journal article
Deletion Mapping a Eukaryotic PromoterProceedings of the National Academy of Sciences of the United States of America
Vol. 78, No. 7, [Part 2: Biological Sciences] [Jul., 1981]
, pp. 4461-4465 [5 pages]
Published By: National Academy of Sciences
//www.jstor.org/stable/10634
Read and download
Log in through your school or library
Alternate access options
For independent researchers
Read Online
Read 100 articles/month free
Subscribe to JPASS
Unlimited reading + 10 downloads
With a personal account, you can read up to 100 articles each month for free.
Get StartedAlready have an account? Log in
Monthly Plan
- Access everything in the JPASS collection
- Read the full-text of every article
- Download up to 10 article PDFs to save and keep
Yearly Plan
- Access everything in the JPASS collection
- Read the full-text of every article
- Download up to 120 article PDFs to save and keep
Abstract
The phenotypes of 24 mutants that successively delete DNA sequences adjacent to the 5′ end of the Saccharomyces cerevisiae [yeast] his3 structural gene are described. Deletions retaining > 155 base pairs before the mRNA coding sequences are phenotypically indistinguishable from the wild-type his3 allele. Deletions having end points between 113 and 65 base pairs before the transcription initiation site express his3 at reduced levels. Mutations retaining < 45 base pairs are indistinguishable from null alleles of the his3 locus. These results indicate [i] that a sequence[s] located 113-155 base pairs upstream from the transcribed region is necessary for wild-type expression and [ii] that the T-A-T-A box [a sequence in front of most eukaryotic genes] is not sufficient for wild-type promoter function. Thus, the yeast his3 promoter region appears large when compared with prokaryotic promoters, suggesting that it may be more complex than a simple site of interaction between RNA polymerase and DNA.
Journal Information
PNAS is the world's most-cited multidisciplinary scientific serial. It publishes high-impact research reports, commentaries, perspectives, reviews, colloquium papers, and actions of the Academy. In accordance with the guiding principles established by George Ellery Hale in 1914, PNAS publishes brief first announcements of Academy Members' and Foreign Associates' more important contributions to research and of work that appears to a Member to be of particular importance.
Publisher Information
The National Academy of Sciences [NAS] is a private, nonprofit organization of the country’s leading researchers. The NAS recognizes and promotes outstanding science through election to membership; publication in its journal, PNAS; and its awards, programs, and special activities. Through the National Academies of Sciences, Engineering, and Medicine, the NAS provides objective, science-based advice on critical issues affecting the nation.
Rights & Usage
This item is part of a JSTOR Collection.
For terms and use, please refer to our Terms and Conditions
Proceedings of the National Academy of Sciences of the United States of America © 1981 National Academy of Sciences
Request Permissions
Despite being one of the first eukaryotic transcriptional regulatory elements identified, the sequence of a native TATA box and its significance remain elusive. Applying criteria associated with TATA boxes we queried several Saccharomyces genomes and arrived at the consensus TATA[A/T]A[A/T][A/G]. Approximately 20% of yeast genes contain a TATA box.
Strikingly, TATA box-containing genes are associated with responses to stress, are highly regulated, and preferentially utilize SAGA rather than TFIID when compared to TATA-less promoters. Transcriptional regulation in yeast appears to be mechanistically bipolar, possibly reflecting a need to balance inducible stress-related responses with constitutive housekeeping functions. The TATA box plays an important role in assembling the transcription machinery at promoters [ ]. Chromatin might keep TATA boxes generally inaccessible until promoter bound activators modify local chromatin structure [ ]. It is now becoming clear that transcriptional activators in coordination with the local chromatin structure recruit the TATA binding protein [TBP] to promoters, most likely through interactions with
coactivator complexes that contact TBP, such as TFIID and SAGA [Abstract
Introduction
,
,
,
]. These complexes then recruit RNA polymerase [pol] II and an entourage of regulatory factors, culminating in gene expression [
]. Turning off gene expression is equally critical, and involves a variety of factors that reverse transcription complex assembly. Two factors in particular, NC2 and Mot1, directly inhibit TBP function.
Not all promoters contain a TATA box. These so-called TATA-less promoters nevertheless require TBP for function [
]. Since promoters that contain a TATA box depend upon it, TATA-containing and TATA-less promoters might direct distinct pathways for assembling TBP. Paradoxically, while the use of a TATA box is conserved among all eukaryotes, a wide variety of sequences can functionally replace a TATA box in yeast, at least in the experimental context of a plasmid-borne promoter [
]. Recently, however, such sequences were found to be poorly conserved among four closely related Saccharomyces species [
], and have not been shown to be functional or widely used in the context of natural chromosomal genes.
The elusiveness of TATA boxes was further illustrated in a de novo search for DNA motifs that are conserved among Saccharomyces species [
]. None of the identified 72 motifs revealed a TATA box. Thus, either A/T richness is the only conserved aspect of a TATA box, or a true TATA box consensus sequence has yet to be defined within its natural chromosomal context. With this latter possibility in mind, we set out to marshal information relevant to TATA box function to achieve the following goals: [1] Identify a TATA box consensus sequence for promoters in their natural chromosomal context; [2] classify each S. cerevisiae gene as either TATA-containing or TATA-less; and [3] determine whether TATA-containing and TATA-less genes have distinct physiological modes of regulation.
Results
The TATA Box Consensus Sequence is TATA[A/T]A[A/T][A/G]
The unusually high A/T-richness of intergenic/promoter regions [63% A+T in Saccharomyces], the lack of a veritable TATA box consensus, and the fact that TBP is required at genes lacking a TATA box make the task of distinguishing functional TATA boxes from irrelevant TATA-like sequence nontrivial. Our strategy centered around the imposition of several loose constraints under which a TATA box sequence is expected to operate, the rationale for which is presented in the Experimental Procedures. These constraints include a maximum sequence length [8 bp], minimal consensus sequence [TA[A/T][A/T]NNNN], confined upstream location [−200 to −50, relative to the ATG start codon] compared to surrounding regions, conserved presence across orthologous upstream regions in four out of four different Saccharomyces species, and association with genes that are particularly sensitive to mutations along TBP's DNA binding surface [coregulation]. 1024 8-mer sequences of the minimal consensus were surveyed as described in the Experimental Procedures. Several were deemed significantly enriched in the promoter region by these location, conservation, and coregulation criteria [Figure 1A]. The rank order of all 1024 sequences can be found in Supplemental Tables S1A–S1C [available at //www.cell.com/cgi/content/full/116/5/699/DC1], and a summary of selected sequences is presented in Table 1. Four sequences were highly significant by each of the three criteria, and two were significant by two criteria.
Figure 1Properties of TATA Boxes and TATA-Like Elements
Show full caption
[A] Venn diagram of 8-mer sequences enriched by location, conservation, and coregulation.
Shown is the relationship of those 8-mer sequences [out of 1024 examined] that were significant by the following criteria: location, enriched in the −200 to −50 region relative to the surrounding −1000 to −200 plus −50 to −1 regions [P-value < 0.001]; conservation, perfectly conserved in both alignment and sequence among four Saccharomyces sensu stricto species [P-value < 0.01]; and coregulation, enriched in the −200 to −50 region of genes sensitive to mutations [V161E and V71E] on TBP's DNA binding surface compared to the same region of insensitive genes [P-value < 0.0001]. Significance values are presented in Supplemental Table S1 [available on Cell website].
[B] Distribution of the TATATAAA and TATATAAT 8-mer sequences along the upstream region of all S. cerevisiae ORFs.
Location is relative to the ATG translational start codon. The frequency [percentage] of occurrence of each sequence in 50 bp intervals is plotted. “Random” denotes the combined average of twenty randomly selected 8-mers.
- View Large Image
- Figure Viewer
- Download Hi-res image
- Download [PPT]
Table 1Rank Order of TATA Box Consensus Sequences Out of 1024 8-Mer Sequences Tested
| TATATAAA | 1 | 1 | 2 |
| TATATAAG | 2 | 3 | 1 |
| TATAAATA | 4 | 2 | 3 |
| TATAAAAG | 3 | 5 | 5 |
| TATATATA | 5 | 29 | 4 |
| TATAAAAA | 6 | 88 | 8 |
| TATAAATG | 279 | 62 | 24 |
| TATATATG | 56 | 307 | 48 |
A complete rank order of all 1024 possible TA[A/T][A/T]NNNN 8-mers in terms of their location [enrichment in the −200 to −50 regions of all Saccharomyces genes], conservation in the orthologous upstream regions of four Saccharomyces species, and coregulation [enrichment among genes sensitive to TBP DNA binding mutants] can be found in Supplemental Table S1 [available on Cell website].
- Open table in a new tab
Consistent with these top-scoring sequences representing bona fide TATA boxes, they are among the highest affinity sites selected by TBP in vitro from a pool of random oligonucleotides [
]. In addition, they are among a variety of sequences that support high levels of transcription in the context of a plasmid-borne reporter gene, in vivo [
,
]. However, no previous study distinguished these six sequences from a variety of other “TATA-like” sequences, which can behave like TATA boxes in certain assays, but are not generally utilized by chromosomal genes.
The top six sequences form a cohort of the eight-member consensus TATA[A/T]A[A/T][A/G]. Two missing members are TATA[A/T]ATG. This sequence ends in an ATG translational start codon, and thus is expected to be underrepresented in promoters. Since it is nevertheless part of the larger consensus that avidly binds TBP, this sequence was included in the TATA consensus, although it is rarely used.
To be certain all possible TATA-like sequences were considered, we used the consensus of the second half of the cohort, allowing the first half to vary [i.e., NNNN[A/T]A[A/T][A/G]] and repeated the analyses. No additional sequence became enriched, by all three criteria, to a significance level comparable to the six identified above [Supplemental Tables S1D–S1F available on Cell website]. The fact that the eight sequences form a complete consensus, and that no other consensus sequence was apparent, indicates that the consensus is fully defined and that there are unlikely to be other major TATA box sequences. Minor variants, however, are likely. In comparison to the 72 previously identified DNA motifs conserved among Saccharomyces species [
], this TATA box consensus ranks fifth [MCS = 20] in terms of conservation, indicating that it is highly conserved. Why it went previously undetected is not clear.
It is striking that the eighth position of the TATA box, which had largely been considered irrelevant, depended very strongly on having a purine at this position. Figure 1B shows the frequency distribution along the intergenic region for TATATAAA and a nonconsensus variant at the eighth position, TATATAAT, which by previous criteria would pass as a TATA box. TATATAAT shows little or no enrichment between −200 and −50 compared to TATATAAA, and thus is not generally utilized as a TATA box.
TATA-Containing Genes Comprise Approximately 19% of the Yeast Genome
Having established a TATA box consensus sequence and implementing strict location and conservation criteria [see Supplemental Data available on Cell website], 763 [∼13%] of the 5714 S. cerevisiae genes [
] were definitively classified as TATA-containing. Less than 2% of all genes contained two or more nonoverlapping TATA boxes. These strict estimates are in contrast to prevailing views that most yeast genes contain TATA boxes and in multiple copies. The discrepancy might be due to biased estimates based upon a small number of well-studied genes, chosen based upon properties that tend to be associated with TATA-containing genes [see below].
To experimentally assess the physiological significance of the assigned TATA boxes, we examined whether TATA-containing genes were particularly sensitive to mutations along TBP's DNA binding surface. Although both TATA-containing and TATA-less promoters are likely to bind TBP, and require TBP for expression, we reasoned that genes having a TATA-box are likely to be more dependent upon TBP/DNA interactions, and thus would be more sensitive to mutations along TBP's DNA binding surface. 5714 S. cerevisiae genes were sorted by their expression sensitivity to dominant mutations [V71E and V161E] along the DNA binding surface of TBP. Sliding 200 gene windows across the data set were examined. For each window, the percentage of genes containing a properly located and conserved TATA box was plotted as a function of average TBP sensitivity [fold change in gene expression]. As shown in Figure 2A, more than 50% of the most TBP-sensitive genes were classified as TATA-containing, while