Summary
Non-random, dynamic three-dimensional organization of the nucleus is important for regulation of gene expression. Numerous studies using chromosome conformation capture strategies have uncovered ensemble organizational principles of individual chromosomes, including organization into active (A) and inactive (B) compartments. In addition, large inactive regions of the genome appear to be associated with the nuclear lamina, the so-called Lamina Associated Domains (LADs). However, the interrelationship between overall chromosome conformation and association of domains with the nuclear lamina remains unclear. In particular, the 3D organization of LADs within the context of the entire chromosome has not been investigated. In this study, we describe “chromosome conformation paints” to determine the relationship in situ between LAD and non-LAD regions of the genome in single cells. We find that LADs organize into constrained and compact regions at the nuclear lamina, and these findings are supported by an integrated analysis of both DamID and Hi-C data. Using a refined algorithm to identify active (A) and inactive (B) compartments from Hi-C data, we demonstrate that the LADs correspond to the B compartment. We demonstrate that in situ single cell chromosome organization is strikingly predicted by integrating both Hi-C and DamID data into a chromosome conformation model. In addition, using the chromosome conformation paints, we demonstrate that LAD (and B-compartment) organization is dependent upon both chromatin state and lamin A/C. Finally, we demonstrate that small regions within LADs escape the repressive regime at the peripheral zone to interact with the A-compartment and are enriched for transcription start sites (TSSs) and active enhancers.
Authors
Teresa R Luperchio, Michael EG Sauria, Xianrong Wong, Marie-Cécile Gaillard, Alice Yamada, James Taylor, Karen L Reddy
Mansuscript
A preprint describing this work is available here
Data Browser
A “track hub” is available for viewing DamID data, LAD calls, compartment scores, ChIP-seq, and RNA-seq datasets used in this manuscript.
Software
LAD Identification: LADs and DiP identification from sequence data can be performed using LADetector.
Hi-C Analysis: All Hi-C analyses including mapping, normalization, compartment scoring and calling, boundary index calculations, and visualization performed using HiFive.
Data Downloads
To be done.