Mammalian transcription factors in yeast: strangers in a familiar land

Directed evolution in mammalian cells

Directed evolution experiments are typically carried out using in vitro systems, bacteria, or yeast-even when the goal is to probe or modulate mammalian biology. Performing directed evolution in systems that do not match the intended mammalian environment severely constrains the scope and functionality of the targets that can be evolved. We review new platforms that are now making it possible to use the mammalian cell itself as the setting for directed evolution and present an overview of frontier challenges and high-impact targets for this approach.

Evaluating the Influence of a G-Quadruplex Prone Sequence on the Transactivation Potential by Wild-Type and/or Mutant P53 Family Proteins through a Yeast-Based Functional Assay

P53, P63, and P73 proteins belong to the P53 family of transcription factors, sharing a common gene organization that, from the P1 and P2 promoters, produces two groups of mRNAs encoding proteins with different N-terminal regions; moreover, alternative splicing events at C-terminus further contribute to the generation of multiple isoforms. P53 family proteins can influence a plethora of cellular pathways mainly through the direct binding to specific DNA sequences known as response elements (REs), and the transactivation of the corresponding target genes. However, the transcriptional activation by P53 family members can be regulated at multiple levels, including the DNA topology at responsive promoters. Here, by using a yeast-based functional assay, we evaluated the influence that a G-quadruplex (G4) prone sequence adjacent to the p53 RE derived from the apoptotic PUMA target gene can exert on the transactivation potential of full-length and N-terminal truncated P53 family α isoforms (wild-type and mutant). Our results show that the presence of a G4 prone sequence upstream or downstream of the P53 RE leads to significant changes in the relative activity of P53 family proteins, emphasizing the potential role of structural DNA features as modifiers of P53 family functions at target promoter sites.

Systematic discovery of novel eukaryotic transcriptional regulators using sequence homology independent prediction

BACKGROUND

The molecular function of a gene is most commonly inferred by sequence similarity. Therefore, genes that lack sufficient sequence similarity to characterized genes (such as certain classes of transcriptional regulators) are difficult to classify using most function prediction algorithms and have remained uncharacterized.

RESULTS

To identify novel transcriptional regulators systematically, we used a feature-based pipeline to screen protein families of unknown function. This method predicted 43 transcriptional regulator families in Arabidopsis thaliana, 7 families in Drosophila melanogaster, and 9 families in Homo sapiens. Literature curation validated 12 of the predicted families to be involved in transcriptional regulation. We tested 33 out of the 195 Arabidopsis putative transcriptional regulators for their ability to activate transcription of a reporter gene in planta and found twelve coactivators, five of which had no prior literature support. To investigate mechanisms of action in which the predicted regulators might work, we looked for interactors of an Arabidopsis candidate that did not show transactivation activity in planta and found that it might work with other members of its own family and a subunit of the Polycomb Repressive Complex 2 to regulate transcription.

CONCLUSIONS

Our results demonstrate the feasibility of assigning molecular function to proteins of unknown function without depending on sequence similarity. In particular, we identified novel transcriptional regulators using biological features enriched in transcription factors. The predictions reported here should accelerate the characterization of novel regulators.

Human transcription factors in yeast: the fruitful examples of P53 and NF-кB

The observation that human transcription factors (TFs) can function when expressed in yeast cells has stimulated the development of various functional assays to investigate (i) the role of binding site sequences (herein referred to as response elements, REs) in transactivation specificity, (ii) the impact of polymorphic nucleotide variants on transactivation potential, (iii) the functional consequences of mutations in TFs and (iv) the impact of cofactors or small molecules. These approaches have found applications in basic as well as applied research, including the identification and the characterisation of mutant TF alleles from clinical samples. The ease of genome editing of yeast cells and the availability of regulated systems for ectopic protein expression enabled the development of quantitative reporter systems, integrated at a chosen chromosomal locus in isogenic yeast strains that differ only at the level of a specific RE targeted by a TF or for the expression of distinct TF alleles. In many cases, these assays were proven predictive of results in higher eukaryotes. The potential to work in small volume formats and the availability of yeast strains with modified chemical uptake have enhanced the scalability of these approaches. Next to well-established one-, two-, three-hybrid assays, the functional assays with non-chimeric human TFs enrich the palette of opportunities for functional characterisation. We review ∼25 years of research on human sequence-specific TFs expressed in yeast, with an emphasis on the P53 and NF-кB family of proteins, highlighting outcomes, advantages, challenges and limitations of these heterologous assays.

Quantitative Analysis of NF-κB Transactivation Specificity Using a Yeast-Based Functional Assay

The NF-κB transcription factor family plays a central role in innate immunity and inflammation processes and is frequently dysregulated in cancer. We developed an NF-κB functional assay in yeast to investigate the following issues: transactivation specificity of NF-κB proteins acting as homodimers or heterodimers; correlation between transactivation capacity and in vitro DNA binding measurements; impact of co-expressed interacting proteins or of small molecule inhibitors on NF-κB-dependent transactivation. Full-length p65 and p50 cDNAs were cloned into centromeric expression vectors under inducible GAL1 promoter in order to vary their expression levels. Since p50 lacks a transactivation domain (TAD), a chimeric construct containing the TAD derived from p65 was also generated (p50TAD) to address its binding and transactivation potential. The p50TAD and p65 had distinct transactivation specificities towards seventeen different κB response elements (κB-REs) where single nucleotide changes could greatly impact transactivation. For four κB-REs, results in yeast were predictive of transactivation potential measured in the human MCF7 cell lines treated with the NF-κB activator TNFα. Transactivation results in yeast correlated only partially with in vitro measured DNA binding affinities, suggesting that features other than strength of interaction with naked DNA affect transactivation, although factors such as chromatin context are kept constant in our isogenic yeast assay. The small molecules BAY11-7082 and ethyl-pyruvate as well as expressed IkBα protein acted as NF-κB inhibitors in yeast, more strongly towards p65. Thus, the yeast-based system can recapitulate NF-κB features found in human cells, thereby providing opportunities to address various NF-κB functions, interactions and chemical modulators.

Evolution of p53 transactivation specificity through the lens of a yeast-based functional assay

Co-evolution of transcription factors (TFs) with their respective cis-regulatory network enhances functional diversity in the course of evolution. We present a new approach to investigate transactivation capacity of sequence-specific TFs in evolutionary studies. Saccharomyces cerevisiae was used as an in vivo test tube and p53 proteins derived from human and five commonly used animal models were chosen as proof of concept. p53 is a highly conserved master regulator of environmental stress responses. Previous reports indicated conserved p53 DNA binding specificity in vitro, even for evolutionary distant species. We used isogenic yeast strains where p53-dependent transactivation was measured towards chromosomally integrated p53 response elements (REs). Ten REs were chosen to sample a wide range of DNA binding affinity and transactivation capacity for human p53 and proteins were expressed at two levels using an inducible expression system. We showed that the assay is amenable to study thermo-sensitivity of frog p53, and that chimeric constructs containing an ectopic transactivation domain could be rapidly developed to enhance the activity of proteins, such as fruit fly p53, that are poorly effective in engaging the yeast transcriptional machinery. Changes in the profile of relative transactivation towards the ten REs were measured for each p53 protein and compared to the profile obtained with human p53. These results, which are largely independent from relative p53 protein levels, revealed widespread evolutionary divergence of p53 transactivation specificity, even between human and mouse p53. Fruit fly and human p53 exhibited the largest discrimination among REs while zebrafish p53 was the least selective.

A-type nuclear lamins act as transcriptional repressors when targeted to promoters

Regions of heterochromatin are often found at the periphery of the mammalian nucleus, juxtaposed to the nuclear lamina. Genes in these regions are likely maintained in a transcriptionally silent state, although other locations at the nuclear periphery associated with nuclear pores are sites of active transcription. As primary components of the nuclear lamina, A- and B-type nuclear lamins are intermediate filament proteins that interact with DNA, histones and known transcriptional repressors, leading to speculation that they may promote establishment of repressive domains. However, no direct evidence of a role for nuclear lamins in transcriptional repression has been reported. Here we find that human lamin A, when expressed in yeast and cultured human cells as a fusion protein to the Gal4 DNA-binding domain (DBD), can mediate robust transcriptional repression of promoters with Gal4 binding sites. Full repression by lamin A requires both the coiled-coil rod domain and the C-terminal tail domain. In human cells, other intermediate filament proteins such as lamin B and vimentin are unable to confer robust repression as Gal4-DBD fusions, indicating that this property is specific to A-type nuclear lamins. These findings indicate that A-type lamins can promote transcriptional repression when in proximity of a promoter.

The tumor suppressor p53 associates with gene coding regions and co-traverses with elongating RNA polymerase II in an in vivo model

Sequence-specific transcriptional regulators function by stably binding cognate DNA sequences followed by recruitment of both general and specialized factors to target gene promoters. The tumor suppressor p53 mediates its anti-oncogenic effect on cells by functioning as a sequence-specific regulator. p53 employs a secondary mechanism to suppress tumor formation by permeabilizing the outer mitochondrial membrane, thereby releasing pro-apoptotic factors. Here, we report a potential third biological function of p53: as a transcriptional elongation factor. Using chromatin immunoprecipitation, we demonstrate that human p53 robustly associates with RNA polymerase II (Pol II), but neither Pol I- nor Pol III-transcribed regions in the budding yeast, Saccharomyces cerevisiae. p53's association with open reading frames is mediated by its physical interaction with elongating Pol II, with which p53 travels in vivo and co-immunoprecipitates in vitro. When similarly expressed, the potent acidic activator VP16 cannot be cross-linked to Pol II coding regions. p53 levels comparable to those found in induced mammalian cells confer synthetic sickness or lethality in combination with deletions in genes encoding transcription elongation factors; p53 likewise confers hypersensitivity to the anti-elongation drug 6-azauracil. Collectively, our results indicate that p53 can physically interact with the transcription elongation complex and influence transcription elongation, and open up new avenues of investigation in mammalian cells.

Functional dissection of sequence-specific NKX2-5 DNA binding domain mutations associated with human heart septation defects using a yeast-based system

Human heart development requires an orderly coordination of transcriptional programs, with the homeodomain protein NKX2-5 being one of the key transcription factors required for the differentiation of mesodermal progenitor cells. Indeed, lack of Nkx2-5 in mice arrests heart development prior to looping, resulting in embryonic lethality. There are 28 germline NKX2-5 mutations identified in humans that are associated with congenital heart disease, and we recently reported multiple somatic mutations in patients with complex cardiac malformations. To address the functional consequences of single and multiple mutations of NKX2-5, we developed a functional assay in the budding yeast Saccharomyces cerevisiae, which could determine transactivation capacity and specificity of expressed NKX2-5 alleles towards targeted response element (RE) sequences. We focused on mutants of the third helix, which provides DNA binding specificity, and characterized mutations that were highly associated with either ventricular (VSD) or atrioventricular (AVSD) septal defects. Individual mutants exhibited partial to complete loss of function and differences in transactivation capacity between the various REs. The mutants also exhibited gene dosage rather than dominant effects on transcription. Surprisingly, all AVSD patients (22/23) had a single K183E mutation in the DNA binding domain, which resulted in transcriptional inactivation. None of the VSD patients had this mutation; yet 14/29 had at least one mutation in the third helix leading to either inactivation or reduction of NKX2-5 transactivation. Therefore, mutations of somatic origin in the binding domains of NKX2-5 were associated specifically with AVSD or VSD and resulted in loss of protein function.

Aym1, a mouse meiotic gene identified by virtue of its ability to activate early meiotic genes in the yeast Saccharomyces cerevisiae

Our understanding of the molecular mechanisms that operate during differentiation of mitotically dividing spermatogonia cells into spermatocytes lags way behind what is known about other differentiating systems. Given the evolutionary conservation of the meiotic process, we screened for mouse proteins that could specifically activate early meiotic promoters in Saccharomyces cerevisiae yeast cells, when fused to the Gal4 activation domain (Gal4AD). Our screen yielded the Aym1 gene that encodes a short peptide of 45 amino acids. We show that a Gal4AD-AYM1 fusion protein activates expression of reporter genes through the promoters of the early meiosis-specific genes IME2 and HOP1, and that this activation is dependent on the DNA-binding protein Ume6. Aym1 is transcribed predominantly in mouse primary spermatocytes and in gonads of female embryos undergoing the corresponding meiotic divisions. Aym1 immunolocalized to nuclei of primary spermatocytes and oocytes and to specific type A spermatogonia cells, suggesting it might play a role in the processes leading to meiotic competence. The potential functional relationship between AYM1 and yeast proteins that regulate expression of early meiotic genes is discussed.

Interaction networks in yeast define and enumerate the signaling steps of the vertebrate aryl hydrocarbon receptor

The aryl hydrocarbon receptor (AHR) is a vertebrate protein that mediates the toxic and adaptive responses to dioxins and related environmental pollutants. In an effort to better understand the details of this signal transduction pathway, we employed the yeast S. cerevisiae as a model system. Through the use of arrayed yeast strains harboring ordered deletions of open reading frames, we determined that 54 out of the 4,507 yeast genes examined significantly influence AHR signal transduction. In an effort to describe the relationship between these modifying genes, we constructed a network map based upon their known protein and genetic interactions. Monte Carlo simulations demonstrated that this network represented a description of AHR signaling that was distinct from those generated by random chance. The network map was then explored with a number of computational and experimental annotations. These analyses revealed that the AHR signaling pathway is defined by at least five distinct signaling steps that are regulated by functional modules of interacting modifiers. These modules can be described as mediating receptor folding, nuclear translocation, transcriptional activation, receptor level, and a previously undescribed nuclear step related to the receptor's Per-Arnt-Sim domain.

Expression of human TFIIA subunits in Saccharomyces cerevisiae identifies regions with conserved and species-specific functions

The transcription factor TFIIA stabilizes the interaction between the TATA-binding protein (TBP) and promoter DNA and facilitates activator function. In yeast, TFIIA is composed of large (TOA1) and small (TOA2) subunits that interact to form a beta-barrel domain and a helix bundle domain. Here we report plasmid shuffle experiments showing that the human subunits (TFIIAalpha/beta, ALF, and TFIIAgamma) are not able to support growth in yeast and that the failure is associated with morphological abnormalities related to cell division. To determine the regions responsible for species specificity, we examined a series of chimeric yeast-human subunits. The results showed that yeast-human hybrids that contained the N-termini of TFIIAgamma or TFIIAalpha/beta were viable, presumably because they could form a functional interspecies alpha-helical bundle. Likewise, a TOA1 hybrid that contained the nonconserved internal region from TFIIAalpha/beta also had no effect on TFIIA function. However, hybrids that contained the acidic region III or C-terminal region IV from TFIIAalpha/beta grew more slowly than the wild-type TOA1 subunit, and if both regions were exchanged, this effect was far more severe. Although these hybrids exchanged sequences which are involved in beta-barrel formation and interactions with TBP, they were all active in a TBP-dependent mobility shift assay. The results suggest that the growth phenotypes of these hybrids might be due to a failure to interact with components of the yeast transcription machinery other than TBP. Finally, we show that sequences from region III of TFIIA large subunits fall into classes that are either highly acidic or that are divergent and nonacidic, and provide the first evidence to suggest that, at least in yeast, this region is important for TFIIA function.

The vertebrate protein CTCF functions as an insulator in Saccharomyces cerevisiae

Insulators are elements that shelter genes from the effects of silencers or enhancers. CTCF is the only vertebrate protein that has a recognized role in transcriptional insulation, but how it exerts its effect is unknown. In an attempt to better understand how CTCF functions, we have used an insulation assay in Saccharomyces cerevisiae. We show that CTCF acts as an insulator in yeast, where it can efficiently block the spreading of repressive telomeric chromatin. We identify two domains of the protein that are responsible for this activity: a short and very potent N-terminal domain, as well as the C-terminus of the protein.