Analysis Of Chromatin Structures And Their Interaction With Long Non-coding RNAs

It is common to consider the human genome as a linear model for genomic analysis.But in reality,the human genome never exists as a linear entity within the cell nucleus.The organized chromatin within the 3D nuclear space are highly associated with the transcription of genes.Additionally,the human genome are extensively(?70%)transcribed,with part of these transcribed regions producing stably expressed protein-coding or non-coding RNA.It is known that the act of transcription can change the local chromatin structure.Moreover,the RNA products also potentially influence the chromatin strucuture.Long non-coding RNA(lnc RNA)are significantly differentially expressed in various cell lines and tissues.It has been proved that some certain lnc RNAs are capable of regulating target genes in trans or in cis.However,it is still difficult to decipher the role of lnc RNAs in the tree-dimensional chromatin structure.In this study,we analysed the interaction relationship between chromatin structures and lnc RNAs and found the key elements related to the chromatin structures.We constructed various models in which different levels of chromatin structures interacted with lnc RNAs in different ways.Works contained in this thesis are listed as follows:(1)We found that enhancer lnc RNAs(elnc RNAs)were highly enriched around chromatin loop anchors(Student's t-test,p=9.42×10^-27),and elnc RNAs can influence chromatin interactions in different ways.We proved that elnc RNAs are highly related to the formation and stablization of chromatin loops.Chromatin loop anchors containing elnc RNAs were significantly rich in transcription factor binding sites(TFBSs)especially for YY1 binding sites which can facilitate the interactions between chromatin loop anchors.We also found that the interaction frequency of elnc RNA-associated enhancer-promoter pairs was significantly stronger than other enhancer-promoter paris(Student's t-test,p=1.47×10^-22).Furthermore,elnc RNA expression levels were positively correlated with the interaction frequency of enhancer–promoter pairs.Intriguingly,elnc RNA-associated enhancer-promoter interaction pairs were not dependent on the architectural proteins as other enhancer-promoter interaction pairs.We clustered enhancer–promoter pairs into ten groups to reflect the different ways in which elnc RNAs could influence enhancer–promoter pairs.These results showed that different enhancer-promoter pairs contain different structuring factors.(2)We constructed a human genomic interaction network and found the frequent interactions between long intergenic non-coding RNA(linc RNA)genes and protein-coding genes,which are highly related to the occupancy of RNA polymerase II on the linc RNA gene.Interestingly,in the human genome interaction networks,the degree of linc RNA genes was significantly higher than that of protein-coding genes(Student's t-test,p<2.79×10^-191).At the same time,the frequency of interactions between linc RNA genes and other genes is positively correlated with the occupancy of RNA polymerase II on the linc RNA gene.The promoter regions of the protein-coding genes interacting with the linc RNA genes are enriched with R-loop structures,indicating that linc RNA may influence the target genes by R-loop structures.These promoters were enriched in more TFBSs.Finally,we used this network to validate and analyze some of the known disease-related genetic interactions.We first proposed that the linc RNA gene PCAT1frequently contacted with the oncogene MYC through the long-distance interaction(genomic distance>600 kbp),which allows the transcription products of PCAT1 to exactly bind on the genomic regions of MYC.(3)Here,we proved that G-quadruplexes had different relationships with different levels of chromatin structures and G-quadruplexes were involved in regulation of the enhancer lnc RNA expression.We found that G-quadruplexes are significantly enriched at boundaries of topological associated domains(TADs).Architectural protein occupancy was highly correlated with the content of G-quadruplexes at TAD boundaries.Moreover,adjacent boundaries containing G-quadruplexes frequently interacted with each other.Similar to CCCTC-binding factor(CTCF)binding sites,G-quadruplexes also showed strong insulation ability in the separation of adjacent regions.Furthermore,G-quadruplex motifs on different strands were associated with the orientation of CTCF binding sites.These findings suggest a potential role for G-quadruplexes in loop extrusion.We found that the formation of G-quadruplexes on enhancers created open chromatin environment and recruited many TFs to facilitate enhancer lnc RNA expression.Because of the elnc RNAs and G-quadruplexes,these enhancers were frequently interacted with promoters(Student's t-test,p=1.71×10^-44).We also proved that G-quadruplexes in different regions of elnc RNA genes can regulate elnc RNA transcription in different ways.(4)This study revealed the different parterns of nucleosomes and chromatin modifications around different gene structures and splicing sites.We constructed the model in which chromatin modifications regulate alternative splicing events.In this study,we focused on the nucleosome occupancy around the human intronless genes.The nucleosome occupancy of the intronless genes was significantly lower than that of intron-containing genes around transcription start sites(TSSs)and transcription termination sites(TTSs).While along gene bodies,the nucleosome occupancy not only showed higher occupancy than that of other regions but also well positioned.We found that this phenomenon is consistent with the nucleosome barrier model.Additionally,this phenomenon was also determined by DNA sequence preferences and DNA methylation.We also found that the inclusion levels of skipped exons(SEs)were negatively correlated with the enrichment of active histone marks.Active chromatin modifications were enriched in the upstream exons of SEs,especially around 5'splicing sites.We inferred that the chromatin modifications around the 5'splicing sites in upstream exon of the SEs could help RNA Polymerase II complex to recruit the effector proteins and facilitate alternative splicing.