The researchers developed a computational pipeline that integrates and analyzes single-cell imaging, microscopy data, and other genome-wide datasets. They called the approach COLA (CO-localization and Looping Analysis). COLA finds DNA looping events in super-resolution microscopy images and links them to activity readouts measured across the genome, like RNA polymerase binding.
Using COLA, the team studied looping events at four genomic loci known to interact long-range. Surprisingly, they found that individual loops rarely last longer than a few minutes, and many are highly dynamic, undergoing rapid assembly and disassembly. These results challenge the prevailing view that genomic contacts reflect stable, persistent chromatin structures that maintain transcriptional programs. Instead, the team proposes a model where loops are transient and form only transiently in response to specific regulatory cues.
"Loops are everywhere in cells, but they're much more transient than anyone has previously appreciated," said co-senior author Gene Yeo, PhD, professor of cellular and molecular medicine and director of the Center for RNA Biology at UC San Diego School of Medicine. "Our findings overturn the assumption that these loops are hardwired features hardwired and imply that loop formation is dynamically regulated."
Cells must tightly regulate how they read instructions from their genomes to produce functional products, such as proteins that perform various cellular functions. The way cells package, arrange, and fold their genomes into three dimensions plays a critical role in gene regulation. These organizational principles determine which segments of DNA are accessed and interpreted by the cellular machinery responsible for reading and transcribing genes.
The most fundamental unit of genome folding involves looping, where distant regions of DNA physically come into contact with each other to help orchestrate gene expression. Loops are thought to bring together proteins that drive gene expression, allowing them to interact and carry out their functions efficiently.
The COLA approach enables researchers to simultaneously visualize loops and other genomic features, capturing spatial and temporal relationships in unprecedented detail. "We now have a method that can directly correlate changes in RNA polymerase occupancy with specific looping changes in single cells," said co-first author Michael Niculescu III, a graduate student in the Yeo Lab.
The researchers say their findings have broad implications for understanding gene expression and genome organization. For instance, they suggest that the dynamic and fluctuating nature of loops could allow cells to rapidly respond to environmental changes or developmental cues. In cancer and neurodegenerative diseases, this flexibility may also allow cells to switch cell identities and reprogram gene expression programs.
"There's a lot we can now do with this tool," said co-first author Matthew Huynh, a postdoctoral fellow in the Yeo Lab. "We can test new hypotheses and probe disease-associated regulatory changes in ways we couldn't before."
Additional co-authors of this study are: Yuzuru Kido, UC San Diego; James McGinnis, UC San Diego; and Nicholas Ingolia, UC Berkeley.
This research was funded in part by the National Institutes of Health (R35 GM143669, T32 GM007240), the National Science Foundation (MCB-2110538), the Simons Foundation (540333), and the Ludwig Institute for Cancer Research.
