By breaking the human genome into millions of pieces and reverse-engineering their arrangement, researchers have produced the highest-resolution picture ever of the genome's three-dimensional structure.
The picture is one of mind-blowing fractal glory, and the technique could help scientists investigate how the very shape of the genome, and not just its DNA content, affects human development and disease.
"It's become clear that the spatial organization of chromosomes is critical for regulating the genome," said study co-author Job Dekker, a molecular biologist at the University of Massachusetts Medical School. "This opens up new aspects of gene regulation that weren't open to investigation before. It's going to lead to a lot of new questions."
As depicted in basic biology textbooks and the public imagination, the human genome is packaged in bundles of DNA and protein on 23 chromosomes, arrayed in a neatly X-shaped form inside each cell nucleus. But that's only true during the fleeting few moments when cells are poised to divide. The rest of the time, those chromosomes exist in a dense and ever-shifting clump. Of course their constituent DNA strings are clumped, too: If the genome could be laid out end-to-end, it'd be six feet long.
For decades, some cell biologists suspected that the genome's compression wasn't just an efficient storage mechanism, but linked to the very function and interaction of its genes. But this wasn't easy to study: Sequencing the genome destroys its shape, and electron microscopes can barely penetrate its active surface. Though its constituent parts are known, the genome's true shape has been a mystery.
In April, a paper published in the Proceedings of the National Academy of Sciences linked patterns of gene activation to their physical proximity on chromosomes. It still provided the most persuasive evidence to date that genome shape matters, even though the researchers' chromosome map was relatively low-resolution. The topography described in the latest research, published Thursday in Science, is far more detailed.
"It's going to change the way that people study chromosomes. It will open up the black box. We didn't know the internal organization. Now we can look at it in high resolution, try to link that structure to the activity of genes, and see how that structure changes in cells and over time," said Dekker.
To determine genome structure without being able to directly see it, the researchers first soaked cell nuclei in formaldehyde, which interacts with DNA like glue. The formaldehyde stuck together genes that are distant from each other in linear genomic sequences, but adjacent to each other in actual three-dimensional genomic space.
The researchers then added a chemical that dissolved the gene-by-gene linear sequence bonds, but left the formaldehyde links intact. The result was a pool of paired genes, something like a frozen ball of noodles that had been sliced into a million fragmentary layers and mixed.
By studying the pairs, the researchers could tell which genes had been near each other in the original genome. With the aid of software that cross-referenced the gene pairs with their known sequences on the genome, they assembled a digital sculpture of the genome. And what a marvelous sculpture it is.
"There's no knots. It's totally unentangled. It's like an incredibly dense noodle ball, but you can pull out some of the noodles and put them back in, without disturbing the structure at all," said Harvard University computational biologist Erez Lieberman-Aiden, also a study co-author.
In mathematical terms, the pieces of the genome are folded into something similar to a Hilbert curve, one of a family of shapes that can fill a two-dimensional space without ever overlapping — and then do the same trick in three dimensions.
How evolution arrived at this solution to the challenge of genome storage is unknown. It might be an intrinsic property of chromatin, the DNA-and-protein mix from which chromosomes are made. But whatever the origin, it's more than mathematically elegant. The researchers also found that chromosomes have two regions, one for active genes and another for inactive genes, and the unentangled curvatures allow genes to be moved easily between them.
Lieberman-Aiden likened the configuration to the compressed rows of mechanized bookshelves found in large libraries. "They're like stacks, side-by-side and on top of each other, with no space between them. And when the genome wants to use a bunch of genes, it opens up the stack. But not only does it open the stack, it moves it to a new section of the library," he said.
The segregation of active and inactive genes adds to evidence that genome structure affects gene function.
"It's a great description of the structure of the nucleus, and if you put that on top of what we did, it forms the big picture," said Steven Kosak, a Northwestern University cell biologist and co-author of the April PNAS paper that linked rough outlines of chromosome arrangement to gene activation. Whereas that study only looked at a few chromosomes, the Science paper "looks at fine resolution over the whole genome," said Kosak.
"Now you can produce these genome maps, and superimpose them with genome-wide analyses of gene expression. You can really start asking how changes in spatial organization relate to changes in genes turning on and off," said Tom Misteli, a National Cancer Institute cell biologist who studies how glitches in chromosome structure may turn cells cancerous. Neither Misteli nor Kosak were involved in the Science study.
Connecting genome shape to gene function could also help explain the connection between genes and disease, which remain largely unexplained by traditional, sequence-focused genomics.
"It's perfectly reasonable and almost inevitable that the 3-D structure of DNA is going to influence how it functions," said Teri Manolio, director of the National Human Genome Research Institute's Office of Population Genomics.
Researchers also want to study how genome shape is altered. That appears to happen constantly during the transition from stem cell to adult cell, and then during cell function.
"How much variation is there in structure across cell types? What controls it? Exactly how important is it? We don't know," said Dekker. "This is a new area of science."
Image: From Science, a two-dimensional Hilbert curve, and the three-dimensional shape of a genome.
See Also:
- To Understand the Blueprint of Life, Crumple It
- Mapping HIV Genome's Shape, Not Just Sequence
- Beyond the Genome
- The Human Genome Is So 2003
Citation: "Comprehensive Mapping of Long-Range Interactions Reveals Folding Principles of the Human Genome." Erez Lieberman-Aiden, Nynke L. van Berkum, Louise Williams, Maxim Imakaev, Tobias Ragoczy, Agnes Telling, Ido Amit, Bryan R. Lajoie, Peter J. Sabo, Michael O. Dorschner, Richard Sandstrom, Bradley Bernstein, M. A. Bender, MarkGroudine, Andreas Gnirke, John Stamatoyannopoulos, Leonid A. Mirny, Eric S. Lander, Job Dekker. Science, Vol. 326 No. 5950, October 9, 2009.
Brandon Keim's Twitter stream and reportorial outtakes; Wired Science on Twitter. Brandon is currently working on a book about ecosystem and planetary tipping points.
