Our team is interested in the process of transcription and epigenetic control of mammalian cells with an important focus on T cell differentiation. An important aspect of our research relies to understanding the role of the Carboxy-Terminal Domain of RNA Polymerase II (Pol II) in transcriptional regulation (axis 1).
During differentiation, important epigenetic and transcriptional changes occur to allow stem cells to modulate expression of their original genetic program, resulting in dramatic transitions in cellular functions and properties. Enhancers and promoters of both coding and noncoding genes are major genomic modules, crucial for regulating this process (axes 2 and 3). Deciphering their mechanism of action represents therefore an essential task for the future years and is also a common theme in our 3 research topics.
Axis 1Role of the residues of the CTD of mammalian Pol II in the transcription cycle.
Pol II CTD and its post-translational modifications play an essential role in the process of recruitment, initiation, elongation of transcription as well as in RNA splicing, processing or stability. The CTD is composed of a repetition of a heptapeptide Y1S2P3T4S5P6S7. Importantly, Tyr1, Ser2, Ser5, Thr4 and Ser7 residues can all be modified by phosphorylation. The phosphorylated Ser5P and Ser2P are well described for their role in transcription initiation and elongation respectively, whereas Ser7P is believed to play a role in snRNA maturation. Less is known for the role of other residues in the CTD heptad. We previously characterized Thr4P and Tyr1P residues genome-wide, showing their potential importance in termination and promoter bidirectional transcription respectively (Figure 1A; (Descostes, Elife 2014 May 9:e02105, Hintermair, EMBO J. 2012 31, 2784).
We now pursue our investigations on various CTD mutants combining functional genomics and proteomics to better define the link between transcription and RNA processing and stability. Most recently, we have demonstrated the Tyr1 residues are essential in the control of termination process at both 5’ and 3’ ends of genes (Figure 1B). When these residues are mutated, a super-pervasive Pol II is generated that can transcribe over hundreds of kb in the genome (Shah, Maqbool et al, Mol. Cell 2017 69, 48). Furthermore, we provide evidence that interaction with the Mediator and Integrator complexes might be involved in this CTD-dependent control of termination.
Figure 1: Pol II CTD modifications during transcription cycle.
A- Model of Pol II transcription cycle integrating location of the CTD phosphoisoforms. While Ser5P associates with early step of transcription (initiation, pausing), Ser2P shows higher enrichment at gene bodies and after 3’ends. Our work also shed light on the roles and locations of Thr4P and Tyr1P. (Descostes, Elife 2014 May 9:e02105, Hintermair, EMBO J. 2012 31, 2784).
B- Tyr1 residues of the CTD are involved in the termination of in mammalian cells. The image shows an example of the massive read-through transcriptional phenotype due to Tyrosine mutations. This read-though can spread up to hundreds of kb both at 5’ and 3’ ends of the genes (Shah, Maqbool et al, Mol. Cell 2017 69, 48).
Axis 2Role of T-cell enhancers during differentiation.
We previously showed that transcription at enhancers is a strong hallmark of cell identity and tissue-specific expression in developing T-cells (Koch, NSMB 2011 18, 956). Importantly, we also showed the existence of novel genomic structures at both promoters and enhancers defined by large Transcription Initiation Platforms (TIPs) that share many features in common with the more recently defined Super-Enhancers. Investigating the role of more specific transcription factors, we demonstrated the essential role of Ets1 transcription factor in T cell lineage commitment and its dual association to both nucleosome-occupied and nucleosome-depleted regions (Cauchy, NAR 2016 44, 3567). In collaborative works, we also contributed to establish a novel genome-wide enhancer assay (Vanhille, Nat. Com. 2015 6, 6905), described a novel role for enhancers in promoter-isoform selection of the NFATc1 gene in thymocytes (Klein-Hessling, Nat. com. 2016 7, 11841) and to show that lineage-specific enhancers activate self-renewal genes in mouse macrophages and ES cells (Soucie, Science 2016 12, 351). We have now extended our enhancer investigations to the whole CD4 T-cell differentiation path and interrogate novel enhancer functions and the question of why genes use distinct sets of enhancers during differentiation.
Finally, we also investigate the process of enhancer neo-generation through small DNA insertion in the genome of cancer cells issued from T-ALL patients (Navarro, Nat. com. 2015 6:6905). Our current model at the TAL1 locus includes activation of an oncogene through derepression of the Tal1 oncogene. The Polycomb repressive complex does not seem to be able to fulfill its function following small mono-allelic insertion upstream of the gene. We are now investigating the mechanism of this derepression at this and other oncogenes.
Figure 2: Dynamic of enhancers and promoters in developing T cells.
A- Both promoters and enhancers recruit RNA Polymerase II and can be transcribed over large areas defining Transcription Initiation Platforms (TIPs). TIPs are a strong hallmark of cell identity. Promoter and enhancer TIPs have similar active epigenetic marking but transcription elongation marks are generally absent from enhancer sites (Koch, NSMB 2011 18, 956).
B- Dynamics of enhancer (H3K27ac) and promoter (H3K4me3) epigenetic marking at the Runx3 locus. Runx3 is a transcription factor essential for lineage to CD4+ or CD8+ T cell subtypes. Our genome-wide analyses support a model in which the number of enhancers associated to one gene correlates the likelihood of isoform expression during differentiation (unpublished).
Axis 3Determinants of promoter identity in mammalian cells.
As a follow-up on our previous studies in which we showed that CpG island-containing promoters open chromatin independently of transcription (Fenouil, Gen. Res. 2012 22, 2399), we are seeking the determinants of nucleosome exclusion CpG islands (CGIs) containing promoters. In previous works, we contributed to characterize their role in replication (Cayrou, Gen. Res. 2015, 25:1873) or the expression of long upstream transcripts at promoters (Lepoivre, BMC Gen. 2013 23;14:914). We are now proposing that DNA secondary structures might play essential role within CGIs to open chromatin at both active and inactive promoters. We thus develop an effort to isolate these structures in vivo through the development of innovative technologies and are making the link between their formation, the processes of transcription initiation and activation and that of promoter repression.
A- Active promoters are ranked by growing CpG content and their nucleosome depletion (middle panels) trend follows that of the CGIs length (left panels).
B- Inhibition of transcription does not result in promoter closure. Human B cells were treated with a-amanitin for the indicated time, resulting in transcription inhibition and Pol II degradation. While Pol II disappears from the promoter (ChIP-seq), nucleosome depletion is only modestly impaired (MNase-seq).
The research program we develop articulates these 3 axes and aims to apply some aspects of promoter/enhancer conceptual knowledge, to the understanding of T-ALL leukemia.