Mediator Is Essential for Small Nuclear and Nucleolar RNA Transcription in Yeast

Yeast methods.Med8, Med14, and Med3 were tagged with 3×FLAG-MNase-HIS3MX6 using pGZ109 (31). The auxin degron parental strain was generated by transformation of pSB2273 (kindly provided by Matthew Miller), encoding the GPD1 promoter-driven OsTIR1, into the HIS3 locus in a wild-type BY4705 strain. MED14 was tagged with AID using pL260 (kindly provided by Matthew Miller), encoding 3×V5-IAA7-kanMX6. SUA7 was tagged with 3×FLAG using pFA6a-6xGLY-3xFLAG-hphMX4 (a gift from Mark Hochstrasser; Addgene plasmid number 20755). Strain details are given in Table 2. S. cerevisiae cells were grown in yeast extract-peptone-dextrose (YPD) at 30°C, and S. pombe cells were grown in yeast extract-sucrose (YES) at 32°C under constant agitation.

ChEC-seq.ChEC-seq was performed as described previously (30, 59) with a 1-min calcium treatment, except for the size selection step. RNase-treated ChEC DNA was brought up to 200 μl with 10 mM Tris, pH 8.0, and 160 μl Solid-phase reversible-immobilization (SPRI) beads (60) was added (we used a 0.8:1 bead/sample ratio, whereas a 2.5:1 ratio was used in the original protocol). The sample was pipetted up and down 10 times to mix and incubated at room temperature for 5 min. Beads were collected on a magnetic rack for 2 min, and the supernatant was collected for DNA isolation. Sequencing libraries were prepared at the Indiana University Center for Genomics and Bioinformatics (CGB) using a NEBNext Ultra II library preparation kit for Illumina. Libraries were sequenced in paired-end mode on the Illumina NextSeq 500 platform at CGB. Read lengths were 79 bp for Med8 and 42 bp for Med14 and Med3.

ChIP-seq.A 100-ml culture at an optical density at 600 nm (OD₆₀₀) of >0.8 was split into two flasks. One culture was treated with DMSO and the other was treated with 500 μM 3-IAA dissolved in DMSO for 30 min. After treatment, the cultures were fixed with formaldehyde (final concentration, 1%) for 10 min at room temperature and then quenched with glycine (final concentration, 125 mM) for 5 min. After washing with phosphate-buffered saline (PBS), fixed cell pellets were preserved at −80°C. Thawed pellets were later spheroplasted by resuspension in PBS along with 90 μl of 5 mg/ml Zymolyase for 5 min at 37°C. All ChIP solutions henceforth were supplemented with a protease inhibitor cocktail. After gentle centrifugation and washing with 1 ml PBS, ∼100 μl spheroplasts was lysed with the addition of 50 μl lysis buffer (50 mM Tris, pH 8.0, 150 mM NaCl, 10 mM EDTA, 3% SDS) at room temperature for 5 min. Lysates were then diluted to 1.5 ml with 1,350 μl of dilution buffer (20 mM Tris, pH 8.0, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100) and transferred to a Covaris milliTUBE on ice. Sonication of diluted ChIP lysate was performed in a Covaris S220 sonicator, with parameters of 150-W peak power, 30% duty cycle, and 200 cycles/burst for a total sonication time of 150 s (on for 30 s and cool for 15 s 5 times). Thereafter, samples were centrifuged for 10 min at 21,000 × g and 4°C. The soluble lysates were then transferred to new tubes along with ∼30 μl of Sigma FLAG beads (catalog number M8823) that had been preblocked with bovine serum albumin, and the immunoprecipitation was performed for 3 h at 4°C. The beads were then washed 2 times with 1 ml of fresh cold dilution buffer and 1 time with 1 ml of LiCl wash buffer (20 mM Tris, pH 8.0, 250 mM LiCl, 1 mM EDTA, 1% Triton, 0.1% Nonidet P-40) with rotation at room temperature, with each wash lasting 3 min. For extraction and de-cross-linking, washed beads were resuspended in 250 μl elution buffer (50 mM Tris, pH 8.0, 250 mM NaCl, 10 mM EDTA, 1% SDS) with 60 μg proteinase K and incubated overnight at 1,200 rpm and 65°C in an Eppendorf ThermoMixer. The next morning, ChIP DNA was extracted with 200 μl phenol-chloroform-isoamyl alcohol, treated with 10 μg RNase A for 20 min at 37°C, reextracted with phenol-chloroform, and precipitated with linear acrylamide in >80% ethanol at −80°C overnight. Precipitated DNA was pelleted for 20 min at 4°C and 21,000 × g, washed with room temperature 75% ethanol, respun for 5 min, dried at room temperature, and finally resuspended in 20 μl T low-E buffer (50 mM Tris, 0.1 mM EDTA). ChIP-seq libraries were prepared and sequenced as described above for the ChEC-seq libraries. Read lengths were 42 bp for the first pair of DMSO/3-IAA replicates and 38 bp for the second replicate pair.

4-Thiouracil labeling and purification of newly synthesized RNAs.A 200-ml culture of strain GZY219 (Med14-AID) at an OD₆₀₀ of 0.6 to 0.8 was split into two 100-ml cultures. One culture was treated with DMSO and the other was treated with 500 μM 3-IAA dissolved in DMSO for 30 min immediately prior to 4tU labeling. RNA labeling was performed for 6 min by adding 4-thiouracil (Sigma) to the cultures to a final concentration of 5 mM. Labeling of S. pombe cells, used as the spike in for normalization across samples, was accomplished similarly to the labeling of S. cerevisiae cells, except that the cells were grown at 32°C in YES medium. Labeled S. cerevisiae and S. pombe cells were mixed in a 3:1 ratio prior to total RNA extraction using a RiboPure yeast kit (Ambion, Life Technologies) following the instructions of the manufacturer. Then, total RNA samples were treated with a Turbo DNA-free kit according to manufacturer guidelines (Ambion, Life Technologies). All experiments were performed using two independent biological replicates.

For biotinylation of 4tU-labeled RNAs (newly synthesized RNAs), 200 μg of total RNA was heated at 60°C for 10 min, followed by cooling for 5 min on ice. Two hundred microliters of 1 mg/ml EZ-Link HPDP {(3aS,4S,6aR)-hexahydro-2-oxo-N-[6-[[1-oxo-3-(2-pyridinyldithio)propyl]amino]hexyl]-1H-thieno[3,4-d]imidazole-4-pentanamide}-biotin (Pierce) and 100 μl of biotinylation buffer (100 mM Tris-HCl, pH 7.5, 10 mM EDTA) were added to the total RNA, the final reaction volume was adjusted to 1 ml with diethyl pyrocarbonate (DEPC)-treated water (Sigma), and the reaction mixture was incubated for 3 h at room temperature. To remove unbound biotin after biotinylation, an equal volume of chloroform was added to the reaction mix and the aqueous and organic phases were separated by centrifugation at 17,000 × g for 10 min at 4°C. Subsequently, the RNAs contained in the aqueous phase were precipitated with 1/10 volume of 5 M NaCl and 2.5 volumes of isopropanol and resuspended in 100 μl DEPC-treated water.

Recovered RNA was heated for 10 min at 65°C and cooled for 5 min on ice. One hundred microliters of μMACs streptavidin beads (Miltenyi Biotec) was added to the RNA to allow binding of the biotinylated newly synthesized RNAs to the streptavidin beads. The mix was incubated for 90 min at room temperature. Purification of labeled RNA was then carried out using a μMACS streptavidin starting kit (Miltenyi Biotec). Prior to loading of the samples on the columns, the columns were equilibrated with 1 ml of washing buffer (100 mM Tris-HCl, pH 7.5, 10 mM EDTA, 1 M NaCl, 0.1% Tween 20). Samples were passed through the columns twice and washed five times with increasing volumes of washing buffer (600, 700, 800, 900, and 1,000 μl). Finally, newly synthesized RNAs were eluted twice with 0.1 M dithiothreitol. The eluate was precipitated overnight in 1/10 volume of 3 M sodium acetate, 3 volumes of 100% ethanol, and 20 μg of RNA-grade glycogen. Newly synthesized RNAs were pelleted, washed in ice-cold 70% ethanol, and resuspended in 15 μl of DEPC-treated water.

qRT-PCR.cDNA synthesis was performed using random hexamers and Transcriptor reverse transcriptase (Roche) following the manufacturer's guidelines. One microgram of total RNA or 5 μl of newly synthesized RNA was used for reverse transcription. qRT-PCR experiments were performed using SYBR green I master mix and a LightCycler 480 instrument II (Roche). All primers used are listed in Table 3. cDNA samples were diluted 20-fold prior to the qPCR experiments, except for RDN25 and RDN58 (which were diluted 1,000-fold). After qRT-PCR, raw values were adjusted to the level of S. pombe act1⁺ (actin) expression. Samples were analyzed in technical triplicate from two independent biological replicates. Graphical representations of the results are shown as a comparison between the control (DMSO-treated, set to 1) conditions and 3-IAA-treated conditions.

Data analysis. (i) sn/snoRNA gene list.We used the YeastMine QueryBuilder tool (https://yeastmine.yeastgenome.org/) to generate a list of sn/snoRNA genes. We first retrieved lists of items with data type “snRNA” (6 items) or “snoRNA” (77 items). For snRNA genes, the RNAPIII-transcribed snR6 gene was removed and the overlapping snR7-L and snR7-S genes were condensed into the long isoform. For snoRNA genes, we used the overlapping features function of YeastMine to identify snoRNAs overlapping open reading frames (ORFs) (13/77). Two snoRNA genes, snR43 and snR73, were found to overlap a deleted and dubious ORF, respectively, and so were retained in the final list. The RNAPIII-transcribed snR52 gene was removed from the final list. We also included the RNAPII-transcribed TLC1 gene, encoding telomerase RNA. This resulted in a list containing 70 genes (4 snRNAs, 65 snoRNAs, and TLC1). For the purposes of our TSS-based analyses, polycistronic snoRNA clusters (snR190-snR128, snR41-snR70-snR51, snR67-snR53, snR57-snR55-snR61, and snR78-snR77-snR76-snR75-snR74-snR73-snR72) were considered single genes, with the start coordinate of the first snoRNA in the cluster being taken to be the cluster's TSS. This condensation resulted in a list of 58 loci (Table 1).

(ii) ChEC-seq.Paired-end Med8, Med14, and Med3 ChEC-seq and 1-min REB1 promoter-driven free MNase data (Sequence Read Archive accession number SRR1947784) were aligned to the sacCer3 genome build with the Bowtie2 program (61) using default settings plus “-I 10 -X 700 --no-unal --dovetail –no-discordant --no-mixed.” The above-described free MNase data set was used because it was generated in the W1588-4C background, which is congenic to the W303 background in which the Mediator ChEC-seq data sets were generated except that a weak RAD5 mutation is corrected (62). Alignment sequence alignment and mapping (SAM) files were used to make tag directories with the HOMER program (http://homer.ucsd.edu) (63). The HOMER “annotatePeaks.pl” tool was used to average the signal and generate matrices for heatmap visualization. Average plots were generated with GraphPad Prism software (version 7), and heatmaps were generated with the Java TreeView program (64). Med14 peaks were called using the HOMER “findPeaks” tool with the flags “-style factor -fdr 0.05 -F 2 -L 2 -gsize 1.2495e7,” with the free MNase data set being used as a control. These parameters require that a peak meet a 5% false discovery rate threshold and be enriched at least 2-fold over the signal for free MNase and the local background. We detected 1,448 Med14 peaks with this approach (see Data Set S2 in the supplemental material). The Med14 peak nearest each sn/snoRNA TSS was determined with the BEDTools “closest” tool (65), and each peak was visually inspected for its position relative to the corresponding sn/snoRNA gene. We required that a peak be no more than 500 bp upstream of an sn/snoRNA TSS to be considered associated with that gene. The Med17 signal was quantified from bp −500 to −100 relative to the TSS using the HOMER “annotatePeaks.pl” tool. Fifty-two of 58 (89.7%) sn/snoRNA genes had >2-fold enrichment of the Med17 signal over that for the corresponding free MNase sample (Sequence Read Archive accession number SRX1755637) in this window. Lists of SAGA- and TFIID-dependent genes that had annotated TSSs and that encoded verified ORFs were from our previous work (31), and we quantified the signal at the 50 most highly Mediator-occupied SAGA-dominated gene UASs and the 500 most highly Mediator-occupied TFIID-dominated gene UASs, as determined by the signal in the WT strain.

(iii) ChIP-seq.Paired-end spike-in RNAPII ChIP-seq data (Rpb3-3×FLAG) (24) were obtained from the Gene Expression Omnibus (GEO) database (GEO accession number GSE97081) and aligned to both the sacCer3 and EF2 (S. pombe) genome builds with the Bowtie2 program as described above for ChEC-seq data. Tag directories were then created with HOMER. For genome browser visualization, we downloaded spike-in-normalized wig files from GEO; these tracks thus offer independent confirmation of our systematic analyses of these data. For quantification of sn/snoRNA RNAPII occupancy, the total raw RNAPII signal in an 83-bp window downstream of the TSS (corresponding to the length of the shortest analyzed monocistronic snoRNA gene, snR79) was determined using the HOMER “annotatePeaks.pl” tool. The obtained values were then multiplied by a spike-in normalization factor (N), the log₂(3-IAA/DMSO) ratio for each gene in each replicate was calculated, and the ratios were averaged. For comparison, we analyzed the 1,000 most highly transcribed mRNA genes, as determined by the average spike-in-normalized signal in the two DMSO-treated replicates of each auxin depletion experiment using a 100-bp window downstream of the TSS. The spike-in normalization factor, N, was calculated as 10,000/number of reads mapped to the S. pombe genome and used by HOMER for generation of a tag directory. Paired-end TFIIB ChIP-seq data were processed as described above for the ChEC-seq data. The TFIIB signal was quantified from −150 to +50 relative to the TSSs of sn/snoRNA genes and the 1,000 most TFIIB-occupied mRNA promoters, as determined by the average number of reads per million (RPM)-normalized signal in the −150 to +50 window; the log₂(3-IAA/DMSO) ratio for each gene in each replicate was calculated; and the ratios were averaged. Single-end RNAPII ChIP-seq data (Rpb1) from WT and tail deletion strains (39) were obtained from the Sequence Read Archive (accession number SRP047524) and aligned to sacCer3 with Bowtie2 using default parameters plus “--no-unal.” Tag directories were created, and the RNAPII signal was quantified as described above for the spike-in RNAPII ChIP-seq data, except that normalization of the number of RPM was used.

(iv) ChIP-chip.MetaMediator bedGraph files representing the average ChIP-chip signal for 12 Mediator subunits were obtained from the supplemental material of the work of Jeronimo et al. (34). The data were quantified as described above for ChEC-seq, except that the -bedGraph option of HOMER annotatePeaks.pl was used to determine the signal. bedGraph files of TFIIB ChIP-chip data from WT and Med18-FRB strains treated with rapamycin and WT and tail deletion strains (34) were obtained from GEO (accession number GSE81107) and quantified as described above for TFIIB ChIP-seq, except that “-bedGraph” was used.

Data availability.Sequencing data have been deposited with the Gene Expression Omnibus database (accession number GSE112721).