcription rate and Sen1 activity support a model whereby kinetic competition between elongating Pol II and Sen1 helicase establishes the temporal and spatial window for early Pol II termination.

Figure 1. The Fast Pol II rpb1-E1103G Mutant Increases Readthrough at snoRNA Genes

(A) ChIP assay to measure in vivo elongation rate. Graphs show occupancy of Pol II subunit Rpb3 in WT (YSB2716) and rpb1-E1103G (YSB2717) strains containing a GAL1-YLR454 construct, before (0 min) and after glucose addition (2, 4, 8 min) to shut off GAL1 promoter-driven transcription. Primer locations are indicated in gene schematic. Rpb3 occupancy values are expressed relative to the 0 min timepoint (pre-glucose) after normalization to a non-transcribed telomeric region. Error bars represent standard error from two independent experiments. (B) Quantitative Reverse Transcription PCR analysis (RT-qPCR) of SNR13 and SNR33 readthrough transcript levels in WT (GRY3020), rpb1-N488D (GRY3027), and rpb1-E1103G (GRY3028) strains grown at 30°C relative to ADH1 control transcripts. Error bars represent standard error from three independent experiments.

Figure 2. Termination Window Mapping of Pre-Processed SNR33 Transcripts in WT, Slow, and Fast RNA Pol II strains

(A) Schematic illustrating Termination Window Mapping analysis. Pre-processed terminated snoRNA transcripts (pre-snoRNAs) and cryptic unstable transcripts (CUTs) accumulate in rrp6Δ cells and offer a visual readout of the termination window. (B) Northern blot analysis of SNR33 from RPB1 (WT, YSB2697), slow rpb1-N488D (YSB2698), and fast rpb1-E1103G (YSB2699) strains also containing rrp6Δ. Cells were grown at 30°C in the presence or absence of 6-AU (75 μg/ml). Relative densitometry comparisons of pre-snR33 transcripts from indicated lanes are shown below. (C) Northern blot analysis of SNR33 was performed as in part B, with the addition of isogenic Sen1 (YSB2859) and sen1-1 (YSB2860) strains containing rrp6Δ grown at 30°C. Before electrophoresis, RNA was treated with RNase H in the absence or presence of oligo(dT). R1-R5; readthrough/extended transcripts that increase in the sen1-1 background. (D) Northern blot analysis of SNR33 from RRP6 strains carrying RPB1 (GRY3020), slow rpb1-N488D (GRY3027), and fast rpb1-E1103G (GRY3028) alleles. (E) Northern blot analysis of SNR33 from an rrp6Δ (YSB2697) strain grown at 20°C, 30°C, and 37°C. See also Figure S1.

Figure 3. Sen1 Mutations Have Opposite Effects on Growth of rpb1-N488D and rpb1-E1103G Strains

(A) Schematic of Sen1 C-terminal helicase domains (residues 1357-1822) with locations of sen1-E1597K and sen1-1 (sen1-G1747D) mutations. (B) Cell growth curves of WT (YSB2700), sen1-1 (YSB2704), sen1-E1597K (YSB2707) strains; (C) rpb1-N488D (YSB2701), sen1-1/rpb1-N488D (YSB2705), sen1-E1597K/rpb1-N488D (YSB2708) strains; (D) rpb1-E1103G (YSB2702), sen1-1/rpb1-E1103G (YSB2706), sen1-E1597K/rpb1-E1103G (YSB2709) strains grown in YPD media at 30°C as measured by optical density at 600 nm (OD₆₀₀₎ from a starting OD of ~0.05. Values were plotted on a log2 scale and the calculated doubling time (Td) of each strain is shown in parentheses. (E) Spot growth assay of Rpb1 and Sen1 single and double mutant strains indicated in (B–D) grown on YPD media plates for 2 days at 30°C. (F) Spot growth assay of strains indicated in (B–E) grown on 2% galactose/SC-uracil/x-gal media plates for 4 days at 30°C.

Figure 4. Slow and Fast Pol II Mutants Differentially Affect snoRNA Termination in sen1 Mutant Strains

(A) Northern blot analysis of SNR13 with RNA from WT (YSB2700), rpb1-N488D (YSB2701), rpb1-E1103G (YSB2702), sen1-1 (YSB2704), sen1-1/rpb1-N488D (YSB2705), sen1-1/rpb1-E1103G (YSB2706), sen1-E1597K (YSB2707), sen1-E1597K/rpb1-N488D (YSB2708), and sen1-E1597K/rpb1-E1103G (YSB2709) strains grown at 30°C. The graphs show the relative levels of SNR13 readthrough transcripts (upper graph) and truncated snR13 transcripts (lower graph) relative to Pol III-transcribed SCR1 transcripts in arbitrary units (a.u.). Error bars represent standard error from two independent experiments. (B) Northern blot analysis of SNR3 with RNA from indicated strains grown at 30°C. R1–R5; readthrough/extended transcripts that increase in sen1-1 and sen1-E1597K backgrounds, truncated; truncated snR3 transcript. (C) Northern blot analysis of SNR31 from indicated strains grown at 30°C. R1–R4; readthrough/ extended transcripts that increase in sen1-1 and sen1-E1597K backgrounds. See also Figure S2.

Figure 5. Termination Window Mapping of Pre-Processed SNR33 and Other Non-coding Transcripts in WT, Slow, and Fast RNA Pol II strains

(A) Northern blot analysis of SNR33 with RNA from WT (YSB2697), rpb1-N488D (YSB2698), rpb1-E1103G (YSB2699), sen1-1 (YSB2711), sen1-1/rpb1-N488D (YSB2712), sen1-1/rpb1-E1103G (YSB2713), sen1-E1597K (YSB2714), and sen1-E1597K/rpb1-N488D (YSB2715), strains containing rrp6Δ grown at 30°C and 20°C. R1-R5; readthrough/extended transcripts that increase in sen1-1 and sen1-E1597K backgrounds. (B, C, D) Relative densitometry comparisons of pre-snR33 transcripts from indicated lanes in part A. (E) Northern blot analysis of CUT060 in indicated rrp6Δ strains grown at 30°C and 20°C. Shown here is the part of the gel corresponding to the CUT060 readthrough transcript (R.T.); the full blot including the terminated but un-degraded CUT transcripts is in Figure S3D. ACT1 mRNA serves as a loading control and the relative amount of readthrough is quantitated below each lane in arbitrary densitometry units (a.u.). (F) Northern blot analysis of SUT447 in indicated rrp6Δ strains grown at 30°C and 20°C. R1, R2; readthrough/extended transcripts that increase in sen1-1 and sen1-E1597K backgrounds. eSUT447 is an extended SUT transcript (Marquardt et al., 2011). See also Figure S3.

Figure 6. Model of Kinetic Competition Between Pol II and Sen1

(A) A termination window, likely prescribed by Nrd1 and Nab3 RNA binding sites in the transcript and CTD phosphorylation patterns, in which normal Pol II elongation coupled with Sen1 bound to Pol II CTD and the RNA transcript leads to efficient termination (Chinchilla et al., 2012). (B) Slow Pol II elongation shows window contraction with increased earlier termination events. (C) Faster elongation in rpb1-E1103G leads to window expansion with more downstream termination events within the window and increased readthrough. (D) Cells expressing both faster Pol II and mutations in Sen1 exhibit higher levels of window expansion and readthrough than single mutants and synergistic growth defects, indicative of a greater desynchronization between fast moving Pol II and less functional Sen1 molecules. (E) Cells expressing both the slower Pol II (rpb1-N488D) and sen1 mutations have a less extended termination window and partial suppression of sen1 mutant phenotypes due to the slower elongation rate making it easier for the partially defective Sen1 to catch the polymerase.