Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

SAGA mediates transcription from the TATA-like element independently of Taf1p/TFIID but dependent on core promoter structures in Saccharomyces cerevisiae

  • Kiyoshi Watanabe,

    Roles Investigation, Methodology, Writing – original draft

    Affiliation Molecular and Cellular Biology Laboratory, Graduate School of Medical Life Science, Yokohama City University, Yokohama, Kanagawa, Japan

    ⨯
  • Tetsuro Kokubo

    Roles Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing

    kokubo@yokohama-cu.ac.jp

    Affiliation Molecular and Cellular Biology Laboratory, Graduate School of Medical Life Science, Yokohama City University, Yokohama, Kanagawa, Japan

    ⨯

Abstract

In Saccharomyces cerevisiae, core promoters of class II genes contain a TATA element, either a TATA box (TATA[A/T]A[A/T][A/G]) or TATA-like element (1 or 2 bp mismatched version of the TATA box). The TATA element directs the assembly of the preinitiation complex (PIC) to ensure accurate transcriptional initiation. It has been proposed the PIC is assembled by two distinct pathways in which TBP is delivered by TFIID or SAGA, leading to the widely accepted model that these complexes mediate transcription mainly from TATA-like element- or TATA box-containing promoters, respectively. Although both complexes are involved in transcription of nearly all class II genes, it remains unclear how efficiently SAGA mediates transcription from TATA-like element-containing promoters independently of TFIID. We found that transcription from the TATA box-containing AGP1 promoter was greatly stimulated in a Spt3p-dependent manner after inactivation of Taf1p/TFIID. Thus, this promoter provides a novel experimental system in which to evaluate SAGA-mediated transcription from TATA-like element(s). We quantitatively measured transcription from various TATA-like elements in the Taf1p-dependent CYC1 promoter and Taf1p-independent AGP1 promoter. The results revealed that SAGA could mediate transcription from at least some TATA-like elements independently of Taf1p/TFIID, and that Taf1p-dependence or -independence is highly robust with respect to variation of the TATA sequence. Furthermore, chimeric promoter mapping revealed that Taf1p-dependence or independence was conferred by the upstream activating sequence (UAS), whereas Spt3p-dependent transcriptional stimulation after inactivation of Taf1p/TFIID was specific to the AGP1 promoter and dependent on core promoter regions other than the TATA box. These results suggest that TFIID and/or SAGA are regulated in two steps: the UAS first specifies TFIID or SAGA as the predominant factor on a given promoter, and then the core promoter structure guides the pertinent factor to conduct transcription in an appropriate manner.

Citation: Watanabe K, Kokubo T (2017) SAGA mediates transcription from the TATA-like element independently of Taf1p/TFIID but dependent on core promoter structures in Saccharomyces cerevisiae. PLoS ONE 12(11): e0188435. https://doi.org/10.1371/journal.pone.0188435

Editor: Laszlo Tora, Institute of Genetics and Molecular and Cellular Biology, FRANCE

Received: October 3, 2017; Accepted: November 7, 2017; Published: November 27, 2017

Copyright: © 2017 Watanabe, Kokubo. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This study was supported by grants from the Japan Society for the Promotion of Science (15K06953) and the AMED-CREST program (JPMJCR11J4) of the Japan Agency for Medical Research and Development, AMED. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

In eukaryotes, general transcription factors (GTFs), Mediator, and RNA polymerase II (pol II) assemble on the core promoter to form a preinitiation complex (PIC) that directs accurate transcriptional initiation [1–5]. In the first step of PIC assembly, TBP is recruited to the core promoter as a subunit of TFIID or via physical association with the SPT module of the SAGA (Spt-Ada-Gcn5-acetyltransferase) complex [6–9]. TFIID and SAGA are structurally related large multi-protein complexes that mediate basal and/or activated transcription [2, 10–12]. They share five Taf subunits that form a scaffold for the assembly of other complex-specific subunits [9, 13–16], and the two complexes define two distinct PIC assembly pathways [4, 8, 17].

Genome-wide studies revealed that SAGA-dominated promoters (i.e., those whose transcriptional activities are primarily supported by SAGA rather than by TFIID) prefer the TATA box (TATAWAWR; W = A/T, R = A/G), whereas TFIID-dominated promoters prefer the TATA-like element (a 1 or 2 bp mismatched version of the TATA box) [7, 8]. According to the nomenclature proposed by Rhee and Pugh [8], the TATA element includes the TATA box and the TATA-like element. More recent studies indicate that TFIID and SAGA are both involved in transcription of nearly all class II genes [18–21]. Consistent with this, it is well established that TFIID can mediate transcription from various types of promoters, both in vivo and in vitro, regardless of whether they contain the TATA box or not [22–28]. However, it remains unclear how efficiently SAGA can mediate transcription from TATA-less promoters that do not contain the TATA box but instead contain TATA-like elements or other less well-characterized core promoter elements (CEs) [25, 29, 30], especially in a manner independent of TFIID.

One reason for this ambiguity is the scarcity of in vitro experiments that scrutinize the requirements of the TATA sequence for SAGA-dependent transcription. In an in vitro SAGA-dependent transcription system, SAGA can mediate transcription from the TATA-containing HIS4 promoter [31] but not from the TATA-less RPS5 promoter [24]. However, it remains to be determined whether SAGA can mediate transcription from a HIS4 promoter containing a TATA-like element (off-consensus TATA) or a RPS5 promoter containing a consensus TATA box under the same conditions. Other studies showed that several sequences isolated as active CEs in a random screen using the gal-his3 hybrid promoter [32] have significant transcriptional activities in vitro, even if they do not contain a consensus TATA box [33], but it remains unclear whether the in vitro transcription system used in the latter experiment is SAGA-dependent.

Another reason for the ambiguity is the functional redundancy between TFIID and SAGA [7, 34]. In general, SAGA-dependent promoters are highly regulated [7, 35] and require the TATA box for transcriptional activation [35, 36], as observed for the GAL1 promoter [37–40]. Consistent with this, mutational studies of the partially (i.e., subunit-specifically) SAGA-dependent [11] but TFIID-independent [41, 42] TR element of the HIS3 promoter revealed a rather stringent requirement of the TATA sequence for transcription [43, 44]. However, the TATA-less TRP3 promoter [45] is regulated not only by TFIID [7, 41, 42] but also by SAGA in a manner very similar to the HIS3-TR promoter [11]. Based on these findings, it is certain that TFIID plays an essential role in transcription from the TATA-less TRP3 promoter, but not from the HIS3-TR promoter. However, these observations imply that, in a SAGA-defective strain, TFIID could alter its role in transcription to generate similar transcriptional profiles for these two promoters. If this is the case, it will be important to carefully evaluate how efficiently SAGA mediates transcription from TATA-less promoters under a condition in which the contribution of TFIID to transcription is minimized, e.g., by functional inactivation of TFIID-specific Taf(s).

Recent studies suggest that transcriptional attenuation, e.g., due to a malfunction of the transcriptional machinery, could be compensated by down-regulation of mRNA degradation [18, 20, 21, 46–49]. Presumably due to such a buffering effect, only limited genome-wide defects can be observed when steady-state mRNA levels are measured in TFIID- or SAGA-defective strains [7, 50]. In fact, measurements of nascent mRNA levels revealed that TFIID and SAGA are both globally required for transcription of nearly all class II genes, regardless of whether they contain the TATA box [20, 21]. These results indicate that the observed defects in the production of steady-state mRNA may not result from reduced transcription per se but rather from the combined effects of a decrease in transcription and an increase in mRNA stabilization. Therefore, TFIID- and SAGA-dominated promoters cannot be distinguished simply by the transcriptional requirements for these two factors [20, 21]. Despite these recent advances, it remains unclear how TFIID and SAGA mediate transcription from the two types of promoters.

Previously [51], we randomized the TATAα element in the Taf1p/TFIID-dependent CYC1 promoter to search for transcriptionally active CEs. The active 601 sequences (from 4,781 clones) obtained in that screen were tentatively classified into nine groups: classes I (TATAWAWR; consensus TATA box), II (TATAWAD; D = A/G/T), III (GAAAA), IV (TATAWKW; K = G/T), V (TTAAAW), VI (Wx6), VII (TATATCWD), VIII (Wx5 other than TATATCWD), and IX (others) [51]. Consistent with the findings of genome-wide studies [7, 8], we found that the Taf1p/TFIID-dependent CYC1 promoter could use various CE sequences that did not match the consensus TATA box [51].

In this study, to determine how efficiently SAGA mediates transcription from TATA-less promoters independently of Taf1p/TFIID, we examined the transcriptional activities of several CE sequences belonging to class I, II, V, or VI in the Taf1p/TFIID-independent AGP1 promoter under conditions in which Taf1p was functionally inactivated by temperature shift and residual transcription was entirely Spt3p/SAGA-dependent. Furthermore, we also mapped the determinant(s) of Taf1p/TFIID-dependence and -independence of the CYC1 or AGP1 promoters, respectively, as well as those of Spt3p/SAGA-dependent transcriptional stimulation of the AGP1 promoter after inactivation of Taf1p/TFIID. The results indicated that the function of TFIID and/or SAGA is regulated in two steps, in which the upstream activating sequence (UAS) and core promoter play specific roles and probably operate in a sequential manner.

Materials and methods

Yeast strains

Standard techniques were used for yeast growth and transformation [52]. Yeast strains used in this study are listed in S1 Table. Sequences of oligonucleotides used for strain or plasmid construction are listed in S2 Table.

All strains used in this study, except those depicted in S1 and S3 Figs, were derived from Y22.1, which carries a deletion of the chromosomal TAF1 coding region and the wild-type TAF1 gene in a URA3-based low-copy-number vector (pYN1) [53]. YTK2741 [54] and YTK3778 [51] were generated from Y22.1 by replacing pYN1 with pM1169 (HA-tagged wild-type TAF1/pRS314) [55] and pM1746 (HA-tagged taf1-N568Δ/pRS314) [51], respectively. BY4741 and Y04228 were obtained from Euroscarf, and YTK16396/16397/16398/16399 were described previously [51].

To create YTK17952 and YTK17974, the two sub-fragments containing AGP1 [-780 – -111 bp] (primers: TK8267-TK8678/template: genomic DNA of BY4741; hereafter abbreviated as TK8267-TK8678/BY4741) or AGP1 [-135 – -1 bp] (TK12517-TK8268/BY4741) were first amplified by PCR using the primer pair/template (genomic DNA or plasmid) as described above in parenthesis, and then fused with TK8267-TK8268 to generate a 0.84 kb fragment (S3 Table). The original TATA box (TATATAAG) of the AGP1 promoter was replaced with a different one (TATATAAA) derived from the CYC1 promoter. Subsequently, the other two sub-fragments containing VTC1 [-320 – -11 bp] + LEU2 [56] (TK7872-TK6582/YTK16396) or VTC1 [+1 – +220 bp] (TK9030-TK9171/BY4741) were amplified by PCR. Finally, these two sub-fragments and the aforementioned 0.84 kb fragment were fused with TK7872-TK9171 to generate a 3.6 kb fragment that was used for transformation of YTK2741 and YTK3778, yielding YTK17952 and YTK17974, respectively (S3 Table). Similar to YTK17952, YTK17954/17964/17966/17968/17970/17956/17958/17960/17962/17972/17586/17584/17591/17593/17582 were generated from YTK2741 by transforming PCR fragments amplified using primers TK12518/13002/12524/12525/12526/12519/12520/12521/ 12522/12527/12926/12925/12928/12929/12924, respectively, instead of TK12517 (S3 Table). Furthermore, similar to YTK17974, YTK17976/17978/17980/17982/17984/ 17599/17597/17604/17606/17595 were generated from YTK3778 by transforming PCR fragments amplified using primers TK12518/12519/12520/12521/12522/12926/ 12925/12928/12929/12924, respectively, instead of TK12517 (S3 Table).

To create YTK17590 and YTK17603, the three sub-fragments containing VTC1 [-320 – -11 bp] + LEU2 [56] (TK7872-TK6582/YTK16396), AGP1 [-780 – -1 bp] (TK8267-TK8268/BY4741), or VTC1 [+1 – +220 bp] (TK9030-TK9171/BY4741) were first amplified by PCR, and then fused with TK7872-TK9171 to generate a 3.6 kb fragment that was used for transformation of YTK2741 and YTK3778, yielding YTK17590 and YTK17603, respectively (S3 Table).

To create YTK16408/16410/16412/16414/16400/16402/16404/16406/16416 and YTK16401/16403/16405/16407, the three sub-fragments containing VTC1 [-530 – -11 bp] (TK10267-TK7873/BY4741), LEU2 (TK12262-TK6582/pUG73 [56]), or the CYC1 promoter [-400 – -1 bp] + VTC1 [+1 – +430 bp] (TK10036-TK4283/genomic DNA derived from each strain that had been isolated by a previous screen [51]) were first amplified by PCR, and then fused by TK10267-TK4283 to generate a 3.6 kb fragment that was used for transformation of YTK2741 and YTK3778, respectively (S3 Table).

To create YTK17566 and YTK17578, the two sub-fragments containing VTC1 [-400 – -11 bp] + LEU2 marker [56] + the CYC1 promoter [-400 – -124 bp] (TK8260-TK10257/YTK16396), or the CYC1 promoter [-123 – -1 bp] + VTC1 [+1 – +320 bp] (TK12922-TK7875/YTK16396), were first amplified by PCR, and then fused with TK8260-TK7875 to generate a 3.4 kb fragment that was used for transformation of YTK2741 and YTK3778, respectively (S3 Table). Similar to YTK17566, YTK17562/17560/17564/17568/17558 were generated from YTK2741 by transforming PCR fragments amplified using primers TK12920/12919/12921/12923/ 12918, respectively, instead of TK12922 (S3 Table). Furthermore, similar to YTK17578, YTK17574/17572/17576/17580/17570 were generated from YTK3778 by transforming PCR fragments amplified using primers TK12920/12919/12921/12923/12918, respectively, instead of TK12922 (S3 Table).

To create YTK17748 and YTK17784, the fragments containing VTC1 [-170 – -11 bp] + LEU2 [56] + AGP1UAS [-780 – -301 bp] + CYC1core [-230 – -1 bp] + VTC1 [+1 – +150 bp] were amplified by PCR (TK2496-TK10408/pM8010 carrying AGP1UAS+CYC1core [TATATAAA]) to generate a 3.3 kb fragment that was used for transformation of YTK2741 and YTK3778, respectively. Similarly, YTK17750/17766/17769 and YTK17786/17802/17804 were generated from YTK2741 and YTK3778, respectively, using pM8011 (AGP1UAS+CYC1core [TAGCGCAA])/ pM8019 (CYC1UAS+AGP1core [TATATAAA])/ pM8020 (CYC1UAS+ AGP1core [TAGCGCAA]) as a template instead of pM8010.

The strains used in S1 and S3 Figs were derived from BY4741 or BY4742. YTK11411, YTK11705, and YTK11708 were described previously [57]. YTK11705 (TAF1) and Y04228 (Δspt3) were crossed and dissected to generate YTK11871 (TAF1 spt3). Similarly, YTK11708 (taf1-N568Δ) and Y04228 (Δspt3) were crossed and dissected to generate YTK11873 (taf1-N568Δ Δspt3). YTK13039 was generated from YTK11871 by replacing HIS3-marked pM4770/TAF1 with URA3-marked pYN1/TAF1. To create YTK18955 and YTK18964, the fragments containing VTC1 [-170 – -11 bp] + LEU2 [56] + AGP1 promoter [-780 – +1 bp] + VTC1 [+1 – +150 bp] were amplified by PCR (TK2496-TK10408/YTK17952) to generate a 3.3 kb fragment that was used for transformation of YTK11411 and YTK13039, respectively. YTK18974 and YTK18986 were generated from YTK18955 by replacing pYN1/TAF1 with pM4770/TAF1 and pM4773/taf1-N568Δ, respectively. Similarly, YTK18998 and YTK19010 were generated from YTK18964 by replacing pYN1/TAF1 with pM4770/TAF1 and pM4773/taf1-N568Δ, respectively.

Construction of plasmids

pYN1, pM4770, and pM4773 were described previously [53, 57]. To create pM8010 containing AGP1UAS+CYC1core [TATATAAA], the two sub-fragments containing VTC1 [-320 – -11 bp] + LEU2 [56] + AGP1UAS [-780 – -301 bp] (TK13578-TK13612/YTK17590), or CYC1core [-230 – -1 bp] + VTC1 [+1 – +814 bp] (TK13619-TK13579/YTK16396), and a linearized vector (TK12873-TK13577/pBluescript II KS+) were first amplified by PCR, and then ligated together using the In-Fusion HD cloning kit (TaKaRa). pM8011 containing AGP1UAS+CYC1core [TAGCGCAA] was created in the same way, except that YTK16398 was used as a template instead of YTK16396.

To create pM8019 containing CYC1UAS+AGP1core [TATATAAA], the two sub-fragments containing VTC1 [-320 – -11 bp] + LEU2 [56] + CYC1UAS [-400 – -231 bp] (TK13578-TK13618/YTK16396), or AGP1core [-300 – -1 bp] + VTC1 [+1 – +814 bp] (TK13614-TK13579/YTK17952), and a linearized vector (TK12873-TK13577/pBluescript II KS+) were first amplified by PCR, and then ligated together using the In-Fusion HD cloning kit. pM8020 containing CYC1UAS+AGP1core [TAGCGCAA] was created in the same way, except that YTK17954 was used as a template instead of YTK17952. These plasmids (i.e., pM8010/8011/8019/8020) were used as PCR templates for the construction of eight yeast strains (i.e., YTK17748/17750/17766/17769/17784/17786/7802/17804) as described above.

Northern blot analysis

Northern blot analysis was performed as described previously [58]. Specifically, total RNA (20 μg) was isolated from the indicated strains grown to logarithmic phase at 25°C in YPD media or further incubated at 37°C for the indicated period of time. RNA was subjected to electrophoresis, blotted onto a membrane, and hybridized with the gene-specific probes.

For detection of VTC1, CYC1, AGP1, RPS5, and SCR1, DNA fragments were amplified by PCR from yeast genomic DNA, purified, and 32P-labeled by random priming with the Klenow fragment (TOYOBO). The PCR primers used were as follows: VTC1, TK9030-TK9013; CYC1, TK9727-TK9745; AGP1, TK8195-TK8196; RPS5, TK493-TK494; and SCR1, TK9507-TK10081.

Results

A reporter system for analyzing Taf1p/TFIID- and/or Spt3p/SAGA- dependent CE function

We first sought to use the CYC1 promoter to measure the activities of SAGA-dependent transcription from various CE sequences isolated by a random screen, in which this promoter was used as a backbone [51]. Consistent with a previous study [59], we confirmed that the Δspt3 mutation significantly decreased CYC1 transcription (S1 Fig). However, the effect of the taf1 mutation was stronger than that of Δspt3: the former abolished the transient increase of CYC1 transcription occurring at a later phase after temperature shift, whereas the latter failed to do so (S1 Fig). Therefore, it is likely that TFIID plays a more predominant and indispensable role in CYC1 transcription than SAGA. If so, this promoter would not be adequate for the purpose of examining how efficiently SAGA mediates transcription from TATA-like element(s) in a manner independent of TFIID.

Fortuitously, we found that the AGP1 promoter was more suitable for our purpose than the CYC1 promoter because transcription of this gene, which encodes a low-affinity and broad-specificity amino acid permease [60, 61], decreased transiently after the temperature shift from 25°C to 37°C, and recovered in the taf1 mutant strain to a greater extent than in the wild-type strain (S1 Fig). Importantly, AGP1 transcription was abolished almost completely under the same conditions in the taf1 Δspt3 double mutant strain, indicating that the transcriptional stimulation observed after the temperature shift in the taf1 mutant strain must be supported by Spt3p/SAGA (S2 Fig). This situation is in stark contrast to CYC1 transcription, which predominantly depends on Taf1p/TFIID, although it may be facilitated by Spt3p/SAGA (S1 and S2 Figs).

Two possible models could explain the function of TFIID and SAGA (S2C Fig). Namely, given that SAGA and TFIID bind to the UAS and core promoter, respectively [20, 21], SAGA may function upstream of (left panel) or in parallel to TFIID (right panel). Recent genome-wide studies [20, 21] support a sequential model (left panel), as transcription of nearly all class II genes depends on both TFIID and SAGA, whereas previous genome-wide studies [7, 34] supported a parallel model (right panel), as severe transcriptional defects were observed only when both factors were impaired. Currently, it remains unknown whether there are distinct types of promoters on which TFIID and SAGA function differently, as described here. Nevertheless, to determine how efficiently SAGA mediates transcription from TATA-like element(s) independently of TFIID, parallel-type promoters would be more adequate than sequential-type promoters because SAGA is more directly involved in transcription on the former than on the latter. The results described in S1 Fig suggest that the CYC1 and AGP1 promoters belong to the sequential and parallel types, respectively (S2A and S2B Fig). Therefore, we decided to measure Spt3p/SAGA-dependent transcriptional activities of various CE sequences using the AGP1 promoter.

To this end, we first tested whether Spt3p/SAGA-dependent transcription from the AGP1 promoter could be recapitulated using the VTC1 reporter system (Fig 1) [51, 62]. In these experiments, promoter activities were assessed by Northern blot analyses for VTC1 mRNA, instead of by toluidine blue (TB) staining for accumulated polyphosphate in the vacuole, because Northern blotting enables more accurate measurement of transcription [25]. The results revealed that transcriptional recovery of the reporter gene driven by the AGP1 promoter was stronger in the taf1 strain than in the wild type after the temperature shift, consistent with the expression kinetics of the endogenous AGP1 mRNA measured in the same cells (Fig 1A and 1C). We also confirmed that this transcriptional stimulation of the AGP1 promoter after the temperature shift was dependent on Spt3p/SAGA in the VTC1 reporter strain (S3 Fig). Furthermore, as expected based on our previous study [51], we confirmed that Taf1p/TFIID-dependent transcription from the CYC1 promoter could also be recapitulated using this system, as the expression kinetics of the VTC1 and endogenous CYC1 mRNAs were almost similar to each other (Fig 1A and 1B). It should be noted that the transient decrease in the level of VTC1 (reporter) mRNA after the temperature shift was greater than that of endogenous CYC1 mRNA (Fig 1A and 1B). Although the cause of this difference is currently unknown, it may be related to differences in the stabilities of these two mRNAs. Collectively, these observations indicate that this system is suitable for analyzing CE function under conditions in which transcriptional activities depend predominantly on Taf1p/TFIID (CYC1 promoter) or Spt3p/SAGA (AGP1 promoter).

thumbnail
Download:
Fig 1. Effect of the taf1 mutation on the expression kinetics of the CYC1 and AGP1 promoters, tested using the VTC1 reporter system.

A. Northern blot analyses to measure expression of the reporter (VTC1) or control genes (CYC1, AGP1, RPS5, and SCR1) in TAF1 (hereafter, WT) or temperature-sensitive taf1-N568Δ (hereafter, taf1) strains carrying a reporter driven by the CYC1 (left panel) or AGP1 (right panel) promoter, as indicated above the blots. The strains used were YTK16396, YTK16397, YTK17952, and YTK17974 (S1 Table). These strains were grown to logarithmic phase at 25°C in YPD media (0 min) or further incubated at 37°C for the period of time indicated below the blots (5, 10, 15, 30, 60, 90, or 120 min). Gene-specific probes are indicated on the right. B. Raw data shown in the left panel of A were quantified and presented graphically. Values for each transcript (left: VTC1; center: CYC1; right: RPS5) from WT (closed squares connected by a solid line) or taf1 (closed circles connected by a dashed line) strains were normalized against the level of SCR1 (pol III transcript), and are presented relative to the corresponding values in the WT at 0 min (i.e., the VTC1/SCR1, CYC1/SCR1, and RPS5/SCR1 ratios in the leftmost lane for the WT strain are defined as 1 in each graph). RPS5 mRNA was used as a control to confirm that the temperature shift experiment was conducted appropriately. C. Raw data shown in the right panel of A, summarized as described in B.

https://doi.org/10.1371/journal.pone.0188435.g001

Spt3p/SAGA mediates transcription from TATA-less promoters

Previously, we showed that, in comparison with many other promoters in the yeast genome, the CYC1 promoter more strongly prefers active CE sequences belonging to class II or class V [51]. Hence, we measured the transcriptional activities of several class II or V sequences in the CYC1 or AGP1 promoters (Fig 2). In the CYC1 promoter, the activities of these sequences as measured by Northern blot analyses (Fig 2A) were well correlated with previous measurements made by TB staining (S4 Fig) [51]. Notably, all of the four class II sequences (II-#1, 2, 3, 4) and at least two class V sequences (V-#1, 4) exhibited significantly (p value < 0.01, one-way ANOVA followed by post-hoc analysis with Tukey's honest significant difference test) stronger CE activities than the negative control (TAGCGCAAW; lane 2) even in the AGP1 promoter (Fig 2). Furthermore, these class II or V sequences had slightly weaker activities (calculated relative to the activity of the consensus TATA box; TATATAAAWI-#1) when they were tested in AGP1 than in CYC1 (compare black and gray bars in Fig 2B). These observations suggest that SAGA can mediate transcription not only from the TATA box but also from class II or V sequences, albeit less efficiently for the class V sequences. Alternatively, considering that both factors are involved in transcription from the AGP1 promoter (S2 Fig), it remains possible that SAGA mediates transcription only from the TATA box, whereas TFIID does so from class II or V sequences.

thumbnail
Download:
Fig 2. Transcriptional activities of several class II and class V core promoter element (CE) sequences in the CYC1 and AGP1 promoters.

A. Northern blot analyses to examine expression of the reporter (VTC1) or control genes (RPS5 and SCR1) in a WT strain carrying a reporter driven by the CYC1 or AGP1 promoter, as indicated on the left, in which the TATA box was replaced with one of several CE sequences of class II (TATAWAD, lanes 3–6) or V (TTAAAW, lanes 7–11), as indicated above the blots. TATATAAA (consensus TATA box, lane 1) or TAGCGCAA (lane 2) sequences were used as positive and negative controls, respectively. The portion of each sequence matching the consensus sequence of each class (i.e., TATAWAWR, TATAWAD, and TTAAAW) is underlined. “W” at the 3’-end of each CE sequence above the blot corresponds to “A” or “T” in the CYC1 or AGP1 promoter, respectively (i.e., the ninth base following the thickly underlined 8 bp sequence in S5 Fig). CYC1 promoter-containing strains were YTK16396, 16398, 16408, 16410, 16412, 16414, 16400, 16402, 16404, 16406, and 16416. AGP1 promoter-containing strains were YTK17952, 17954, 17964, 17966, 17968, 17970, 17956, 17958, 17960, 17962, and 17972 (S1 Table). These strains were grown to logarithmic phase at 25°C in YPD media. Gene-specific probes were indicated on the right. B. Raw data shown in A were quantified and presented graphically. Values for each transcript (left: VTC1; right: RPS5) from strains carrying CYC1 (black bars) or AGP1 (gray bars) promoter-driven reporters were normalized against the level of SCR1 (pol III transcript) and are presented relative to the value for the TATA box (TATATAAAWI-#1, lane 1). Each bar represents the average of biological triplicates, with standard deviation. One representative experiment of three is shown in A. Asterisks in the left panel (lanes 8, 9, and 11) indicate statistically insignificant differences (p > 0.01, one-way ANOVA followed by post-hoc analysis with Tukey's honest significant difference test) relative to the value in the negative control (lane 2).

https://doi.org/10.1371/journal.pone.0188435.g002

To clarify this issue, we measured transcriptional activities of the four class V sequences (V-#1, 2, 3, 4) in the CYC1 or AGP1 promoters under the condition in which Taf1p/TFIID function was intact (WT) or impaired (taf1) (Fig 3). The results showed that functional impairment of Taf1p/TFIID decreased the CE activities of these four sequences in the CYC1 promoter (Fig 3A and 3B) but not in the AGP1 promoter (Fig 3A and 3C). As expected, RPS5 transcription was greatly reduced in the taf1 strain (control), confirming that Taf1p/TFIID function was impaired as expected under this experimental condition. Therefore, we conclude that Spt3p/SAGA can mediate transcription from the TATA-less promoters, at least those containing class V sequences such as TATTTAAAWV-#1 (lane 21 in Fig 3C) or TACTTAAAWV-#4 (lane 24 in Fig 3C), independently of Taf1p/TFIID.

thumbnail
Download:
Fig 3. Taf1-dependence or -independence of transcription from several class V CE sequences in the CYC1 and AGP1 promoters.

A. Northern blot analyses to examine expression of the reporter (VTC1) or control genes (RPS5 and SCR1) in WT or taf1 strains carrying a reporter driven by the CYC1 or AGP1 promoter in which the TATA box was replaced by one of several CE sequences of class V (TTAAAW, lanes 3–6, 9–12, 15–18, and 21–24), as indicated above the blots. TATATAAA (consensus TATA box: lanes 1, 7, 13, and 19) and TAGCGCAA (lanes 2, 8, 14, and 20) sequences were used as positive and negative controls, respectively. The portion of each sequence matching the consensus sequence of each class (i.e., TATAWAWR, TTAAAW) is underlined. “W” at the 3’-end of each CE sequence is as described in Fig 2A. CYC1 promoter-containing strains were YTK16396, 16398, 16400, 16402, 16404, 16406, 1639, 16399, 16401, 16403, 16405, and 16407. AGP1 promoter-containing strains were YTK17952, 17954, 17956, 17958, 17960, 17962, 17974, 17976, 17978, 17980, 17982, and 17984 (S1 Table). These strains were grown to logarithmic phase at 25°C in YPD media (indicated as “25°C” on the left), or further incubated at 37°C for 2 hours (lanes 1–12) or 1 hour (lanes 13–24) (indicated as “37°C” on the left). Gene-specific probes are indicated on the right. B. Raw data shown in the left panel of A were quantified and presented graphically. Values for each transcript (left: VTC1; right: RPS5) from WT or taf1 strains cultured at 25°C (black bars) or 37°C (gray bars) were normalized against the level of SCR1 (pol III transcript), and are presented relative to the value for the TATA box at 25°C (i.e., the VTC1/SCR1 or RPS5/SCR1 values for TATATAAAWI-#1 indicated to the left of lane 1). C. Raw data shown in the right panel of A are summarized as described in B.

https://doi.org/10.1371/journal.pone.0188435.g003

Taf1p/TFIID-dependence or -independence of transcription cannot be altered by variation of the TATATANN sequence

Previous studies showed that whether the creation of a TATA box could decrease the Taf1p-dependence of transcription from TATA-less promoters depended on promoter context, e.g., it could restore transcription of the TUB2 promoter [58] but not that of the RPS5 promoter [63, 64] in taf1 strains. Moreover, another study reported that randomization of a very short (2 bp) flanking sequence of the TATA box greatly affected transcription from the ENO2 and PDC1 promoters [29], supporting the idea that TATA function depends on promoter context. Furthermore, the two TATA boxes in the CYC1 promoter, i.e., TATAß [TATATATA] and TATAα [TATATAAA], were shown to be functionally different [65], even though this difference could not be recapitulated in the VTC1 reporter system [51]. As shown in Figs 1–3 and S3 Fig, the TATATAAAWI-#1 sequence derived from the CYC1 promoter was used as the TATA box even for the AGP1 promoter whose original TATA box was TATATAAG (S5 Fig). These two TATA boxes could support Taf1p-independent transcription from the AGP1 promoter (Figs 1 and 3; S1 and S3 Figs). However, given that Taf1p-dependence [58, 63, 64] or some as-yet-uncharacterized function [29, 65] of the TATA box depends on the promoter context or the sequence itself, as described above, it remains possible that some minor variation of the TATA sequence may influence the Taf1p-dependence or -independence of transcription from the CYC1 and AGP1 promoters.

To explore this possibility, we measured transcriptional activities of seven TATATANN sequences in each promoter in cells in which Taf1p/TFIID function was intact (WT) or impaired (taf1) (Fig 4). The TATATANN sequence was chosen for these studies because the difference between TATAß [TATATATA] and TATAα [TATATAAA] is located within the two 3’-terminal bases “NN”. Specifically, the same set of four TATATANN sequences (NN: AG [I-#2], AA [I-#1], AC [II-#4C], AT [II-#4]) were tested in both promoters, whereas the remaining three differed slightly between promoters, as follows: NN = TA [I-#3], GA [II-#1A], or CA [VI-#1CA] for the CYC1 promoter; and NN = TG [I-#4], GG [II-#1G], or CG [VI-#1CG] for the AGP1 promoter. Thus, the most 3’-terminal nucleotide (A or G) was replaced by the original one (NN: AA in CYC1 and AG in AGP1) (S5 Fig). Note that the identification tags in square brackets were assigned according to CE class [i.e., I (TATAWAWR), II (TATAWAD), or VI (Wx6)], as well as to the content of the dinucleotide sequence (NN), to enable discernment of which sequences were identical or different in Figs 2, 3 and 4.

thumbnail
Download:
Fig 4. Taf1-dependence or -independence of transcription from several TATATANN sequences of classes I, II, and VI in the CYC1 and AGP1 promoters.

A. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (CYC1, AGP1, RPS5, and SCR1) in WT or taf1 strains carrying a reporter driven by the CYC1 or AGP1 promoter in which the TATA box was replaced with one of several TATATANN sequences of class I (TATAWAWR: lanes 1, 2, 4, 9, 10, 12, 17, 18, 20, 25, 26, and 28), II (TATAWAD: lanes 5–7, 13–15, 21–23, and 29–31), or VI (Wx6: lanes 8, 16, 24, and 32), as indicated above the blots. TATATAAA (CYC1-TATA; lanes 1, 9) / TATATAAG (AGP1-TATA; lanes 17, 25) and TAGCGCAA (lanes 3, 11, 19, 27) sequences were used as positive and negative controls, respectively. The portion of each sequence matching the consensus sequence of each class (i.e., TATAWAWR, TATAWAD, Wx6) is underlined. “W” at the 3’-end of each CE sequence is as described in Fig 2A. Identification tags for each sequence are indicated as a superscript according to CE class (i.e., I, II, or VI) and the dinucleotide sequence NN at the 3’-end of TATATANN to enable discernment of which sequences are identical or different among Figs 2, 3 and 4. CYC1 promoter-containing strains were YTK16396, 17566, 16398, 17562, 17560, 17564, 17568, 17558, 16397, 17578, 16399, 17574, 17572, 17576, 17580, and 17570. AGP1 promoter-containing strains were YTK17590, 17952, 17954, 17586, 17584, 17591, 17593, 17582, 17603, 17974, 17976, 17599, 17597, 17604, 17606, and 17595 (S1 Table). These strains were grown to logarithmic phase at 25°C in YPD media (indicated as “25°C” on the left) or further incubated at 37°C for 2 hours (lanes 1–16) or 1 hour (lanes 17–32) (indicated as “37°C” on the left). Gene-specific probes were indicated at the right. “endo” indicates CYC1 (left panel) or AGP1 (right panel) probes. B. Raw data shown in the left panel of A were quantified and presented graphically. Values for each transcript (left: VTC1; center: CYC1; right: RPS5) from WT or taf1 strains cultured at 25°C (black bars) or 37°C (gray bars) were normalized against the level of SCR1 (pol III transcript) and are presented relative to the value for the TATA box at 25°C (i.e., the VTC1/SCR1, CYC1/SCR1, or RPS5/SCR1 ratios of TATATAAAWI-#1 indicated to the left of lane 1). C. Raw data shown in the right panel of A are summarized as described in B, except that the values are presented relative to the value for TATATAAGWI-#2 at 25°C indicated to the left of lane 17.

https://doi.org/10.1371/journal.pone.0188435.g004

The results, categorized by class (I, II, and VI) (Fig 4), showed that transcription from all active CE sequences in the CYC1 promoter was Taf1p-dependent, while that from those in the AGP1 promoter was Taf1p-independent. Namely, no TATATANN sequences, as far as we examined here, could alter the Taf1p-dependence or -independence of transcription from these two promoters. Therefore, we conclude that Spt3p/SAGA can mediate transcription from the AGP1 promoter carrying CE sequences belonging not only to class V (Fig 3) but also to class II or VI (Fig 4) independently of Taf1p/TFIID. Notably, in the AGP1 promoter the TATATACGWVI-#1CG sequence (lanes 24 and 32) exhibited weak activity that was nonetheless significantly stronger than that of the negative control (TAGCGCAAW) (lanes 19 and 27). TATATAC-containing sequences were rarely isolated in the original screen [51], as inferred from the class II consensus sequence (i.e., TATAWAD). Consistent with this, in the CYC1 promoter the TATATACAWVI-#1CA sequence (lanes 8 and 16) did not exhibit stronger activity than the negative control (lanes 3 and 11). These observations suggest that TATATAC might be a Spt3p/SAGA-specific CE motif.

Taf1p/TFIID-dependence or -independence is conferred by the UAS

Previous work showed that the Taf1p-dependence of transcription could be conferred by the core promoter [64, 66], the UAS [67], or both [22, 63], depending on the promoter context. The results described above indicated that Taf1p-dependence or -independence is highly robust with respect to variation of the TATA sequence, at least in the CYC1 or AGP1 promoters (Fig 4). Hence, we next sought to determine which region, e.g., the core promoter or UAS, could confer Taf1p-dependence or -independence on these two promoters. For this purpose, we constructed two additional chimeric promoters, i.e., AGP1UAS+CYC1core and CYC1UAS+AGP1core, by connecting AGP1UAS [-780 –-301 bp] or CYC1UAS [-400 –-231 bp] to CYC1core [-230 –+1 bp] or AGP1core [-300 –+1 bp], respectively [68, 69]. Transcriptional activities of these two chimeric promoters, as well as those of the original CYC1 and AGP1 promoters, were measured in cells in which Taf1p/TFIID function was intact (WT) or impaired (taf1) (Fig 5). The results clearly showed that the AGP1UAS+CYC1core promoter was more active than the AGP1 promoter, and that both promoters were expressed in a Taf1p-independent manner (lanes 1–2 and 9–10). However, it was unclear whether the CYC1UAS+AGP1core promoter was expressed in a Taf1p-dependent manner (lanes 5–6), similar to the CYC1 promoter (lanes 13–14), because the activity of the CYC1UAS+AGP1core promoter was weaker than that of the CYC1 promoter under these experimental conditions.

thumbnail
Download:
Fig 5. Taf1p-dependence or -independence of transcription is conferred by the upstream activating sequence (UAS) of the CYC1 or AGP1 promoters.

A. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (RPS5 and SCR1) in WT or taf1 strains carrying a reporter driven by the CYC1 promoter (i.e., CYC1UAS [-400 –-231 bp] + CYC1core [-230 –+1 bp], lanes 13–16), the AGP1 promoter (i.e., AGP1UAS [-780 –-301 bp] + AGP1core [-300 –+1 bp], lanes 1–4), or chimeric promoters such as CYC1UAS+AGP1core (lanes 5–8) or AGP1UAS+CYC1core (lanes 9–12), as indicated above the blots. Nucleotide coordinates are relative to the A (+1) of the initiation codon ATG. All four promoters contain TATATAAA (lanes 1–2, 5–6, 9–10, 13–14) or TAGCGCAA (lanes 3–4, 7–8, 11–12, 15–16) sequences at the TATA site, as indicated below the index for “UAS” and “core”. TAF1 strains were YTK17952, 17954, 17766, 17769, 17748, 17750, 16396, and 16398; taf1 strains were YTK17974, 17976, 17802, 17804, 17784, 17786, 16397, and 16399 (S1 Table). These strains were grown to logarithmic phase at 25°C in YPD media (indicated as “25” above the blot) or further incubated at 37°C for 2 hours (indicated as “37” above the blot). Gene-specific probes are indicated at the right. B. Raw data shown in A were quantified and presented graphically. Values for each transcript (left: VTC1; right: RPS5) derived from WT (black bars) or taf1 (gray bars) strains were normalized against the level of SCR1 (pol III transcript), and are presented relative to the value for the TATA box of the AGP1 promoter at 25°C (i.e., the VTC1/SCR1 or RPS5/SCR1 values indicated to the left of lane 1).

https://doi.org/10.1371/journal.pone.0188435.g005

To clarify this point, we measured the activities of these promoters in media containing raffinose instead of glucose. As expected, transcription of the two promoters containing CYC1UAS was stimulated by raffinose, presumably due to release from glucose repression (compare lanes 1–4 and 9–12 in S6A and S6B Fig) [70]. Next, the activity of the CYC1UAS+AGP1core promoter was measured again in cells in which Taf1p/TFIID function was intact (WT) or impaired (taf1) (S6C and S6D Fig). Unexpectedly, transcription from this promoter was not stimulated by raffinose in the taf1 strain, even at 25°C (lanes 5 and 13 in S6C and S6D Fig). However, transcription from the CYC1 promoter could be stimulated by raffinose in the taf1 strain at 25°C, but not at 37°C (data not shown). These observations suggest that activation of either AGP1core or CYC1core by CYC1UAS is Taf1p-dependent, but the AGP1core requires more integral Taf1p function than the CYC1core. More importantly, the results obtained after optimizing the experimental conditions (e.g., using fresh probe and/or exposing for a longer time) clearly showed that the CYC1UAS+AGP1core promoter could be expressed in a Taf1p-dependent manner (lanes 5–6 in S6C and S6D Fig), similar to the CYC1 promoter (lanes 13–14 in Fig 5). Based on these findings, we conclude that the Taf1p-dependence or -independence of these two promoters was conferred by CYC1UAS or AGP1UAS, respectively.

Core promoter sequences other than the TATA box may determine the function of Spt3p/SAGA on the AGP1 promoter after inactivation of Taf1p/TFIID

As described above, transcription from the AGP1 promoter was stimulated to a greater extent in the taf1 strain than in the wild type after the temperature shift (Fig 1). In addition, the chimeric AGP1UAS+CYC1core promoter, which is more active than the AGP1 promoter, was also expressed in a Taf1p-independent manner (Fig 5). Hence, we next investigated whether transcription from the AGP1UAS+CYC1core promoter exhibited similar kinetics to that of the AGP1 promoter after the temperature shift (Fig 6). Surprisingly, the results revealed that the expression kinetics of these two promoters were quite different. Namely, stronger taf1 strain-specific transcriptional stimulation after the temperature shift was reproducibly observed for the AGP1 promoter (Fig 6A and 6B), but not for the AGP1UAS+CYC1core promoter (Fig 6A and 6C). Transcription from the AGP1UAS+CYC1core promoter was instead weakened by the taf1 mutation under the same conditions (Fig 6A and 6C).

thumbnail
Download:
Fig 6. Spt3p-dependent transcriptional stimulation after the temperature shift in the taf1 strain occurs specifically on the AGP1 core promoter.

A. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (AGP1, RPS5, and SCR1) in WT or taf1 strains carrying a reporter driven by the AGP1 (left panel) or chimeric AGP1UAS+CYC1core (right panel) promoter. Experiments were conducted as described in Fig 1A. The strains used were YTK17952 and 17974 (left panel) and YTK17748 and 17784 (right panel). B. Raw data shown in the left panel of A were quantified and presented graphically as described in Fig 1B. Note that transcriptional stimulation in the taf1 strain after the temperature shift was highly reproducible, i.e., the results were consistent with those shown in Fig 1C. C. Raw data shown in the right panel of A were quantified and presented graphically as described in Fig 1B. Note that transcriptional stimulation of the CYC1 core promoter after the temperature shift was not observed in the taf1 strain.

https://doi.org/10.1371/journal.pone.0188435.g006

Fig 3 shows that SAGA could mediate transcription from two class V CE sequences, i.e., TATTTAAAWV-#1 and TACTTAAAWV-#4. To determine whether the TATA box is required for Spt3p/SAGA-dependent transcriptional stimulation in the taf1 strain after the temperature shift, we examined the expression kinetics of transcription from these two CEs, as described above for the AGP1 and AGP1UAS+CYC1core promoters (Fig 7A–7C). The results revealed a significant taf1 strain-specific transcriptional stimulation for both CEs, particularly when the relative ratios of the value obtained at 60 min to that at 15 min were compared between WT and taf1 strains (Fig 7D). Therefore, we conclude that, although Taf1p/TFIID-independence of AGP1 transcription is conferred by AGP1UAS, Spt3p/SAGA-dependent transcriptional stimulation of this promoter after inactivation of Taf1p/TFIID is directed by AGP1core sequences other than the TATA element.

thumbnail
Download:
Fig 7. Spt3p-dependent transcriptional stimulation after the temperature shift in the taf1 strain occurs on AGP1 core promoters whose TATA boxes were replaced with class V CE sequences.

A. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (RPS5 and SCR1) in WT or taf1 strains carrying a reporter driven by the AGP1 promoter in which the TATA box was replaced with TATTTAAA (left panel) or TACTTAAA (right panel) sequences. Experiments were conducted as described in Fig 1A. The strains used were YTK17956 and 17978 (left panel), and YTK17962 and 17984 (right panel). B. Raw data shown in the left panel of A were quantified and presented graphically as described in Fig 1B, except that measurement of an endogenous gene was omitted. C. Raw data shown in the right panel of A were quantified and presented graphically as described in Fig 1B, except that the measurement of an endogenous gene was omitted. D. Normalized VTC1/SCR1 ratios at 15 min (open bars) or 60 min (gray bars) after the temperature shift, adapted from the plots in Figs 1C (first four lanes), 6B (second four lanes), 7B (third four lanes) and 7C (fourth four lanes), are summarized in this figure to confirm that Spt3p-dependent transcriptional stimulation occurs on the AGP1 promoter regardless of whether it contains a TATA box (TATATAAA) or a TATA-like element (TATTTAAA or TACTTAAA). In each set, the values are presented relative to those of the WT at 0 min (defined as 1). Note that the ratios of the value at 60 min to the value at 15 min are always higher in the taf1 strain than in the WT, indicating that Spt3p-dependent transcriptional stimulation after the temperature shift occurred on all promoters tested here.

https://doi.org/10.1371/journal.pone.0188435.g007

Discussion

Previously, we showed that the chimeric UASGAL+CYC1 promoter on a plasmid could be activated by various types of activation domains (e.g., those derived from Abf1p, Gal4p, Gcn4p, Adr1p, Rap1p, TAND1, EBNA2, or VP16) in a Taf1p-independent manner, unless the TATA sequence was mutated [58, 63]. Consistent with this, other groups also reported that the CYC1 promoter belongs to a SAGA-dependent and TFIID-independent class [59, 66]. However, the criteria used in those studies were not especially stringent: one group tested the Taf1p-dependence of transcription only under mild conditions (e.g., 30°C in the taf1 strain) [66], whereas the other group did not test the Taf1p-dependence of transcription itself, but instead examined Taf1p occupancy on this promoter [59]. On the other hand, genome-wide expression analyses demonstrated that the CYC1 promoter is TFIID-dominated and not SAGA-dominated [7]. In a more recent study, we showed that the CYC1 promoter was Taf1p-dependent at the VTC1 reporter locus, but not at the endogenous CYC1 locus [51]. During the course of the current study, however, we noticed that the data for CYC1 mRNA derived from the CYC1 locus obtained at that time, cited as “data not shown” in the original reference [51], were contaminated with the signal from SCR1 RNA due to insufficient washing of the Northern blot prior to reprobing. After repeating the same experiment several times, we are now convinced that the CYC1 promoter is Taf1p-dependent even at the endogenous CYC1 locus (S1 Fig), whereas it is Taf1p-independent in the context of the chimeric UASGAL+CYC1 promoter on a plasmid. Although it remains unclear why only the chimeric UASGAL+CYC1 promoter exhibited Taf1p-independence, it is possible that Taf1p/TFIID function is specifically required for activity of the CYC1 promoter in the chromatin environment. In fact, recent studies suggest that the +1 nucleosome or its repositioning may differentially affect the two PIC assembly pathways mediated by TFIID and SAGA [8, 71, 72]. Consistent with this, TBP binding to the CYC1 promoter is stimulated by activators when it is on a plasmid [73] but not when it is integrated into the genome [74], suggesting that TBP preloading and pol II release from the poised state upon activation, which are characteristic features of this promoter [59, 75], depend on the chromatin environment and Taf1p/TFIID function.

SAGA mediates transcription from the TATA-like element independently of TFIID

In this study, we set up a VTC1 reporter system using the AGP1 promoter to test for Spt3p/SAGA-dependent transcription from various types of CE sequences without the involvement of Taf1p/TFIID. In this system, following the temperature shift to inactivate the mutated version of Taf1p/TFIID, transcription transiently decreased and then recovered to a much greater extent in the taf1 strain than in the WT in a Spt3p/SAGA-dependent manner (Fig 1, S3 Fig). Because an increase in mRNA level strongly implies de novo mRNA synthesis, the accumulation of larger amounts of mRNA in the taf1 strain could be regarded as a bona fide product of Spt3p/SAGA-mediated transcription. Thus, even though the amount of newly synthesized mRNA [18, 46–49, 72, 76] was not measured, this system minimizes the possibility that mRNA remained due to elevated stability without any increase in transcription. Using this novel system, we found that several CE sequences belonging not only to class I (consensus TATA box; TATATAAAWI-#1, TATATAAGWI-#2, TATATATGWI-#4) but also to class II (TATAWAD; TATATAGGWII-#1G, TATATAACWII-#4C, TATATAATWII-#4), class V (TTAAAW; TATTTAAAWV-#1, TACTTAAAWV-#4), or class VI (Wx6; TATATACGWVI-#1CG) (the region of each sequence matching the consensus in each class is underlined) were transcriptionally active in the AGP1 promoter even after inactivation of Taf1p/TFIID. These observations provide strong evidence that Spt3p/SAGA mediates transcription from the TATA-like element independently of Taf1p/TFIID.

Very recent genome-wide studies showed that TFIID and SAGA bind to the core promoter region and UAS, respectively, and mediate transcription from the promoters of nearly all class II genes [19–21]. In these studies, factor binding was monitored by ChEC-seq (chromatin endogenous cleavage coupled with high-throughput sequencing) [77], whereas transcriptional activity was assessed by measuring the amount of newly synthesized mRNA. By contrast, the longstanding and widely accepted view that TFIID and SAGA function predominantly in transcription from TATA-less and TATA box-containing promoters, respectively, was established based on measurements of steady-state mRNA levels [7]. As noted above, a buffering effect maintains global cytoplasmic mRNA levels upon transcriptional impairment [18, 20, 21, 46–49]. Thus, transcriptional defects of TATA-less or TATA box-containing promoters are likely to be restored unevenly by such an effect in TFIID or SAGA-defective strains. More specifically, because the defects of TATA-less promoters may be restored more completely in a SAGA-defective strain than in a TFIID-defective strain, the defects would be observable specifically in the TFIID-defective strain, whereas the defects of TATA box-containing promoters may be restored in the opposite fashion. If so, the mechanisms by which TFIID and SAGA mediate transcription differ for these two types of promoters, at least from the standpoint of “restorability” by such a buffering effect. Consistent with this, SAGA yielded stronger ChEC-seq signals at more upstream regions relative to the initiation site on TATA box-containing promoters than on TATA-less promoters [21]. Furthermore, depletion of Taf4p/TFIID from the nucleus decreased the amounts of newly synthesized mRNA to a greater extent from TATA-less promoters than from TATA box-containing promoters [20]. Therefore, as a next step, it will be crucial to more precisely determine the mechanisms by which TFIID and SAGA mediate transcription from these two types of promoters. The CYC1 and AGP1 promoters would provide useful model systems for this purpose because TFIID and SAGA function very differently on these two promoters, even if they both originally contained TATA boxes (S2 Fig).

SAGA mediates transcription in a manner dependent on a core promoter structure other than the TATA box

Our mapping experiments using chimeric promoters revealed that the Taf1p-dependence and -independence of transcription from the CYC1 and AGP1 promoters, respectively, was conferred by the UAS rather than the core promoter (Fig 5, S6 Fig). Similarly, Taf1p-dependence was previously mapped to the UAS by comparing the LexA-RPS5core (Taf1p-independent) and RPS5UAS-RPS5core (Taf1p-dependent) promoters [67]. By contrast, Taf1p-dependence was mapped to the core promoter based on comparison between RPS5UAS-ADH1core (Taf1p-independent) and RPS5UAS-RPS5core (Taf1p-dependent) promoters [64, 67]. Comparison of the four chimeric promoters 2xGAL4 sites-GAL1core (Taf1p-independent), RPS5UAS-GAL1core (Taf1p-independent), 2xGAL4 sites-RPS5core (Taf1p-dependent), and RPS5UAS-RPS5core (Taf1p-dependent) also supported the latter conclusion that the core promoter is responsible for Taf1p-dependence [66]. Alternatively, Taf1p-dependence could be also conferred by the combination of the UAS and core promoter [22, 63].

In these studies, Taf1p-dependence of activation by RPS5UAS was observed only for the TATA-less RPS5core [64, 66, 67], but not for the TATA-containing ADH1core [64, 67], GAL1core [66], or CYC1core [63]. A recent study showed that transcriptional repression of ribosomal protein genes (RPGs) under stressed conditions (including temperature shift, often used for inactivation of temperature-sensitive Taf proteins) occurs through the eviction of PIC and the Hmo1p-Ifh1p-Sfp1p complex, followed by repositioning of the +1 nucleosome to the upstream PIC assembly site [78, 79]. Because RPG transcription usually recovers within 1 hour after the temperature shift (Fig 1), Taf1p-dependence of RPS5core activation by RPS5UAS is likely to result from competitive binding of TFIID to overcome the repressive effect by repositioning the +1 nucleosome. Given that Taf1p was also recruited on ADH1core by RPS5UAS [67], Taf1p-independent ADH1core/GAL1core/CYC1core activation by RPS5UAS could be explained by a model in which, once TFIID is evicted by RPS5UAS under temperature stress condition, TBP alone delivered by SAGA binds to these TATA-containing core promoters, probably because the TATA sites are vacant due to inappropriate repositioning of the +1 nucleosome. Consistent with this model, ADH1core activation by RPS5UAS is dependent on Taf6p, a common subunit of TFIID and SAGA [67]. Similar to RPS5 transcription, a transient decrease in activity and Taf1p-dependent recovery after the temperature shift was also observed for the CYC1 promoter (Fig 1). However, in the case of this promoter, Taf1p-dependence was mapped to CYC1UAS instead of CYC1core. Furthermore, CYC1core and AGP1core both contain TATA boxes, and minor variation in the TATA sequence did not affect the Taf1p-dependence or -independence of these two promoters (Figs 3 and 4). Therefore, the mechanisms responsible for conferring Taf1p-dependence on the RPS5 and CYC1 promoters must be quite different.

Intriguingly, Spt3p/SAGA-dependent transcriptional stimulation of the AGP1 promoter in the taf1 strain after the temperature shift was specific to AGP1core (Fig 6). Furthermore, this transcriptional stimulation was not compromised by the substitution of the TATA box with either of the two TATA-like elements (Fig 7), indicating that a region other than the TATA box is a critical determinant for the function of Spt3p/SAGA on this promoter. Two possible models could explain these observations. SAGA could specifically recognize AGP1core to stimulate transcription after inactivation of TFIID. Alternatively, SAGA itself could be non-selective for CYC1core or AGP1core, but inactivated Taf1p/TFIID could somehow inhibit SAGA specifically on CYC1core so that SAGA cannot stimulate transcription from that sequence. Unfortunately in this regard, Taf1p occupancy was very low at the CYC1 promoter [8, 59]. In addition, SAGA is usually recruited to the UAS, and its occupancy at the core promoter is difficult to detect, probably due to its highly dynamic nature [6, 18, 20, 21, 39, 80, 81]. Therefore, it may be difficult to distinguish between these two possibilities.

A recent study demonstrated that the difference in “regulatability” (regulatory amplitude conferred by the activator) [82] between TFIID-dominated and SAGA-dominated genes was due to the core promoter type rather than the number or type of activators [72]. Notably, the TATA box is not the sole determinant of core promoter type in regard to regulatability [72]. Consistent with this, another recent study using an in vitro transcription system suggested that some other features, in addition to the TATA box, may determine the difference in the ability of TBP to substitute for TFIID on TATA box-containing vs. TATA-less promoters [28]. In addition, the core promoters of TFIID- and SAGA-dominated genes have different free-energy landscapes [83, 84]. Considering that TFIID and SAGA bind to the promoters of nearly all class II genes [20, 21], it is likely that the function of these two factors is regulated by the core promoter structure at certain post-recruitment step(s).

In summary, we showed here that AGP1UAS activated CYC1core and AGP1core in a Taf1p/TFIID-independent manner (Fig 5, S6 Fig) via different mechanisms (Fig 6). Similarly, CYC1UAS activated CYC1core and AGP1core in a Taf1p/TFIID-dependent manner (Fig 5, S6 Fig), also via different mechanisms, with the AGP1core activation requiring more integral function of Taf1p/TFIID than the CYC1core activation (data not shown). Therefore, we propose that the functions of TFIID and SAGA are regulated in two steps: first, the UAS specifies TFIID or SAGA as the predominant factor on a given promoter, and then the core promoter structure guides this factor to conduct transcription in an appropriate manner. Future studies should seek to reveal the detailed mechanisms by which TFIID and SAGA mediate transcription from TATA box-containing and TATA-less promoters, as well as the physiological significance of the promoter-specific functions of these two related transcription complexes.

Supporting information

S1 Fig. Effect of the taf1 and/or Δspt3 mutation on the expression kinetics of the endogenous CYC1 and AGP1 promoters.

A. Northern blot analyses to examine expression of CYC1, AGP1, or control genes (RPS5 and SCR1) in four strains: TAF1 SPT3 (indicated as WT in B; YTK11705), taf1-N568Δ SPT3 (indicated as taf1 in B; YTK11708), TAF1 Δspt3 (indicated as Δspt3 in B; YTK11871), or taf1-N568Δ Δspt3 (indicated as taf1 Δspt3 in B; YTK11873). The analyses were conducted as described in Fig 1A. B. Raw data shown in A were quantified and presented graphically as described in Fig 1B.

https://doi.org/10.1371/journal.pone.0188435.s001

(EPS)

S2 Fig. Model of the roles of TFIID and SAGA on the CYC1 and AGP1 promoters.

A. Model of the roles of TFIID and SAGA on the CYC1 promoter. Transcription from the CYC1 promoter was significantly weakened in the taf1 strain after the temperature shift (II), whereas it was also weakened in the Δspt3 strain, but in a temperature shift-independent manner (III) (S1 Fig). Notably, transcription increased slightly at a later phase after the temperature shift in the Δspt3 strain. Furthermore, transcription from this promoter was almost abolished in the taf1 Δspt3 strain (IV) (S1 Fig), indicating that TFIID plays a predominant and more indispensable role in CYC1 transcription, whereas SAGA plays a supportive and less indispensable role, i.e., it assists TFIID or other factors in transcription from this promoter in a WT strain (I). B. Model of the roles of TFIID and SAGA on the AGP1 promoter. Transcription from the AGP1 promoter was stimulated after the temperature shift in the taf1 strain (II) but not in the Δspt3 strain (III), whereas it was almost abolished in the taf1 Δspt3 strain (IV) (S1 Fig). These observations indicate that TFIID and SAGA play redundant and/or antagonistic roles in AGP1 transcription in the WT strain (I). Importantly, it is possible to determine whether Spt3p/SAGA mediates transcription from various CE sequences by assaying their transcriptional activities under condition II. C. Two possible models of the function of TFIID and SAGA. SAGA may function upstream of TFIID (left panel) or in parallel to TFIID (right panel). Parallel-type promoters (right panel) were more suitable for our purpose than sequential-type promoters (left panel) because SAGA is more directly involved in transcription on the former than on the latter.

https://doi.org/10.1371/journal.pone.0188435.s002

(EPS)

S3 Fig. Effect of a taf1 and/or Δspt3 mutation on the expression kinetics of the AGP1 promoter when tested using the VTC1 reporter system.

A. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (AGP1, RPS5, and SCR1) in four strains: TAF1 SPT3 (indicated as WT in B; YTK18974), taf1-N568Δ SPT3 (indicated as taf1 in B; YTK18986), TAF1 Δspt3 (indicated as Δspt3 in B; YTK18998), or taf1-N568Δ Δspt3 (indicated as taf1 Δspt3 in B; YTK19010). Analyses were conducted as described in Fig 1A. B. Raw data shown in A were quantified and presented graphically as described in Fig 1B.

https://doi.org/10.1371/journal.pone.0188435.s003

(EPS)

S4 Fig. Transcriptional profiles of several class II or V sequences tested in this study, measured previously by toluidine blue (TB) staining.

A. Four CE sequences (marked with asterisks at the left) belonging to class II (TATATAGCA [#1], TATATATCA [#2], GCTATAAAA [#3], and TATATAATA [#4] isolated by a previous screen [51]) were tested in Figs 2 or 4 as TATATAGCWII-#1, TATATATCWII-#2, GCTATAAAWII-#3, and TATATAATWII-#4, respectively. The portion of each sequence matching the consensus TATAWAD is underlined in the preceding sentence or indicated by a red rectangle in the figure. This figure is adapted from S2 Fig of the original study [51]. The color code (left panel) and normalized isolation frequency (right panel) are as described previously [51]. B. Five CE sequences (marked with asterisks at the left) belonging to class V (TATTTAAAA [#1], GTATTAAAA [#2], TTTTTAAAA [#3], TACTTAAAA [#4], and CATTTAAAA [#5]) isolated in a previous screen [51] were tested in Figs 2 or 3 as TATTTAAAWV-#1, GTATTAAAWV-#2, TTTTTAAAWV-#3, TACTTAAAWV-#4, and CATTTAAAWV-#5, respectively. The portion of each sequence matching the consensus TTAAAW is underlined in the preceding sentence or indicated by a red rectangle in the figure. This figure is adapted from S3 Fig of the original study [51]. The color code (left panel) and normalized isolation frequency (right panel) are as described previously [51].

https://doi.org/10.1371/journal.pone.0188435.s004

(EPS)

S5 Fig. Nucleotide sequences of the CYC1 and AGP1 promoters.

A. Nucleotide sequence of the CYC1 promoter. The positions of the two TATA boxes and two TATA-like elements are underlined: #1 (TATAβ), #2 (TATAα), #3, and #4. Each TATA box or TATA-like element was replaced with a specific sequence, denoted as “m”, to disrupt its transcriptional activity [51, 65]. The substituted nucleotides that differ from those in the WT are shown in bold italic font. The initiation codon ATG is marked with an open square along with the number +1 (indicating A of ATG). The arrows above the sequence indicate the transcriptional start site(s) (TSSs), which depend on TATAβ located at site #2 (marked with a thick underline). In this study, TATAβ was replaced with various CE sequences belonging to class I, II, V, or VI. B. Nucleotide sequence of the AGP1 promoter. The position of the TATA box in the AGP1 promoter is thickly underlined. This element was modified as described in A. Note that it was also replaced with TATAβ (TATATAAA) of the CYC1 promoter and used as a positive control in all experiments, except in Fig 4, in which TATATAAG (original TATA box) was also tested in parallel. TSS and ATG are marked as described in A.

https://doi.org/10.1371/journal.pone.0188435.s005

(EPS)

S6 Fig. Taf1p-dependence of transcription from the CYC1 promoter is conferred by the upstream activating sequence (UAS).

A. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (RPS5 and SCR1) in the WT strain carrying the reporter driven by the CYC1 (lanes 1, 2, 9, and 10), AGP1 (lanes 7, 8, 15, and 16), CYC1UAS+AGP1core (lanes 3, 4, 11, and 12), or AGP1UAS+CYC1core (lanes 5, 6, 13, and 14) promoters, as indicated at the top. All four promoters contain TATATAAA (odd numbered lanes) or TAGCGCAA (even-numbered lanes) sequences at the TATA site, as indicated below the index for “UAS” and “core”. Total RNA (20 μg) was isolated from these eight strains (i.e., YTK16396/16398/17766/17769/17748/17750/17952/17954) grown to a logarithmic phase at 25°C (lanes 1–8) or 30°C (lanes 9–16) in rich media containing 2% glucose (lanes 1–8) or 4% raffinose (lanes 9–16), blotted onto the membrane, and hybridized with the gene-specific probes indicated at the right. B. Raw data shown in A were quantified and presented graphically. Values for each transcript (VTC1/left graph, RPS5/right graph) were normalized against the level of SCR1 (pol III transcript) and presented relative to the value for the TATA box of the CYC1 promoter (lane 1). C. Northern blot analyses to examine the expression of the reporter (VTC1) or control genes (RPS5 and SCR1) in WT or taf1 strains carrying a reporter driven by the chimeric CYC1UAS+AGP1core promoter in which the TATA box was substituted with TATATAAA (lanes 1–2, 5–6, 9–10, 13–14) or TAGCGCAA (lanes 3–4, 7–8, 11–12, 15–16) as indicated. Total RNA (20 μg) was isolated from these four strains (i.e., YTK17766/17769/17802/17804) grown to a logarithmic phase at 25°C in rich media containing 2% glucose (lanes 1, 3, 5, and 7) or 4% raffinose (lanes 9, 11, 13, and 15) (indicated as “25” above the blot), or further incubated at 37°C for 2 hours in each type of media (even-numbered lanes; indicated as “37” above the blot), blotted onto a membrane, and hybridized with the gene-specific probes indicated at the right. Autoradiogram hybridized with a VTC1 probe and exposed for a longer time is also presented with an asterisk at the left. D. Raw data shown in C were quantified and presented graphically. Values for each transcript (VTC1/left graph, RPS5/right graph) derived from WT (black bars) or taf1 (gray bars) strains were normalized against the level of SCR1 (pol III transcript) and are presented relative to the value for the TATA box in the WT strain cultured in glucose-containing media at 25°C (i.e., the VTC1/SCR1 or RPS5/SCR1 values indicated to the left of lane 1).

https://doi.org/10.1371/journal.pone.0188435.s006

(EPS)

S1 Table. Saccharomyces cerevisiae strains used in this study.

https://doi.org/10.1371/journal.pone.0188435.s007

(DOC)

S2 Table. Oligonucleotides used in this study.

https://doi.org/10.1371/journal.pone.0188435.s008

(DOC)

S3 Table. PCR primers used for the construction of yeast strains in this study.

https://doi.org/10.1371/journal.pone.0188435.s009

(XLSX)

Acknowledgments

We thank Dr. Y. Ohyama for constructing YTK11871, YTK11873, and YTK13039. We also thank Drs. K. Kasahara, S. Ki, K. Kataoka, N. Takai, and Y. Ohyama, and other members of our laboratory, for advice and helpful discussions.

References

  1. 1. Sainsbury S, Bernecky C, Cramer P. Structural basis of transcription initiation by RNA polymerase II. Nat Rev Mol Cell Biol. 2015;16(3):129–43. pmid:25693126.
  2. 2. Grunberg S, Hahn S. Structural insights into transcription initiation by RNA polymerase II. Trends Biochem Sci. 2013;38(12):603–11. pmid:24120742; PubMed Central PMCID: PMCPMC3843768.
  3. 3. Hahn S, Young ET. Transcriptional regulation in Saccharomyces cerevisiae: transcription factor regulation and function, mechanisms of initiation, and roles of activators and coactivators. Genetics. 2011;189(3):705–36. pmid:22084422; PubMed Central PMCID: PMC3213380.
  4. 4. Sikorski TW, Buratowski S. The basal initiation machinery: beyond the general transcription factors. Curr Opin Cell Biol. 2009;21(3):344–51. pmid:19411170; PubMed Central PMCID: PMCPMC2692371.
  5. 5. Thomas MC, Chiang CM. The general transcription machinery and general cofactors. Crit Rev Biochem Mol Biol. 2006;41(3):105–78. pmid:16858867.
  6. 6. Bhaumik SR, Green MR. Differential requirement of SAGA components for recruitment of TATA-box-binding protein to promoters in vivo. Mol Cell Biol. 2002;22(21):7365–71. pmid:12370284.
  7. 7. Huisinga KL, Pugh BF. A genome-wide housekeeping role for TFIID and a highly regulated stress-related role for SAGA in Saccharomyces cerevisiae. Mol Cell. 2004;13(4):573–85. pmid:14992726.
  8. 8. Rhee HS, Pugh BF. Genome-wide structure and organization of eukaryotic pre-initiation complexes. Nature. 2012;483(7389):295–301. pmid:22258509; PubMed Central PMCID: PMC3306527.
  9. 9. Han Y, Luo J, Ranish J, Hahn S. Architecture of the Saccharomyces cerevisiae SAGA transcription coactivator complex. EMBO J. 2014;33(21):2534–46. pmid:25216679; PubMed Central PMCID: PMCPMC4283410.
  10. 10. Timmers HT, Tora L. SAGA unveiled. Trends Biochem Sci. 2005;30(1):7–10. pmid:15653319.
  11. 11. Belotserkovskaya R, Sterner DE, Deng M, Sayre MH, Lieberman PM, Berger SL. Inhibition of TATA-binding protein function by SAGA subunits Spt3 and Spt8 at Gcn4-activated promoters. Mol Cell Biol. 2000;20(2):634–47. pmid:10611242
  12. 12. Bhaumik SR. Distinct regulatory mechanisms of eukaryotic transcriptional activation by SAGA and TFIID. Biochim Biophys Acta. 2011;1809(2):97–108. pmid:20800707; PubMed Central PMCID: PMCPMC3018551.
  13. 13. Trowitzsch S, Viola C, Scheer E, Conic S, Chavant V, Fournier M, et al. Cytoplasmic TAF2-TAF8-TAF10 complex provides evidence for nuclear holo-TFIID assembly from preformed submodules. Nat Commun. 2015;6:6011. pmid:25586196; PubMed Central PMCID: PMCPMC4309443.
  14. 14. Bieniossek C, Papai G, Schaffitzel C, Garzoni F, Chaillet M, Scheer E, et al. The architecture of human general transcription factor TFIID core complex. Nature. 2013;493(7434):699–702. pmid:23292512.
  15. 15. Setiaputra D, Ross JD, Lu S, Cheng DT, Dong MQ, Yip CK. Conformational flexibility and subunit arrangement of the modular yeast Spt-Ada-Gcn5 acetyltransferase complex. J Biol Chem. 2015;290(16):10057–70. pmid:25713136; PubMed Central PMCID: PMCPMC4400322.
  16. 16. Durand A, Bonnet J, Fournier M, Chavant V, Schultz P. Mapping the deubiquitination module within the SAGA complex. Structure. 2014;22(11):1553–9. pmid:25441028.
  17. 17. Zanton SJ, Pugh BF. Full and partial genome-wide assembly and disassembly of the yeast transcription machinery in response to heat shock. Genes Dev. 2006;20(16):2250–65. Epub 2006/08/17. doi: 20/16/2250 [pii] pmid:16912275; PubMed Central PMCID: PMC1553208.
  18. 18. Bonnet J, Wang CY, Baptista T, Vincent SD, Hsiao WC, Stierle M, et al. The SAGA coactivator complex acts on the whole transcribed genome and is required for RNA polymerase II transcription. Genes Dev. 2014;28(18):1999–2012. pmid:25228644; PubMed Central PMCID: PMCPMC4173158.
  19. 19. Grunberg S, Henikoff S, Hahn S, Zentner GE. Mediator binding to UASs is broadly uncoupled from transcription and cooperative with TFIID recruitment to promoters. EMBO J. 2016;35(22):2435–46. pmid:27797823; PubMed Central PMCID: PMCPMC5109241.
  20. 20. Warfield L, Ramachandran S, Baptista T, Devys D, Tora L, Hahn S. Transcription of Nearly All Yeast RNA Polymerase II-Transcribed Genes Is Dependent on Transcription Factor TFIID. Mol Cell. 2017. pmid:28918900.
  21. 21. Baptista T, Grunberg S, Minoungou N, Koster MJE, Timmers HTM, Hahn S, et al. SAGA Is a General Cofactor for RNA Polymerase II Transcription. Mol Cell. 2017. pmid:28918903.
  22. 22. Mencia M, Moqtaderi Z, Geisberg JV, Kuras L, Struhl K. Activator-specific recruitment of TFIID and regulation of ribosomal protein genes in yeast. Mol Cell. 2002;9(4):823–33. pmid:11983173.
  23. 23. Garbett KA, Tripathi MK, Cencki B, Layer JH, Weil PA. Yeast TFIID serves as a coactivator for Rap1p by direct protein-protein interaction. Mol Cell Biol. 2007;27(1):297–311. pmid:17074814.
  24. 24. Kamenova I, Warfield L, Hahn S. Mutations on the DNA binding surface of TBP discriminate between yeast TATA and TATA-less gene transcription. Mol Cell Biol. 2014;34(15):2929–43. pmid:24865972; PubMed Central PMCID: PMCPMC4135567.
  25. 25. Sugihara F, Kasahara K, Kokubo T. Highly redundant function of multiple AT-rich sequences as core promoter elements in the TATA-less RPS5 promoter of Saccharomyces cerevisiae. Nucleic Acids Res. 2011;39(1):59–75. pmid:20805245; PubMed Central PMCID: PMC3017598.
  26. 26. Bjornsdottir G, Myers LC. Minimal components of the RNA polymerase II transcription apparatus determine the consensus TATA box. Nucleic Acids Res. 2008;36(9):2906–16. pmid:18385157.
  27. 27. Auty R, Steen H, Myers LC, Persinger J, Bartholomew B, Gygi SP, et al. Purification of active TFIID from Saccharomyces cerevisiae. Extensive promoter contacts and co-activator function. J Biol Chem. 2004;279(48):49973–81. pmid:15448131.
  28. 28. Donczew R, Hahn S. Mechanistic differences in transcription initiation at TATA-less and TATA-containing promoters. Mol Cell Biol. 2017. pmid:29038161.
  29. 29. Lubliner S, Regev I, Lotan-Pompan M, Edelheit S, Weinberger A, Segal E. Core promoter sequence in yeast is a major determinant of expression level. Genome Res. 2015;25(7):1008–17. pmid:25969468; PubMed Central PMCID: PMCPMC4484384.
  30. 30. Lubliner S, Keren L, Segal E. Sequence features of yeast and human core promoters that are predictive of maximal promoter activity. Nucleic Acids Res. 2013;41(11):5569–81. pmid:23599004; PubMed Central PMCID: PMCPMC3675475.
  31. 31. Fishburn J, Mohibullah N, Hahn S. Function of a eukaryotic transcription activator during the transcription cycle. Mol Cell. 2005;18(3):369–78. pmid:15866178.
  32. 32. Singer VL, Wobbe CR, Struhl K. A wide variety of DNA sequences can functionally replace a yeast TATA element for transcriptional activation. Genes Dev. 1990;4(4):636–45. Epub 1990/04/01. pmid:2163345.
  33. 33. Ponticelli AS, Struhl K. Analysis of Saccharomyces cerevisiae his3 transcription in vitro: biochemical support for multiple mechanisms of transcription [published erratum appears in Mol Cell Biol 1990 Oct;10(10):5600]. Mol Cell Biol. 1990;10(6):2832–9. pmid:2188101
  34. 34. Lee TI, Causton HC, Holstege FC, Shen WC, Hannett N, Jennings EG, et al. Redundant roles for the TFIID and SAGA complexes in global transcription. Nature. 2000;405(6787):701–4. pmid:10864329.
  35. 35. Basehoar AD, Zanton SJ, Pugh BF. Identification and distinct regulation of yeast TATA box-containing genes. Cell. 2004;116(5):699–709. pmid:15006352.
  36. 36. Tora L, Timmers HT. The TATA box regulates TATA-binding protein (TBP) dynamics in vivo. Trends Biochem Sci. 2010;35(6):309–14. pmid:20176488.
  37. 37. Bhaumik SR, Green MR. SAGA is an essential in vivo target of the yeast acidic activator Gal4p. Genes Dev. 2001;15(15):1935–45. pmid:11485988.
  38. 38. Dudley AM, Rougeulle C, Winston F. The Spt components of SAGA facilitate TBP binding to a promoter at a post-activator-binding step in vivo. Genes Dev. 1999;13(22):2940–5. Epub 1999/12/02. pmid:10580001; PubMed Central PMCID: PMC317152.
  39. 39. Larschan E, Winston F. The S. cerevisiae SAGA complex functions in vivo as a coactivator for transcriptional activation by Gal4. Genes Dev. 2001;15(15):1946–56. pmid:11485989.
  40. 40. Ruden DM, Ma J, Ptashne M. No strict alignment is required between a transcriptional activator binding site and the "TATA box" of a yeast gene. Proc Natl Acad Sci U S A. 1988;85(12):4262–6. pmid:3132708; PubMed Central PMCID: PMCPMC280407.
  41. 41. Moqtaderi Z, Bai Y, Poon D, Weil PA, Struhl K. TBP-associated factors are not generally required for transcriptional activation in yeast [see comments]. Nature. 1996;383(6596):188–91. pmid:8774887
  42. 42. Moqtaderi Z, Keaveney M, Struhl K. The histone H3-like TAF is broadly required for transcription in yeast. Mol Cell. 1998;2(5):675–82. pmid:9844639
  43. 43. Chen W, Struhl K. Saturation mutagenesis of a yeast his3 "TATA element": genetic evidence for a specific TATA-binding protein. Proc Natl Acad Sci U S A. 1988;85(8):2691–5. Epub 1988/04/01. pmid:3282236; PubMed Central PMCID: PMC280064.
  44. 44. Harbury PA, Struhl K. Functional distinctions between yeast TATA elements. Mol Cell Biol. 1989;9(12):5298–304. Epub 1989/12/01. pmid:2685558; PubMed Central PMCID: PMC363694.
  45. 45. Martens JA, Brandl CJ. GCN4p activation of the yeast TRP3 gene is enhanced by ABF1p and uses a suboptimal TATA element. J Biol Chem. 1994;269(22):15661–7. pmid:8195216.
  46. 46. Schulz D, Pirkl N, Lehmann E, Cramer P. Rpb4 subunit functions mainly in mRNA synthesis by RNA polymerase II. J Biol Chem. 2014;289(25):17446–52. pmid:24802753; PubMed Central PMCID: PMCPMC4067182.
  47. 47. Rodriguez-Molina JB, Tseng SC, Simonett SP, Taunton J, Ansari AZ. Engineered Covalent Inactivation of TFIIH-Kinase Reveals an Elongation Checkpoint and Results in Widespread mRNA Stabilization. Mol Cell. 2016;63(3):433–44. pmid:27477907; PubMed Central PMCID: PMCPMC5122673.
  48. 48. Haimovich G, Medina DA, Causse SZ, Garber M, Millan-Zambrano G, Barkai O, et al. Gene expression is circular: factors for mRNA degradation also foster mRNA synthesis. Cell. 2013;153(5):1000–11. pmid:23706738.
  49. 49. Sun M, Schwalb B, Schulz D, Pirkl N, Etzold S, Lariviere L, et al. Comparative dynamic transcriptome analysis (cDTA) reveals mutual feedback between mRNA synthesis and degradation. Genome Res. 2012;22(7):1350–9. pmid:22466169; PubMed Central PMCID: PMCPMC3396375.
  50. 50. Lenstra TL, Benschop JJ, Kim T, Schulze JM, Brabers NA, Margaritis T, et al. The specificity and topology of chromatin interaction pathways in yeast. Mol Cell. 2011;42(4):536–49. pmid:21596317; PubMed Central PMCID: PMCPMC4435841.
  51. 51. Watanabe K, Yabe M, Kasahara K, Kokubo T. A Random Screen Using a Novel Reporter Assay System Reveals a Set of Sequences That Are Preferred as the TATA or TATA-Like Elements in the CYC1 Promoter of Saccharomyces cerevisiae. PLoS One. 2015;10(6):e0129357. pmid:26046838; PubMed Central PMCID: PMCPMC4457894.
  52. 52. Lundblack V. 13. Saccharomyces cerevisiae. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, et al., editors. Current protocols in molecular Biology. 2. New York: John Wiley & Sons; 1998. p. 13.0.1-..9.
    • 53. Kokubo T, Swanson MJ, Nishikawa JI, Hinnebusch AG, Nakatani Y. The yeast TAF145 inhibitory domain and TFIIA competitively bind to TATA- binding protein. Mol Cell Biol. 1998;18(2):1003–12. pmid:9447997
    • 54. Takahata S, Kasahara K, Kawaichi M, Kokubo T. Autonomous function of the amino-terminal inhibitory domain of TAF1 in transcriptional regulation. Mol Cell Biol. 2004;24(8):3089–99. pmid:15060133.
    • 55. Takahata S, Ryu H, Ohtsuki K, Kasahara K, Kawaichi M, Kokubo T. Identification of a novel TATA element-binding protein binding region at the N terminus of the Saccharomyces cerevisiae TAF1 protein. J Biol Chem. 2003;278(46):45888–902. pmid:12939271.
    • 56. Gueldener U, Heinisch J, Koehler GJ, Voss D, Hegemann JH. A second set of loxP marker cassettes for Cre-mediated multiple gene knockouts in budding yeast. Nucleic Acids Res. 2002;30(6):e23. pmid:11884642; PubMed Central PMCID: PMC101367.
    • 57. Ohyama Y, Kasahara K, Kokubo T. Saccharomyces cerevisiae Ssd1p promotes CLN2 expression by binding to the 5'-untranslated region of CLN2 mRNA. Genes Cells. 2010;15(12):1169–88. Epub 2010/10/28. pmid:20977549.
    • 58. Tsukihashi Y, Miyake T, Kawaichi M, Kokubo T. Impaired core promoter recognition caused by novel yeast TAF145 mutations can be restored by creating a canonical TATA element within the promoter region of the TUB2 gene. Mol Cell Biol. 2000;20(7):2385–99. pmid:10713163
    • 59. Lee SK, Fletcher AG, Zhang L, Chen X, Fischbeck JA, Stargell LA. Activation of a poised RNAPII-dependent promoter requires both SAGA and mediator. Genetics. 2010;184(3):659–72. pmid:20048049; PubMed Central PMCID: PMC2845336.
    • 60. Schreve JL, Sin JK, Garrett JM. The Saccharomyces cerevisiae YCC5 (YCL025c) gene encodes an amino acid permease, Agp1, which transports asparagine and glutamine. J Bacteriol. 1998;180(9):2556–9. pmid:9573211; PubMed Central PMCID: PMCPMC107201.
    • 61. Iraqui I, Vissers S, Bernard F, de Craene JO, Boles E, Urrestarazu A, et al. Amino acid signaling in Saccharomyces cerevisiae: a permease-like sensor of external amino acids and F-Box protein Grr1p are required for transcriptional induction of the AGP1 gene, which encodes a broad-specificity amino acid permease. Mol Cell Biol. 1999;19(2):989–1001. pmid:9891035; PubMed Central PMCID: PMCPMC116030.
    • 62. Ki S, Sugihara F, Kasahara K, Tochio H, Okada-Marubayashi A, Tomita S, et al. A novel magnetic resonance-based method to measure gene expression in living cells. Nucleic Acids Res. 2006;34(6):e51. pmid:16598072.
    • 63. Tsukihashi Y, Kawaichi M, Kokubo T. Requirement for yeast TAF145 function in transcriptional activation of the RPS5 promoter that depends on both core promoter structure and upstream activating sequences. J Biol Chem. 2001;276(28):25715–26. Epub 2001/05/05. M102416200 [pii]. pmid:11337503.
    • 64. Shen WC, Green MR. Yeast TAF(II)145 functions as a core promoter selectivity factor, not a general coactivator. Cell. 1997;90(4):615–24. Epub 1997/08/22. doi: S0092-8674(00)80523-1 [pii]. pmid:9288742.
    • 65. Li WZ, Sherman F. Two types of TATA elements for the CYC1 gene of the yeast Saccharomyces cerevisiae. Mol Cell Biol. 1991;11(2):666–76. pmid:1846668; PubMed Central PMCID: PMC359717.
    • 66. Cheng JX, Floer M, Ononaji P, Bryant G, Ptashne M. Responses of four yeast genes to changes in the transcriptional machinery are determined by their promoters. Curr Biol. 2002;12(21):1828–32. pmid:12419182.
    • 67. Li XY, Bhaumik SR, Zhu X, Li L, Shen WC, Dixit BL, et al. Selective recruitment of TAFs by yeast upstream activating sequences. Implications for eukaryotic promoter structure. Curr Biol. 2002;12(14):1240–4. pmid:12176335.
    • 68. Guarente L, Lalonde B, Gifford P, Alani E. Distinctly regulated tandem upstream activation sites mediate catabolite repression of the CYC1 gene of S. cerevisiae. Cell. 1984;36(2):503–11. pmid:6319028.
    • 69. Abdel-Sater F, Iraqui I, Urrestarazu A, Andre B. The external amino acid signaling pathway promotes activation of Stp1 and Uga35/Dal81 transcription factors for induction of the AGP1 gene in Saccharomyces cerevisiae. Genetics. 2004;166(4):1727–39. pmid:15126393; PubMed Central PMCID: PMCPMC1470804.
    • 70. Guarente L, Ptashne M. Fusion of Escherichia coli lacZ to the cytochrome c gene of Saccharomyces cerevisiae. Proc Natl Acad Sci U S A. 1981;78(4):2199–203. pmid:6264467; PubMed Central PMCID: PMCPMC319311.
    • 71. Nocetti N, Whitehouse I. Nucleosome repositioning underlies dynamic gene expression. Genes Dev. 2016;30(6):660–72. pmid:26966245; PubMed Central PMCID: PMCPMC4803052.
    • 72. de Jonge WJ, O'Duibhir E, Lijnzaad P, van Leenen D, Groot Koerkamp MJ, Kemmeren P, et al. Molecular mechanisms that distinguish TFIID housekeeping from regulatable SAGA promoters. EMBO J. 2017;36(3):274–90. pmid:27979920; PubMed Central PMCID: PMCPMC5286361.
    • 73. Li XY, Virbasius A, Zhu X, Green MR. Enhancement of TBP binding by activators and general transcription factors. Nature. 1999;399(6736):605–9. pmid:10376604
    • 74. Kuras L, Struhl K. Binding of TBP to promoters in vivo is stimulated by activators and requires Pol II holoenzyme. Nature. 1999;399(6736):609–13. pmid:10376605
    • 75. Martens C, Krett B, Laybourn PJ. RNA polymerase II and TBP occupy the repressed CYC1 promoter. Mol Microbiol. 2001;40(4):1009–19. pmid:11401707.
    • 76. Miller C, Schwalb B, Maier K, Schulz D, Dumcke S, Zacher B, et al. Dynamic transcriptome analysis measures rates of mRNA synthesis and decay in yeast. Mol Syst Biol. 2011;7:458. pmid:21206491; PubMed Central PMCID: PMCPMC3049410.
    • 77. Zentner GE, Kasinathan S, Xin B, Rohs R, Henikoff S. ChEC-seq kinetics discriminates transcription factor binding sites by DNA sequence and shape in vivo. Nat Commun. 2015;6:8733. pmid:26490019; PubMed Central PMCID: PMCPMC4618392.
    • 78. Lai WKM, Pugh BF. Understanding nucleosome dynamics and their links to gene expression and DNA replication. Nat Rev Mol Cell Biol. 2017;18(9):548–62. pmid:28537572.
    • 79. Reja R, Vinayachandran V, Ghosh S, Pugh BF. Molecular mechanisms of ribosomal protein gene coregulation. Genes Dev. 2015;29(18):1942–54. pmid:26385964; PubMed Central PMCID: PMCPMC4579351.
    • 80. van Werven FJ, van Teeffelen HA, Holstege FC, Timmers HT. Distinct promoter dynamics of the basal transcription factor TBP across the yeast genome. Nat Struct Mol Biol. 2009;16(10):1043–8. Epub 2009/09/22. doi: nsmb.1674 [pii] pmid:19767748.
    • 81. Venters BJ, Wachi S, Mavrich TN, Andersen BE, Jena P, Sinnamon AJ, et al. A comprehensive genomic binding map of gene and chromatin regulatory proteins in Saccharomyces. Mol Cell. 2011;41(4):480–92. pmid:21329885; PubMed Central PMCID: PMC3057419.
    • 82. Kubik S, Bruzzone MJ, Shore D. TFIID or not TFIID, a continuing transcriptional SAGA. EMBO J. 2017;36(3):248–9. pmid:28069707; PubMed Central PMCID: PMCPMC5286358.
    • 83. Afek A, Lukatsky DB. Genome-wide organization of eukaryotic preinitiation complex is influenced by nonconsensus protein-DNA binding. Biophys J. 2013;104(5):1107–15. pmid:23473494; PubMed Central PMCID: PMCPMC3870814.
    • 84. Afek A, Lukatsky DB. Nonspecific protein-DNA binding is widespread in the yeast genome. Biophys J. 2012;102(8):1881–8. pmid:22768944; PubMed Central PMCID: PMCPMC3328700.