Rat1p maintains RNA polymerase II CTD phosphorylation balance.

In S. cerevisiae, the 5'-3' exonuclease Rat1p partakes in transcription termination. Although Rat1p-mediated RNA degradation has been suggested to play a role for this activity, the exact mechanisms by which Rat1p helps release RNA polymerase II (RNAPII) from the DNA template are poorly understood. Here we describe a function of Rat1p in regulating phosphorylation levels of the C-terminal domain (CTD) of the largest RNAPII subunit, Rpb1p, during transcription elongation. The rat1-1 mutant exhibits highly elevated levels of CTD phosphorylation as well as RNAPII distribution and transcription termination defects. These phenotypes are all rescued by overexpression of the CTD phosphatase Fcp1p, suggesting a functional relationship between the absence of Rat1p activity, elevated CTD phosphorylation, and transcription defects. We also demonstrate that rat1-1 cells display increased RNAPII transcription kinetics, a feature that may contribute to the cellular phenotypes of the mutant. Consistently, the rat1-1 allele is synthetic lethal with the rpb1-E1103G mutation, causing increased RNAPII speed, and is suppressed by the rpb2-10 mutation, causing slowed transcription. Thus, Rat1p plays more complex roles in controlling transcription than previously thought.

INTRODUCTION

The transcription cycle of RNAPII comprises three main phases—initiation, elongation, and termination, each of which is associated with specific phosphorylation patterns of the C-terminal domain (CTD) of Rpb1p. The “CTD code” proposes that differential CTD phosphorylation regulates the recruitment of transcription and mRNA processing factors to the RNAPII machinery and is best understood for phosphorylation of serine2 (Ser2P) and serine5 (Ser5P) residues. During initiation, the CTD is hypophosphorylated, which allows binding of transcription initiation factors and assembly of the preinitiation complex (PIC). Promoter escape and early transcription elongation occur concomitantly with Ser5 phosphorylation, which facilitates pre-mRNA 5′-end capping and the stable transition from initiation to elongation (promoter release) (Rasmussen and Lis 1993; Komarnitsky et al. 2000; Schroeder et al. 2000; Kim et al. 2010). Ser5P is gradually removed during transcription elongation, due to the actions of the CTD phosphatases Ssu72p and Rtr1p and is only completely removed at the polyadenylation (pA) site (Kim et al. 2010; Bataille et al. 2012). Ser2P, which is absent during transcription initiation, gradually increases along transcription units and remains high until transcription termination. Ser2P plays a central role in coupling pre-mRNA processing to transcription elongation, and substitution of CTD-Ser2 residues to alanine impairs splicing and 3′ end processing in mammalian cells (Gu et al. 2013). A major Ser2 kinase in Saccharomyces cerevisiae is Ctk1p, whose inactivation leads to strongly decreased Ser2P levels, a severely impaired recruitment of 3′end processing factors and an accumulation of RNAPII at the 3′ends of genes with weak pA sites (Ahn et al. 2004; Kim et al. 2010). Ser2P is removed during or after transcription termination by the CTD-phosphatase Fcp1p to prepare RNAPII for new initiation events (Kobor et al. 1999; Cho et al. 2001; Mandal et al. 2002). In addition, Fcp1p dampens Ser2P levels during normal transcription elongation as fcp1 mutants display increased Ser2P levels inside genes (Archambault et al. 1997; Kobor et al. 1999; Cho et al. 2001; Ghosh et al. 2008).

Apart from coupling transcription with RNA processing, Ser2P may also influence the speed by which RNAPII proceeds through chromatin. For example, mammalian RNAPII, harboring Ser2 to alanine substitutions, shows decreased elongation rates (Gu et al. 2013). Furthermore, in human cells the speed of RNAPII increases toward the gene 3′ end, which could be related to increasing Ser2P (Danko et al. 2013). Similarly in S. cerevisiae, mutation of the multipurpose transcription factor Sub1p leads to altered CTD phosphorylation levels and cotranscriptional recruitment of CTD kinases/phosphatases, which is paralleled by decreased transcription elongation rates (Calvo and Manley 2005; Garcia et al. 2010). Thus, although mechanistic details need to be elucidated, a link between CTD phosphorylation and transcription elongation appears to exist.

The mechanistic details underlying transcription termination are not well worked out. A current model postulates that Rat1p targets the 5′ end of the nascent downstream RNA fragment appearing after pA site cleavage. This causes, in an unknown fashion, the dissociation of RNAPII from the DNA template (Kuehner et al. 2011). RNAPII transcription stalling at or near the pA site may help in this process. CTD phosphorylation also impacts transcription termination. Most importantly, Ser2P facilitates binding of the pre-mRNA 3′end processing-factor and transcription termination-factor Pcf11p and the Rat1p-cofactor Rtt103p (Lunde et al. 2011). The cotranscriptional recruitment of the Ser2P kinase Ctk1p ceases around the pA site, suggesting that Ser2P is lost in the downstream region (Mayer et al. 2010). However, to what extent Ser2P, or its removal, is functionally important for transcription termination in vivo is unclear. In addition to its role in transcription termination at gene 3′ ends, Rat1p also aids in terminating transcription events inside the gene body, provided that they are engaged with inefficiently capped pre-mRNA (Jimeno-Gonzalez et al. 2010).

The majority of the in vivo studies of Rat1p function rely on the thermosensitive rat1-1 mutant, the activity of which is poorly characterized. It contains a single mutation (Y657C) situated outside of the enzyme's exonuclease domain, and since the Rat1-1p protein is stable at the restrictive temperature, it may retain some exonuclease activity. Even so, growth and transcription termination phenotypes of rat1-1 cells are not rescued by the coexpression of the catalytically inactive rat1-D325A point mutant (Kim et al. 2004; Luo et al. 2006; Mayer et al. 2010). This has been taken as evidence that 5′-3′ exonucleolysis is essential for Rat1p function and that this activity is, at least partly, compromised in the rat1-1 background.

Here, we find that overexpression of Fcp1p suppresses the thermosensitivity of rat1-1. Surprisingly, rat1-1 cells not only exhibit a transcription termination defect but also display highly elevated CTD Ser2P levels as well as decreased RNAPII occupancy within genes. These phenotypes are all partially restored by Fcp1p overexpression. High Ser2P levels in rat1-1 cells are not due to decreased Fcp1p levels around transcribed chromatin but rather to an increased cotranscriptional recruitment of Ctk1p. Finally, rat1-1 cells show increased transcription elongation rates. The data suggest that Rat1p plays a CTD-modulatory role during transcription elongation, which needs to be considered when interpreting molecular phenotypes of the rat1-1 mutant.

RESULTS AND DISCUSSION

Increased phosphorylation of the Rpb1p CTD in rat1-1 cells

To explore mechanisms used by Rat1p to promote transcription termination, we screened for multicopy suppressors of the rat1-1 mutant at its nonpermissive temperature of 34°C (see Materials and Methods). In addition to the RAT1 gene, we found that overexpression of the CTD Ser2P phosphatase, Fcp1p, was able to rescue rat1-1 thermosensitivity at 34°C (Fig. 1A). This suggested that reduction of CTD Ser2P phosphorylation levels might help overcome growth-limiting defects of the rat1-1 mutant. We therefore analyzed the global CTD phosphorylation status of Rpb1p in the rat1-1 background with or without Fcp1p overexpression. Western blotting analysis of whole-cell extracts using antibodies recognizing CTD-Ser2P, -Ser5P, or -Ser7P residues demonstrated elevated Ser2P and Ser5P levels in the rat1-1 strain, whereas Ser7P levels were not affected (Fig. 1B). Both Ser2P and Ser5P levels were decreased by Fcp1p overexpression.

FIGURE 1.

Excess Fcp1p rescues growth deficiency of rat1-1 cells. (A) Tenfold dilution series of wt and rat1-1 strains transformed with pRS425 or pRS425-FCP1 plasmids as indicated and spotted onto AA-Leu plates. Strains were grown for 3 d at 30°C or 34°C. (B) Rpb1p CTD phosphorylation status in whole-cell extracts measured by Western blotting analysis. Strains from A were grown at 30°C in AA-Leu medium to an OD600 of 0.6, followed by a subsequent incubation for 1 h at 34°C, before being harvested for protein extraction. Membranes were probed with H14, H5, 3E8, 3E10, 4E12, Rpb1p (Y-80), or Fcp1p antibodies as indicated.

Fcp1p overexpression reduces elevated CTD-Ser2P levels at active chromatin in rat1-1 cells and partly suppresses related RNAPII distribution phenotypes

We next tested whether increased CTD phosphorylation in rat1-1 cells was also manifested during active transcription and, if so, which impact excess Fcp1p might have. For this analysis we utilized the Ser2P antibody H5, whose epitope is increased in rat1-1 cells and dampened to roughly wt levels upon Fcp1p overexpression (Fig. 1B). Notably, H5 has highest affinity for a CTD Ser2P epitope containing neighboring Ser5P (Chapman et al. 2007). Chromatin immunoprecipitation (ChIP) experiments using H5 as well as Rpb1p (Y-80) antibodies were carried out and immunoprecipitated chromatin was interrogated by PCR amplicons distributed at different positions along the reporter gene GAL1-YLR454W (Fig. 2A; Mason and Struhl 2005). In line with previous reports (Komarnitsky et al. 2000; Bataille et al. 2012), we found that in wt conditions, Ser2P levels were low at the promoter, increased in the coding region, and declined immediately after RNAPII passage of the pA site (Fig. 2B, left panel). In contrast, rat1-1 cells displayed abnormally high levels of Ser2P inside the YLR454W gene body, whereas the presence of this epitope disappeared downstream from the pA site. Such increased Ser2P levels were not simply due to the presence of more chromatin-associated RNAPII in rat1-1 cells as evidenced by ChIP assays conducted with the phosphorylation-insensitive Y-80 antibody (Fig. 2C, left panel). Indeed, the Ser2P/Rpb1p signal ratio increased robustly (up to sixfold) upon Rat1p mutation (Fig. 2D, left panel). A similar phenotype is observed at the PMA1 gene (Supplemental Fig. S1A,B), which, together with the Western blot analysis, suggests that increased CTD phosphorylation is a hallmark of transcription in rat1-1 cells. As predicted, Ser2P levels were reduced upon Fcp1p overexpression, which restored high Ser2P ChIP signals of rat1-1 cells to close to wt levels (Fig. 2B,D, cf. right and left panels).

FIGURE 2.

Transcriptional phenotypes of the rat1-1 mutation are rescued by Fcp1p overexpression. (A) Schematic representation of the GAL1::YLR454W system. Promoter is shaded gray, and the transcript unit is white. The transcription start site (TSS) is depicted by an arrow; the major pA site, by a vertical line. Distances (kb) of ChIP amplicons to the reporter TSS are indicated. (B–E) ChIP experiments at the GAL1::YLR454W reporter gene. Cells containing the indicated plasmids were grown in AA-Ura-Leu/2% raffinose at 30°C to an OD600 of 0.6. The GAL1 promoter was induced with 2% galactose for 3 h at 30°C and rat1-1 inactivated by an additional incubation for 1 h at 34°C before fixation. ChIP values are normalized to the signal at the YLR454W promoter region (P) in wt cells containing empty plasmid. Averages and standard deviations of all ChIP data were calculated from three independent biological experiments, each subjected to triplicate quantitative PCR analysis. Data from cells containing empty plasmids (left panels) and cells containing high-copy FCP1 plasmids (right panels) are averages from the same experiments and only split for better visualization. (B) Rpb1 Ser2P distribution measured by the H5 antibody. (C) Total Rpb1p distribution measured by the Y-80 antibody. (D) Ratio between Ser2P and Rpb1p ChIP values in wt and rat1-1 strains. Positions 8.5 and 9 in the wt strain are set to zero since the Rpb1 ChIP levels used for normalization are at background levels in wt cells. (E) ChIP values from C normalized separately for each strain to reveal differences in local changes of Rpb1p distribution at GAL1::YLR454W. (Left) Promoter escape shows occupancy at fragments 2 and 4 normalized to promoter p. (Middle) pA site stalling shows occupancy at fragments 8.2 normalized to fragment 8. (Right) Read-through is fragments 8.5 and 9 normalized to fragment 8.2 separately for each strain.

Which molecular phenotypes of rat1-1 cells may be associated with elevated CTD-Ser2P levels? Consistent with previous reports (Kim et al. 2004; West et al. 2004; Luo et al. 2006; Jimeno-Gonzalez et al. 2010), the rat1-1 mutation caused RNAPII transcription read-through of the YLR454 pA site (amplicons 8.5 and 9) (Fig. 2C, left panel, and E, right panel). Interestingly, RNAPII accumulated to similar levels near the pA site (8.2) in the wt and rat1-1 strains, despite lower levels within the YLR454W gene body of rat1-1 cells (2 to 8 regions) (Fig. 2C, left panel). A possible interpretation of these data is that RNAPII stalls in the vicinity of the pA site to be targeted by Rat1p-dependent “torpedo” termination (Kim et al. 2004; Luo et al. 2006). Failure to terminate efficiently in the rat1-1 context leads to an enhanced accumulation of stalled RNAPII versus transcribing polymerases within the gene body (Fig. 2E, middle and right panel). In addition, only in mutant cells can RNAPII escape the stalling site to “read-through” and reach downstream regions 8.5 and 9.

A strong rat1-1–dependent phenotype could also be observed at the YLR454W 5′ end; while RNAPII immunoprecipitation (IP) efficiencies at the promoter were similar in the two backgrounds, only ∼40%–50% of RNAPII levels were detected along the YLR454W gene body in rat1-1 compared with wt cells (Fig. 2C, left panel, cf. amplicon p with 2, 4, 6, and 8; Fig. 2E, left panel). A similar phenotype was described for the PMA1 gene in previous reports (Kim et al. 2004; Kawauchi et al. 2008). This effect may be explained by the faster transcription elongation rates in the rat1-1 mutant (see below). Importantly, Fcp1p overexpression partially suppressed both the decreased levels of RNAPII in the gene body (ratio between promoter and gene body ChIP amplicons) (Fig. 2E, left panel) and the transcription read-through (ChIP signals at amplicons 8.5 and 9) (Fig. 2E, right panel) phenotype of the rat1-1 mutant, without markedly changing levels of RNAPII “pA site stalling” (8.2) (Fig. 2E, middle panel).

To rule out that the observed effects would be specific for the GAL1-YLR454W reporter, we also analyzed RNAPII distribution at various positions along the PMA1 gene (Supplemental Fig. S1A). Gratifyingly, both decreased polymerase in the gene relative to the promoter (Supplemental Fig. S1C, cf. 5 to middle amplicons), and transcription termination (Supplemental Fig. S1C, cf. 3′ UTR and T1 amplicons) phenotypes of rat1-1 cells were observed at the PMA1 locus. Note that RNAPII pA site-stalling and read-through effects are difficult to distinguish for the PMA1 gene due to the presence of multiple pA sites spread over a 450-bp region (Ozsolak et al. 2010). Even so, a clear suppression of both gene 5′- and 3′-end defects could be measured upon Fcp1p overexpression (Supplemental Fig. S1B).

We previously reported that rat1-1 thermosensitivity is also suppressed by the rpb1-N488D allelic background conferring slow transcription rates (Jimeno-Gonzalez et al. 2010). The molecular mechanism underlying this suppression was not elucidated. Interestingly, introduction of the rpb1-N488D mutation caused decreased Ser2P levels over the YLR454W reporter gene in both the wt and rat1-1 mutant contexts (Supplemental Fig. S2A, cf. left and right panels). This effect was not due to a change in RNAPII levels in the rpb1-N488D background (Supplemental Fig. S2B). Since both suppressors of rat1-1 thermosensitivity, the rpb1-N488D allele and high-copy FCP1, are linked to decreased CTD phosphorylation, this may underlie the growth defect of rat1-1 cells. Indeed, the rpb1-N488D allele also suppressed the decreased RNAPII levels inside the gene and transcription read-through phenotypes of rat1-1 (Supplemental Fig. 2C, left and right panels; Jimeno-Gonzalez et al. 2010). Taken together, these data suggest that abnormally high Ser2P levels contribute to altered transcription elongation and termination in rat1-1 cells. Reducing Ser2P levels by excess Fcp1p or the rpb1-N488D mutant can correct these defects. The increased levels of pA-site-stalled RNAPII in rat1-1 cells (8.2/8) (Fig. 2E; Supplemental S2C, middle panel) are not decreased by FCP1 overexpression or in the rpb1-N488D mutant. Hence, unusually high levels of Ser2P appear to specifically impact a subset of RNAPII that either fails to stall or is capable of resuming transcription after stalling.

Rat1p displaces Ctk1p

Increased CTD phosphorylation in rat1-1 cells and its suppression by extracopy Fcp1p could be explained by defective recruitment of Fcp1p in the absence of wt Rat1p activity. To test this possibility, we conducted YLR454W ChIP using myc-tagged Fcp1p, which yielded specific signals at amplicons within the gene (Supplemental Fig. S3A). Unexpectedly, the relative ChIP efficiency of Fcp1-myc at the YLR454 and PMA1 genes was not decreased but was rather comparable to that of Rpb1p in both wt and rat1-1 cells at all measured positions (Fig. 3A; Supplemental Fig. S3B). Thus, Fcp1p recruitment per transcription event is not affected by the rat1-1 mutation. While it is possible that the catalytic capacity of Fcp1p may be compromised in the rat1-1 background, we deemed it more likely that increased recruitment of a relevant kinase may explain the observed phenotype. We therefore chose to monitor levels of the main Ser2P kinase, Ctk1p, using HA-tagged Ctk1p. Indeed, elevated Ctk1-HA occupancy per RNAPII could be observed in rat1-1 relative to wt cells at the YLR454W and PMA1 genes (Fig. 3B,C; Supplemental Fig. S3B). Hence, high Ser2P levels at chromatin in rat1-1 cells are likely due to increased recruitment of Ctk1p. This suggests a possible competition between cotranscriptional recruitment of Rat1p and Ctk1p. Indeed, Ctk1p becomes excluded from chromatin when RNAPII reaches the pA site where Rat1p levels peak (Fig. 3, cf. 8.2 amplicon in B, left panel, and D; Kim et al. 2004; Luo et al. 2006; Mayer et al. 2010). Notably, Ctk1p disappearance in that region is less efficient in rat1-1 cells; at the 8.2 position, a threefold increase of Ctk1p per RNAPII can be detected compared with the wt strain (Fig. 3C). The molecular basis for the competition between Rat1p and Ctk1p is unclear, but we speculate that they may compete for a common binding site, possibly on the CTD.

FIGURE 3.

Ctk1-HA recruitment at GAL1::YLR454W is increased in rat1-1 cells. Fcp1-myc (A), Ctk1-HA (B), and Rat1-TAP (D) distributions along the GAL1::YLR454W gene as measured by ChIP (solid lines). Parallel Rpb1p ChIP values in the indicated strains are shown as dotted lines. Cells were grown as described in the legend of Figure 2B. ChIP values are relative to the wt signal at the YLR454W promoter region (P). (C) Ctk1-HA ChIP normalized to Rpb1p in wt and rat1-1 strains. Position 8.5 in the wt strain is set to zero since Rpb1 ChIP levels are at background in wt cells (see Fig. 2B). Averages and standard deviations of all ChIP data were calculated from three independent biological experiments, each subjected to triplicate quantitative PCR analysis.

RNAPII elongation rate is increased in the rat1-1 mutant

Rat1p partakes in pre-mRNA 5′ capping quality control and causes premature transcription termination of RNAPII engaged with uncapped RNA (Jiao et al. 2010; Jimeno-Gonzalez et al. 2010). Under such conditions, the rat1-1 mutation leads to increased RNAPII levels inside gene bodies. Hence, the decreased RNAPII occupancy within gene bodies of rat1-1 cells, observed here without any impaired capping, is not explained by this role of Rat1p. On the other hand, Gu et al. (2013) recently found that substitution of CTD-Ser2 residues with alanines slows down transcription in mammalian cells. Thus, we speculated that CTD hyperphosphorylation in rat1-1 cells might cause an increased transcription elongation rate. In fact, this could help explain the lowered levels of gene body RNAPII in rat1-1 cells; i.e. the same amount of RNAPII is loaded at the promoter but spends less time transcribing along the coding region. To investigate this possibility, we monitored transcription kinetics of the GAL1-YLR454W system, where transcription rate can be extrapolated from the time it takes RNAPII to leave the gene template after turning off transcription by the addition of glucose to the culture (Mason and Struhl 2005). Hence, such a time course of the last wave of transcription was conducted by Rpb1p ChIP analysis at different positions along the gene. Indeed, RNAPII took a shorter time to leave the YLR454W gene in the rat1-1 compared with the wt context (Fig. 4A). Therefore, RNAPII speed is higher in rat1-1 cells.

FIGURE 4.

Increased RNAPII transcription rate in rat1-1 cells. (A) Transcription elongation rate assay measuring the last wave of transcription. Cells were grown in AA-Ura/2% raffinose at 30°C to an OD600 of 0.6. The GAL1 promoter was then induced with 2% galactose for 3 h at 30°C. After an additional incubation for 1 h at 34°C, the promoter was turned off by the addition of 2% glucose to the cultures, and samples were cross-linked at the indicated time points. Values correspond to Rpb1p ChIP signals, obtained with the indicated amplicons, at each time point, and normalized to signals in galactose (time point 0). Averages and standard deviations of all ChIP data were calculated from three independent biological experiments, each subjected to triplicate quantitative PCR analysis. (B,C) Growth capabilities measured by 10-fold dilution series of the indicated strains spotted onto YM-1 plates. Strains were grown for 2 d at 30°C or 34°C as indicated.

If an altered RNAPII elongation rate in the rat1-1 strain contributes to its growth defect, one might expect genetic interactions with other mutants affecting RNAPII speed. Indeed, combining the rat1-1 allele with the rpb1-E1103G mutation, which causes a faster transcription rate in vivo and in vitro (Malagon et al. 2006; Hazelbaker et al. 2013), resulted in an enhanced growth defect (Fig. 4B). Conversely, the rpb1-N488D or rpb2-10 alleles, which both cause a slower RNAPII elongation rate in vivo and in vitro (Powell and Reines 1996; Mason and Struhl 2005; Jimeno-Gonzalez et al. 2010), suppressed rat1-1 thermosensitivity (Fig. 4C; Jimeno-Gonzalez et al. 2010). As a slowdown of RNAPII transcription downstream from the pA site is believed to be a key step during the transcription termination process, an increased elongation rate may contribute to the termination defect observed in rat1-1 cells. Consistently, rescued growth of rat1-1 cells by introducing rpb1 alleles slowing RNAPII progress is paralleled by a correction of the rat1-1 transcription termination phenotype (Jimeno-Gonzalez et al. 2010).

Rat1p is widely accepted to be involved in transcription termination. In addition, the enzyme is required for quality control of RNA 5′-end capping (Jiao et al. 2010; Jimeno-Gonzalez et al. 2010). Previous reports all suggest a causal role for exonucleolytic attack of nascent RNA 5′ ends by Rat1p. Consistently, coexpression of a Rat1p catalytic dead mutant, in the rat1-1 strain, did not rescue CTD hyperphosphorylation (Supplemental Fig. S4), and exonucleolysis thus appears to also be required for the molecular phenotypes described here. As CTD phosphorylation impacts all steps of the transcription cycle, we suggest that the RNAPII CTD hyperphosphorylation phenotype of rat1-1 cells is taken into account when molecular phenotypes upon Rat1p inactivation are considered. How altered CTD phosphorylation status impinges on the various roles of Rat1p in transcription will be the focus of further investigation.

MATERIALS AND METHODS

Yeast strains and plasmids

Yeast manipulations and growth conditions were as previously described (Malagon et al. 2006). All strains used in this study are derived from a S288C background and listed in Table 1. The GAL1::YLR454W reporter was introduced by crossing with the strain BY4741 GAL1p::YLR454W (Mason and Struhl 2005). To isolate suppressors of rat1-1, we used the severe thermosensitivity phenotype conferred by the rat1-1/rpb1-E1103G double mutant at 34°C. The strain Y2823 was transformed with a genomic library (a kind gift of François Lacroute) cloned into the URA3 containing multicopy plasmid pFL44L (Bonneaud et al. 1991). Out of approximately 60,000 transformants, 14 clones suppressed the lack of growth at 34°C after reisolation of the original clones and retransformation. Sequencing of the clones identified 13 identical insertions containing the RAT1 gene (cXV coordinates 717670-721951; plasmid pRES21), while one clone carried a genomic fragment (cXIII 818791-822802) containing the FCP1 gene (plamid pRES7). Suppression was also observed in the rat1-1 single mutant, and no effect of RAT1 or FCP1 overexpression was observed for the rpb1-E1103G single mutant. Hence, FCP1 is a high-copy suppressor of rat1-1. All results shown here were obtained using the FCP1 gene subcloned into the multicopy LEU2 containing vector pRS425. The plasmid expressing the rat1 exonuclease mutant pRS315 rat1-D235A was kindly supplied by Steve Buratowski.

TABLE 1.

Yeast strains

ChIP analysis

ChIP reactions were performed as previously described (Jimeno-Gonzalez et al. 2006) by cross-linking with 1% formaldehyde for 20 min at room temperature. IPs were carried out using the following reagents: Rpb1p, Sepharose–protein A beads (GE Healthcare) and the rabbit polyclonal antibody Rpb1-y80 (Santa Cruz Biotechnology); Rpb3, Dynabeads M-280 sheep anti-Mouse IgG (Invitrogen) and the mouse monoclonal antibody anti-POLR2C (Abcam); Rat1-TAP, Sepharose-IgG beads (GE Healthcare); Rpb1p CTD-Ser2P, Dynabeads M-280 sheep anti-Mouse IgM (Invitrogen) and the mouse monoclonal antibody H5 (Covance); Fcp1-myc, Dynabeads M-280 sheep anti-Mouse IgG (Invitrogen) and the mouse monoclonal antibody c-myc (Santa Cruz); and Ctk1-HA, Dynabeads M-280 sheep anti-Mouse IgG (Invitrogen) and the mouse monoclonal antibody anti-HA (Santa Cruz). Purified DNA was analyzed as previously described (Jimeno-Gonzalez et al. 2010). Oligonucleotide sequences used for qPCR analysis of the GAL-YLR454W reporter gene are described by Jimeno-Gonzalez et al. (2010) and for PMA1 are as follows: 5, TCAGCTCATCAGCCAACTCAAG and CGTCGACACCGTGATTAGATTG; middle, TTGCCAGCTGTCGTTACCAC and TCGACACCAGCCAAGGATTC; 3′ UTR, TCTCTGGATGGTACTTTTTCTTTCTTG and TGCGTGTTGTGAATTGTGCTC; and T1, GCGCCCATACAGACA and CTTGTAGAATGGCCT.

Antibodies for Western blotting

The following antibodies were used for Western blotting analysis: mouse monoclonal antibody H14 (Covance); rat monoclonal antibodies 3E8, 4E12, and 3E10 (a gift from Dirk Eick); and rabbit polyclonal anti-Fcp1p (a gift from Steve Buratowski). The other antibodies are as described for ChIP.

SUPPLEMENTAL MATERIAL

Supplemental material is available for this article.