Science Spotlight

An answer to the age-old question: TFIID or not TFIID?

From the Hahn Lab, Basic Sciences Division

The process of gene expression, in which the information in DNA is used to make proteins, is a universal requirement for all cells. The first step of this process—transcription of DNA into RNA—is performed by a molecular machine known as RNA polymerase, in concert with a host of accessory proteins termed transcription factors. In yeast, essentially all transcription was thought to require one of two factors, TFIID or SAGA, which are each in turn comprised of over ten subunits. TFIID was thought to control “housekeeping” genes, which are always expressed, while SAGA has been associated with genes whose expression is highly regulated. However, “several published findings were not entirely consistent with this model” says Dr. Steve Hahn, a member in the Basic Sciences Division. In work published last month in Molecular Cell, Hahn and colleagues from his laboratory, in collaboration with the Henikoff Lab and the Laszlo Tora Lab in France, took advantage of newly available experimental methods to definitively establish which yeast genes are dependent on TFIID vs. SAGA.

In yeast, the transcription factor TFIID binds at the promoter of nearly every gene, regardless of previous classifications. Image provided by Dr. Steve Hahn.

Because all TFIID subunits are essential for cellular survival, it is not possible to simply delete TFIID from the genome and ask which genes are no longer transcribed. Previous studies attempted to circumvent this problem in a number of ways, such as by using temperature-sensitive versions of TFIID that become non-functional when yeast are switched to a higher growth temperature. However, it has since been shown that heat shock itself can reduce genome-wide transcription, making it difficult to identify effects that are specific to TFIID depletion. Moreover, earlier experiments measured the transcriptional roles of TFIID and SAGA by measuring steady-state mRNA levels, which is problematic because inhibition of transcription factors with genome-wide roles stabilizes most mRNAs.

Thanks to recently developed techniques, it is now possible to rapidly deplete essential proteins without the need for heat-sensitive alleles. Armed with this technology, the Hahn Lab set out to revisit the role of TFIID in genome-wide transcription. First, the researchers constructed mutant alleles of TFIID subunits, called Tafs, in which the Taf protein is fused to the plant protein IAA7. The IAA7 tag confers a unique property: in the presence of the small molecule auxin and a plant enzyme called Tir1, any protein that IAA7 is attached to will be quickly degraded. This approach is highly specific because there are no auxin-responsive proteins in non-plant cells. Thus, by integrating Tir1 into the genome of yeast strains harboring IAA7-tagged Taf alleles, the authors could trigger rapid Taf destruction, and thereby compromise TFIID function, simply by adding auxin to the growth medium.

With these strains in hand, the next step was to measure genome-wide transcription in the presence or absence of auxin. Recent advances in sequencing technology have made it possible to specifically quantify newly synthesized RNA molecules using a technique known as native RNA polymerase chromatin immunoprecipitation. As a result, the authors could isolate only those RNA molecules that are actively being transcribed, a key improvement over previous approaches.

Dr. Hahn and his colleagues observed that depletion of several TFIID subunits, including Taf7 and Taf11, or subunits of another essential transcription factor called Mediator, globally reduced transcription of nearly every gene in the yeast genome. Interestingly, this conclusion held regardless of genes’ previous classification as TFIID- or SAGA-dependent. Genes that were shown in previous studies to have less TFIID physically associated with their promoters were not as strongly affected, but their expression still decreased significantly upon TFIID depletion. In addition, the presence of a TATA box within the promoter of a given gene, which was previously thought to indicate SAGA dependence, did not confer independence from TFIID.

The authors validated their results with a complementary approach known as “anchor away”, in which target proteins are fused to a rapamycin-binding tag and thus rapidly exported from the nucleus upon rapamycin treatment. Because transcription factors operate within the nucleus, using this strategy with Taf proteins renders them non-functional in an inducible manner. Upon depletion of the TFIID component Taf4, the researchers again observed repression of nearly all transcription. Genes previously classified as SAGA-dependent had a lower median fold change than those considered TFIID-dominated, but the reduction in transcription was significant across the board.

Finally, the authors wanted to make sure that the reason their results differed from published studies was not due to differences in growth conditions. So far, their experiments had all been done in YPD, a standard laboratory medium composed of yeast extract, peptone and glucose. To simulate conditions used in previous studies, the experiments were repeated in YPD medium with a heat shock, as well as in synthetic complete glucose medium. Overall, the observed changes in transcription were smaller in these conditions, but the conclusion was the same: transcription of essentially every gene is dependent on TFIID.

In an accompanying paper published in the same issue of Molecular Cell, the Hahn lab and their collaborators showed that SAGA localizes to regulatory regions of nearly all yeast genes, upstream of promoters and TFIID binding sites. Thus, SAGA and TFIID are likely both present at actively transcribed genes, where each can make independent contributions to gene expression. Future studies will focus on understanding how TFIID recognizes promoters and what features determine whether transcription of a given gene is constitutive or highly regulated.



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.” Molecular Cell. 2017 Oct 5;68(1):118-129.e5. doi: 10.1016/j.molcel.2017.08.014.

Baptista T, Grunberg S, Minoungou N, Koster MJE, Timmers HTM, Hahn S, Devys D, Tora L. “SAGA Is a General Cofactor for RNA Polymerase II Transcription.” Molecular Cell. 2017 Oct 5;68(1):130-143.e5. doi: 10.1016/j.molcel.2017.08.016.

This research was supported by the National Institutes of Health, the European Research Council and multiple agencies in France (Marie Curie Initial Training Networks, Association pour la Recherche sur le Cancer, Agence Nationale de la Recherche)