Final Report Summary - DYNACLOCK (Dynamic protein-DNA interactomes and circadian transcription regulatory networks in mammals)
Circadian clocks orchestrate organisms’ daily rhythms in behavior and physiology. Such 24-hour rhythms are pervasive at all levels of biological organization, from complex behaviors, to organ specific physiology, down to the scale of cell-autonomous oscillators. The ERC starting grant DynaClock was organized around three main subprojects to investigate the temporal and genome conformation dynamics of protein-DNA interactomes underlying circadian transcription in mammals, and to study the kinetics by which individual mammalian genes transcribe their gene products. To tackle these questions, we followed a multi-disciplinary approach, including molecular biology experiments, computational analyses and biophysical modeling.
The first main objective was to understand the temporal dynamics of protein-DNA interactomes underlying circadian rhythms in gene expression in mammalian tissues, and how these determine circadian transcriptional outputs. In this context we have successfully performed time resolved location analysis for circadian activating transcription factors in mouse cells and liver tissue. These data allowed to perform a comparative study of transcription factor binding across different cell types in mouse, and also with human. Specifically, comparative analyses of ChIP-Seq data in mouse liver and fibroblasts revealed that only a fraction of BMAL1/CLOCK binding sites are shared between different cell types and that many sites are bound, active and accessible in a cell type specific way. Moreover, comparison between mouse and human indicated that binding sites are conserved to a similar extent between mouse and human as between two mouse cell types. Using this approach, we identified a putative new player in the clock, namely the ZEB1 repressor, whose loss of function in the cells shows a circadian phenotype reminiscent of reduced BMAL1 activity. The second major aim was to study whether the spatial organization of chromatin plays is reorganized in a daily manner, whether this might involve the circadian clock, and what the function consequences might be. To investigate these questions, we performed 4C-sequencing (4C-seq) experiments on a set of core clock genes as well as genes involved in rhythmic metabolic functions in the liver. We identified time-varying chromatin reorganization and DNA loops between circadian gene promoters and surrounding cis- genomic regions. Importantly, this fluctuating chromatin structure depended on a functional molecular clock. Moreover, 4C-seq experiments performed in kidney of WT animals supported that DNA looping was strongly correlated with the transcription state. Altogether, our data revealed time varying and clock-dependent chromatin structure implicated in the regulation of circadian gene expression, and thus provide new insight on the role of the mammalian clock to gene regulation. The third major objective consisted in the development of experimental and computation methods to investigate transcriptional bursting of single allele, endogenous mammalian genes in individual cells. We demonstrated for the first time that transcriptional bursts are ubiquitous in mammalian gene transcription, and that the temporal patterns of transcription activity are highly gene specific and these patterns were markedly altered by sequence modifications of cis-regulatory sequences, while chromatin state seemed to have a lesser effect. We also studied how the fine kinetics of transcriptional bursting of a single mammalian gene responds to different physiological stimuli. Finally we extended our mathematical models to identify minimal models of promoter cycles, which inform on the number and durations of rate-limiting steps responsible for refractory periods. We found that the structure of promoter cycles was gene specific and independent on genomic location.
The first main objective was to understand the temporal dynamics of protein-DNA interactomes underlying circadian rhythms in gene expression in mammalian tissues, and how these determine circadian transcriptional outputs. In this context we have successfully performed time resolved location analysis for circadian activating transcription factors in mouse cells and liver tissue. These data allowed to perform a comparative study of transcription factor binding across different cell types in mouse, and also with human. Specifically, comparative analyses of ChIP-Seq data in mouse liver and fibroblasts revealed that only a fraction of BMAL1/CLOCK binding sites are shared between different cell types and that many sites are bound, active and accessible in a cell type specific way. Moreover, comparison between mouse and human indicated that binding sites are conserved to a similar extent between mouse and human as between two mouse cell types. Using this approach, we identified a putative new player in the clock, namely the ZEB1 repressor, whose loss of function in the cells shows a circadian phenotype reminiscent of reduced BMAL1 activity. The second major aim was to study whether the spatial organization of chromatin plays is reorganized in a daily manner, whether this might involve the circadian clock, and what the function consequences might be. To investigate these questions, we performed 4C-sequencing (4C-seq) experiments on a set of core clock genes as well as genes involved in rhythmic metabolic functions in the liver. We identified time-varying chromatin reorganization and DNA loops between circadian gene promoters and surrounding cis- genomic regions. Importantly, this fluctuating chromatin structure depended on a functional molecular clock. Moreover, 4C-seq experiments performed in kidney of WT animals supported that DNA looping was strongly correlated with the transcription state. Altogether, our data revealed time varying and clock-dependent chromatin structure implicated in the regulation of circadian gene expression, and thus provide new insight on the role of the mammalian clock to gene regulation. The third major objective consisted in the development of experimental and computation methods to investigate transcriptional bursting of single allele, endogenous mammalian genes in individual cells. We demonstrated for the first time that transcriptional bursts are ubiquitous in mammalian gene transcription, and that the temporal patterns of transcription activity are highly gene specific and these patterns were markedly altered by sequence modifications of cis-regulatory sequences, while chromatin state seemed to have a lesser effect. We also studied how the fine kinetics of transcriptional bursting of a single mammalian gene responds to different physiological stimuli. Finally we extended our mathematical models to identify minimal models of promoter cycles, which inform on the number and durations of rate-limiting steps responsible for refractory periods. We found that the structure of promoter cycles was gene specific and independent on genomic location.