Periodic Reporting for period 4 - ChromArch (Single Molecule Mechanisms of Spatio-Temporal Chromatin Architecture)
Reporting period: 2019-11-01 to 2021-10-31
Chromatin packaging into the nucleus of eukaryotic cells is highly complex. It not only serves to condense the genomic content into restricted space, but mainly to encode epigenetic traits ensuring temporally controlled and balanced transcription of genes and coordinated DNA replication and repair. Over the years it became evident that regulatory traits are not only encrypted along the one-dimensional sequence of DNA, but within the three-dimensional arrangement of chromatin. This non-random chromatin organization is of utmost importance for the correct read-out and control of genetic information. Transcription of a gene for example can be significantly increased by long-range chromatin interactions between enhancers and promoters, and misarrangements of chromatin structures are associated with severe diseases. Many open question related to chromatin oranization and gene regulation are unanswered. How long does it take to form chromatin loops, how long do functional connections, for example enhancer-promoter interactions, persist and what are the molecular mechanisms of loop formation and gene transcription?
We aimed at unveiling molecular mechanisms of chromatin organization and gene regulation by quantitative in vivo and in vitro single molecule experiments. We used single cell and single molecule fluorescence microscopy to measure the dynamics of organizational structures within chromatin and to study the molecular mechanisms of biomolecules mediating chromatin topology and gene regulation in the nucleus. Our goal was to enhance the mechanistic understanding of three-dimensional chromatin architecture and the regulatory underpinnings of gene regulation and to inspire experiments on the potential therapeutic utility of controlled modification of regulatory traits mediating chromatin topology and gene regulation.
To achieve our goals, we developed several new single molecule imaging, analysis and simulation methodologies. These enabled us to track individual fluorescent molecules in living cells and live zebrafish embryos, to extract kinetic parameters such as diffusion coefficients, bound fractions and residence times and to simulate their effect on gene regulation and gene regulatory networks. We found that architectural proteins interact dynamically with chromatin in the range of minutes, which sets upper limits on the time that architectural features might persist. For a molecular motor we observed a novel role in dynamic organisation of chromatin and the transciription process. We further found that transcription repressors and activators modify the frequency of gene transcription events and identified a role of transcription factor residence time on chromatin in transcription regulation. Moreover, we observed that transcription factor binding is modulated by nuclear volume during zebrafish development. Overall, our experiments reveal a previously unappreciated regulatory effect of biomolecular interaction times on chromatin organization and gene transcription, which might be exploited for therapeutic purposes.
We aimed at unveiling molecular mechanisms of chromatin organization and gene regulation by quantitative in vivo and in vitro single molecule experiments. We used single cell and single molecule fluorescence microscopy to measure the dynamics of organizational structures within chromatin and to study the molecular mechanisms of biomolecules mediating chromatin topology and gene regulation in the nucleus. Our goal was to enhance the mechanistic understanding of three-dimensional chromatin architecture and the regulatory underpinnings of gene regulation and to inspire experiments on the potential therapeutic utility of controlled modification of regulatory traits mediating chromatin topology and gene regulation.
To achieve our goals, we developed several new single molecule imaging, analysis and simulation methodologies. These enabled us to track individual fluorescent molecules in living cells and live zebrafish embryos, to extract kinetic parameters such as diffusion coefficients, bound fractions and residence times and to simulate their effect on gene regulation and gene regulatory networks. We found that architectural proteins interact dynamically with chromatin in the range of minutes, which sets upper limits on the time that architectural features might persist. For a molecular motor we observed a novel role in dynamic organisation of chromatin and the transciription process. We further found that transcription repressors and activators modify the frequency of gene transcription events and identified a role of transcription factor residence time on chromatin in transcription regulation. Moreover, we observed that transcription factor binding is modulated by nuclear volume during zebrafish development. Overall, our experiments reveal a previously unappreciated regulatory effect of biomolecular interaction times on chromatin organization and gene transcription, which might be exploited for therapeutic purposes.
Method development:
Single molecule tracking is well suited to unravel the functioning of architectural proteins and transcription factors within the complex environment of a live cell. In particular, information on kinetic parameters such as diffusion coefficients, bound fractions, residence times and on subpopulations of molecules within different functional states may be obtained. We developed several methodologies to facilitate analysis of and to extract kinetic parameters from single molecule data.
TrackIt is a user-friendly software to identify tracks of molecules from single molecule fluorescence microscopy movies and to apply various analysis schemes yielding kinetic information. TrackIt was published in Kuhn and Hettich et al., Sci Rep 2021.
GRID is an algorithm which enables extracting kinetic rates from single molecule fluorescence time series. For example, a spectrum of dissociation rates may be obtained from survival time distributions of bound molecules. GRID was published in Reisser and Hettich et al., Sci Rep 2020.
BIRD is an algorithm to extract kinetic rates of gene bursting from distributions of RNA molecules, measured in single cells on a single RNA molecule level. BIRD was published in Popp et al., Nucl Acids Res 2021.
SPLIT is an algorithm enabling separating single molecule tracks into components of different kinetic behaviour, e.g. diffusive and directed motion. SPLIT was publishd in Große-Berkenbusch et al., bioRxiv 2020.
CaiNet is a framework to simulate gene regulatory networks at molecular detail and to infer kinetic parameters from steady state measurements. CaiNet was published in Hettich et al., bioRxiv 2021.
We further developed the DNA binding domain of TALEs as tool to study temporal aspects of gene regulation and chromatin architecture (Clauß and Popp et al., Nucl Acids Res 2017).
ITM is an illumination scheme suited to classify bound states of molecules into long and short binding regimes and to map the spatial distribution of those classes obtained from a single cell. ITM was published in Reisser et al., Nat Comms 2018.
While live cell single molecule tracking reveals unprecedented insights into the functioning of cells, studying isolated cells limits the range of possible biological findings. We thus developed single molecue tracking of biomolecules in the natural environment of a live developing organism. This methodology was published in Reisser et al., Nat Comms 2018.
Scientific findings:
The architectural protein CTCF is involved in organizing topologically associating domains and chromatin loops. To gain a better understanding of its mechanism of action, we measured the time that CTCF is bound to chromatin within different periods of the cell cycle. We observed three subpopulations of CTCF exhibiting significantly different residence times. Our data suggest that CTCF scans DNA unspecifically in search for different subsets of specific target sites and provide information on the timescales over which topologically associating domains might be restructured (< 15 min). We further observed a drop of specific interactions with chromatin in S-phase, indicating that specific interactions need to be dissolved for replication to proceed. This work was published in Agarwal et al., Biophys. J 2017.
The molecular motor myosin VI mainly works in the cytoplasm in transportation and tethering tasks, but also localizes to the nucleus. We observed micrometer long tracks of myosin VI in the nucleus and found this motor contributes to chromosomal rearrangements. This opens the possibility that myosin VI performs a transportation function also in the nucleus. The findings were published in Große-Berkenbusch et al., bioRxiv 2020. We further revealed that myosin VI in the nucleus acts as the molecular anchor that holds RNA polymerase II in high density clusters. Perturbation of myosin VI leads to the disruption of RNA polymerase II localisation, chromatin organization and subsequently a decrease in gene expression. MVI thus has a fundamental role in the spatial regulation of gene expression. This work was published in Hari-Gupta et al., bioRxiv 2020.
Transcription of most genes is heterogeneous in time, with periods of active transcription and long periods of transcriptional quiescence. We questioned the role of target site residence time of transcription repressors and activators in regulating the timing of gene transcription. We found that both repression and activation of transcription is more efficient the longer the residence time of the transcription factor is. Moreover, a transcription factor solely alters the frequency of transcriptional bursts. The findings were published in Clauß and Popp et al., Nucl Acids Res 2017 and Popp et al., Nucl Acids Res 2021. We further observed that the residence time of the transcription factor SRF is prolonged upon transcription stimulation, and this correlates with the presence of a cofactor (published in Hipp et al., PNAS 2019). In another set of experiments we found that the mediator of Notch signalling, RBPJ, and its pancreas-specific paralog RBPJL differ in their DNA residence time, which might explain their differential modes of action (Pan et al., Cancers 2021). We further revealed that non-specific DNA binding properties of transcription factors regulate their search efficiency and occupancy of specific genomic sites (Raccaud et al., Nat Comms 2019). Using a simple state-based theoretical model that coarse-grains facilitated diffusion, we calculated that experimentally measured non-specific residence times allow for optimally fast target site search, but generally lead to low occupation frequencies of the specific target site (Hettich et al., J Theor Biol 2018).
In further experiments we studied transcription factor binding during early zebrafish developmental and observed that the frequency of specific transcription factor - DNA interactions increased after each cell devision cycle. This increase correlated to a decrease of the nuclear volume during early cell divisions, suggesting that a mechanism based on the law of mass action might facilitate transcription facor binding during early zebrafish development. This mechanism might contribute to timing of zygotic genome activation, an important developmental phase. This work was published in Reisser et al., Nat Comms 2018.
Single molecule tracking is well suited to unravel the functioning of architectural proteins and transcription factors within the complex environment of a live cell. In particular, information on kinetic parameters such as diffusion coefficients, bound fractions, residence times and on subpopulations of molecules within different functional states may be obtained. We developed several methodologies to facilitate analysis of and to extract kinetic parameters from single molecule data.
TrackIt is a user-friendly software to identify tracks of molecules from single molecule fluorescence microscopy movies and to apply various analysis schemes yielding kinetic information. TrackIt was published in Kuhn and Hettich et al., Sci Rep 2021.
GRID is an algorithm which enables extracting kinetic rates from single molecule fluorescence time series. For example, a spectrum of dissociation rates may be obtained from survival time distributions of bound molecules. GRID was published in Reisser and Hettich et al., Sci Rep 2020.
BIRD is an algorithm to extract kinetic rates of gene bursting from distributions of RNA molecules, measured in single cells on a single RNA molecule level. BIRD was published in Popp et al., Nucl Acids Res 2021.
SPLIT is an algorithm enabling separating single molecule tracks into components of different kinetic behaviour, e.g. diffusive and directed motion. SPLIT was publishd in Große-Berkenbusch et al., bioRxiv 2020.
CaiNet is a framework to simulate gene regulatory networks at molecular detail and to infer kinetic parameters from steady state measurements. CaiNet was published in Hettich et al., bioRxiv 2021.
We further developed the DNA binding domain of TALEs as tool to study temporal aspects of gene regulation and chromatin architecture (Clauß and Popp et al., Nucl Acids Res 2017).
ITM is an illumination scheme suited to classify bound states of molecules into long and short binding regimes and to map the spatial distribution of those classes obtained from a single cell. ITM was published in Reisser et al., Nat Comms 2018.
While live cell single molecule tracking reveals unprecedented insights into the functioning of cells, studying isolated cells limits the range of possible biological findings. We thus developed single molecue tracking of biomolecules in the natural environment of a live developing organism. This methodology was published in Reisser et al., Nat Comms 2018.
Scientific findings:
The architectural protein CTCF is involved in organizing topologically associating domains and chromatin loops. To gain a better understanding of its mechanism of action, we measured the time that CTCF is bound to chromatin within different periods of the cell cycle. We observed three subpopulations of CTCF exhibiting significantly different residence times. Our data suggest that CTCF scans DNA unspecifically in search for different subsets of specific target sites and provide information on the timescales over which topologically associating domains might be restructured (< 15 min). We further observed a drop of specific interactions with chromatin in S-phase, indicating that specific interactions need to be dissolved for replication to proceed. This work was published in Agarwal et al., Biophys. J 2017.
The molecular motor myosin VI mainly works in the cytoplasm in transportation and tethering tasks, but also localizes to the nucleus. We observed micrometer long tracks of myosin VI in the nucleus and found this motor contributes to chromosomal rearrangements. This opens the possibility that myosin VI performs a transportation function also in the nucleus. The findings were published in Große-Berkenbusch et al., bioRxiv 2020. We further revealed that myosin VI in the nucleus acts as the molecular anchor that holds RNA polymerase II in high density clusters. Perturbation of myosin VI leads to the disruption of RNA polymerase II localisation, chromatin organization and subsequently a decrease in gene expression. MVI thus has a fundamental role in the spatial regulation of gene expression. This work was published in Hari-Gupta et al., bioRxiv 2020.
Transcription of most genes is heterogeneous in time, with periods of active transcription and long periods of transcriptional quiescence. We questioned the role of target site residence time of transcription repressors and activators in regulating the timing of gene transcription. We found that both repression and activation of transcription is more efficient the longer the residence time of the transcription factor is. Moreover, a transcription factor solely alters the frequency of transcriptional bursts. The findings were published in Clauß and Popp et al., Nucl Acids Res 2017 and Popp et al., Nucl Acids Res 2021. We further observed that the residence time of the transcription factor SRF is prolonged upon transcription stimulation, and this correlates with the presence of a cofactor (published in Hipp et al., PNAS 2019). In another set of experiments we found that the mediator of Notch signalling, RBPJ, and its pancreas-specific paralog RBPJL differ in their DNA residence time, which might explain their differential modes of action (Pan et al., Cancers 2021). We further revealed that non-specific DNA binding properties of transcription factors regulate their search efficiency and occupancy of specific genomic sites (Raccaud et al., Nat Comms 2019). Using a simple state-based theoretical model that coarse-grains facilitated diffusion, we calculated that experimentally measured non-specific residence times allow for optimally fast target site search, but generally lead to low occupation frequencies of the specific target site (Hettich et al., J Theor Biol 2018).
In further experiments we studied transcription factor binding during early zebrafish developmental and observed that the frequency of specific transcription factor - DNA interactions increased after each cell devision cycle. This increase correlated to a decrease of the nuclear volume during early cell divisions, suggesting that a mechanism based on the law of mass action might facilitate transcription facor binding during early zebrafish development. This mechanism might contribute to timing of zygotic genome activation, an important developmental phase. This work was published in Reisser et al., Nat Comms 2018.
Our single molecule experiments on architectural proteins and transcription factors in live cells and live organisms and our theoretical considerations revealed important insights into the mechanistic underpinnings of chromatin organization and gene regulation:
- chromatin architecture is dynamic and chromatin loops may reorganize within a few minutes
- long-range chromatin reorganization might be actively driven by molecular motors
- clusters of RNA polymerase II involved in gene transcription are stabilized by the molecular motor myosin VI
- specific DNA residence time of transcription factors is a regulatory factor of transcription and may predict the transcriptional activity of a gene
- non-specific DNA binding properties of transcription factors are determinants of target site search and occupancy
- transcription of a gene may be descibed by an enhanced three-state model, in which transcription factor binding needs to be followed by several successive events before transcription is possible
- the nucleus plays a role as decreasing reaction volume during early embryo development. Intranuclear biomolecules are thus concentrated up during development, with tremendous impact on the fraction of biomolecules entering a complexed state.
- reflected light sheet microscopy enables single molecule tracking in the natural environment of live developing embryos for several hours
- chromatin architecture is dynamic and chromatin loops may reorganize within a few minutes
- long-range chromatin reorganization might be actively driven by molecular motors
- clusters of RNA polymerase II involved in gene transcription are stabilized by the molecular motor myosin VI
- specific DNA residence time of transcription factors is a regulatory factor of transcription and may predict the transcriptional activity of a gene
- non-specific DNA binding properties of transcription factors are determinants of target site search and occupancy
- transcription of a gene may be descibed by an enhanced three-state model, in which transcription factor binding needs to be followed by several successive events before transcription is possible
- the nucleus plays a role as decreasing reaction volume during early embryo development. Intranuclear biomolecules are thus concentrated up during development, with tremendous impact on the fraction of biomolecules entering a complexed state.
- reflected light sheet microscopy enables single molecule tracking in the natural environment of live developing embryos for several hours