Kinetic CRAC uncovers a role for Nab3 in determining gene expression profiles during stress

RNA-binding proteins play a key role in shaping gene expression profiles during stress, however, little is known about the dynamic nature of these interactions and how this influences the kinetics of gene expression. To address this, we developed kinetic cross-linking and analysis of cDNAs (χCRAC), an ultraviolet cross-linking method that enabled us to quantitatively measure the dynamics of protein–RNA interactions in vivo on a minute time-scale. Here, using χCRAC we measure the global RNA-binding dynamics of the yeast transcription termination factor Nab3 in response to glucose starvation. These measurements reveal rapid changes in protein–RNA interactions within 1 min following stress imposition. Changes in Nab3 binding are largely independent of alterations in transcription rate during the early stages of stress response, indicating orthogonal transcriptional control mechanisms. We also uncover a function for Nab3 in dampening expression of stress-responsive genes. χCRAC has the potential to greatly enhance our understanding of in vivo dynamics of protein–RNA interactions.

strain (Nab3 panels), demonstrating that these changes are a direct result of Nab3 depletion.
We conclude that with our Nab3 depletion conditions we can faithfully reproduce previously published work.

Supplementary note 2 Nab3 and regulation of ENO1 transcription initiation
As shown in a schematic overview of ENO1 transcription-regulation (not to scale; Supplementary Fig. 12), two CUTs are detected upstream of the gene, which is activated by two upstream activating sequences (UAS1, UAS2) that bind multiple factors [4][5][6] . ENO1 also contains an upstream repressor sequence element (URS) in the promoter, located between -226 and -125 bp from the TSS. This URS has a directionality as reverting this element relieves inhibition of ENO1 expression when cells are grown on glucose 7 . Experimental evidence has been published for three URSbinding factors: i) Reb1, that associates with a region not essential for, but enhancing URS function 8 and binds the UAS2 as well 4,8 . Deletion of Reb1 had no major effect on ENO1 expression 9,10 ; ii) the BUF-complex, consisting of Rfa1 and Rfa2 -which participate in yeast DNA replication as ssDNA binding proteins -that recognizes the repressor-of-CAR1-like sequence (TaGCCaCCTC) at the 5' end of the region essential for URS activity 11,12 . Repression is possibly mediated by Ume6 13 , which recruits histone deacetylase Rpd3p and chromatin-remodeling factor Isw2p 14 . Long-range DNA-looping via Ume6 is associated with transcriptional repression 15 . Although Ume6 had no influence, Isw2 was found to downregulate ENO1 expression 9,10 and can repress transcription of cryptic RNAs 16 . iii) The basic-domain helix-loop-helix (bHLH) protein Sgc1 (Tye7) binds to the E-box motif (CAnnTG) in the 3' region of the URS 17 , possibly as a (hetero) dimer 18 , and, depending on surrounding sequences, could bend DNA 19 .
As shown in Supplementary Fig. 8c and Fig. 6d, we observed increased Nab3 binding to RNAs overlapping the region spanning the TATA-box, a long pyrimidine-rich region including three Nab3 binding motifs (UCUU; CUUG) and up to the TSS when cells were grown on glucose, but significantly less so after removal of the carbon source. These products increased in number after depletion of Nab3 (Fig. 6e) and carry poly-A tails (Fig. 6d), indicative for poly-adenylation by the TRAMP-complex and needed for degradation by the exosome. The 5'UTR-derived products also overlap the two CUTs, that are only observed in the absence of exosome components 21,22 .
CUT166, which initiates from the 5' end of the URS has also been detected by CRAC using Cbc1, a component of the cap-binding complex 23 and by transcript isoform sequencing (TIFseq) 24 (Supplementary Fig. 8).
The role of the URS and the associated protein factors suggest that these are the primary regulators of ENO1 transcription by controlling productive transcription initiation from its TSS, ~40 nt upstream of the ATG. They appear to do this (possibly indirectly), by stimulation of CUT166 transcription, but not by promoting alternative TSS choice. For cells growing on glucose we did not observe (also after depleting Nab3) a marked change in the distribution of Pol II; all the associated RNAs had their 5' end around the mapped TSS 25,26 as confirmed by TIFseq 24 (Supplemental Fig. 8). Furthermore, after removal of glucose, the levels of transcripts derived from the 5'UTR did not alter over time (also when Nab3 had been depleted; Fig. 6e), whereas those reflective of ENO1 transcription increased more than 2-fold (Fig. 6f). Overall, our data suggest that Pol II still finds the TSS but that under repressive conditions this happens slowly. When repression of ENO1 is lifted, although transcription of CUT166 still occurs, formation of an active transcription initiation complex at the TSS of ENO1 is promoted. This could be realized by the release of repressive factors such as Sgc1. The level of induced transcription increases so rapidly that NNS-guided degradation -although increasing as well according to enhanced Nab3-crosslinking -does not have an overall impact on steady state levels. Possibly, NNS-mediated termination of transcripts will happen on Pol II complexes that are slowed down or do not elongate properly, so that ongoing ENO1 transcription will not be affected.

Yeast strains and media
Saccharomyces cerevisiae strain BY4741 (MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0) was used as the main parental strain 29 . The HTP (HIS6-TEV-2xProtA) carboxyl-tagged strains (Calmodulin binding peptide-TEV-2xProtA) were generated by PCR as described 30,31 . Strains used in this study are listed in Supplementary Table 1. For the anchor away experiments we fused Nab3 to the FRB domain and integrated a HTP-tag at the 3' end of the Rpo21 gene in the HHY168 strain 1 . Rapamycin was added to the media at a final concentration of 1 µg/ml. For the glucose deprivation experiments, strains were grown in synthetic medium lacking tryptophan (Formedium) in the presence (SD-TRP) or absence of glucose (S-TRP). For the PAR-CLIP experiments, cells were grown in synthetic glucose-containing medium lacking tryptophan and uracil (SD-URA-TRP). Strains carrying changes in cross-linked Nab3motifs were generated by site-directed mutagenesis in two steps using the Delitto-Perfetto system 32 ; first a small deletion was generated covering the YBR085C-A Nab3 motifs by insertion of the cassette from plasmid pGSHU, which was then replaced by a gBlock (Integrated DNA Technologies) containing the mutations. Resultant strains were checked by sequencing and for normal growth on glycerol.

Western blot analyses
Western blot analysis was performed using the polyclonal rabbit anti-TAP antibody from Thermo Fisher (CAB1001), which recognizes the spacer between the TEV cleavage site and the six histidines. Blots were incubated with the antibody (1:5000 dilution) in blocking buffer (5% nonfat milk powder, 0.1% Tween-20 and phosphate buffer saline (PBS)) for one hour at room temperature (diluted 1:5000 in blocking buffer. Following two five minute washes with PBS-0.1% Tween, the blots were then incubated with goat anti-rabbit antibodies (Thermo Fisher (31466) 1:5000 in blocking buffer) for one hour at room temperature. Proteins were visualized using the Pearce enhanced chemiluminescence solutions as described by the manufacturer's procedures.

Quantitative RT-PCR
Cells were grown in SD-TRP to an OD 600 of 0.4, harvested by filtration and then shifted to S-TRP. Cells were harvested before the shift (0) and 20, 40 minutes after the shift. RNA was extracted using the Guanidium thiocyanide method 33 or the masterpure yeast RNA purification kit (Epicentre) and quantitative RT-PCR was carried out using the Agilent Brilliant III SYBR master mix, using oligonucleotides listed in Supplementary

Northern blot analysis
Total RNA was resolved on a 1.25% Agarose Bis-Tris (pH 7) gel and transferred to nitrocellulose. Northern blotting was performed using UltraHyb hybridization buffer according to the manufacturer's procedures (Ambion). The snR13 oligo sequence used for hybridization is provided in Supplementary invddT-ACACrGrArCrGrCrUrCrUrUrCrCrGrArUrCrUrNrNrNrArCrArGrUrGrN

Glucose* No Glucose (4 minutes) No Glucose (18 minutes)
*) Genes in italics might be identified due to transcripts originating from a preceding snoRNA gene (in brackets) that is not properly terminated in absence of Nab3. ‡ ) See also Supplementary Fig. 5b (Northern blot probed for snR13) and Supplementary Fig. 6c. The machine also has a tilting mechanism to allow the bag in the UV chamber to empty completely. After transferring the RNA to a nitrocellulose membrane, the blot was probed with an anti-sense snR13 oligonucleotide (Supplementary Table 2). After about 60 minutes of rapamycin treatment the amount of 3' extended snR13 species reached its maximum level. Therefore, for subsequent depletion experiments a 60-minute rapamycin incubation was used.

Supplementary Figure 6. Nuclear depletion of Nab3 results in the accumulation of 3' extended CUTs and snoRNAs.
(a) Distribution of reads that mapped to CUTs and snoRNAs around the 3' end of the features (x-axis). Cells expressing the FRB-tagged Nab3 and the parental strain (Nab3) were grown in glucose to exponential phase and rapidly shifted to medium lacking glucose. Cells were harvested before the shift (glucose panels) or 14 (Nab3) to 18 minutes (Nab3-FRB) after the shift (no glucose panels). Reads mapped to each features were divided over 400 bins (1nt per bin) and for each bin the fraction of total reads that mapped to each bin was calculated. These numbers were then averaged (y- Shown are genome browser images of CRAC data generated using the anchor-away strain expressing an HTP-tagged Rpo21. Cells were grown in glucose to exponential phase and incubated with ethanol (-) or rapamycin (RAP; +) for one hour. A fraction of the cells was harvested (glucose samples) and the rest was shifted to medium lacking glucose for 14 minutes. These data demonstrate that the drug rapamycin does not influence Pol II transcription of these genes.