Transcriptional Regulation in the G1-S Cell Cycle Stage in Fungi: Insights through Computational Analysis

The transcription factor complexes Mlu1-box binding factor (MBF) and Swi4/6 cell cycle box binding factor (SBF) regulate the cell cycle in Saccharomyces cerevisiae. They activate hundreds of genes and are responsible for nor-mal cell cycle progression from G1 to S phase. We investigated the conservation of MBF and SBF binding sites during fungal evolution. Orthologs of S. cerevisiae targets of these transcription factors were identified in 37 fungal species and their upstream regions were analyzed for putative transcription factor binding sites. Both groups displayed enrichment in specific putative regulatory DNA sequences in their upstream regions and showed different preferred upstream motif loca-tions, variable patterns of evolutionary conservation of the motifs and enrichment in unique biological functions for the regulated genes. The results indicate that despite high sequence similarity of upstream DNA motifs putatively associated with G1-S transcriptional regulation by MBF and SBF transcription factors, there are important upstream sequence feature differences that may help differentiate the two seemingly similar regulatory modes. The incorporation of upstream motif sequence comparison, positional distribution and evolutionary variability of the motif can complement functional infor-mation about roles of the respective gene products and help elucidate transcriptional regulatory pathways and functions.

Download Full-text

Ash1 and Tup1 Dependent Repression of the Saccharomyces cerevisiae HO promoter Requires Activator-Dependent Nucleosome Eviction

10.1101/2020.09.29.318063 ◽

2020 ◽

Author(s):

Emily J. Parnell ◽

Timothy J. Parnell ◽

Chao Yan ◽

Lu Bai ◽

David J. Stillman

Keyword(s):

Saccharomyces Cerevisiae ◽

Cell Cycle ◽

Transcription Factor ◽

Transcriptional Regulation ◽

Dna Sequences ◽

Binding Sites ◽

Genome Wide ◽

Intergenic Regions ◽

Mother Cells ◽

Activation Cycle

ABSTRACTTranscriptional regulation of the Saccharomyces cerevisiae HO gene is highly complex, requiring a balance of multiple activating and repressing factors to ensure that only a few transcripts are produced in mother cells within a narrow window of the cell cycle. Here, we show that the Ash1 repressor associates with two DNA sequences that are usually concealed within nucleosomes in the HO promoter and recruits the Tup1 corepressor and the Rpd3 histone deacetylase, both of which are required for full repression in daughters. Genome-wide ChIP identified greater than 200 additional sites of co-localization of these factors, primarily within large, intergenic regions from which they could regulate adjacent genes. Most Ash1 binding sites are in nucleosome depleted regions (NDRs), while a small number overlap nucleosomes, similar to HO. We demonstrate that Ash1 binding to the HO promoter does not occur in the absence of the Swi5 transcription factor, which recruits coactivators that evict nucleosomes, including the nucleosomes obscuring the Ash1 binding sites. In the absence of Swi5, artificial nucleosome depletion allowed Ash1 to bind, demonstrating that nucleosomes are inhibitory to Ash1 binding. The location of binding sites within nucleosomes may therefore be a mechanism for limiting repressive activity to periods of nucleosome eviction that are otherwise associated with activation of the promoter. Our results illustrate that activation and repression can be intricately connected, and events set in motion by an activator may also ensure the appropriate level of repression and reset the promoter for the next activation cycle.

Download Full-text

Ash1 and Tup1 dependent repression of the Saccharomyces cerevisiae HO promoter requires activator-dependent nucleosome eviction

PLoS Genetics ◽

10.1371/journal.pgen.1009133 ◽

2020 ◽

Vol 16 (12) ◽

pp. e1009133

Author(s):

Emily J. Parnell ◽

Timothy J. Parnell ◽

Chao Yan ◽

Lu Bai ◽

David J. Stillman

Keyword(s):

Saccharomyces Cerevisiae ◽

Cell Cycle ◽

Transcription Factor ◽

Transcriptional Regulation ◽

Dna Sequences ◽

Binding Sites ◽

Genome Wide ◽

Intergenic Regions ◽

Mother Cells ◽

Activation Cycle

Transcriptional regulation of the Saccharomyces cerevisiae HO gene is highly complex, requiring a balance of multiple activating and repressing factors to ensure that only a few transcripts are produced in mother cells within a narrow window of the cell cycle. Here, we show that the Ash1 repressor associates with two DNA sequences that are usually concealed within nucleosomes in the HO promoter and recruits the Tup1 corepressor and the Rpd3 histone deacetylase, both of which are required for full repression in daughters. Genome-wide ChIP identified greater than 200 additional sites of co-localization of these factors, primarily within large, intergenic regions from which they could regulate adjacent genes. Most Ash1 binding sites are in nucleosome depleted regions (NDRs), while a small number overlap nucleosomes, similar to HO. We demonstrate that Ash1 binding to the HO promoter does not occur in the absence of the Swi5 transcription factor, which recruits coactivators that evict nucleosomes, including the nucleosomes obscuring the Ash1 binding sites. In the absence of Swi5, artificial nucleosome depletion allowed Ash1 to bind, demonstrating that nucleosomes are inhibitory to Ash1 binding. The location of binding sites within nucleosomes may therefore be a mechanism for limiting repressive activity to periods of nucleosome eviction that are otherwise associated with activation of the promoter. Our results illustrate that activation and repression can be intricately connected, and events set in motion by an activator may also ensure the appropriate level of repression and reset the promoter for the next activation cycle.

Download Full-text

DeepGRN: Prediction of transcription factor binding site across cell-types using attention-based deep neural networks

10.21203/rs.3.rs-19323/v3 ◽

2021 ◽

Author(s):

Chen Chen ◽

Jie Hou ◽

Xiaowen Shi ◽

Hua Yang ◽

James A. Birchler ◽

...

Keyword(s):

Transcription Factor ◽

Transcription Factors ◽

Binding Site ◽

Transcription Factor Binding Site ◽

Dna Sequences ◽

Binding Sites ◽

Transcription Factor Binding ◽

Parallel Sequencing ◽

Factor Binding Site ◽

Factor Binding

Abstract BackgroundDue to the complexity of the biological systems, the prediction of the potential DNA binding sites for transcription factors remains a difficult problem in computational biology. Genomic DNA sequences and experimental results from parallel sequencing provide available information about the affinity and accessibility of genome and are commonly used features in binding sites prediction. The attention mechanism in deep learning has shown its capability to learn long-range dependencies from sequential data, such as sentences and voices. Until now, no study has applied this approach in binding site inference from massively parallel sequencing data. The successful applications of attention mechanism in similar input contexts motivate us to build and test new methods that can accurately determine the binding sites of transcription factors.ResultsIn this study, we propose a novel tool (named DeepGRN) for transcription factors binding site prediction based on the combination of two components: single attention module and pairwise attention module. The performance of our methods is evaluated on the ENCODE-DREAM in vivo Transcription Factor Binding Site Prediction Challenge datasets. The results show that DeepGRN achieves higher unified scores in 6 of 13 targets than any of the top four methods in the DREAM challenge. We also demonstrate that the attention weights learned by the model are correlated with potential informative inputs, such as DNase-Seq coverage and motifs, which provide possible explanations for the predictive improvements in DeepGRN.ConclusionsDeepGRN can automatically and effectively predict transcription factor binding sites from DNA sequences and DNase-Seq coverage. Furthermore, the visualization techniques we developed for the attention modules help to interpret how critical patterns from different types of input features are recognized by our model.

Download Full-text

Nuclear microenvironments modulate transcription from low-affinity enhancers

eLife ◽

10.7554/elife.28975 ◽

2017 ◽

Vol 6 ◽

Cited By ~ 45

Author(s):

Albert Tsai ◽

Anand K Muthusamy ◽

Mariana RP Alves ◽

Luke D Lavis ◽

Robert H Singer ◽

...

Keyword(s):

Transcription Factor ◽

Drosophila Melanogaster ◽

Transcription Factors ◽

Dna Sequences ◽

Transcriptional Activation ◽

Binding Sites ◽

General Feature ◽

Inverse Relationship ◽

Chromosomal Location ◽

Affinity Binding

Transcription factors bind low-affinity DNA sequences for only short durations. It is not clear how brief, low-affinity interactions can drive efficient transcription. Here, we report that the transcription factor Ultrabithorax (Ubx) utilizes low-affinity binding sites in the Drosophila melanogaster shavenbaby (svb) locus and related enhancers in nuclear microenvironments of high Ubx concentrations. Related enhancers colocalize to the same microenvironments independently of their chromosomal location, suggesting that microenvironments are highly differentiated transcription domains. Manipulating the affinity of svb enhancers revealed an inverse relationship between enhancer affinity and Ubx concentration required for transcriptional activation. The Ubx cofactor, Homothorax (Hth), was co-enriched with Ubx near enhancers that require Hth, even though Ubx and Hth did not co-localize throughout the nucleus. Thus, microenvironments of high local transcription factor and cofactor concentrations could help low-affinity sites overcome their kinetic inefficiency. Mechanisms that generate these microenvironments could be a general feature of eukaryotic transcriptional regulation.

Download Full-text

DeepGRN: prediction of transcription factor binding site across cell-types using attention-based deep neural networks

BMC Bioinformatics ◽

10.1186/s12859-020-03952-1 ◽

2021 ◽

Vol 22 (1) ◽

Author(s):

Chen Chen ◽

Jie Hou ◽

Xiaowen Shi ◽

Hua Yang ◽

James A. Birchler ◽

...

Keyword(s):

Transcription Factor ◽

Transcription Factors ◽

Binding Site ◽

Transcription Factor Binding Site ◽

Dna Sequences ◽

Binding Sites ◽

Transcription Factor Binding ◽

Parallel Sequencing ◽

Factor Binding Site ◽

Factor Binding

Abstract Background Due to the complexity of the biological systems, the prediction of the potential DNA binding sites for transcription factors remains a difficult problem in computational biology. Genomic DNA sequences and experimental results from parallel sequencing provide available information about the affinity and accessibility of genome and are commonly used features in binding sites prediction. The attention mechanism in deep learning has shown its capability to learn long-range dependencies from sequential data, such as sentences and voices. Until now, no study has applied this approach in binding site inference from massively parallel sequencing data. The successful applications of attention mechanism in similar input contexts motivate us to build and test new methods that can accurately determine the binding sites of transcription factors. Results In this study, we propose a novel tool (named DeepGRN) for transcription factors binding site prediction based on the combination of two components: single attention module and pairwise attention module. The performance of our methods is evaluated on the ENCODE-DREAM in vivo Transcription Factor Binding Site Prediction Challenge datasets. The results show that DeepGRN achieves higher unified scores in 6 of 13 targets than any of the top four methods in the DREAM challenge. We also demonstrate that the attention weights learned by the model are correlated with potential informative inputs, such as DNase-Seq coverage and motifs, which provide possible explanations for the predictive improvements in DeepGRN. Conclusions DeepGRN can automatically and effectively predict transcription factor binding sites from DNA sequences and DNase-Seq coverage. Furthermore, the visualization techniques we developed for the attention modules help to interpret how critical patterns from different types of input features are recognized by our model.

Download Full-text

LMO2 Regulates DNA Replication in Hematopoietic Cells

Blood ◽

10.1182/blood.v126.23.553.553 ◽

2015 ◽

Vol 126 (23) ◽

pp. 553-553

Author(s):

Marie-Claude Sincennes ◽

Magali Humbert ◽

Benoit Grondin ◽

Christophe Cazaux ◽

Veronique Lisi ◽

...

Keyword(s):

Cell Cycle ◽

Transcription Factor ◽

Transcription Factors ◽

Cell Differentiation ◽

Dna Replication ◽

Cell Fate ◽

Binding Sites ◽

Erythroid Differentiation ◽

S Phase

Abstract Oncogenic transcription factors are major drivers in acute leukemias. These oncogenes are believed to subvert normal cell identity via the establishment of gene expression programs that dictate cell differentiation and growth. The LMO2 oncogene, which is commonly activated in T-cell acute lymphoblastic leukemia (T-ALL), has a well-established function in transcription regulation. We and others previously demonstrated that LMO1 or LMO2 collaborate with the SCL transcription factor to activate a self-renewal program that converts non self-renewing progenitors into pre-leukemic stem cells. Here we demonstrate a non-transcriptional role of LMO2 in controlling cell fate by directly promoting DNA replication, a hitherto unrecognized mechanism that might also account for its oncogenic properties. To address the question whether LMO2 controls other functions via protein-protein interactions, we performed a proteome-wide screen for LMO2 interaction partners in Kit+ Lin- cells. In addition to known LMO2-interacting proteins such as LDB1 and to proteins associated with transcription, we unexpectedly identified new interactions with three essential DNA replication enzymes, namely minichromosome 6 (MCM6), DNA polymerase delta (POLD1) and DNA primase (PRIM1). First, we show that in Kit+ hematopoietic cells (TF-1), all components of the pre-replication complex co-immunoprecipitate with LMO2 but not with SCL, suggesting a novel SCL-independent function. Second, LMO2 is recruited to DNA replication origins in these cells together with MCM5. Third, tethering LMO2 to synthetic DNA sequences is sufficient to transform these into origins of replication. Indeed, we show by DNA capture that LMO2 fused to the DNA binding domain of GAL4 is sufficient to recruit DNA replication proteins to GAL4 binding sites on DNA. In vivo, this recruitment is sufficient to drive DNA replication in a manner which is dependent on the integrity of the GAL4 binding sites. These results provide unambiguous evidence for a role of LMO2 in directly controlling DNA replication. Cell cycle and cell differentiation are tightly coordinated during normal hematopoiesis, both during erythroid differentiation and during thymocyte development. We next addressed the functional importance of LMO2 in these two lineages. Erythroid cell differentiation proceeds through different stages from the CD71+Ter119- to the CD71-Ter119+. These stages are also distinguishable by morphological criteria. We observe that LMO2 protein levels directly correlate with the proportion of cells in S phase, i.e. both LMO2 levels and the proportions of cycling cells decrease with terminal erythroid differentiation. Strikingly, lowering LMO2 levels in fetal liver erythroid progenitors via shRNAs decreases the proportion of cells in S phase and arrests Epo-dependent cell growth. Despite a drastic decrease in the numbers of erythroid precursors, these cells differentiate readily to the CD71-Ter119+ stage. Therefore, LMO2 levels dictate cell fate in the erythroid lineage, by favoring DNA replication at the expense of terminal maturation. Conversely, ectopic expression in thymocytes induces DNA replication and drives cells into cell cycle, causing differentiation blockade. Our results define a novel role for the oncogenic transcription factor LMO2 in directly promoting DNA synthesis. To our knowledge, this is the first evidence for a non-transcriptional function of the LMO2 oncogene that drives cell cycle at the expense of differentiation, favouring progenitor cell expansion in the thymus, and causing T-ALL when ectopically expressed in the T lineage. We propose that the non-transcriptional control of DNA replication uncovered here for LMO2 may be a more common function of oncogenic transcription factors than previously appreciated. Disclosures No relevant conflicts of interest to declare.

Download Full-text

Prediction of transcription factor binding site across cell-types using attention-based deep neural networks

10.21203/rs.3.rs-19323/v1 ◽

2020 ◽

Author(s):

Chen Chen ◽

Jie Hou ◽

Xiaowen Shi ◽

Hua Yang ◽

James A. Birchler ◽

...

Keyword(s):

Transcription Factor ◽

Transcription Factors ◽

Binding Site ◽

Transcription Factor Binding Site ◽

Dna Sequences ◽

Binding Sites ◽

Transcription Factor Binding ◽

Parallel Sequencing ◽

Factor Binding Site ◽

Factor Binding

Abstract Background Due to the complexity of the biological systems, the prediction of the potential DNA binding sites for transcription factors remains a difficult problem in computational biology. Genomic DNA sequences and experimental results from parallel sequencing provide available information about the affinity and accessibility of genome and are commonly used features in binding sites prediction. The attention mechanism in deep learning has shown its capability to learn long-range dependencies from sequential data, such as sentences and voices. Until now, no study has applied this approach in binding site inference from massively parallel sequencing data. The successful applications of attention mechanism in similar input contexts motivate us to build and test new methods that can accurately determine the binding sites of transcription factors. Results In this study, we propose a novel tool (named DeepGRN) for transcription factors binding site prediction based on the combination of two components: single attention module and pairwise attention module. The performance of our methods is evaluated on the ENCODE-DREAM in vivo Transcription Factor Binding Site Prediction Challenge datasets. The results show that DeepGRN achieves higher unified scores in 6 of 13 targets than any of the top four methods in the DREAM challenge. We also demonstrate that the attention weights learned by the model are correlated with potential informative inputs, such as DNase-Seq coverage and motifs, which provide possible explanations for the predictive improvements in DeepGRN. Conclusions DeepGRN can automatically and effectively predict transcription factor binding sites from DNA sequences and DNase-Seq coverage. Furthermore, the visualization techniques we developed for the attention modules help to interpret how critical patterns from different types of input features are recognized by our model.

Download Full-text

Tissue-specific actions of Pax6 on proliferation-differentiation balance in the developing forebrain are Foxg1-dependent

10.1101/374074 ◽

2018 ◽

Cited By ~ 1

Author(s):

Idoia Quintana-Urzainqui ◽

Zrinko Kozić ◽

Soham Mitra ◽

Tian Tian ◽

Martine Manuel ◽

...

Keyword(s):

Gene Expression ◽

Cell Cycle ◽

Transcription Factor ◽

Transcriptional Regulation ◽

Transcription Factors ◽

Cell Cycle Gene ◽

Tissue Specific ◽

Proliferation And Differentiation ◽

Multiple Regions ◽

Cell Cycle Gene Expression

SummaryDifferences in the growth and maturation of diverse forebrain tissues depends on region-specific transcriptional regulation. Individual transcription factors act simultaneously in multiple regions that develop very differently, raising questions about the extent to which their actions vary regionally. We found that the transcription factor Pax6 affects the transcriptomes and the balance between proliferation and differentiation in opposite directions in murine diencephalon versus cortex. We tested several possible mechanisms to explain Pax6’s tissue-specific actions and found that the presence of the transcription factor Foxg1 in cortex but not diencephalon was most influential. We found that Foxg1 is responsible for many of the differences in cell cycle gene expression between diencephalon and cortex. In cortex lacking Foxg1, Pax6’s action on the balance of proliferation versus differentiation became diencephalon-like. Our findings reveal a mechanism for generating regional forebrain diversity in which the actions of one transcription factor completely reverse the actions of another.

Download Full-text

Interaction of Sp1 with the growth- and cell cycle-regulated transcription factor E2F.

Molecular and Cellular Biology ◽

10.1128/mcb.16.4.1659 ◽

1996 ◽

Vol 16 (4) ◽

pp. 1659-1667 ◽

Cited By ~ 238

Author(s):

J Karlseder ◽

H Rotheneder ◽

E Wintersberger

Keyword(s):

Cell Cycle ◽

Transcription Factor ◽

Transcription Factors ◽

Binding Sites ◽

Promoter Activity ◽

Binding Motif ◽

Transcription Machinery ◽

Transcription Factor E2f

Within the region around 150 bp upstream of the initiation codon, which was previously shown to suffice for growth-regulated expression, the murine thymidine kinase gene carries a single binding site for transcription factor Sp1; about 10 bp downstream of this site, there is a binding motif for transcription factor E2F. The latter protein appears to be responsible for growth regulation of the promoter. Mutational inactivation of either the Sp1 or the E2F site almost completely abolishes promoter activity, suggesting that the two transcription factors interact directly in delivering an activation signal to the basic transcription machinery. This was verified by demonstrating with the use of glutathione S-transferase fusion proteins that E2F and Sp1 bind to each other in vitro. For this interaction, the C-terminal part of Sp1 and the N terminus of E2F1, a domain also present in E2F2 and E2F3 but absent in E2F4 and E2F5, were essential. Accordingly, E2F1 to E2F3 but not E2F4 and E2F5 were found to bind sp1 in vitro. Coimmunoprecipitation experiments showed that complexes exist in vivo, and it was estabilished that the distance between the binding sites for the two transcription factors was critical for optimal promoter activity. Finally, in vivo footprinting experiments indicated that both the sp1 and E2F binding sites are occupied throughout the cell cycle. Mutation of either binding motif abolished binding of both transcription factors in vivo, which may indicate cooperative binding of the two proteins to chromatin-organized DNA. Our data are in line with the hypothesis that E2F functions as a growth- and cell cycle regulated tethering factor between Sp1 and the basic transcription machinery.

Download Full-text

Reciprocal regulation of the Epstein-Barr virus BamHI-F promoter by EBNA-1 and an E2F transcription factor.

Molecular and Cellular Biology ◽

10.1128/mcb.14.11.7144 ◽

1994 ◽

Vol 14 (11) ◽

pp. 7144-7152 ◽

Cited By ~ 40

Author(s):

N S Sung ◽

J Wilson ◽

M Davenport ◽

N D Sista ◽

J S Pagano

Keyword(s):

Cell Cycle ◽

Transcription Factor ◽

Transcription Factors ◽

Epithelial Cells ◽

Binding Site ◽

Binding Sites ◽

Epstein Barr Virus ◽

Barr Virus ◽

E2f Transcription Factor ◽

Epstein Barr

The Epstein-Barr virus BamHI-F promoter (Fp) is one of three used to transcribe the EBNA latency proteins, in particular, EBNA-1, the only viral gene product needed for episomal replication. Fp is distinguished by possession of the only EBNA-1 binding sites (the Q locus) in the Epstein-Barr virus genome outside oriP. Activity of Fp is negatively autoregulated by interaction of EBNA-1 at two sites in the Q locus, which is situated downstream of the RNA start site. We demonstrate in transient assays that this EBNA-1-mediated repression of Fp can be overcome by an E2F transcription factor which interacts with the DNA at a site centered between the two EBNA-1 binding sites within the Q locus. An E2F-1 fusion protein protects the sequence 5'-GGATGGCGGGTAATA-3' from DNase I digestion, and a DNA probe containing this sequence binds an E2F-specific protein complex from cell extracts, although this region is only loosely homologous with known consensus binding sites for E2F transcription factors. In mobility shift assays, E2F can displace the binding of EBNA-1 from the Q locus but not from oriP, where the E2F binding site is not present. E2F also activates expression of Fp in epithelial cells. These findings identify a potentially new binding site for members of the E2F family of transcription factors and suggest that such a factor is important for expression of EBNA-1 in lymphoid and epithelial cells by displacing EBNA-1 from the Q locus. In addition, the possibility that Fp activity is under cell cycle control is raised. Since the supply of functional E2F varies during the cell cycle and since in these assays overexpression of E2F can overcome repression of Fp by EBNA-1, control of transcription of EBNA-1 mRNA by cell cycle regulatory factors may help to bring about ordered replication of episomes.

Download Full-text