The discrepancy between chromatin factor location and effect

Optimized reporters for multiplexed detection of transcription factor activity

In any given cell type, dozens of transcription factors (TFs) act in concert to control the activity of the genome by binding to specific DNA sequences in regulatory elements. Despite their considerable importance in determining cell identity and their pivotal role in numerous disorders, we currently lack simple tools to directly measure the activity of many TFs in parallel. Massively parallel reporter assays (MPRAs) allow the detection of TF activities in a multiplexed fashion; however, we lack basic understanding to rationally design sensitive reporters for many TFs. Here, we use an MPRA to systematically optimize transcriptional reporters for 86 TFs and evaluate the specificity of all reporters across a wide array of TF perturbation conditions. We thus identified critical TF reporter design features and obtained highly sensitive and specific reporters for 60 TFs, many of which outperform available reporters. The resulting collection of "prime" TF reporters can be used to uncover TF regulatory networks and to illuminate signaling pathways.

HIGHLIGHTS

Systematic design and optimization of transcriptional reporters for 86 TFsCharacterization of TF-specific reporter design optimization rulesEvaluation of reporter TF-specificity across a wide array of TF perturbationsIdentification of a collection of 60 "prime" TF reporters with optimized performance.

NetProphet 3: a machine learning framework for transcription factor network mapping and multi-omics integration

MOTIVATION

Many methods have been proposed for mapping the targets of transcription factors (TFs) from gene expression data. It is known that combining outputs from multiple methods can improve performance. To date, outputs have been combined by using either simplistic formulae, such as geometric mean, or carefully hand-tuned formulae that may not generalize well to new inputs. Finally, the evaluation of accuracy has been challenging due to the lack of genome-scale, ground-truth networks.

RESULTS

We developed NetProphet3, which combines scores from multiple analyses automatically, using a tree boosting algorithm trained on TF binding location data. We also developed three independent, genome-scale evaluation metrics. By these metrics, NetProphet3 is more accurate than other commonly used packages, including NetProphet 2.0, when gene expression data from direct TF perturbations are available. Furthermore, its integration mode can forge a consensus network from gene expression data and TF binding location data.

AVAILABILITY AND IMPLEMENTATION

All data and code are available at https://zenodo.org/record/7504131#.Y7Wu3i-B2x8.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

An integrated SAGA and TFIID PIC assembly pathway selective for poised and induced promoters

Genome-wide, little is understood about how proteins organize at inducible promoters before and after induction and to what extent inducible and constitutive architectures depend on cofactors. We report that sequence-specific transcription factors and their tethered cofactors (e.g., SAGA [Spt-Ada-Gcn5-acetyltransferase], Mediator, TUP, NuA4, SWI/SNF, and RPD3-L) are generally bound to promoters prior to induction ("poised"), rather than recruited upon induction, whereas induction recruits the preinitiation complex (PIC) to DNA. Through depletion and/or deletion experiments, we show that SAGA does not function at constitutive promoters, although a SAGA-independent Gcn5 acetylates +1 nucleosomes there. When inducible promoters are poised, SAGA catalyzes +1 nucleosome acetylation but not PIC assembly. When induced, SAGA catalyzes acetylation, deubiquitylation, and PIC assembly. Surprisingly, SAGA mediates induction by creating a PIC that allows TFIID (transcription factor II-D) to stably associate, rather than creating a completely TFIID-independent PIC, as generally thought. These findings suggest that inducible systems, where present, are integrated with constitutive systems.

Predicting which genes will respond to transcription factor perturbations

The ability to predict which genes will respond to the perturbation of a transcription factor serves as a benchmark for our systems-level understanding of transcriptional regulatory networks. In previous work, machine learning models have been trained to predict static gene expression levels in a biological sample by using data from the same or similar samples, including data on their transcription factor binding locations, histone marks, or DNA sequence. We report on a different challenge-training machine learning models to predict which genes will respond to the perturbation of a transcription factor without using any data from the perturbed cells. We find that existing transcription factor location data (ChIP-seq) from human cells have very little detectable utility for predicting which genes will respond to perturbation of a transcription factor. Features of genes, including their preperturbation expression level and expression variation, are very useful for predicting responses to perturbation of any transcription factor. This shows that some genes are poised to respond to transcription factor perturbations and others are resistant, shedding light on why it has been so difficult to predict responses from binding locations. Certain histone marks, including H3K4me1 and H3K4me3, have some predictive power when located downstream of the transcription start site. However, the predictive power of histone marks is much less than that of gene expression level and expression variation. Sequence-based or epigenetic properties of genes strongly influence their tendency to respond to direct transcription factor perturbations, partially explaining the oft-noted difficulty of predicting responsiveness from transcription factor binding location data. These molecular features are largely reflected in and summarized by the gene's expression level and expression variation. Code is available at https://github.com/BrentLab/TFPertRespExplainer.

Histone acetyltransferase Gcn5 regulates gene expression by promoting the transcription of histone methyltransferase SET1

Many chromatin modifying factors regulate gene expression in an as-yet-unknown indirect manner. Revealing the molecular basis for this indirect gene regulation will help understand their precise roles in gene regulation and associated biological processes. Here, we studied histone modifying enzymes that indirectly regulate gene expression by modulating the expression of histone methyltransferase, Set1. Through unbiased screening of the histone H3/H4 mutant library, we identified 13 histone substitution mutations with reduced levels of Set1 and H3K4 trimethylation (H3K4me3) and 2 mutations with increased levels of Set1 and H3K4me3, which concentrate at 3 structure clusters. Among these substitutions, the H3K14A mutant substantially reduces SET1 transcription and H3K4me3. H3K14 is acetylated by histone acetyltransferase Gcn5 at SET1 promoter, which then promotes SET1 transcription to maintain normal H3K4me3 levels. In contrast, the histone deacetylase Rpd3 deacetylates H3K14 to repress SET1 transcription and hence reduce H3K4me3 levels, establishing a dynamic crosstalk between H3K14ac and H3K4me3. By promoting the transcription of SET1 and maintaining H3K4me3 levels, Gcn5 regulates the transcription of a subset gene in an indirect manner. Collectively, we propose a model wherein Gcn5 promotes the expression of chromatin modifiers to regulate histone crosstalk and gene transcription.

GLP-1 Notch-LAG-1 CSL control of the germline stem cell fate is mediated by transcriptional targets lst-1 and sygl-1

Stem cell systems are essential for the development and maintenance of polarized tissues. Intercellular signaling pathways control stem cell systems, where niche cells signal stem cells to maintain the stem cell fate/self-renewal and inhibit differentiation. In the C. elegans germline, GLP-1 Notch signaling specifies the stem cell fate, employing the sequence-specific DNA binding protein LAG-1 to implement the transcriptional response. We undertook a comprehensive genome-wide approach to identify transcriptional targets of GLP-1 signaling. We expected primary response target genes to be evident at the intersection of genes identified as directly bound by LAG-1, from ChIP-seq experiments, with genes identified as requiring GLP-1 signaling for RNA accumulation, from RNA-seq analysis. Furthermore, we performed a time-course transcriptomics analysis following auxin inducible degradation of LAG-1 to distinguish between genes whose RNA level was a primary or secondary response of GLP-1 signaling. Surprisingly, only lst-1 and sygl-1, the two known target genes of GLP-1 in the germline, fulfilled these criteria, indicating that these two genes are the primary response targets of GLP-1 Notch and may be the sole germline GLP-1 signaling protein-coding transcriptional targets for mediating the stem cell fate. In addition, three secondary response genes were identified based on their timing following loss of LAG-1, their lack of a LAG-1 ChIP-seq peak and that their glp-1 dependent mRNA accumulation could be explained by a requirement for lst-1 and sygl-1 activity. Moreover, our analysis also suggests that the function of the primary response genes lst-1 and sygl-1 can account for the glp-1 dependent peak protein accumulation of FBF-2, which promotes the stem cell fate and, in part, for the spatial restriction of elevated LAG-1 accumulation to the stem cell region.

Stem cell systems are central to tissue development, homeostasis and regeneration, where niche to stem cell signaling pathways promote the stem cell fate/self-renewal and inhibit differentiation. The evolutionarily conserved GLP-1 Notch signaling pathway in the C. elegans germline is an experimentally tractable system, allowing dissection of control of the stem cell fate and inhibition of meiotic development. However, as in many systems, the primary molecular targets of the signaling pathway in stem cells is incompletely known, as are secondary molecular targets, and this knowledge is essential for a deep understanding of stem cell systems. Here we focus on the identification of the primary transcriptional targets of the GLP-1 signaling pathway that promotes the stem cell fate, employing unbiased multilevel genomic approaches. We identify only lst-1 and sygl-1, two of a number of previously reported targets, as likely the sole primary mRNA transcriptional targets of GLP-1 signaling that promote the germline stem cell fate. We also identify secondary GLP-1 signaling RNA and protein targets, whose expression shows dependence on lst-1 and sygl-1, where the protein targets reinforce the importance of posttranscriptional regulation in control of the stem cell fate.

Dual threshold optimization and network inference reveal convergent evidence from TF binding locations and TF perturbation responses

A high-confidence map of the direct, functional targets of each transcription factor (TF) requires convergent evidence from independent sources. Two significant sources of evidence are TF binding locations and the transcriptional responses to direct TF perturbations. Systematic data sets of both types exist for yeast and human, but they rarely converge on a common set of direct, functional targets for a TF. Even the few genes that are both bound and responsive may not be direct functional targets. Our analysis shows that when there are many nonfunctional binding sites and many indirect targets, nonfunctional sites are expected to occur in the cis-regulatory DNA of indirect targets by chance. To address this problem, we introduce dual threshold optimization (DTO), a new method for setting significance thresholds on binding and perturbation-response data, and show that it improves convergence. It also enables comparison of binding data to perturbation-response data that have been processed by network inference algorithms, which further improves convergence. The combination of dual threshold optimization and network inference greatly expands the high-confidence TF network map in both yeast and human. Next, we analyze a comprehensive new data set measuring the transcriptional response shortly after inducing overexpression of a yeast TF. We also present a new yeast binding location data set obtained by transposon calling cards and compare it to recent ChIP-exo data. These new data sets improve convergence and expand the high-confidence network synergistically.

Does genome surveillance explain the global discrepancy between binding and effect of chromatin factors?

Knocking out a chromatin factor often does not alter the transcription of its binding targets. What explains the observed disconnect between binding and effect? We hypothesize that this discrepancy could be associated with the role of chromatin factors in maintaining genetic and epigenetic integrity at promoters, and not necessarily with transcription. Through re-analysis of published datasets, we present several lines of evidence that support our hypothesis and deflate the popular assumptions. We also tested the hypothesis through mutation accumulation assays on yeast knockouts of chromatin factors. Altogether, the proposed hypothesis presents a simple explanation for the global discord between chromatin factor binding and effect. Future work in this direction might fortify the hypothesis and elucidate the underlying mechanisms.

Global role for coactivator complexes in RNA polymerase II transcription

SAGA and TFIID are related transcription complexes, which were proposed to alternatively deliver TBP at different promoter classes. Recent genome-wide studies in yeast revealed that both complexes are required for the transcription of a vast majority of genes by RNA polymerase II raising new questions about the role of coactivators.

The SAGA/TREX-2 subunit Sus1 binds widely to transcribed genes and affects mRNA turnover globally

BACKGROUND

Eukaryotic transcription is regulated through two complexes, the general transcription factor IID (TFIID) and the coactivator Spt-Ada-Gcn5 acetyltransferase (SAGA). Recent findings confirm that both TFIID and SAGA contribute to the synthesis of nearly all transcripts and are recruited genome-wide in yeast. However, how this broad recruitment confers selectivity under specific conditions remains an open question.

RESULTS

Here we find that the SAGA/TREX-2 subunit Sus1 associates with upstream regulatory regions of many yeast genes and that heat shock drastically changes Sus1 binding. While Sus1 binding to TFIID-dominated genes is not affected by temperature, its recruitment to SAGA-dominated genes and RP genes is significantly disturbed under heat shock, with Sus1 relocated to environmental stress-responsive genes in these conditions. Moreover, in contrast to recent results showing that SAGA deubiquitinating enzyme Ubp8 is dispensable for RNA synthesis, genomic run-on experiments demonstrate that Sus1 contributes to synthesis and stability of a wide range of transcripts.

CONCLUSIONS

Our study provides support for a model in which SAGA/TREX-2 factor Sus1 acts as a global transcriptional regulator in yeast but has differential activity at yeast genes as a function of their transcription rate or during stress conditions.

SAGA Is a General Cofactor for RNA Polymerase II Transcription

Prior studies suggested that SAGA and TFIID are alternative factors that promote RNA polymerase II transcription with about 10% of genes in S. cerevisiae dependent on SAGA. We reassessed the role of SAGA by mapping its genome-wide location and role in global transcription in budding yeast. We find that SAGA maps to the UAS elements of most genes, overlapping with Mediator binding and irrespective of previous designations of SAGA or TFIID-dominated genes. Disruption of SAGA through mutation or rapid subunit depletion reduces transcription from nearly all genes, measured by newly-synthesized RNA. We also find that the acetyltransferase Gcn5 synergizes with Spt3 to promote global transcription and that Spt3 functions to stimulate TBP recruitment at all tested genes. Our data demonstrate that SAGA acts as a general cofactor required for essentially all RNA polymerase II transcription and is not consistent with the previous classification of SAGA and TFIID-dominated genes.

Coactivators and general transcription factors have two distinct dynamic populations dependent on transcription

SAGA and ATAC are two distinct chromatin modifying co‐activator complexes with distinct enzymatic activities involved in RNA polymerase II (Pol II) transcription regulation. To investigate the mobility of co‐activator complexes and general transcription factors in live‐cell nuclei, we performed imaging experiments based on photobleaching. SAGA and ATAC, but also two general transcription factors (TFIID and TFIIB), were highly dynamic, exhibiting mainly transient associations with chromatin, contrary to Pol II, which formed more stable chromatin interactions. Fluorescence correlation spectroscopy analyses revealed that the mobile pool of the two co‐activators, as well as that of TFIID and TFIIB, can be subdivided into “fast” (free) and “slow” (chromatin‐interacting) populations. Inhibiting transcription elongation decreased H3K4 trimethylation and reduced the “slow” population of SAGA, ATAC, TFIIB and TFIID. In addition, inhibiting histone H3K4 trimethylation also reduced the “slow” populations of SAGA and ATAC. Thus, our results demonstrate that in the nuclei of live cells the equilibrium between fast and slow population of SAGA or ATAC complexes is regulated by active transcription via changes in the abundance of H3K4me3 on chromatin.

Past Roadblocks and New Opportunities in Transcription Factor Network Mapping

One of the principal mechanisms by which cells differentiate and respond to changes in external signals or conditions is by changing the activity levels of transcription factors (TFs). This changes the transcription rates of target genes via the cell's TF network, which ultimately contributes to reconfiguring cellular state. Since microarrays provided our first window into global cellular state, computational biologists have eagerly attacked the problem of mapping TF networks, a key part of the cell's control circuitry. In retrospect, however, steady-state mRNA abundance levels were a poor substitute for TF activity levels and gene transcription rates. Likewise, mapping TF binding through chromatin immunoprecipitation proved less predictive of functional regulation and less amenable to systematic elucidation of complete networks than originally hoped. This review explains these roadblocks and the current, unprecedented blossoming of new experimental techniques built on second-generation sequencing, which hold out the promise of rapid progress in TF network mapping.

The SAGA coactivator complex acts on the whole transcribed genome and is required for RNA polymerase II transcription

The SAGA coactivator complex contains distinct chromatin-modifying activities and is recruited by DNA-bound activators to regulate the expression of a subset of genes. Bonnet et al. discovered that SAGA acetylates the promoters and deubiquitinates the transcribed region of all expressed genes. SAGA also plays a critical role for RNA polymerase II recruitment at all expressed genes. This study uncovers a new function for SAGA as a bona fide cofactor for all RNA polymerase II transcription.

The SAGA (Spt–Ada–Gcn5 acetyltransferase) coactivator complex contains distinct chromatin-modifying activities and is recruited by DNA-bound activators to regulate the expression of a subset of genes. Surprisingly, recent studies revealed little overlap between genome-wide SAGA-binding profiles and changes in gene expression upon depletion of subunits of the complex. As indicators of SAGA recruitment on chromatin, we monitored in yeast and human cells the genome-wide distribution of histone H3K9 acetylation and H2B ubiquitination, which are respectively deposited or removed by SAGA. Changes in these modifications after inactivation of the corresponding enzyme revealed that SAGA acetylates the promoters and deubiquitinates the transcribed region of all expressed genes. In agreement with this broad distribution, we show that SAGA plays a critical role for RNA polymerase II recruitment at all expressed genes. In addition, through quantification of newly synthesized RNA, we demonstrated that SAGA inactivation induced a strong decrease of mRNA synthesis at all tested genes. Analysis of the SAGA deubiquitination activity further revealed that SAGA acts on the whole transcribed genome in a very fast manner, indicating a highly dynamic association of the complex with chromatin. Thus, our study uncovers a new function for SAGA as a bone fide cofactor for all RNA polymerase II transcription.

Mapping yeast transcriptional networks

The term “transcriptional network” refers to the mechanism(s) that underlies coordinated expression of genes, typically involving transcription factors (TFs) binding to the promoters of multiple genes, and individual genes controlled by multiple TFs. A multitude of studies in the last two decades have aimed to map and characterize transcriptional networks in the yeast Saccharomyces cerevisiae. We review the methodologies and accomplishments of these studies, as well as challenges we now face. For most yeast TFs, data have been collected on their sequence preferences, in vivo promoter occupancy, and gene expression profiles in deletion mutants. These systematic studies have led to the identification of new regulators of numerous cellular functions and shed light on the overall organization of yeast gene regulation. However, many yeast TFs appear to be inactive under standard laboratory growth conditions, and many of the available data were collected using techniques that have since been improved. Perhaps as a consequence, comprehensive and accurate mapping among TF sequence preferences, promoter binding, and gene expression remains an open challenge. We propose that the time is ripe for renewed systematic efforts toward a complete mapping of yeast transcriptional regulatory mechanisms.

Bridge-induced chromosome translocation in yeast relies upon a Rad54/Rdh54-dependent, Pol32-independent pathway

While in mammalian cells the genetic determinism of chromosomal translocation remains unclear, the yeast Saccharomyces cerevisiae has become an ideal model system to generate ad hoc translocations and analyze their cellular and molecular outcome. A linear DNA cassette carrying a selectable marker flanked by perfect homologies to two chromosomes triggers a bridge-induced translocation (BIT) in budding yeast, with variable efficiency. A postulated two-step process to produce BIT translocants is based on the cooperation between the Homologous Recombination System (HRS) and Break-Induced Replication (BIR); however, a clear indication of the molecular factors underlying the genetic mechanism is still missing. In this work we provide evidence that BIT translocation is elicited by the Rad54 helicase and completed by a Pol32-independent replication pathway. Our results demonstrate also that Rdh54 is involved in the stability of the translocants, suggesting a mitotic role in chromosome pairing and segregation. Moreover, when RAD54 is over-expressed, an ensemble of secondary rearrangements between repeated DNA tracts arise after the initial translocation event, leading to severe aneuploidy with loss of genetic material, which prompts the identification of fragile sites within the yeast genome.

Taking into account nucleosomes for predicting gene expression

The eukaryotic genome is organized in a chain of nucleosomes that consist of 145-147 bp of DNA wrapped around a histone octamer protein core. Binding of transcription factors (TF) to nucleosomal DNA is frequently impeded, which makes it a challenging task to calculate TF occupancy at a given regulatory genomic site for predicting gene expression. Here, we review methods to calculate TF binding to DNA in the presence of nucleosomes. The main theoretical problems are (i) the computation speed that is becoming a bottleneck when partial unwrapping of DNA from the nucleosome is considered, (ii) the perturbation of the binding equilibrium by the activity of ATP-dependent chromatin remodelers, which translocate nucleosomes along the DNA, and (iii) the model parameterization from high-throughput sequencing data and fluorescence microscopy experiments in living cells. We discuss strategies that address these issues to efficiently compute transcription factor binding in chromatin.