Research Group Liu
Plant chromatin structure and dynamics
Dr. Chang Liu
ZMBP, Allgemeine Genetik
Auf der Morgenstelle 32
We focus on investigating the 3D-chromatin structure in plants, as well as its biological meaning in terms of transcriptional regulation in response to developmental cues and environmental changes. From a genome structure point of view, by combining high throughput sequencing, computing, and modern molecular biology approaches, much of our work has been focused on integrative genome-wide analysis on the interplay among chromatin folding, transcriptome, and epigenetic marks. Nowadays we are able to study chromatin structure at an unprecedented resolution, and our recent findings in Arabidopsis thaliana chromatin packing have revealed many unique properties. We welcome talented and motivated people to further explore this emerging field with us.
Figure shown above: Genome-wide interaction map of A. thaliana seedlings at 20 kb resolution. Elements represent normalized contact strength. Arrows highlight selected features. (i) The centromeric and pericentromeric regions are tightly packed and depleted for interactions with other regions located on chromosome arms. This corresponds to the observation that pericentromeric heterochromatin appears as bright dots in A. thaliana nuclei stained with DAPI. (ii) Telomeres from different chromosomes are close to each other. This phenomenon has been found previously with FISH studies, which revealed that telomeres often gather around the nucleolus. (iii) Strong interactions among a subset of interstitial heterochromatin regions.
The packing of genomic DNA inside nucleus enables long range DNA contact. Aside from stochastic collisions, there are DNA interactions happened for reasons, and they are very likely part of the transcription regulatory network. We are interested in elucidating the molecular mechanisms by which gene looping and chromatin compartmentalization in the nuclear space are achieved. For more information related to chromatin loops in Arabidopsis, please visit: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4971768/
Figure shown above: Identification of interactions between sub-kb chromatin segments in Arabidopsis. A genomic region divided into segments (gray ovals) based on the location of restriction enzyme cutting sites (vertical ticks) is shown. All chromatin interactions within this region are shown above as lines connecting the corresponding segments and colored according to the q-values. The box highlights chromatin loops at the FLC locus.
The nuclear space is not a homogeneous biochemical environment. A large number of cases have been demonstrated that the transcriptional activity of a gene is linked to its positioning in the nuclear space. Following the discovery of Lamina-Associated Domains (LADs), which are transcriptionally repressed chromatin regions, the non-random positioning of chromatin at the nuclear periphery and its biological relevance have been intensively studied in animals. However, it has remained unknown that whether in plants similar chromatin organization exists at the nuclear periphery. We have developed a method employing restriction enzyme-mediated ChIP (RE-ChIP), permitting identification of the non-random domain organization of chromatin at the peripheral zone of Arabidopsis nuclei on a genome-wide level. Our results indicate that chromatin regions tethered at the nuclear periphery are enriched for silenced protein-coding genes, transposable elements and classic heterochromatic marks, which collectively define a repressed environment. In addition, these domains highly correlate with our genome-wide chromatin interaction dataset (Hi-C) and largely explain the patterns of chromatin compartment revealed from Hi-C maps. In this regard, ongoing experiments are focusing on elucidating the molecular mechanism by which chromatin anchoring at plant nuclei periphery happens. For more information related to chromatin positioning at the nuclear periphery in Arabidopsis, please visit: https://www.ncbi.nlm.nih.gov/pubmed/28385710
Figure shown above: Genome-wide identification of non-random organization of the Arabidopsis chromatin at the nuclear periphery. (A) Localization of the NUP1:GFP protein. (B) Normalized sequencing coverage (50 kb window size) of chromosome 5 from NUP1:GFP RE-ChIP-seq. (C) Boxplot of expression of genes enriched (green) and not enriched (gray) at the nuclear periphery.
We have generated Hi-C maps of other plant species including Oryza sativa (rice) and several members of the Brassicaceae family. Together with the previous Hi-C experiment on Arabidopsis thaliana, these datasets provide us the very first step towards understanding the biological principles underlying plant genome structures in 3D space. It is surprising that the rice genome forms thousands of TADs (Topologically Associated Domains), which have been known as the fundamental structural units of the higher-order chromatin organization in animals. On the other hand, at a local level, the Arabidopsis genome does not display conspicuous TADs. We are very interested to know why and how the rice genome forms TADs. For more information, please visit https://www.nature.com/articles/s41477-017-0005-9
Figure shown above: Comparison of rice (panels A and C) and A. thaliana (panels B and D) chromatin packing patterns. For the local level comparison, 1 Mb regions from rice (C) and A. thaliana (D) are shown, respectively. TADs (Topologically Associated Domains) can be easily spotted in the rice genome, but this is not the case in Arabidopsis.