Corrected 17 February 2012; See Last Page Ge Prize Essay

Corrected 17 February 2012; See Last Page Ge Prize Essay

ESSAY CORRECTED 17 FEBRUARY 2012; SEE LAST PAGE GE PRIZE ESSAY Exploring how the genome folds by using Zoom! three-dimensional genome sequencing. Erez Lieberman Aiden rowing up, I watched lowed by restriction, proximity Next, I collaborated with Nynke van Ber- the 1968 film Pow- ligation, and the use of univer- kum, a postdoc in Job Dekker’s lab, to refi ne Gers of Ten. The camera sal polymerase chain reaction this strategy and bring it to life (17). In Hi-C, begins with a couple on a picnic (PCR) primers to amplify junc- cells are cross-linked; DNA is restricted, and then zooms out: to the pic- tions containing a target frag- leaving overhangs; overhangs are fi lled in, nic ground (101 meters), Chi- ment (13). Chromosome con- incorporating a biotinylated nucleotide; and cago (105 m), Earth (107 m), formation capture (3C) replaced the blunt ends are ligated in dilute condi- GE Healthcare and Science and, eventually, the universe spermine and spermidine with tions (Fig A). Because the site of ligation is 26 are pleased to present the (10 m), only to zoom back in prize-winning essay by Erez formaldehyde, enabling dilution marked with biotin, ligation junctions—evi- until it reaches the interior of a Lieberman Aiden, a regional prior to ligation and markedly dence of physical contact between two loci— proton (10 to 16 m). Breathtak- winner from North America, improving signal (14). Coupling can be purifi ed and sequenced. The result- ing structures emerge at each who is the Grand Prize win- microarray and sequencing tech- ing maps reveal features at three scales: the scale. I realized that if, one day, ner of the GE & Science prize nology to Cullen et al.’s universal nucleus, individual chromosomes, and single I could hold that magical cam- for Young Life Scientists. PCR approach (4C) or to multi- megabases (Fig B). era—examining a phenomenon plexed ligation mediated ampli- At the nuclear scale, we found that loci at a new scale, however briefl y—I would see fi cation (5C) increased throughput (15, 16). on the same chromosome—even opposite things that had not been seen before. But these methods still require target loci. ends—interact more than loci on different on April 10, 2014 Where to look? The fi lm provides a clue: As a student in Eric Lander’s lab, I won- chromosomes. This is the proximity-pattern at 10–6 m, we enter the cell nucleus, but see dered whether this approach could be mod- fi ngerprint of chromosome territories, con- little before reaching the double helix (10–8 m) ified to create a genome-wide proximity firming that Hi-C’s picture resembles the (1, 2). Something is missing. DNA must fold map that would enable zooming in on scales microscope’s. But we can zoom further. further; if it did not, human genomes would between chromosome territories and nucleo- Examining individual chromosomes, we be 2 m long and would not fi t in the cell, much somes. Targeted primers would not work; found that open chromatin interacts more less the nucleus. instead, junctions would be biochemically with open chromatin, and closed with closed. Since 1968, we have learned that DNA marked. With collaborators at the Broad This suggests that open and closed chroma- 2 wraps around histones, packing ~10 base Institute, I outlined such a method: “Hi-C.” tin are segregated during interphase. Unlike www.sciencemag.org pairs into the 10–8-m nucleosome (3–6). A Crosslink DNA Cut with Fill ends Ligate Purify and shear DNA; Sequence using We also know that individual chromosomes restriction and mark pull down biotin paired-ends occupy distinct subnuclear volumes called HindIII enzyme with biotin NheI chromosome territories which pack ~108 base pairs into 10–6 m (7, 8). In between? We lack suitable experiments: x-ray crystallog- raphy and nuclear magnetic resonance work Downloaded from best in vitro with shorter polymers, and con- temporary microscopists can only track a few genomic loci at a time (9, 10). Fortunately, decades of work are enabling new approaches. In 1966, Wang and David- Nuclear scale Chromosome scale Megabase scale son measured “cyclization”—how fast an B Chromosome territories Open Closed Fractal globule oligonucleotide’s sticky ends form loops— to argue that naked DNA folds into shapes resembling a random walk (11). Promoter- enhancer interactions between DNA ends catalyze cyclization in vitro (12). In 1993, Cullen et al.’s nuclear ligation assay (NLA) brought cyclization into the cell by trans- forming in vivo proximity relations between genomic loci into chimeric oligonucleotide sequences. NLA entails stabilizing nuclear structure using spermine and spermidine, fol- Zooming in on the human genome. (A) The Hi-C protocol. (B) The architecture of the genome is repre- Harvard Society of Fellows, Harvard University, Cambridge, sented at three scales. Chromosomes (dark blue, light blue, green) occupy distinct territories (Left), and MA 02138, USA; Visiting Faculty, Google, Inc., Mountain weave back and forth between the open and closed compartments (Middle). At the megabase scale, chro- 17 View, CA 94043, USA. E-mail: [email protected] matin tends to fold into fractal globules (Right). Panels adapted from ( ). AND MIT ADAPTED FROM BANG WONG AND LESLIE GAFFNEY/THE BROAD INSTITUTE OF HARVARD CREDIT: 1222 2 DECEMBER 2011 VOL 334 SCIENCE www.sciencemag.org Published by AAAS ESSAY most epigenetic marks, in which local bio- out disturbing the rest. But after cooking, they able universe (1026 m). We suspect that this chemical changes activate or repress a region become highly knotted, clumping together on is all there is: that in this universe of scales, (18), this compartmentalization is a spatial your fork. Thus, ramen noodles are unstable: humankind lies at the center. Perhaps we are mark: epigenetics through origami. Given enough thermal energy (i.e., cooking) right. But as our forebears—who marked the Last, we zoomed in on a megabase. Sur- they reach a knotted equilibrium. But, like edges of their maps with phrases like “hic sunt prisingly, our data lacked the hallmarks of the genome, when the conditions are right, dracones” (“here be dragons”)—learned time a classic condensed polymer: It resembled the unknotted state can persist for a very long and time again, at the fringes of our maps the neither a random walk, nor its condensed- time: Ramen has an admirably long shelf life. world is full of surprises. So on we must go, polymer cousin, the highly knotted equilib- Sequencing genomes in three dimensions building better cameras. Zoom! rium globule. What could we be looking at? is becoming increasingly common. Several Eventually, I found a paper theorizing methods are emerging (21). For instance, References 1. J. D. Watson, F. H. Crick, Nature 171, 737 (1953). that extremely long polymers initially con- chromatin interaction analysis by paired end- 2. R. E. Franklin, R. E. Gosling, Nature 171, 740 (1953). dense into a confi guration known as the frac- tag sequencing (22) uses antibodies to target 3. A. L. Olins, D. E. Olins, Science 183, 330 (1974). tal globule, (19). This state has remarkable contacts mediated by a particular protein. 4. C. L. F. Woodcock, J. Cell Biol. 59, 368a (1973). 5. R. D. Kornberg, J. O. Thomas, Science 184, 865 (1974). properties: dense, but completely unknot- Spatial maps will help us unravel fundamen- 6. R. D. Kornberg, Science 184, 868 (1974). ted, one can unfold and refold sections with- tal mysteries in biology, such as how the cells 7. C. Rabl, Morphologisches Jahrbuch 10, 214 (1885). out disturbing the rest (and, at 10–7 m, it fi ts in our bodies—genetically identical—per- 8. T. Cremer et al., Hum. Genet. 60, 46 (1982). 9. A. Edwards, Ann. Rev. Biochem. 78, 541 (2009). –6 snugly in the Powers of Ten gap between 10 form different functions; they will also help 10. T. Cremer, C. Cremer, Nat. Rev. Genet. 2, 292 (2001). and 10–8 m). A back-of-the-envelope calcula- us understand cancer, where this differentia- 11. J. C. Wang, N. Davidson, J. Mol. Biol. 19, 469 (1966). tion (subsequently confi rmed by simulations) tion has gone awry. At fi ner scales, we will 12. S. Mukherjee et al., Cell 52, 375 (1988). 13. K. E. Cullen et al., Science 261, 203 (1993). suggested that our data was consistent with a be able to comprehensively map long-range 14. J. Dekker et al., Science 295, 1306 (2002). fractal globule. gene regulatory elements buried in so-called 15. J. Dostie et al., Genome Res. 16, 1299 (2006). Though it may sound abstract, the fractal “junk” DNA. 16. Z. Zhao et al., Nat. Genet. 38, 1341 (2006). 17. E. Lieberman-Aiden et al., Science 326, 289 (2009). globule is easy to explain to graduate students More than just a cinematic device, the 18. T. S. Mikkelsen et al., Nature 448, 553 (2007). because it closely resembles the only food we camera in Powers of Ten is an apt metaphor 19. A. Y. Grosberg et al., J. Phys. France 49, 2095 (1988). 20. R. Konzak, The Book of Ramen (Turtleback Books, Friday can afford: ramen. Ramen is densely packed: for science itself, which progresses by probing Harbor, Washington, 1995). A 4–by–4-inch package contains ~100 feet the world at one scale after another. Humans 21. Z. Duan et al., Nature 465, 363 (2010). of noodles (20). Also, as any grad student are naturally found at 100 m and, after thou- 22. M. J. Fullwood et al., Nature 462, 58 (2009). can tell you, undercooked ramen noodles are sands of years, have managed to explore ~25 Funding: NIH 1DP2OD008540-01 (New Innovator Award).

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    3 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us