Exploring Single-Cell Chromatin Organization with Multiplexed DNA-FISH: Towards an Imaging Platform for Single-Cell Multi-Omics

Abstract

The three-dimensional (3D) organization of chromatin in the nucleus has been an active area of research for many years. Chromatin organization during interphase has especially drawn much attention over the past two decades or so, with results showing that this organization has implications on a diverse set of cellular and nuclear functions. Our understanding in this field has been greatly aided by the development of Hi-C and related technologies, revealing structural features such as topologically associated domains (TADs), A/B compartments and CTCF-anchored loop domains. As I set out on my thesis research, our lab had recently developed an imaging-based method to explore chromatin structure, showing that A/B compartments could be identified as physical structures in individual chromosomes in single cells. Thus, there was much interest in further developing this technology and investigating chromatin structure at all relevant length-scales.

My thesis research has focused on further developing our lab’s imaging-based tools, as well as using these tools to characterize chromatin organization in single cells. In Chapter 2, my colleagues and I combine this approach with a super-resolution microscopy (specifically, stochastic optical reconstruction microscopy, or STORM) and apply this imaging approach to characterize single-chromosome organization at the length-scale relevant to TADs. We then further develop this method to increase its throughput, enabling high-resolution tracing of entire human chromosomes. We revealed the existence of highly variable chromatin domains in single chromosomes that preferentially form boundaries at CTCF/cohesin-binding sites. Interestingly, the depletion of cohesin did not remove these chromatin domains, but without cohesin, the domain boundaries were randomly positioned and no longer preferentially located at CTCF/cohesin-binding sites. We further demonstrated that these chromatin domains preferentially interact with other domains containing similar ensemble A/B compartment makeups, hinting at a physical mechanism for how single cell structures can give rise to ensemble features such as TADs and A/B compartments. In Chapter 3, I focus on a different technical innovation we made. We adapted a barcode-based imaging approach developed by our lab (multiplexed error-robust fluorescent in-situ hybridization, or MERFISH) to image >1,000 DNA loci genome-wide, and proceeded to combine this capability with RNA and protein imaging to create an imaging-based platform for single-cell multi-omics. Using this technique, we find that trans-chromosomal contact frequencies are enriched for A-A, but not B-B, interactions. Additionally, we found that nascent transcription is correlated with the relative local concentrations of A/B chromatin, and that this relationship persists in multiple distinct nuclear environments.