Genetic studies promise to provide insight into the molecular mechanisms underlying type 2 diabetes (T2D). Variants associated with T2D are often located in tissue-specific enhancer clusters or super-enhancers. So far, such domains have been defined through clustering of enhancers in linear genome maps rather than in 3D space. Furthermore, their target genes are often unknown. We have now created promoter capture Hi-C maps in human pancreatic islets. This linked diabetes-associated enhancers with their target genes, often located hundreds of kilobases away. It also revealed >1300 groups of islet enhancers, super-enhancers and active promoters that form 3D hubs, some of which show coordinated glucose-dependent activity. We demonstrate that genetic variation in hubs impacts insulin secretion heritability, and show that hub annotations can be used for polygenic scores that predict T2D risk driven by islet regulatory variants. Human islet 3D chromatin architecture, therefore, provides a framework for interpretation of T2D GWAS signals.

Three-dimensional organization of the genome has emerged as an important player in transcriptional regulation [1][2][3][4][5][6][7] . In mammals, CTCF and the cohesin complex create sub-megabase structures with elevated internal chromatin contact frequencies, called topologically associating domains (TADs) [8][9][10][11] . Although TADs are generally considered important for transcriptional regulation, ablation of TAD organization by disrupting CTCF or the cohesin complex only caused modest gene expression changes [12][13][14][15] . In contrast, CTCF is required for cell cycle regulation 16 , early embryo development and for the formation of various adult cell types 17 . To uncouple the role of CTCF in cell state transitions and cell proliferation we studied the effect of CTCF be efficiently converted by exogenous CEBPA expression into functional induced macrophages (iMacs) with only one cell division on average (Fig. 1a; Supplementary Note 1) 26 . Using this system, we analyzed a time-series of transdifferentiating cells for genome-wide changes in 3D genome organization (in-situ Hi-C), enhancer activity (ChIP-seq of histone modifications), chromatin accessibility (ATAC-seq) and gene expression (RNA-seq).We first determined genome segmentation into A and B compartments on the basis of the first eigenvector values of a principal component analysis (PCA) on the Hi-C correlation matrix ('PC1 values'). Overall, although most of the genome remained stable, around 14% of A or B compartment regions were dynamic during transdifferentiation, showing transcriptional changed correlating with the altered compartmentalization state (Fig. 1b-f, Extended Data Fig. 1a-d; Supplementary Note 2). Next, we used chromosome-wide insulation potential 27 to identify between 3,100-3,300 TAD borders per time point (Fig. 1g). Boundaries were highly reproducible between biological replicates (Jaccard index>0.99) and enriched in binding sites for CTCF (Extended Data Fig. 1e). Genome-wide insulation scores analysed by PCA over time revealed progressive changes, reflecting a transdifferentiation trajectory (Extended Data Fig. 1f). While 70% of TAD borders were stable across all stages, 18% were lost or gained and 12% were transiently altered (Fig. 1g). CTCF binding was significantly more enriched at stable than at dynamic boundaries (Fig. 1h), as observed earlier 28 . Furthermore, while lost borders showed some CTCF occupancy in B cells that decreased in iMacs, gained borders were depleted for CTCF in both cell states (Fig. 1h), indicating CTCF-independent mechanisms driving local insulation. The dynamic rearrangement of TAD borders during transdifferentiation is illustrated by the DDX54 locus (Fig. 1i), in which a new boundary appears in iMacs without apparent changes in CTCF binding. Furthermore, border gain or loss did not correlate with changes in local gene expression (Extended Data Fig. 1g), indicating that transcription is not a driver of the observed changes. However, whereas motif analysis at ATAC-seq peaks within stable borders indee...

Using Hi-C, promoter-capture Hi-C (pCHi-C), and other genome-wide approaches in skeletal muscle progenitors that inducibly express a master transcription factor, Pax7, we systematically characterize at high-resolution the spatio-temporal re-organization of compartments and promoter-anchored interactions as a consequence of myogenic commitment and differentiation. We identify key promoter-enhancer interaction motifs, namely, cliques and networks, and interactions that are dependent on Pax7 binding. Remarkably, Pax7 binds to a majority of super-enhancers, and together with a cadre of interacting transcription factors, assembles feed-forward regulatory loops. During differentiation, epigenetic memory and persistent looping are maintained at a subset of Pax7 enhancers in the absence of Pax7. We also identify and functionally validate a previously uncharacterized Pax7-bound enhancer hub that regulates the essential myosin heavy chain cluster during skeletal muscle cell differentiation. Our studies lay the groundwork for understanding the role of Pax7 in orchestrating changes in the three-dimensional chromatin conformation in muscle progenitors.

Genetic studies promise to provide insight into the molecular mechanisms underlying type 2 diabetes (T2D). Variants associated with T2D are often located in tissue-specific enhancer regions (enhancer clusters, stretch enhancers or super-enhancers). So far, such domains have been defined through clustering of enhancers in linear genome maps rather than in 3Dspace. Furthermore, their target genes are generally unknown. We have now created promoter capture Hi-C maps in human pancreatic islets. This linked diabetes-associated enhancers with their target genes, often located hundreds of kilobases away. It further revealed sets of islet enhancers, superenhancers and active promoters that form 3D higher-order hubs, some of which show coordinated glucose-dependent activity. Hub genetic variants impact the heritability of insulin secretion, and help identify individuals in whom genetic variation of islet function is important for T2D. Human islet 3D chromatin architecture thus provides a framework for interpretation of T2D GWAS signals. 45. View ORCID ProfileMark I McCarthy. Fine-mapping of an expanded set of type 2 diabetes loci to single-variant resolution using high-density imputation and islet-specific epigenome maps. BioRxiv (2018).