Systematic analysis of protein complexes involved in the human RNA polymerase II machinery

La transcription, la maturation d’ARN, et le remodelage de la chromatine sont tous des processus centraux dans l'interprétation de l'information contenue dans l’ADN. Bien que beaucoup de complexes de protéines formant la machinerie cellulaire de transcription aient été étudiés, plusieurs restent encore à identifier et caractériser. En utilisant une approche protéomique, notre laboratoire a purifié plusieurs composantes de la machinerie de transcription de l’ARNPII humaine par double chromatographie d’affinité "TAP". Cette procédure permet l'isolement de complexes protéiques comme ils existent vraisemblablement in vivo dans les cellules mammifères, et l'identification de partenaires d'interactions par spectrométrie de masse. Les interactions protéiques qui sont validées bioinformatiquement, sont choisies et utilisées pour cartographier un réseau connectant plusieurs composantes de la machinerie transcriptionnelle. En appliquant cette procédure, notre laboratoire a identifié, pour la première fois, un groupe de protéines, qui interagit physiquement et fonctionnellement avec l’ARNPII humaine. Les propriétés de ces protéines suggèrent un rôle dans l'assemblage de complexes à plusieurs sous-unités, comme les protéines d'échafaudage et chaperonnes. L'objectif de mon projet était de continuer la caractérisation du réseau de complexes protéiques impliquant les facteurs de transcription. Huit nouveaux partenaires de l’ARNPII (PIH1D1, GPN3, WDR92, PFDN2, KIAA0406, PDRG1, CCT4 et CCT5) ont été purifiés par la méthode TAP, et la spectrométrie de masse a permis d’identifier de nouvelles interactions. Au cours des années, l’analyse par notre laboratoire des mécanismes de la transcription a contribué à apporter de nouvelles connaissances et à mieux comprendre son fonctionnement. Cette connaissance est essentielle au développement de médicaments qui cibleront les mécanismes de la transcription.

The role of the peptidyl prolyl isomerase Rrd1 in the transcriptional stress response

La régulation de la transcription est un processus complexe qui a évolué pendant des millions d’années permettant ainsi aux cellules de s’adapter aux changements environnementaux. Notre laboratoire étudie le rôle de la rapamycine, un agent immunosuppresseur et anticancéreux, qui mime la carence nutritionelle. Afin de comprendre les mécanismes impliqués dans la réponse a la rapamycine, nous recherchons des mutants de la levure Saccaromyces cerevisiae qui ont un phenotype altérée envers cette drogue. Nous avons identifié le gène RRD1, qui encode une peptidyl prolyl isomérase et dont la mutation rend les levures très résistantes à la rapamycine et il semble que se soit associé à une réponse transcriptionelle alterée. Mon projet de recherche de doctorat est d’identifier le rôle de Rrd1 dans la réponse à la rapamycine. Tout d’abord nous avons trouvé que Rrd1 interagit avec l’ARN polymérase II (RNAPII), plus spécifiquement avec son domaine C-terminal. En réponse à la rapamycine, Rrd1 induit un changement dans la conformation du domaine C-terminal in vivo permettant la régulation de l’association de RNAPII avec certains gènes. Des analyses in vitro ont également montré que cette action est directe et probablement liée à l’activité isomérase de Rrd1 suggérant un rôle pour Rrd1 dans la régulation de la transcription. Nous avons utilisé la technologie de ChIP sur micropuce pour localiser Rrd1 sur la majorité des gènes transcrits par RNAPII et montre que Rrd1 agit en tant que facteur d’élongation de RNAPII. Pour finir, des résultats suggèrent que Rrd1 n’est pas seulement impliqué dans la réponse à la rapamycine mais aussi à differents stress environnementaux, nous permettant ainsi d’établir que Rrd1 est un facteur d’élongation de la transcription requis pour la régulation de la transcription via RNAPII en réponse au stress.

Étude structure / fonction des sous-unités catalytiques de l'ARN polymérase II

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Étude de la variante d’histone H2A.Z et du cycle de phosphorylation de l’ARN polymérase II chez Saccharomyces cerevisiae

La chromatine est plus qu’un système d’empaquetage de l’ADN ; elle est le support de toutes les réactions liées à l’ADN dans le noyau des cellules eucaryotes et participe au contrôle de l’accès de l’ARN polymérase II (ARNPolII) à l’ADN. Responsable de la transcription de tous les ARNm des cellules eucaryotes, l’ARNPolII doit, suivant son recrutement aux promoteurs des gènes, transcrire l’ADN en traversant la matrice chromatinienne. Grâce au domaine C-terminal (CTD) de sa sous-unité Rpb1, elle coordonne la maturation de l’ARNm en cours de synthèse ainsi que les modifications de la chromatine, concomitantes à la transcription. Cette thèse s’intéresse à deux aspects de la transcription : la matrice, avec la localisation de la variante d’histone H2A.Z, et la machinerie de transcription avec le cycle de phosphorylation du CTD de l’ARNPolII. Suivant l’introduction, le chapitre 2 de cette thèse constitue un protocole détaillé et annoté de la technique de ChIP-chip, chez la levure Saccharomyces cerevisiae. Cette technique phare dans l’étude in vivo des phénomènes liés à l’ADN a grandement facilité l’étude du rôle de la chromatine dans les phénomènes nucléaires, en permettant de localiser sur le génome les marques et les variantes d’histones. Ce chapitre souligne l’importance de contrôles adéquats, spécifiques à l’étude de la chromatine. Au chapitre 3, grâce à la méthode de ChIP-chip, la variante d’histone H2A.Z est cartographiée au génome de la levure Saccharomyces cerevisiae avec une résolution d’environ 300 paires de bases. Nos résultats montrent que H2A.Z orne un à deux nucléosomes au promoteur de la majorité des gènes. L’enrichissement de H2A.Z est anticorrélé à la transcription et nos résultats suggèrent qu’elle prépare la chromatine pour l’activation des gènes. De plus H2A.Z semble réguler la localisation des nucléosomes. Le chapitre suivant s’intéresse à la transcription sous l’angle de la machinerie de transcription en se focalisant sur le cycle de phosphorylation de l’ARN polymérase II. Le domaine C-terminal de sa plus large sous-unité est formé de répétitions d’un heptapeptide YSPTSPS dont les résidus peuvent être modifiés au cours de la transcription. Cette étude localise les marques de phosphorylation des trois résidus sérine de manière systématique dans des souches mutantes des kinases et phosphatases. Nos travaux confirment le profil universel des marques de phosphorylations aux gènes transcrits. Appuyés par des essais in vitro, ils révèlent l’interaction complexe des enzymes impliqués dans la phosphorylation, et identifient Ssu72 comme la phosphatase de la sérine 7. Cet article appuie également la notion de « variantes » des marques de phosphorylation bien que leur étude spécifique s’avère encore difficile. La discussion fait le point sur les travaux qui ont suivi ces articles, et sur les expériences excitantes en cours dans notre laboratoire.

Étude de la fonction de la protéine RPAP4 et de son association avec l’ARN polymérase II

L’ARN polymérase II (ARNPII), l’enzyme responsable de la transcription des ARN messagers, procède au décodage du génome des organismes vivants. Cette fonction requiert l’action concertée de plusieurs protéines, les facteurs généraux de la transcription, par exemple, formant un réseau d’interactions protéine-protéine, plusieurs étant impliquées dans la régulation de l’ARNPII à différents niveaux. La régulation de la transcription a été largement étudiée durant les quatre dernières décennies. Néanmoins, nous en connaissons peu sur les mécanismes qui régulent l’ARNPII avant ou après la transcription. Dans la première partie de cette thèse, nous poursuivons la caractérisation du réseau d’interactions de l’ARNPII dans la fraction soluble de la cellule humaine, travail qui a débuté précédemment dans notre laboratoire. Ce réseau, développé à partir de la méthode de la purification d’affinité en tandem couplée à la spectrométrie de masse (AP-MS) et à des méthodes d’analyses bioinformatiques, nous amène une foule d’informations concernant la régulation de l’ARNPII avant et après son interaction avec la chromatine. Nous y identifions des protéines qui pourraient participer à l’assemblage de l’ARNPII telles des chaperonnes et les protéines du complexe R2TP/prefoldin-like ainsi que des protéines impliquées dans le transport nucléocytoplasmique. Au centre de ce réseau se trouvent RPAP4, une GTPase qui semble se positionner à l’interface entre ces protéines régulatrices et l’ARNPII. Nous avons donc entamé l’étude la fonction de RPAP4, ce qui nous a menés à la conclusion que RPAP4 est essentielle à l’import nucléaire de l’ARNPII au noyau, où elle exerce sa fonction. Nous avons également montré que les motifs G et GPN sont essentiels à la fonction de RPAP4. Le traitement des cellules avec le bénomyl nous montre aussi que la fonction de RPAP4 et l’import nucléaire de l’ARNPII requièrent l’action des microtubules. La deuxième partie de la thèse s’intéresse à une autre protéine positionnée au centre du réseau, RPAP2. Cette dernière partage plusieurs interactions avec RPAP4. Elle est aussi essentielle à la localisation nucléaire de l’ARNPII et interagit directement avec celle-ci. RPAP4 et RPAP2 étant toutes deux des protéines cytoplasmiques qui font la navette entre le noyau et le cytoplasme, nous présentons des évidences que RPAP4 est impliquée dans l’export nucléaire de RPAP2 pour permettre à celle-ci d’être disponible dans le cytoplasme pour l’import de l’ARNPII dans le noyau. Dans la troisième partie de la thèse, nous étudions plus en profondeur les modifications post-traductionnelles de RPAP4, ce qui nous aide à mieux comprendre sa propre régulation et sa fonction auprès de l’ARNPII. RPAP4 est phosphorylée en mitose par la MAP kinase ERK5. Cette phosphorylation favorise l’interaction entre RPAP4 et RPAP2, ce qui empêche RPAP2 d’interagir avec l’ARNPII pendant la mitose, prévenant du même coup, son interaction avec la chromatine pendant cette phase du cycle cellulaire où la transcription est presque inexistante.

Interactions of G-quadruplex binders and Topoisomerase I inhibitors in cancer cells

Top1-DNA cleavage complexes (Top1ccs) trigger an accumulation of antisense RNAPII transcripts specifically at active divergent CpG-island promoters in a replication independent and Top1 dependent manner, leading to transcription-dependent genome instability and altered transcription regulation. Using different cancer cell lines of colon and osteo origins, we show that they display different sensitivity to CPT and G4 binder that is independent from Top1 level. To look at the interactions between Top1 and G4, we show that co-treatment with G4 binders potentiate the cell cytotoxicity of CPT regardless of the treatment sequences. Potentiation is indicated by a reduced inhibition concentration (IC50) with a more profound cytotoxicity in CPT-resistant cell lines, HCT15 and U2OS, hence, indicating an interaction between Top1inhibitor and G4 binders. Moreover, computational analysis confirmed the present of G4 motifs in genes with CPT-induced antisense transcription. G4 motifs are present mostly 5000 bp upstream from transcription start site and notably lower in genes. Comparisons between genes with no antisense transcription and genes with antisense transcription show that G4 motifs in this region are notably lower in the genes with antisense transcripts. Since CPT increases negative supercoils at promoters of intermediate activity, the formation of G4 is also increased in CPT-treated cells. Suprisingly, formation of G4 is regulated in parallel to the transient stabilization of R-loops, indicating a role in response to CPT-induced stress. G4 formation is highly elevated in Pyridostatin treated cells, which previous study shows increased formation of γH2Ax foci. This effect is also seen in the CPT-resistant cell lines, HCT15, indicating that the formation is a general event in response to CPT. We also show that R-loop formation is greatly increased in Pyridostatin treated cells. In order to study the role of R-loops and G4 structures in Top1cc-dependant repair pathway, we inhibited tyrosyl-phosphodiestrase 1 (TDP-1) using a TDP-1 inhibitor.

Rtr1 is the Saccharomyces cerevisiae homolog of a novel family of RNA polymerase II-binding proteins.

Cells must rapidly sense and respond to a wide variety of potentially cytotoxic external stressors to survive in a constantly changing environment. In a search for novel genes required for stress tolerance in Saccharomyces cerevisiae, we identified the uncharacterized open reading frame YER139C as a gene required for growth at 37 degrees C in the presence of the heat shock mimetic formamide. YER139C encodes the closest yeast homolog of the human RPAP2 protein, recently identified as a novel RNA polymerase II (RNAPII)-associated factor. Multiple lines of evidence support a role for this gene family in transcription, prompting us to rename YER139C RTR1 (regulator of transcription). The core RNAPII subunits RPB5, RPB7, and RPB9 were isolated as potent high-copy-number suppressors of the rtr1Delta temperature-sensitive growth phenotype, and deletion of the nonessential subunits RPB4 and RPB9 hypersensitized cells to RTR1 overexpression. Disruption of RTR1 resulted in mycophenolic acid sensitivity and synthetic genetic interactions with a number of genes involved in multiple phases of transcription. Consistently, rtr1Delta cells are defective in inducible transcription from the GAL1 promoter. Rtr1 constitutively shuttles between the cytoplasm and nucleus, where it physically associates with an active RNAPII transcriptional complex. Taken together, our data reveal a role for members of the RTR1/RPAP2 family as regulators of core RNAPII function.

A Cell Biological Determination of Integrator Subunit Localization

Uridine-rich small nuclear (U snRNAs), with the exception of the U6 snRNA, are RNA polymerase II (RNAPII) transcripts. The mechanism of 3’ cleavage of snRNAs has been unknown until recently. This area was greatly advanced when 12 of the Integrator complex subunits (IntS) were purified in 2005 through their interaction with the C-terminal domain (CTD) of the large subunit (RpbI) of RNAPII. Subsequently, our lab performed a genome-wide RNAi screen that identified two more members of the complex that we have termed IntS13 and IntS14. We have determined that IntS9 and 11 mediate the 3’ cleavage of snRNAs, but the exact function of the other subunits remains unknown. However, through the use of a U7 snRNA-GFP reporter and RNAi knockdown of the Integrator subunits in Drosophila S2 cells, we have shown that all subunits are required for the proper processing of snRNAs, albeit to differing degrees. Because snRNA transcription takes place in the nucleus of the cell, it is expected that all of the Integrator subunits would exhibit nuclear localization, but the knowledge of discrete subnuclear localization (i.e. to Cajal bodies) of any of the subunits could provide important clues to the function of that subunit. In this study, we used a cell biological approach to determine the localization of the 14 Integrator subunits. We hypothesized that the majority of the subunits would be nuclear, however, a few would display distinct localization to the Cajal bodies, as this is where snRNA genes are localized and transcribed. The specific aims and results are: 1. To determine the subcellular localization of the 14 Integrator subunits. To accomplish this, mCherry and GFP tagged clones were generated for each of the 14 Drosophila and human Integrator subunits. Confocal microscopy studies revealed that the majority of the subunits were diffuse in the nucleus, however, IntS3 formed discrete subnuclear foci. Surprisingly, two of the subunits, IntS2 and 7 were observed in cytoplasmic foci. 2. To further characterize Integrator subunits with unique subcellular localizations. Colocalization studies with endogenous IntS3 and Cajal body marker, coilin, showed that these two proteins overlap, and from this we concluded that IntS3 localized to Cajal bodies. Additionally, colocalization studies with mCherry-tagged IntS2 and 7 and the P body marker, Dcp1, revealed that these proteins colocalize as well. IntS7, however, is more stable in cytoplasmic foci than Dcp1. It was also shown through RNAi knockdown of Integrator subunits, that the cytoplasmic localization of IntS2 and 7 is dependent on the expression of IntS1 and 11 in S2 cells.

RNA polymerase II large subunit degradation induced by ecteinascidin 743: Molecular characterization and subsequent rational investigation of antitumor mechanisms in cancers with clinical response

Ecteinascidin 743 (Et-743), which is a novel DNA minor groove alkylator with a unique spectrum of antitumor activity, is currently being evaluated in phase II/III clinical trials. Although the precise molecular mechanisms responsible for the observed antitumor activity are poorly understood, recent data suggests that post-translational modifications of RNA polymerase II Large Subunit (RNAPII LS) may play a central role in the cellular response to this promising anticancer agent. The stalling of an actively transcribing RNAPII LS at Et-743-DNA adducts is the initial cellular signal for transcription-coupled nucleotide excision repair (TC-NER). In this manner, Et-743 poisons TC-NER and produces DNA single strand breaks. Et-743 also inhibits the transcription and RNAPII LS-mediated expression of selected genes. Because the poisoning of TC-NER and transcription inhibition are critical components of the molecular response to Et-743 treatment, we have investigated if changes in RNAPII LS contribute to the disruption of these two cellular pathways. In addition, we have studied changes in RNAPII LS in two tumors for which clinical responses were reported in phase I/II clinical trials: renal cell carcinoma and Ewing's sarcoma. Our results demonstrate that Et-743 induces degradation of the RNAPII LS that is dependent on active transcription, a functional 26S proteasome, and requires functional TC-NER, but not global genome repair. Additionally, we have provided the first experimental data indicating that degradation of RNAPII LS might lead to the inhibition of activated gene transcription. A set of studies performed in isogenic renal carcinoma cells deficient in von Hippel-Lindau protein, which is a ubiquitin-E3-ligase for RNAPII LS, confirmed the central role of RNAPII LS degradation in the sensitivity to Et-743. Finally, we have shown that RNAPII LS is also degraded in Ewing's sarcoma tumors following Et-743 treatment and provide data to suggest that this event plays a role in decreased expression of the Ewing's sarcoma oncoprotein, EWS-Fli1. Altogether, these data implicate degradation of RNAPII LS as a critical event following Et-743 exposure and suggest that the clinical activity observed in renal carcinoma and Ewing's sarcoma may be mediated by disruption of molecular pathways requiring a fully functional RNAPII LS. ^

Structure-function analysis of human Integrator subunit-4

Structure-function analysis of human Integrator subunit 4 Anupama Sataluri Advisor: Eric. J. Wagner, Ph.D. Uridine-rich small nuclear RNAs (U snRNA) are RNA Polymerase-II (RNAPII) transcripts that are ubiquitously expressed and are known to be essential for gene expression. snRNAs play a key role in mRNA splicing and in histone mRNA expression. Inaccurate snRNA biosynthesis can lead to diseases related to defective splicing and histone mRNA expression. Although the 3′ end formation mechanism and processing machinery of other RNAPII transcripts such as mRNA has been well studied, the mechanism of snRNA 3′ end processing has remained a mystery until the recent discovery of the machinery that mediates this process. In 2005, a complex of 14 subunits (the Integrator complex) associated with RNA Polymerase-II was discovered. The 14subunits were annotated Integrator 1-14 based on their size. The subunits of this complex together were found to facilitate 3′ end processing of snRNA. Identification of the Integrator complex propelled research in the direction of understanding the events of snRNA 3’end processing. Recent studies from our lab confirmed that Integrator subunit (IntS) 9 and 11 together perform the endonucleolytic cleavage of the nascent snRNA 3′ end to generate mature snRNA. However, the role of other members of the Integrator complex remains elusive. Current research in our lab is focused on deciphering the role of each subunit within the Integrator complex This work specifically focuses on elucidating the role of human Integrator subunit 4 (IntS4) and understanding how it facilitates the overall function of the complex. IntS4 has structural similarity with a protein called “Symplekin”, which is part of the mRNA 3’end processing machinery. Symplekin has been thoroughly researched in recent years and structure-function correlation studies in the context of mRNA 3’end processing have reported a scaffold function for Symplekin due to the presence of HEAT repeat motifs in its N-terminus. Based upon the structural similarity between IntS4 and Symplekin, we hypothesized that Integrator subunit 4 may be behaving as a Symplekin-like scaffold molecule that facilitates the interaction between other members of the Integrator Complex. To answer this question, the two important goals of this study were to: 1) identify the region of IntS4, which is important for snRNA 3′ end processing and 2) determine binding partners of IntS4 which promote its function as a scaffold. IntS4 structurally consists of a highly conserved N-terminus with 8 HEAT repeats, followed by a nonconserved C- terminus. A series of siRNA resistant N and C-terminus deletion constructs as well as specific point mutants within its N-terminal HEAT repeats were generated for human IntS4 and, utilizing a snRNA transcriptional readthrough GFP-reporter assay, we tested their ability to rescue misprocessing. This assay revealed a possible scaffold like property of IntS4. To probe IntS4 for interaction partners, we performed co-immunoprecipitation on nuclear extracts of IntS4 expressing stable cell lines and identified IntS3 and IntS5 among other Integrator subunits to be binding partners which facilitate the scaffold like function of hIntS4. These findings have established a critical role for IntS4 in snRNA 3′ end processing, identified that both its N and C termini are essential for its function, and mapped putative interaction domains with other Integrator subunits.

Trajectory of DNA in the RNA polymerase II transcription preinitiation complex

By using site-specific protein-DNA photocrosslinking, we define the positions of TATA-binding protein, transcription factor IIB, transcription factor IIF, and subunits of RNA polymerase II (RNAPII) relative to promoter DNA within the human transcription preinitiation complex. The results indicate that the interface between the largest and second-largest subunits of RNAPII forms an extended, ≈240 Å channel that interacts with promoter DNA both upstream and downstream of the transcription start. By using electron microscopy, we show that RNAPII compacts promoter DNA by the equivalent of ≈50 bp. Together with the published structure of RNAPII, the results indicate that RNAPII wraps DNA around its surface and suggest a specific model for the trajectory of the wrapped DNA.

UV-induced inhibition of transcription involves repression of transcription initiation and phosphorylation of RNA polymerase II

Cells from patients with Cockayne syndrome (CS) are hypersensitive to DNA-damaging agents and are unable to restore damage-inhibited RNA synthesis. On the basis of repair kinetics of different types of lesions in transcriptionally active genes, we hypothesized previously that impaired transcription in CS cells is a consequence of defective transcription initiation after DNA damage induction. Here, we investigated the effect of UV irradiation on transcription by using an in vitro transcription system that allowed uncoupling of initiation from elongation events. Nuclear extracts prepared from UV-irradiated or mock-treated normal human and CS cells were assayed for transcription activity on an undamaged β-globin template. Transcription activity in nuclear extracts closely mimicked kinetics of transcription in intact cells: extracts from normal cells prepared 1 h after UV exposure showed a strongly reduced activity, whereas transcription activity was fully restored in extracts prepared 6 h after treatment. Extracts from CS cells exhibited reduced transcription activity at any time after UV exposure. Reduced transcription activity in extracts coincided with a strong reduction of RNA polymerase II (RNAPII) containing hypophosphorylated C-terminal domain, the form of RNAPII known to be recruited to the initiation complex. These results suggest that inhibition of transcription after UV irradiation is at least partially caused by repression of transcription initiation and not solely by blocked elongation at sites of lesions. Generation of hypophosphorylated RNAPII after DNA damage appears to play a crucial role in restoration of transcription. CS proteins may be required for this process in a yet unknown way.

In vitro characterization of mutant yeast RNA polymerase II with reduced binding for elongation factor TFIIS.

We have reported previously the isolation and genetic characterization of mutations in the gene encoding the largest subunit of yeast RNA polymerase II (RNAPII), which lead to 6-azauracil (6AU)-sensitive growth. It was suggested that these mutations affect the functional interaction between RNAPII and transcription-elongation factor TFIIS because the 6AU-sensitive phenotype of the mutant strains was similar to that of a strain defective in the production of TFIIS and can be suppressed by increasing the dosage of the yeast TFIIS-encoding gene, PPR2, RNAPIIs were purified and characterized from two independent 6AU-sensitive yeast mutants and from wild-type (wt) cells. In vitro, in the absence of TFIIS, the purified wt polymerase and the two mutant polymerases showed similar specific activity in polymerization, readthrough at intrinsic transcriptional arrest sites and nascent RNA cleavage. In contrast to the wt polymerase, both mutant polymerases were not stimulated by the addition of a 3-fold molar excess of TFIIS in assays of promoter-independent transcription, readthrough or cleavage. However, stimulation of the ability of the mutant RNAPIIs to cleave nascent RNA and to read through intrinsic arrest sites was observed at TFIIS:RNAPII molar ratios greater than 600:1. Consistent with these findings, the binding affinity of the mutant polymerases for TFIIS was found to be reduced by more than 50-fold compared with that of the wt enzyme. These studies demonstrate that TFIIS has an important role in the regulation of transcription by yeast RNAPII and identify a possible binding site for TFIIS on RNAPII.

Predicting Transcript Production Rates in Yeast With Sparse Linear Models

To provide biological insights into transcriptional regulation, a couple of groups have recently presented models relating the promoter DNA-bound transcription factors (TFs) to downstream gene’s mean transcript level or transcript production rates over time. However, transcript production is dynamic in response to changes of TF concentrations over time. Also, TFs are not the only factors binding to promoters; other DNA binding factors (DBFs) bind as well, especially nucleosomes, resulting in competition between DBFs for binding at same genomic location. Additionally, not only TFs, but also some other elements regulate transcription. Within core promoter, various regulatory elements influence RNAPII recruitment, PIC formation, RNAPII searching for TSS, and RNAPII initiating transcription. Moreover, it is proposed that downstream from TSS, nucleosomes resist RNAPII elongation.

Here, we provide a machine learning framework to predict transcript production rates from DNA sequences. We applied this framework in the S. cerevisiae yeast for two scenarios: a) to predict the dynamic transcript production rate during the cell cycle for native promoters; b) to predict the mean transcript production rate over time for synthetic promoters. As far as we know, our framework is the first successful attempt to have a model that can predict dynamic transcript production rates from DNA sequences only: with cell cycle data set, we got Pearson correlation coefficient Cp = 0.751 and coefficient of determination r2 = 0.564 on test set for predicting dynamic transcript production rate over time. Also, for DREAM6 Gene Promoter Expression Prediction challenge, our fitted model outperformed all participant teams, best of all teams, and a model combining best team’s k-mer based sequence features and another paper’s biologically mechanistic features, in terms of all scoring metrics.

Moreover, our framework shows its capability of identifying generalizable fea- tures by interpreting the highly predictive models, and thereby provide support for associated hypothesized mechanisms about transcriptional regulation. With the learned sparse linear models, we got results supporting the following biological insights: a) TFs govern the probability of RNAPII recruitment and initiation possibly through interactions with PIC components and transcription cofactors; b) the core promoter amplifies the transcript production probably by influencing PIC formation, RNAPII recruitment, DNA melting, RNAPII searching for and selecting TSS, releasing RNAPII from general transcription factors, and thereby initiation; c) there is strong transcriptional synergy between TFs and core promoter elements; d) the regulatory elements within core promoter region are more than TATA box and nucleosome free region, suggesting the existence of still unidentified TAF-dependent and cofactor-dependent core promoter elements in yeast S. cerevisiae; e) nucleosome occupancy is helpful for representing +1 and -1 nucleosomes’ regulatory roles on transcription.