Yeast Nrd1, Nab3, and Sen1 transcriptome-wide binding maps suggest multiple roles in post-transcriptional RNA processing

Abstract

RNA polymerase II transcribes both coding and noncoding genes, and termination of these different classes of transcripts is facilitated by different sets of termination factors. Pre-mRNAs are terminated through a process that is coupled to the cleavage/polyadenylation machinery, and noncoding RNAs in the yeast Saccharomyces cerevisiae are terminated through a pathway directed by the RNA-binding proteins Nrd1, Nab3, and the RNA helicase Sen1. We have used an in vivo cross-linking approach to map the binding sites of components of the yeast non-poly(A) termination pathway. We show here that Nrd1, Nab3, and Sen1 bind to a number of noncoding RNAs in an unexpected manner. Sen1 shows a preference for H/ACA over box C/D snoRNAs. Nrd1, which binds to snoRNA terminators, also binds to the upstream region of some snoRNA transcripts and to snoRNAs embedded in introns. We present results showing that several RNAs, including the telomerase RNA TLC1, require Nrd1 for proper processing. Binding of Nrd1 to transcripts from tRNA genes is another unexpected observation. We also observe RNA polymerase II binding to transcripts from RNA polymerase III genes, indicating a possible role for the Nrd1 pathway in surveillance of transcripts synthesized by the wrong polymerase. The binding targets of Nrd1 pathway components change in the absence of glucose, with Nrd1 and Nab3 showing a preference for binding to sites in the mature snoRNA and tRNAs. This suggests a novel role for Nrd1 and Nab3 in destruction of ncRNAs in response to nutrient limitation.

FIGURE 1.

Euler plots of cross-linked regions. (Left) Overlap between Nrd1, Nab3, and Sen1 cross-linked regions. (Right) Overlap between Rpb2, Sen1, and Nab3 cross-linked regions. The 100 most frequent T-to-C cross-linked sites in each data set were selected for this analysis. A region extending 20 nt in each direction from the cross-linked site was mapped to the genome, and values in the plot represent the number of overlapping or nonoverlapping nucleotides.

FIGURE 2.

In vivo Nrd1, Nab3, and Rpb2 cross-linking to snR13 transcripts. (A) Cross-linking to transcripts from a 1400-nt region of chromosome 4 encoding the SNR13 and TRS31 genes. (B) Higher-resolution mapping of cross-linked reads to the SNR13 downstream region. The number of reads per 10⁷ reads is indicated on the y-axis. The position on the x-axis is the center of the read for the PAR-CLIP experiment but is the 3′ end of the read for oligo(A)-containing reads. Asterisks designate the positions of naturally occurring genomic oligo(A) sequences that were not filtered in our bioinformatic analysis. (C) Sequences of the major cross-linked regions. The arrows in the left panel correspond to the most frequent T-to-C transitions in the Nrd1 (blue) or Nab3 (red) data sets. In the right panel, lines represent the exact position of oligo(A) addition. The size of the line corresponds to the frequency of the reads at that position.

FIGURE 3.

In vivo Nrd1, Nab3, Sen1, and Rpb2 cross-linking to box H/ACA and box C/D snoRNA transcripts. (A,B) Cross-linking to snR11 and snR5 box H/ACA transcripts. (C,D) Cross-linking to snR45 and snR40 box C/D snoRNA transcripts. The y-axis is the number of sequence reads per 10⁷ reads. The x-axis scale is given as the start and end of the snoRNA.

FIGURE 4.

In vivo Nrd1 and Nab3 cross-linking to snoRNA transcripts associated with Rnt1. (A) Nrd1 cross-linking to transcripts from the SNR50 gene. (B) Higher-resolution map showing the positions of T-to-C transitions and oligo(A) sites relative to the Rnt1 cleavage site. (C) Nrd1 and Nab3 cross-linking to snR54 transcripts in the intron of IMD4. (D) Nrd1 cross-linking to sequences in the polycistronic transcript that is processed to yield seven snoRNAs.

FIGURE 5.

In vivo Nrd1 cross-linking to snoRNA transcripts in cells growing in the presence or absence of glucose. (A,B) Cross-linking of Nrd1 in vivo to snR3 and snR11 transcripts in the presence (blue) or absence (red) of glucose. (C) Frequency of Nrd1 cross-linked reads from the SNR11 gene, showing the presence of oligo(A) reads within the mature snR11 sequence. (D) Plot of all reads on the 77 snoRNAs on the genomic plus strand. The y-axis is median frequency over a 25-nt window. The x-axis is plotted relative to the 3′ end of the snoRNA.

FIGURE 6.

In vivo cross-linking to snRNA and telomerase RNA. (A) Nrd1 cross-linking to the 3′ end of U1 RNA. (B) Nrd1 cross-linking to transcripts from the 5′ end of U2 RNA. (C) Nrd1 and Nab3 cross-linking to transcripts from the 3′ end of the TLC1 gene encoding telomerase RNA. Lines below the gene map indicate the positions of PCR amplicons used to quantify TLC1 transcripts. (D) High-resolution cross-linking map showing the position of the mature 3′ end (arrow), the Sm binding site (box), the precise positions of cross-linked U residues (Nrd1 T to C), and the positions where oligo(A) is added (Nrd1 polyA). (E) Quantitative RT-PCR of TLC1 RNA derived from wild-type cells or nrd1-102 cells grown at the nonpermissive temperature. The relative abundance of each target transcript is normalized to the abundance of ACT1 in the same sample.

FIGURE 7.

In vivo Nrd1 and Sen1 cross-linking downstream from protein-coding genes. The distribution of Nrd1 cross-linked RNA is shown in blue and Sen1 in green above the gene map. The position of poly(A) sites derived from RNA sequencing (Ozsolak et al. 2010) is shown in red. (A) Nrd1 and Sen1 cross-linking downstream from the TIS11 (CTH2) mRNA 3′ end. (B,C) Binding of Nrd1 and Sen1 in the 3′ UTR of the AHP1 and URA1 genes. (D) Binding of Nrd1 in the SNA3 3′ UTR. (E) Map of binding sites in the 3′ UTR. (F) Quantitative RT-PCR of RNA derived from wild-type or nrd1-102 cells grown at the nonpermissive temperature. The amplicon is located downstream from the SNA3 polyA site as indicated in D, and its relative abundance was normalized to the abundance of ACT1 in the same sample.

FIGURE 8.

In vivo Nrd1, Nab3, Sen1, and Rpb2 cross-linking to tRNA transcripts. (A,B) Cross-linking to tRNA Pro and Phe, respectively. Sequences below each map show the positions of Nrd1 (blue) and Nab3 (red) cross-links. Boxed sequences represent the mature tRNA. Bold sequence indicates a conserved Nab3 binding site upstream of tRNA Phe.

FIGURE 9.

In vivo cross-linking of Nrd1 to transcripts from Pol III genes in the presence and absence of glucose. (A) Cross-linking to tRNA genes. (B) Cross-linking to the transcript of the RPR1 gene encoding the RNase P RNA.