Condensing DNA in dormancy

From the Tsukiyama lab, Basic Sciences Division

During quiescence, cells cease to proliferate and enter a state of dormancy where they are resistant to stressors and can survive long-term.  In quiescent cells, global gene expression is decreased by widespread transcriptional repression. This change in gene expression is rapidly reversible; how quiescent cells can reinitiate transcription on short notice is thus an important question. Quiescent cells also have highly condensed chromatin, which is generally correlated with reduced transcription and hypothesized to be the contributing factor for repression of transcription during quiescence.  However, this has been hampered by a lack in technology to accurately interrogate higher-order chromatin at a resolution sufficient to draw conclusions about its relationship to transcription.  Basic Sciences Member Dr. Toshio Tsukiyama and his lab, together with collaborators, employed a technology capable of mapping genome-wide chromatin interactions at the resolution of nucleosomes.  They found that the condensin complex induces looping between the boundaries of previously unidentified chromatin domains, thus compacting chromatin to repress transcription during quiescence. Their work was published in a recent issue of Molecular Cell.

Genomic DNA is package into chromatin via a hierarchical series of folds. Much is known about the first level of chromatin compaction, through crystal structures and genome-wide mapping studies of the nucleosome, while less is known about higher-order chromatin structure.  The 3C (chromosome conformation capture) family of techniques (such as Hi-C) has been used to study higher-order chromatin structure.  These techniques involve the crosslinking of chromatin with formaldehyde, followed by digestion and re-ligation to only allow DNA fragments that are covalently linked together to form ligation products. The ligated products contain information about where they originated from in the genomic sequence and where they physically reside, in the 3D organization of the genome. Unfortunately, these techniques are limited by suboptimal resolution, as they rely on restriction digestion that yields genomic DNA fragments several kilobases in length.  To obtain more superior resolution, Dr. Oliver Rando and graduate student Tsung-Han Hsieh at the University of Massachusetts developed Micro-C, where chromatin is fragmented into mononucleosomes using micrococcal nuclease, thus enabling nucleosome-resolution maps of chromosome folding. In this work, the authors used Micro-C XL, an improved method for mapping chromosome folding at mononucleosome resolution. Micro-C XL can be used to capture both short-range chromosome fiber features, as well as higher order features.  Dr. Sarah Swygert, a postdoctoral fellow in the Tsukiyama lab, employed Micro-C XL to interrogate the 3D structure of chromatin in quiescent Saccharomyces cerevisiae cells, compared to non-quiescent S. cerevisiae cells.  In Saccharomyces cerevisiae, a population of pure quiescent cells can be generated for studying quiescence.  Furthermore, they share similar characteristics with mammalian quiescent cells, making them an ideal model to study the molecular underpinnings of quiescence.

Dr. Swygert explained some of the challenges of the technology: “By far, the most difficult part of this process was getting Micro-C XL to work in quiescent cells.  Micro-C is known to be a very challenging method- as far as we know, we are only the second group (after Ollie Rando's group, who invented it) to be able to do it.  And even though yeast are considered an easy model organism, we've found that a lot of what is uncomplicated in exponentially growing cells goes out the window when dealing with quiescence.  So the combination of difficult protocol and difficult model system was something of a nightmare!  We spent a couple of years just painstakingly optimizing each step of the method.” However, the effort paid off. Dr. Swygert: “The ability to examine chromatin structure in cells at single-nucleosome resolution is huge!  It's something that can only be done in yeast right now, which gives us the ability to examine things like how chromatin domains are formed and how they regulate transcription at a level that no one outside of yeast can do,”


Schematic figure of chromatin compaction by condensin in quiescent cells Figure from Dr. Sarah Swygert

The authors report that chromatin loops are formed between boundaries of large chromatin domains during quiescence in S. cerevisiae. A ChIP-seq screen confirmed that condensin, which belong to the structural maintenance of chromatin (SMC) complex family, is responsible for forming these chromatin loops during quiescence.  To investigate if condensin was truly required for chromatin loop formation, the authors generated inducible depletion strains of a condensin subunit, and performed Micro-C XL in quiescent cells formed in the presence or absence of condensin. They found that condensin indeed mediates chromatin compaction in quiescent cells by forming chromatin loops.  Using this system, they further showed by PolII ChIP-Seq that condensin is also required for repression of transcription during quiescence.  To explore if this condensin-dependent chromatin compaction is also conserved in other organisms besides S. cerevisiae, the authors knocked down expression of human condensin in human foreskin fibroblasts, and found that reduced condensin activity in human cells prevents chromatin condensation at quiescence entry.

When asked about next steps, Dr. Swygert revealed two major directions. “First, we are very interested in how condensin is targeted to bind in so many different locations in quiescent cells, so I have been tracking condensin binding during the quiescence entry process and investigating different mechanisms by which it could be recruited.  So far it looks as though transcription itself may be involved in recruiting condensin, which is puzzling as condensin appears to be acting as a transcriptional repressor and yet quiescent cells seem to want it at the promoters of active genes.  Second, now that we have mapped out chromatin structure in a repressive environment at nucleosome-resolution, we are very excited to determine if the chromatin of quiescent cells takes on structural conformations predicted for repressive chromatin by biochemical experiments but have been extremely difficult to observe in cells.  The combination of Micro-C and quiescent yeast cells really gives us an ideal model system to study mechanisms of transcriptional repression in spectacular detail.”

Swygert SG,Kim S,Wu X,Fu T,Hsieh T-H,Rando OJ,Eisenman RN,Shendure J,McKnight JN,Tsukiyama T. 2018. Condensin-Dependent Chromatin Compaction Represses Transcription Globally during Quiescence. Molecular Cell 73:1-14

Funding was provided by the National Institutes of Health and the Howard Hughes Medical Institute.

Research reported in the publication is a collaboration between Fred Hutch/UW Cancer Consortium members Toshio Tsukiyama (Fred Hutch), Robert Eisenman (Fred Hutch), and Jay Shendure (UW).