The macro and micro of chromosome conformation capture

Abstract

The 3D organization of the genome facilitates gene regulation, replication, and repair, making it a key feature of genomic function and one that remains to be properly understood. Over the past two decades, a variety of chromosome conformation capture (3C) methods have delineated genome folding from megabase-scale compartments and topologically associating domains (TADs) down to kilobase-scale enhancer-promoter interactions. Understanding the functional role of each layer of genome organization is a gateway to understanding cell state, development, and disease. Here, we discuss the evolution of 3C-based technologies for mapping 3D genome organization. We focus on genomics methods and provide a historical account of the development from 3C to Hi-C. We also discuss ChIP-based techniques that focus on 3D genome organization mediated by specific proteins, capture-based methods that focus on particular regions or regulatory elements, 3C-orthogonal methods that do not rely on restriction digestion and proximity ligation, and methods for mapping the DNA-RNA and RNA-RNA interactomes. We consider the biological discoveries that have come from these methods, examine the mechanistic contributions of CTCF, cohesin, and loop extrusion to genomic folding, and detail the 3D genome field's current understanding of nuclear architecture. Finally, we give special consideration to Micro-C as an emerging frontier in chromosome conformation capture and discuss recent Micro-C findings uncovering fine-scale chromatin organization in unprecedented detail. This article is categorized under: Gene Expression and Transcriptional Hierarchies > Regulatory Mechanisms Gene Expression and Transcriptional Hierarchies > Gene Networks and Genomics.

Keywords: 3C technologies; 3D genome; Hi-C; Micro-C; chromosome conformation capture; genomic organization; nuclear architecture.

FIGURE 1

Timeline and comparison of major chromosome conformation capture techniques. (a) Chronological development of chromosome conformation capture technologies colored by type of method. Major observational, mechanistic, or biological discoveries are listed above the timeline. (b) Comparison of landmark 3C‐based methods and the resolutions at which their datasets are typically analyzed. The typical resolution ranges for these technologies are historically grounded and may widen or shift with the inclusion of recent advances in methodology. Resolutions at which key features of chromatin organization typically manifest are shown on the right

FIGURE 2

Micro‐C captures finer‐scale features of chromatin organization than Hi‐C. (Top row) High‐resolution Hi‐C (Bonev et al., 2017) and Micro‐C (Hsieh et al., 2020) datasets generated from wild‐type mESCs are visually juxtaposed at various scales of chromosomal organization. The data is binned at different resolutions using HiGlass, with the visualization of any particular feature requiring bins of a finer resolution than the size of the feature. Contact heatmaps of the whole genome, compartments, topologically associating domains (TADs), and loops are shown. The checkerboard pattern in the second column of plots indicates separation into A/B compartments. The markers in the third column of plots indicate TADs, while the markers in the fourth column of plots indicate corner peaks or “dots” specifying loops. (Bottom row) Fine‐scale resolution maps of the same Hi‐C and Micro‐C datasets. Markers identify microarchitecture, such as enhancer‐promoter (E–P) or promoter‐promoter (P–P) loops, stripes, and domains, visible in Micro‐C but not discernable in Hi‐C, and genes within the region are annotated below. This figure is inspired by fig. 1d and S1d in Hsieh et al. (2020)

FIGURE 3

The Micro‐C protocol. Major steps in the Micro‐C method are shown. Cells are first chemically fixed using formaldehyde and cross‐linked using a protein–protein cross‐linker (shown as bright green jagged lines) such as disuccinimidyl glutarate (DSG) or ethylene glycol bis(succinimidyl succinate) (EGS). MNase digestion cleaves DNA into mononucleosomes and is inactivated by EGTA. The resulting DNA ends are blunted, polished, and labeled with biotin. DNA is proximity ligated, cross‐links are reversed, and nonligated products are removed through streptavidin purification. The Micro‐C library is prepared for paired‐end sequencing by sequencing adapter ligation and PCR amplification. Data processing (not shown) can be performed with a data analysis pipeline similar to Hi‐C

FIGURE 4

Current model for mammalian 3D genomic organization. Features of nuclear architecture are shown across scales of organization. (Top left panel) Chromosomes, hundreds of Mb in length, occupy distinct territories within the nucleus. Compartments up to of tens of Mb in length distinguish preferentially self‐interacting domains of more transcriptionally active DNA (Compartment A) from less active DNA (Compartment B) and may interact across chromosomes. Nuclear bodies further define large‐scale hubs of spatial organization, with denser, less active DNA clustering along the nuclear periphery and around the nucleolus and more accessible and active DNA clustering around nuclear speckles. (Top right panel) Topologically associating domains (TADs) arise at and below the Mb scale within each chromosome, with subTADs nesting within larger parent TADs. CTCF loops are identified by CTCF anchors at the point of contact at the base of a loop. (Bottom right panel) Loop formation is primarily driven by cohesin‐mediated loop extrusion, which is halted by convergent CTCF sites. CTCF's binding motif and components of the cohesin complex are shown. (Bottom left panel) Microarchitecture on the scale of hundreds of bp up to tens of kb consists of a diverse array of P–P and E–P linkages, small CTCF‐ and cohesin‐mediated domains, bundles of repressive chromatin, and other gene domains. This panel is largely inspired by fig. 7f of Hsieh et al. (2020)