1 January 2010 $10

AAAS, publisher of Science, thanks the sponsors and supporters of the

2010 Annual Meeting

Bridging Science and Society
1822 February San Diego
Presenting Sponsor

Tijuana
National Estuarine Research Reserve

River

In addition generous funding for AAAS Awards is provided by the Kavli Foundation and Affymetrix.

Be part of a proven equation:

Your Organization + AAAS Annual Meeting = Global Visibility
Call today about the benefits of sponsorships currently available:
Jill C. Perla AAAS Meetings Direct Dial: (202) 326-6736 E-mail: [email protected]

cell sciences

ultra pure cytokines

Produced in barley, these proteins are animal, bacterial, and viral free, and are ultra pure, with extremely low endotoxin.

www.cellsciences.com

Cell Sciences offers innovative, unique growth factors and hard-to-produce recombinant proteins, bypassing the use of bacterial or animal cell systems. These ultra pure proteins contain no contamination from other growth factors and negligible amounts of endotoxin.

Ultra pure cytokines & growth factors

FGF1, human FGF2, human FLT3 ligand, human GCSF, human IFNA2, human IFN gamma, human IGF1, human IL1-alpha, human IL2, human IL3, human IL4, human IL5, human IL6, human IL7, human IL9, human IL16, human IL22, human KGF, human M-CSF, human NRG1/HRG beta 2, human SCF, mouse SF20/IL25, human TNF-alpha, human TNF-beta, human VEGF121, human VEGF165, human

Background: barley endosperm

The host organism, barley, with its specialized endosperm storage tissue, provides many unique features including proficient protein machinery, with eukaryotic folding, and a distinct route for long-term protein protection and storage. A biochemically inert environment, void of endotoxins, low protease activity and secondary metabolite content, and a simple protein profile, aid in downstream processing. Barley has also a G.R.A.S. (generally recognized as safe) status from the FDA.

Cell Sciences ultra pure growth factors and cytokines are produced
for use in basic and applied medical scientific research, cell culture media and diagnostics. serum free animal, bacterial & viral free extremely low endotoxin (<0.005 ng/ug) highly biologically active easier regulatory clearance perfect for cell culture, drug development, stem cell research, animal research for use in all in vitro cellular studies for use in all in vivo animal studies

CELL SCIENCES INC 480 NEPONSET STREET, BUILDING 12A, CANTON, MA 02021 [email protected] TOLL FREE: (888) 769-1246 TEL: (781) 828-0610 FAX: (781) 828-0542 WEB WWW.CELLSCIENCES.COM

San Diego
18-22 February

Join us in

Presidents Address
Peter C. Agre, M.D.
AAAS President, and Director, Malaria Research Institute, Johns Hopkins Bloomberg School of Public Health
Agre shared the 2003 Nobel Prize in Chemistry with Roderick MacKinnon of Rockefeller University for the discovery of aquaporins, the key proteins that transport water across cell membranes. Not long after receiving the Nobel Prize, Agre began working to extend his studies of aquaporins to malaria, addressing the question of whether or not aquaporins could be exploited as a means of treating or preventing the disease. Initial results led his laboratory to focus on malaria as its primary area of study. As director of the Malaria Research Center, he oversees 19 Hopkins faculty members who concentrate on advancing basic science to develop new methods in malaria prevention and treatment. Agre is a member of the National Academy of Sciences (NAS), chair of the NAS Committee on Human Rights, and a Fellow of AAAS and the American Academy of Arts and Sciences. He received his M.D. degree from Johns Hopkins University.

The place for celebrations is the

AAAS Annual Meeting

25 Years of science education reform through AAAS Project 2061 50 Years of accomplishments in higher education and academic research by the University of California, San Diego 60 Years of discovery through support from the U.S. National Science Foundation 350 Years of scientific achievement and endeavor by the Royal Society, the worlds oldest science academy And celebrate one of the greatest inventions of the 20th century the laser

Plenary Speakers
Carol W. Greider, Ph.D.
Daniel Nathans Professor and Director, Department of Molecular Biology and Genetics, and Professor of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD Title To Be Determined
Greider, one of the worlds pioneering researchers on the structure of telomeres, was awarded the 2009 Nobel Prize in physiology or medicine by the Royal Swedish Academy of Sciences along with Elizabeth Blackburn and Jack W. Szostak. While a 23-year-old graduate student at the University of California, Berkeley, working together with Blackburn, Greider discovered the enzyme telomerase and later, in her own lab, she cloned its RNA component. This work laid the foundation for studies that have linked telomerase and telomeres to human cancer

and age-related degenerative disease. It represents another example of curiosity-driven basic research that has direct medical implications. Greider obtained her Ph.D. degree in molecular biology from UC Berkeley in 1987.

Eric S. Lander, Ph.D.

Director, The Broad Institute of MIT and Harvard University, and Co-Chair, Presidents Council of Advisors on Science and Technology (PCAST) Science and Technology in the First Year of the New Administration
Lander is widely known as one of the driving forces behind todays revolution in genomics, the study of all of the genes in an organism and how they function together in health and disease. He also is co-chair of President Obamas council of science and technology advisers. PCAST is an advisory group of the nations leading scientists and engineers who directly advise the President and make policy recommendations in the many areas where understanding of science, technology, and innovation is key to strengthening the economy and forming policy. Lander also was one of the principal leaders of the Human Genome Project and is a member of both the National Academy of Sciences and Institute of Medicine. He is also an AAAS Fellow. Lander earned his B.A. degree in mathematics from Princeton University and Ph.D. degree in mathematics from Oxford University as a Rhodes Scholar.

executive officer of the Monterey Bay Aquarium Research Institute. Her biography includes a broad range of research interests and numerous honors and awards. Her research has ranged from studies of ocean island volcanism in French Polynesia to continental break-up in the Western United States to uplift of the Tibet Plateau. She also spent 3 years with the USGS in California working on earthquake prediction. She is a member of the National Academy of Sciences and a Fellow of AAAS. McNutt earned her Ph.D. degree in earth sciences at the Scripps Institution of Oceanography.

Barry C. Barish, Ph.D.

Director, Global Design Effort for the International Linear Collider (ILC), and Linde Professor of Physics, emeritus, California Institute of Technology, Pasadena

Lecture Title To Be Determined

Among Barishs noteworthy experiments were those performed at Fermilab using high-energy neutrino collisions. These experiments were among the first to observe the weak neutral current, a linchpin of electroweak unification theories. Today he directs the ILC, the highest priority future project for particle physics worldwide that promises to complement the Large Hadron Collider at CERN in exploring the TeV energy scale. In the 1980s, Barish initiated an ambitious international effort to build a sophisticated underground detector which provided some key evidence that neutrinos have mass. In 1994, he became principal investigator of the Laser Interferometer Gravitational-Wave Observatory (LIGO) project. As director of the LIGO Laboratory from 1997 to 2005, he led a team of scientists who built two facilities to detect and study gravitational waves from astrophysical sources. Barish is a member of the National Academy of Sciences and is a Fellow of AAAS. He earned his Ph.D. degree in experimental high energy physics at the University of California, Berkeley.

Marcia McNutt, Ph.D.

Director, U.S. Geological Survey, and Science Adviser to the Secretary, U.S. Department of the Interior (invited) Science Below the Sea
McNutts appointment in 2009 marked a milestone for USGS she is the first female director in the agencys 130-year history. She heads a multidisciplinary organization that focuses on biology, geography, geology, geospatial information, and water, and is dedicated to studying the landscape, natural resources, and natural hazards. Most recently she served as president and chief

Register at:

www.aaas.org/meetings

GE Healthcare Life Sciences

Interesting what a little imagination can do

Imagination has always inspired the scientic mind. At GE Healthcare Life Sciences, the same imagination inspires us to provide the most complete range of products and solutions available. Everything from innovative research system platforms, such as KTA, Biacore and IN Cell Analyzer, to everyday lab essentials from our Whatman and Amersham brands, to a full range of products for bioprocessing. Scientists around the world rely on these brands to deliver reproducible results, with the highest quality, that ultimately helps improve their productivity. At GE Healthcare Life Sciences, our focus is on helping scientists achieve even more, faster. Its a commitment we have in our genes. And all this is backed by the service, support and investment for the future that being part of GE can bring. Want to set your imagination free and do more? Why not talk with us today. Visit www.gelifesciences.com
I KTA I Amersham I Biacore I IN Cell Analyzer I Whatman I GE Service I

KTA, Amersham, Biacore, Capto, MabSelect, MicroCal, Sephadex and Whatman are trademarks of GE Healthcare companies. 2009 General Electric Company All rights reserved. First published September 2009 GE Healthcare Bio-Sciences AB, Bjrkgatan 30, 751 84 Uppsala, Sweden GE15-09

CONTENTS
EDITORIAL
12

Volume 327

Issue 5961

BOOKS ET AL.
33 34

Promoting Scientific Standards

Bruce Alberts

Lizards in an Evolutionary Tree

J. B. Losos, reviewed by R. G. Gillespie

NEWS OF THE WEEK

18 19 20 21 22 22 23 24 25

Dont Be Such a Scientist R. Olson; Am I Making Myself Clear? C. Dean,

reviewed by P. Kareiva

Obama Backs New Launcher and Bigger NASA Budget Senate Looms as Bigger Hurdle After Copenhagen Protests by Staff, Advisers Rattle Australian Synchrotron From Sciences Online Daily News Site U.K. Physicists Cry Foul at Major Budget Cuts Errors in Chemistry Claims Cast Doubt on Reactome Paper From the Science Policy Blog Recession Hits Some Sciences Hard at Florida State University Embryo Ruling Keeps Stem Cell Research Legal

34

Browsings

POLICY FORUM
36

Opportunities for Research and NIH

F. S. Collins
page 26

PERSPECTIVES
38

Molecular Donuts and Donut Holes

K. H. Whitmire >> Report p. 72

39 41

Serendipitous Astronomy
K. R. Lang

Janus Catalysts Direct Nanoparticle Reactivity

D. J. Cole-Hamilton >> Report p. 68

42

NEWS FOCUS
26

Some Enzymes Just Need a Space of Their Own

S. Kang and T. Douglas >> Report p. 81

In the Afterglow of the Big Bang

>> Science Podcast
43

Brain Activity to Rely On?

D. S. Schwarzkopf and G. Rees >> Report p. 97

LETTERS
30

Bushmeat Hunting and Climate: An Indirect Link

P. A. Jansen et al.

45

Retrospective: Rossiter H. Crozier (19432009)

J. J. Boomsma and P. Pamilo

Gray Wolves Not Out of the Woods Yet

J. T. Bruskotter et al.

Patents: A Threat to Innovation?

M. Manocaran

REVIEW
46

Response
S. J. H. Graham and M. J. Higgins

Lipid Rifts As a Membrane-Organizing Principle

D. Lingwood and K. Simons

page 34

Let Top Students Go Forth and Prosper

G. Madhavan and B. A. Oakley

CONTENTS continued >>

COVER Crystal structure of a molybdenum oxide nanowheel, 2.6 nanometers in diameter, around a smaller molybdenum oxide cluster. Miras et al. (page 72; related Perspective page 38) used a controlled-flow reactor to show that the central core serves as a transient template for the self-assembly of the nanowheel and is ultimately ejected to yield a hollow finished product.
Image: Leroy Cronin, Ryo Tsunashima, Haralampos Miras/ University of Glasgow

DEPARTMENTS
10 13 16 17 100 101

This Week in Science Editors Choice Science Staff Random Samples New Products Science Careers

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

CONTENTS

BREVIA
51

75

Masquerade: Camouflage Without Crypsis

J. Skelhorn et al. Caterpillars masquerading as twigs are misidentified by chick predators as inanimate objects, rather than remaining undetected. >> Science Podcast

Synchronous Deglacial Overturning and Water Mass Source Changes

N. L. Roberts et al. Large-scale ocean circulation changed in different ways during a millennial-scale climate event.

78

RESEARCH ARTICLE
52

Human Genome Sequencing Using Unchained Base Reads on Self-Assembling DNA Nanoarrays
R. Drmanac et al. A low-cost sequencing technique advances us closer to the goal of the $1000 human genome.

Dendritic Mechanisms Underlying Rapid Synaptic Activation of Fast-Spiking Hippocampal Interneurons

H. Hu et al. Potassium channel enrichment in the dendrites of hippocampal basket cells defines a mechanism of neural network function.
81

Structure and Mechanisms of a Protein-Based Organelle in Escherichia coli

S. Tanaka et al. Structures of the shell proteins from a bacterial organelle help to explain how it functions in metabolizing ethanolamine. >> Perspective p. 42

page 33

REPORTS
58

An Unusually Fast-Evolving Supernova

D. Poznanski et al. The distinctive properties of this supernova suggest that it is of a kind predicted by theory but not previously observed.
84

The Tasmanian Devil Transcriptome Reveals Schwann Cell Origins of a Clonally Transmissible Cancer
E. P. Murchison et al. Sequencing the Tasmanian devil facial tumor disease provides a potential diagnostic marker. >> Science Podcast

60

Polarization-Induced Hole Doping in WideBand-Gap Uniaxial Semiconductor Heterostructures

J. Simon et al. A compositional gradient of two semiconductors creates an electronic polarization that ionizes and activates dopant atoms.
88

O-Mannosyl Phosphorylation of Alpha-Dystroglycan Is Required for Laminin Binding

T. Yoshida-Moriguchi et al. A posttranslational sugar modification required to prevent certain dystrophies is identified and characterized.

64

Translocation of Single-Stranded DNA Through Single-Walled Carbon Nanotubes

H. Liu et al. Transfer of DNA by electrophoresis through some carbon nanotubes is accompanied by giant current pulses.
91

page 64

68

Solid Nanoparticles That Catalyze Biofuel Upgrade Reactions at the Water/Oil Interface
S. Crossley et al. Oxide nanoparticles bearing carbon nanotubes and functionalized with palladium act as both emulsifiers and catalysts. >> Perspective p. 41
94

The Rate and Molecular Spectrum of Spontaneous Mutations in Arabidopsis thaliana

S. Ossowski et al. Rapid sequencing technologies allow a more accurate calculation of the mutation rate for plants.

Targeted 3 Processing of Antisense Transcripts Triggers Arabidopsis FLC Chromatin Silencing

F. Liu et al. A backward transcript of the FLOWERING LOCUS C gene of Arabidopsis is involved in regulation of the sense-strand transcription.

72

Unveiling the Transient Template in the Self-Assembly of a Molecular Oxide Nanowheel

H. N. Miras et al. Use of a flow reactor reveals a key intermediate in the formation of a molybdenum oxide nanostructure. >> Perspective p. 38
97

Reproducibility Distinguishes Conscious from Nonconscious Neural Representations

A. Schurger et al. Analysis of functional magnetic resonance imaging data reveals that neural activation patterns are more reproducible for seen versus unseen objects. >> Perspective p. 43

pages 42 & 81

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

CONTENTS

SCIENCEONLINE
SCIENCEXPRESS
www.sciencexpress.org

SCIENCESIGNALING
www.sciencesignaling.org The Signal Transduction Knowledge Environment

Detection of Gamma-Ray Emission from the Vela Pulsar Wind Nebula with AGILE
A. Pellizzoni et al. Pulsar wind nebulae could account for some of the yet unidentified Galactic gamma-ray sources. 10.1126/science.1183844
238U/235U

RESEARCH ARTICLE: Regulation of Epidermal Growth Factor Receptor Trafficking by Lysine Deacetylase HDAC6
Y. Lissanu Deribe et al. HDAC6 sets a brake that slows down the delivery of activated epidermal growth factor receptors to the degradative compartment.

Variations in Meteorites: Extant 247Cm and Implications for Pb-Pb Dating

G. A. Brennecka et al. Variable abundances of meteorite isotopes may require correcting the lead-based age of the solar system by 5 million years. 10.1126/science.1180871

RESEARCH ARTICLE: Tumor Suppression by PTEN Requires the Activation of the PKR-eIF2 Phosphorylation Pathway
Z. Mounir et al. PTEN provides a link between tumor suppression and the inhibition of protein synthesis independently of its regulation of PI3K signaling.

SCIENCESIGNALING Coupling exocytosis with endocytosis.

Darwinian Evolution of Prions in Cell Culture

J. Li et al. When propagated in vitro, prion strains demonstrate adaptability and selection. 10.1126/science.1183218

RESEARCH ARTICLE: Measurement and Clinical Monitoring of Human Lymphocyte Clonality by Massively Parallel V-D-J Pyrosequencing
S. D. Boyd et al. Rapid sequencing of immune receptor loci can provide direct detection and tracking of immune diversity.

RESEARCH ARTICLE: Cbl Controls EGFR Fate by Regulating Early Endosome Fusion
G. D. Visser Smit et al. The E3 ubiquitin ligase Cbl mediates the fusion of early endosomes necessary to target EGFR for lysosomal degradation.

Protein PRDM9 Is a Major Determinant of Meiotic Recombination Hotspots in Humans and Mice
F. Baudat et al. 10.1126/science.1183439

REPORT: Seasonal Influenza Vaccine Provides Priming for A/H1N1 Influenza Immunization
G. Del Giudice et al. A vaccine currently used to protect against H1N1 is more effective when co-injected with an adjuvant and seasonal flu vaccine.

Drive Against Hotspot Motifs in Primates Implicates the PRDM9 Gene in Meiotic Recombination
S. Myers et al. 10.1126/science.1182363

PERSPECTIVE: Channeling CalciumA Shared Mechanism for Exocytosis-Endocytosis Coupling

S. S. Vogel Exocytotic insertion of calcium channels can couple exocytosis with endocytosis.

SCIENCEPODCAST
www.sciencemag.org/multimedia/podcast Free Weekly Show Download the 1 January Science Podcast to hear about caterpillars masquerading as twigs, the chief scientist of Russias Space Research Institute, and clues to cancer in Tasmanian devils.

Prdm9 Controls Activation of Mammalian Recombination Hotspots

E. D. Parvanov et al. A chromatin-modifying enzyme functions in the determination of recombination loci within the genome. 10.1126/science.1181495

SCIENCECAREERS
www.sciencecareers.org/career_magazine Free Career Resources for Scientists

A Scientists Infectious Enthusiasm

S. Webb Benjamin tenOever is an unconventional virologist who is working to make his discoveries clinically relevant.

ORIGINSBLOG
blogs.sciencemag.org/origins A History of Beginnings

SCIENCENOW
www.sciencenow.org Highlights From Our Daily News Coverage

Coming to America: Doing a Postdoc in the U.S.

L. Laursen The logistics can be intimidating, but a postdoc in the United States is rewarding.

SCIENCEINSIDER
blogs.sciencemag.org/scienceinsider
Science Policy News and Analysis

When Fire Approaches, Chimps Keep Their Cool

Primates clear an important hurdle in humanlike mastery of fire.

Seasick? Try Controlling Your Breathing

Changing your breathing pattern may delay the onset of nausea at sea.

SCIENCETRANSLATIONAL MEDICINE
www.sciencetranslationalmedicine.org Integrating Medicine and Science

Hand SizeNot SexDetermines Sense of Touch

Womens fingers are more sensitive, but thats just because they are smaller than mens are.
CREDIT: YANA GREENMAN/SCIENCE

PERSPECTIVE: PPAR ActivationA Potential Treatment for Pulmonary Hypertension

G. Hansmann and R. T. Zamanian A new target is identified to thwart heart failure.

SCIENCE (ISSN 0036-8075) is published weekly on Friday, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue, NW, Washington, DC 20005. Periodicals Mail postage (publication No. 484460) paid at Washington, DC, and additional mailing offices. Copyright 2010 by the American Association for the Advancement of Science. The title SCIENCE is a registered trademark of the AAAS. Domestic individual membership and subscription (51 issues): $146 ($74 allocated to subscription). Domestic institutional subscription (51 issues): $910; Foreign postage extra: Mexico, Caribbean (surface mail) $55; other countries (air assist delivery) $85. First class, airmail, student, and emeritus rates on request. Canadian rates with GST available upon request, GST #1254 88122. Publications Mail Agreement Number 1069624. Printed in the U.S.A. Change of address: Allow 4 weeks, giving old and new addresses and 8-digit account number. Postmaster: Send change of address to AAAS, P.O. Box 96178, Washington, DC 200906178. Single-copy sales: $10.00 current issue, $15.00 back issue prepaid includes surface postage; bulk rates on request. Authorization to photocopy material for internal or personal use under circumstances not falling within the fair use provisions of the Copyright Act is granted by AAAS to libraries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided that $20.00 per article is paid directly to CCC, 222 Rosewood Drive, Danvers, MA 01923. The identification code for Science is 0036-8075. Science is indexed in the Readers Guide to Periodical Literature and in several specialized indexes.

PERSPECTIVE: 2A-Adrenergic Receptors in the Genetics, Pathogenesis, and Treatment of Type 2 Diabetes
S. B. Liggett Results originating from rat genetics make a compelling case for the ADRA2A locus and type 2 diabetes in humans.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

EDITED BY STELLA HURTLEY

Be-Deviled Cancer >>

Recently, a deadly transmissible cancer has emerged in Tasmanian devils, the largest existing marsupial carnivore. This disease, devil facial tumor disease (DFTD), leads to the growth of large facial tumors that frequently metastasize to internal organs. DFTD is thought to be transmitted by biting, and leads to death of affected animals within months, usually by obstructing the animals ability to feed. Consequently, in the last 10 years Tasmanian devil numbers have dropped by about 60%. There are no genetic tests, vaccines, or treatments available for this disease, and without intervention, models predict that DFTD could cause extinction of Tasmanian devils in the wild within 50 years. Several lines of evidence suggest that DFTD is transmitted as a clonal allograft, whereby the cancer cells themselves are the agents of tumor transmission. Murchison et al. (p. 84) examined this hypothesis in detail by genotyping 25 tumor-host pairs from around Tasmania at 14 microsatellite loci and at a variable mitochondrial polymorphism. DFTD tumors were indeed found to be genetically distinct from their hosts and almost completely genetically identical to one another, supporting the idea of transmission by allograft.

Lipid Rafts Come of Age

Living cells are surrounded by cellular membranes composed of lipids and proteins. Much attention has been paid to the biogenesis and sorting of membrane proteins. The dynamics and sorting of lipids have been much more difficult to study. Lingwood and Simons (p. 46) review the evidence for, and the role played by, socalled lipid raftslaterally segregated regions within membranes enriched for particular lipids and proteins.

ratios in the dendrites of the interneurons. These dendritic mechanisms can explain the highfrequency firing and precise timing of basket cells seen in network activity in vivo.

Activating Stubborn Dopants

Many applications of semiconductor light-emitting diodes and lasers, such as reading optical disks, benefit from shorter wavelengths, but this requires materials with larger energy gaps between their valance and conduction bands. The electronic conductivity of these materials often has to be increased by doping with impurity atoms. However, in nitride materials, such as GaN and AlGaN, hole doping with acceptor atoms such as Mg is ineffective at room temperature. Simon et al. (p. 60) grew a gradient of AlGaN on the surface of GaN and found that the polarization of the layer could field-ionize the acceptor dopants efficiently at room temperature. The heterostructure was used successfully in a lightemitting diode that emits in the ultraviolet.

in which the nanotube is metallic, the ionic conductivity is anomalously higher than that expected from the bulk resistivity of the electrolyte. This high conductivity was exploited for the transport of single-stranded DNA, which was accompanied by large but transient increases in the ion current.

Metal in the Middle

Biphasic reaction mixtures allow the isolation of sensitive products, which can form in one solvent and then shift rapidly into another, protected from side reactions and interfering by-products. However, ensuring that catalysts remain in the proper phase can be challenging, and surfactants added to induce efficient mixing often prove difficult to separate from the product stream. Crossley et al. (p. 68; see the Perspective by Cole-Hamilton) tackle these issues by preparing easily recoverable amphiphilic nanoparticles that simultaneously stabilize aqueous-organic emulsions and catalyze organic reactions. The particles combine hydrophobic nanotubes with hydrophilic oxides, causing them to accumulate at water-hydrocarbon interfaces. Depositing palladium on specific portions of the particles surfaces thus localizes the metal in one or both phases, facilitating hydrogenation of several compounds of interest in biofuel refining.

Dendrites Shape Interneuron Firing

Basket cells, a group of fast-spiking inhibitory interneurons, play an important part in the function of neuronal networks. The mechanisms underlying the high temporal precision and short latency of basket cell activity are unclear. Hu et al. (p. 52, published online 3 December) investigated dendrite functions in fast-spiking hippocampal basket cells and found that action potentials are initiated in the axon and propagate back into the dendrites without activity dependence but with strongly reduced amplitude. This is very different from what has been observed previously in widely investigated pyramidal cell dendrites, probably due to the high potassium to sodium conductance

Carbon Nanotube Bridge for DNA Transport

The nanoporosity of carbon nanotubes has been exploited in the control of molecular transport for example, in creating membranes. Liu et al. (p. 64) fabricated devices in which one singlewalled carbon nanotube connects two fluid reservoirs. In some of these devices, apparently those VOL 327 SCIENCE

The Depths of the Changes

Over the course of the past glacial cycle, there have been two major types of rapid, large climate warming events: shorter-lived warm intervals lasting on the order of 1000 years and the

10

1 JANUARY 2010

www.sciencemag.org

CREDITS (TOP TO BOTTOM): SAVE THE TASMANIAN DEVIL PROGRAM; HU ET AL.

This Week in Science

last glacial-interglacial transition. Although both involved dramatic changes in large-scale ocean circulation, the extent to which those changes were similar is unclear. Roberts et al. (p. 75) analyzed the neodymium isotopic composition of the Fe-Mn oxide coatings of planktonic foraminifera and reconstructed patterns of Atlantic Ocean circulation during Heinrich event 1, a rapid global climate fluctuation about 14,000 years ago involving the destruction of Northern Hemisphere ice shelves and the last deglaciation. While both the source of deep water and the whole-ocean overturning rate shifted rapidly and synchronously during the last deglacial transition, only upper ocean circulation strength was affected during Heinrich event 1.

Call for Papers

Evolution in Action
Rates of evolution in gene and genome sequences have been estimated, but these estimates are subject to error because many of the steps of evolution over the ages are not directly measurable or are hidden under subsequent changes. Ossowski et al. (p. 92) now provide a more accurate measurement of how often spontaneous mutations arise in a nuclear genome. Mutations arising over 30 generations were compared by sequencing DNA from individual Arabidopsis thaliana plants. UV- and deamination-induced mutagenesis appeared to bias the type of mutations found.

Si Science Signaling
Science Signaling, from the publisher of Science, AAAS, features top-notch, peerreviewed, original research weekly. Submit your manuscripts in the following areas of cellular regulation: Biochemistry Bioinformatics Cell Biology Development Immunology Microbiology Molecular Biology Neuroscience Pharmacology Physiology and Medicine Systems Biology Subscribing to Science Signaling ensures that you and your lab have the latest cell signaling resources. For more information visit www.ScienceSignaling.org

We Are Stardust
Supernovae form as the result of stellar explosions and are classified according to the properties of their spectra. Poznanski et al. (p. 58, published online 5 November) present a peculiar supernova that is characterized by extremely fast temporal evolution and unusual spectroscopic features, such that it defies classification. SN2002bj appears to be a member of a new class of supernovae, possibly formed by a helium detonation on a white dwarf ejecting a small envelope of material.

Modifying Protein Modification

Alpha-dystroglycan (-DG) is a cell-surface receptor that anchors the basal lamina to the sarcolemma by binding proteins containing laminin-G domains. This binding is essential for protecting muscle from contraction-induced injury, and defective binding is thought to cause a subclass of congenital muscular dystrophy (CMD) in humans. Mutations in six (putative) glycosyltransferase genes have been identified in patients with CMD, suggesting that glycosylation of -DG may confer the ability to bind laminin. Despite extensive efforts for over 20 years, the actual laminin-binding moiety has remained unclear. Now, Yoshida-Moriguchi et al. (p. 88) have identified a phosphorylated O-mannosyl glycan on -DG. This modification occurred in the Golgi via an unidentified kinase and was required for the maturation of -DG into its laminin-binding form.

Chief Scientic Editor

Flowery Regulator
Control of gene transcription is multilayered, depending on transcription factors, epigenetic mechanisms, and interactions with small RNA molecules. Liu et al. (p. 94, published online 3 December) have now found that for the FLOWERING LOCUS C (FLC) gene of the plant Arabidopsis, a backwards transcript of the gene conspires with 3 RNA-processing tools and histone demethylation to regulate the transcription of the protein-coding gene. The 3 -processing events require the antisense, not the sense, RNA transcript. It is then the sense transcript that, in the end, regulates onset of flowering.

Michael B. Yaffe, M.D., Ph.D.

Associate Professor, Department of Biology Massachusetts Institute of Technology

Editor

Nancy R. Gough, Ph.D.

AAAS

Bacterial Compartmentalization
CREDIT: 2005 TONY PIRO

Submit your research at: www.sciencesignaling.org/ about/help/research.dtl

In diverse bacteria, reactions that involve toxic or volatile metabolites are carried out by enzymes inside proteinaceous microcompartments. Tanaka et al. (p. 81; see the Perspective by Kang and Douglas) now report high-resolution crystal structures for four homologous proteins that are constituents of the shell that sequesters the metabolism of ethanolamine in bacteria. While the structures have similar overall folds, they have distinctive structural features that provide insight into how they build the shell and participate in microcompartment function. www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

EDITORIAL

Promoting Scientic Standards

THE SCIENTIFIC ENTERPRISE IS BUILT ON A FOUNDATION OF TRUST. AS KENNETH SHINE AND I

10.1126/science.1185983

*B. Alberts, K. Shine, Science 266, 1660 (1994). www.nap.edu/catalog/12192.html.

12

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

CREDITS: (TOP) TOM KOCHEL; (LEFT) Y. GREENMAN/SCIENCE

emphasized15 years ago in this journal, if science is to ourish and attain its appropriate role in aiding human progress, It is incumbent upon all of us in the scientic community to help provide a research environment that, through its adherence to high ethical standards and creative productivity, will attract and retain individuals of outstanding intellect and character to one of societys most important professions.* Journals such as Science occupy a special place in the maintenance of scientic standards. As an inuential gatekeeper to the peer-reviewed literature across the natural and social sciences, what Science decides to publish helps to dene scientic excellence for scientists. And with remarkable frequency, the broader media uses our selections to decide which scientic advances to convey to the public, adding to our profound sense of responsibility. For these reasons, the chief editors of the journals Science, Nature, and the Proceedings of the National Academy of Sciences have been working together to consider how to improve our procedures, so as to help make science as productive as possible in serving both scientists and the greater society. As a start, we have focused on two critical authorship issues. First, to discourage honorary authorships, we agreed that before acceptance, each author will be required to identify his or her contribution to the research (see www.sciencemag.org/about/authors). Sciences policy is specically designed to support the authorship requirements presented in On Being a Scientist: Third Edition, published by the U.S. National Academy of Sciences. That report emphasizes the importance of an intellectual contribution for authorship and states that Just providing the laboratory space for a project or furnishing a sample used in the research is not sufcient to be included as an author. Second, Science will require that the senior author for each laboratory or group conrm that he or she has personally reviewed the original data generated by that unit, ascertaining that the data selected for publication in specic gures and tables have been appropriately presented. Thus, for example, a researcher who prepares a digitally processed gure displaying an assortment of electrophoretic gel separations will need to present all of the original gel data to a specied senior author, who must certify that this has been done when the manuscript is returned for revision. In this way, Science aims to identify a few senior authors who collectively take responsibility for all of the data presented in each published paper. Traditionally, a single individual has been asked to accept this responsibility. But the former requirement has become increasingly unrealistic, considering that a large fraction of publications now contain contributions from groups with very different expertiseand that half of the papers published in 2009 by Science had authors from more than one nation. One issue not yet resolved is what scientic journals might do to encourage good mentoring practices by experienced scientists. Many universities now require that their young faculty members choose one or more mentors among the senior faculty. These mentors then use the wisdom and connections developed from their decades of experience to help the younger scientist in whatever ways are requested, including decisions that involve ethical standards. Being a good mentor resembles being a good parent: It involves a great deal of listening and help with problem solving and requires mutual respect and trust. Should the acknowledgments section of a publication specically list any mentoring that made a major contribution to the research? Could a special mentor search function on PubMed (and on other literature compilation Web sites) then help to reward mentors? Effective mentoring is critical to the future success of science, and as scientists remain active to more advanced ages, it provides a meaningful way to end a career. Scientists everywhere can and should do more to promote it. Bruce Alberts

Bruce Alberts is Editorin-Chief of Science.

EDITORSCHOICE
EDITED BY GILBERT CHIN AND JAKE YESTON

PHYSICS

Microwave Manipulation
Optical lattices populated by neutral atoms are a good candidate for storing quantum information. Normally, internal degrees of freedom such as the hyperfine state are used to create the basic information unit, the qubit. However, atoms also possess motional degrees of freedom; for example, the confinement of atoms in the lattice wells creates quantized vibrational states. These motional degrees of freedom are usually controlled by introducing time-dependent lattice potentials. Now, Frster et al. use microwave fields to effect transitions between vibrational levels of opposite hyperfine states of Cs atoms. Atoms in the two hyperfine states are loaded into two lattices spatially offset from each other. This arrangement enables transitions between different vibrational states, but the probability depends on the overlap of the (offset) wave functions. If the lattice is deep, transitions only happen between neighboring, slightly offset wells; if it is shallow, the offset can be increased and the atom becomes delocalized. Effective initialization into the lowest vibrational state is achieved, and Rabi oscillations between arbitrary vibrational states are demonstrated. This approach may lead to full control of quantum transport, likely a necessity for processing quantum information in this system. JS
Phys. Rev. Lett. 103, 233001 (2009).
CELL BIOLOGY

Plasmodium in comparison to laboratory cultures revealed differences in gene expression profiles. Acharya et al. have analyzed the protein expression profiles of two species of Plasmodium that were isolated from the blood of patients; they identified about 100 proteins, some of which had not been found in laboratory cultures and could make promising drug or vaccine targets. HP
Proteomics Clin. Appl. 3, 1314 (2009).
NEUROSCIENCE

The Next Top Model

Consumers may be familiar with high-end graphic processing components in video game consoles, such as the PlayStation3, or as a consequence of outfitting personal computers ordered online with NVIDIA graphics cards; these advances in hardware have also attracted the attention of procurement officials in the military services. In the academic realm, Pinto et al. have harnessed the power of clustered graphics processors to assess the relative performance of machine vision models of object recognition. The availability of massively parallel processing power at reasonable cost allowed them to explore, in 1 week versus 2 years, sizable regions of parameter space by varying the number of filters, the learning rate, and so forth. They generated a library of 7500 models that were trained on individually rendered objects during an unsupervised learning phase, and then screened on the basis of recognizing cars versus planes, which were presented in a range of orientations and on a variety of backgrounds. The top-ranked models were then evaluated broadly across other objects and on one of the toughest recognition tasksphotographs of human facesand compared to a number of sophisticated algorithms, which yielded a small set of parameter values that were associated with high object recognition accuracy. GJC
PLoS Comput. Biol. 5, e1000579 (2009).

BIOMEDICINE

Sealed with a Platelet

CREDITS (LEFT TO RIGHT): ADAPTED BY NAT. MED. 10.1038/NM.2060 (2009); PINTO ET AL., PLOS COMPUT. BIOL. 5, E1000579 (2009)

The fetal circulatory system has distinctive anatomical features because the fetus obtains oxygen through the placenta rather than through its lungs. Before birth, a blood vessel called the ductus arteriosus (DA) allows blood to bypass the nonfunctional fetal lungs by connecting the pulmonary artery, which supplies blood to the lungs, with the aorta, which supplies blood to the rest of the body. This vessel normally closes a day or two after birth, but in some newborns, it remains open and can lead to lifethreatening complications. Studying newborn mice, Echtler et al. make the surprising observation that plateletscells noted for their role in blood clottingwere recruited to the lumen of the DA within 20 minutes after birth of the mice; when platelet production or function was disrupted, the DA failed to close completely, leading to abnormal patterns of blood flow. The recruited platelets play a dual role in DA closureby forming a physical plug that seals the lumen of the constricted DA and by altering the behavior of other cell types involved in blood vessel remodeling. PAK
Nat. Med. 10.1038/nm.2060 (2009).

In the Wild
Malaria is one of the most prevalent infectious diseases and kills around 900,000 people per year. It is caused by parasites of the genus Plasmodium, which are transmitted to humans by mosquitoes and enter red blood cells, causing fever and, if left untreated, death. Human pathogens of all kinds can develop resistance to the most effective drugs, such as artemisinin, so there is a constant need to identify new compounds. Animal models of malaria have proven problematic to establish, and most studies have used laboratory cultures of human blood cells to grow the parasites. While important insights into the life cycle and pathogenic action of Plasmodium have come from these in vitro studies, a recent study of clinically isolated samples of SCIENCE VOL 327

www.sciencemag.org

1 JANUARY 2010

13

AAAS, publisher of Science, presents

Nearly 700 speakers who will convene in San Diego

Who do you know? Who do you want to meet?
Dalia Abdel-Salam Octavio AburtoOropeza Arun Agrawal Peter C. Agre Olanrewaju Aina A. Hadi Al Khalili Bruce M. Alberts Jonathan Alger Luca Alinovi Brad Allenby Abdalla Alnajjar David Anderson Richard Askey Lloyd Axworthy Stefan Bachu Peter Backus Jeffrey Bada Maureen A. Baginski Chunli Bai Barbara A. Baird David Balton Christopher Barty George Basile David Baulcombe Julia Baum Richard D. Beger Frances Beinecke Keith Bell Philip Bell Arden L. Bement Eshel Ben-Jacob Kennette Benedict Mark L. Benthien Andrea Bertozzi Bernard Bigot Linda Billings Christine A. Biron Richard Bissell Pratim Biswas Robert Biswas-Diener Cecilia Bitz Erich Bloch Sally Blower Ray Boeman Steve Bograd William Bond Jacques Bouchard Jean-Phillipe Bouchaud Philip Bourne W. Thomas Boyce Maxwell T. Boykoff Alan Boyle Shannon Bradley Steven J. Brams Lewis M. Branscomb John Bransford James Brase James Brewer William B. Bridges Larry Brilliant Edward Briscoe Lars Bromley Daniel G. Brown Marilyn Brown Peter Bruns Harvey Bryan Jed Z. Buchwald James Buizer Martin Bunzl Dennis R. Burton John Bwarie Robert L. Byer Ken Caldeira Meg Caldwell Tyrone Cannon Lon R. Cardon Lynda T. Carlson Sebastian Carney Mark Carr Michael Carr Sean M. Carroll Jean-Laurent Casanova Gail Cassell Kenneth Cassman Thomas Casten Carlos Castillo-Chavez Claudia Cevenini Oliver Chadwick Christopher Chambers Frances Champagne Grantley D. Charles R. Alta Charo Hongda Chen Zhong Chen Donghong Cheng So-Min Cheong Kevin Chipman Patrick Christie Sarah P. Chu Ratana Chuenpagdee Yoon Chung Joshua E. Cinner Constance Citro KC Claffy Andrew Clark Christopher Clark Douglas Clark Elizabeth Clarkson Kathryn Clay Beatriz C. Clewell Kenneth Coale Theo Colborn Arthur L. Coleman E. William Colglazier Brianno Coller Scott L. Collins Vicki Colvin Rita R. Colwell Susan G. Conard Cassie Conley Tara Connelly Pamela G. Conrad William C. Copeland Marie Coppola Clemencia Cosentino Christopher Costello Donna Cox Peter Crane Robert P. Crease Larry Crowder Ian Crute Yi Cui Lisa M. Curran Cynthia Cwik Bruce Dale George Daley Al Darzins Peter Daszak Jon Michael Davis Andre de Gouvea Jose Antonio de la Pea Maryellen de Mars Michele de Nigris Julio de Paula Gerda de Vries Pierre Degond Richard DeMillo Nancy Denslow Kemal Dervis Keith Devlin Roseanne Diab Mariette DiChristina Catherine Didion Ed Diener George Dimopoulos Carl Djerassi Walter Doerfler Cheng Donghong Simon Donner Fanny Douvere Frank D. Drake John W. Drake Anna Dreber Kelvin Droegemeier Sean Drummond Catherine Dulac Kevin N. Dunbar Greg J. Duncan Maitreya Dunham William Dunlop Mike Dunne Jim Dunwell Joachim Eble Paul H. Edelman Robert Edwards Scott Edwards Bud Ehler Charles N. Ehler Hajo Eicken Nadia El-Awady Lynn E. Elfner Kirsten Ellenbogen John Ellis Michael Emerman Karen Emmorey Robert Engle Saskia Estupin-Day E. Margaret Evans Victoria J. Fabry James Fahn J. Doyne Farmer Richard A. Feely Nina Fefferman Suzanne E. Fenton Luigi Ferrucci Phillip Finck Diane Fischer Martin Fisher Richard Flavell James R. Fleming Lamont Flowers Jo Foden James Fowler Marye Anne Fox Michelle Fox Harrison S. Fraker Robert T. Fraley Catherine Franche Uta Francke Marcos Frank Andrew G. Fraser Kimberley Freeman William R. Freudenburg Uwe Fritsche Monique Frize Bruce A. Fuchs Erin M. Furtak Abraham Gademu Fred H. Gage Michele Garfinkel S. James Gates John Gearhart James Paul Gee Pallab Ghosh Peter J. Gianaros Holly K. Gibbs John Gilleland Kristina Gjerde Malti Goel Stig Goethe Edward J. Goetzl Simon Goldbard Madge Goldman Lawrence Goldstein Vladimir Golitsyn Patrick Gonzales Federico Goodsaid Dmitry A. Gordenin Stephen Grand Robert Grant Giorgio Gratta Lou Gray Henry Greely Stanley Gregory Carol W. Greider Tina Grieco-Calub Melinda Grier Huw Griffiths Kirsten Grorud-Colvert Tony H. Grubesic Seth Guikema Lauren Gunderson David H. Guston Evan Hadingham Victoria Hale Amy Laura Hall Charles Hall Jason Hall-Spencer Christopher G. Halliday Benjamin Halpern Diane F. Halpern Larry A. Hammond Marah J. Hardt Tracy Hart Mohamed H.A. Hassan Egill Hauksson Robert M. Hazen Bob Hebner Jeff Hecht James Heckman Richard Heinberg Dirk Helbing John F. Helliwell Andrea Henderson Socrates Herrera Alejandro Herrero Lawrence A. Herzog Mark Hildebrand Amy Hill John Hocevar Keith Hodgson Jules A. Hoffmann Marika M. Holland Hugo Hollanders Philip J. Holmes Greg Horowit Paul Horowitz Steve Horton Paul Horwitz Derek Houston Kenneth Huber Terry Hughes Arti Hurria

Start connecting now www.aaas.org/meetings

2010 AAAS Annual Meeting

18-22 FebruarySan Diego
Jo Husbands Ian D. Hutcheon Susanne Ihsen Ioannis Ikessides Barbara Illman Irwin Jacobs Jennifer Jacquet Rudolf Jaenisch David Jentsch Chris Johnson Jean Johnson Kristina Johnson Lucile M. Jones Thomas H. Jordan Cliff Joslyn Gerald F. Joyce Tina Kaarsberg Kathryn Kailikole Jim Kakalios Dan Kammen Carrie Kappel David Keith Jamie Lewis Keith Willett Kempton Diane Ketelhut Suk Joon Kim Lonnie King Nicole King David Klahr George Klein Joanie Kleypas Erik Klopfer Michael Klymkowsky Marcelo Knobel Leon Kochian Karen Kohfeld Kasper Kok Richard D. Kolodner Maria Kozhevnikov Nina Kraus Meg A. Krawchuk Tim Kring Michael Lssig Myanna Lahsen Sherman Lai Dolores J. Lamb Zeph Landau Eric S. Lander Julia Lane Neal Lane Daniel J. Lang Chris Langdon Tom Laughlin Lester Lave Stephan Lechner June Seung Lee Wim Leemans Vladimir Lefebvre Anthony Leggett Jack R. Leibowitz Sarah E. Lester Teresa Lettieri Bruce V. Lewenstein Nathan Lewis Rebecca Lewison Tim Ley XiaoChun Li Barbara Libby Mike Lindstrom Marcia Linn Bjrn-Ola Linnr Andy Lipkis Chris Llewellyn-Smith David Lobell Kathryn Long Christopher G. Lowe Stefano Luccioli Stephen Luoni Klaus Lutzenkirchen Jonathan Lynch John Lynham Sonja Lyubomirsky Patricia L. Mabry Allison Macfarlane Anne Jane MacLachlan Francis Macrina Katherine Magnuson Natalie Mahowald Arunava Majumdar Shirley M. Malcom Bradley Malin Stanley Maloy Geoffrey Manley Margaret Mann Christopher Manning Lori Marino Elizabeth Marshall Daniel Martell Laura Martin Joan Martinez-Alier Krzysztof Maruszewski Wieslaw Maslowski Walter E. Massey Ivan Matic Andrew D. Maynard Sarah McCaffrey William McClintock Kathleen McDermott Bruce S. McEwen Mark McGranaghan Amy McGuire Marcia McNutt Sara C. Mednick Julia E. Melkers Dennis Meredith Frederick L. Merrill Stephen A. Merrill Vera Michalchik Edward L. Miles Jon D. Miller Meredith Minkler Erik Jap Molenaar Megan Mollenhauer Suresh Moolgavkar James L. Moore Alejandro Moreno Granger Morgan John Morton Edward Moses Mehdi Moussaid Rongping Mu Sead Muftic Daniel R. Muhs Margaret Murnane Jason Murray Charles C. Muscoplat Steven Musser Lynn Nadel Hiroshi Nagano Yuko Nagano Rosamond Naylor Jeff Nesbit Claudia Neuhauser Robert Nicholls Mary D. Nichols Peter Nicholson H.E. Tsuneo Nishida Hendrik H. Nollens Doug Nowacek Ruth Nowjack-Raymer Arthur J. Nozak Ron ODor Michael OHare Onche Odeh Jay Odell Shigehiro Oishi Carol Oliver Stephen G. Oliver Donald W. Olson Lennart Olsson Naomi Oreskes Elinor Ostrom Dorcas Otieno Jennifer Ouellette Bernhard O. Palsson Stephen R. Palumbi Becky Parker John Parmentola Ron Pate Aniruddh D. Patel Heather B. Patisaul Ari Patrinos Tad Patzek Tanya T. Paull Daniel Pauly David N. Payne Ivette Perfecto Sidney Perkowitz Donald K. Perovich Ted Peters Keith Pezzoli Julia Phillips William D. Phillips Steward T.A. Pickett Ellen K. Pikitch Stewart L. Pimm Huang Ping Richard Plevin Geoffrey S. Plumlee Richard Pollnac Wendy Pollock Shirley Pomponi Keith Porter Oscar Porter Jeffrey P. Prestemon Jules Pretty Kenneth Prewitt Huw Price Patti Price Susanna Priest Luisa Prista Joseph M. Prospero Margaret Race Michael Rafii Lee Rainie Baldev Raj Regina Rameika John S. Ramsdell Philip J. Rasch Andrew Rassweiler David H. Raulet Gerrit Rauws Stephen Raverty Catherine Redgwell Daniel C. Reed Knut Reinert Diana Reiss Mary Relling Felix Rembold Ortwin Renn Curt Rice Liesl Riddle David Ridling John Robinson Dan Roden Oscar Romo Margot Roosevelt Jo Ellen Roseman Andrew Rosenberg Susan M. Rosenberg Cynthia Rosenzweig Ivor Royston Carol Rubin Edward S. Rubin Mary Ruckelshaus Norah Rudin Rosaura Ruiz John Rummel Ivan Rusyn Donald G. Saari James Sallis Wayne Samuelson James N. Sanchirico Carmen Sapienza Hiram Sarabia Daniel Sarewitz Carter Savage Akira Sawa Anna Lee Saxenian Daniel L. Schacter Harry N. Scheiber Roland Schenkel Gottfried Schlaug Stephen H. Schneider Sylvia Schwaag Serger Kellogg Schwab Lori Schwacke Brian Schwartz Megan Schwarzman Eugenie C. Scott Richard Sears Edward Seidel Robert Semper Ismail Serageldin Juliet P. Shaffer John Sheehan John Shepherd Kinya Shimizu Thomas F. Shipley Olive Shishana Paul Shuch Rajesh Shukla Anthony Siegman Ellen V. Sigal Brian Silliman Jacob Silverman Finbarr Sloane Larry Smarr Brian Smith Douglas Smith Harold Smith Laurel Smith-Doerr V. Scott Solberg Robert Solow Ralf Sommerfeld Kevin St. Martin H. Eugene Stanley Melanie Stegman Jeff Stein Dirk Steinke Adin Steinsaltz Jane Stevens Martin Storskdieck Steffanie Strathdee Thomas W. Swetnam Toshiki Tajima Kazuhiko Takeuchi Heather Tallis Paul Tanenbaum Margaret Taylor Volker ter Meulen Sharon Terry Claes Thegerstrm David Tilman George E. Tita Ewen C. Todd George Tolley Jasna Tomic Thomas P. Tomich Eric J. Topol Aline Tribollet Loucas Tsakalakos Mary Turnipseed John Tyson Christina Ullenius David Ulph Hans G. Ulrich Vishal S. Vaidya William J. Valdez Andre Van Steirteghem John G. Vandenbergh Hal R. Varian Jay Vavra Enriqueta Velarde Stephanie Venn-Watson Craig Venter James Verdin Joseph M. Vinetz Robert C. von Borstel Tomoaki Wada Graham C. Walker Matthew Walker John Walsh Fenyu Wang Patsy Wang-Iverson Masataka Watanabe Alice Waters Les Watling Michelle Watt Frank Wattenberg Raoul Weiler Martin Weiss Scott Weiss Irving Weissman David W. Weller Kevin D. Wells Dan Werthimer John Wettlaufer Maurice Whelan Alan White Crow White Thomas I. White Lyndell Whitley Norm Whitton Douglas J. Wiebe Arnim Wiek Harry Wiland Peter A. Wilderer Brian Williams David R. Williams Fiona Williams Jennifer M. Wilson Richard Wilson Cynthia Winston Owen Witte Nathan Wolfe Susan Wolfinbarger Gordon Woo John Wood Nanette Woonton Boris Worm David Wright John Wright William Wulf Victor Yakovenko Daniel Yankelovich Masaru Yarime Kimberley Yates Banhar Yeap Gary Yohe Jeong-Ro Yoon Christina Young Stanley Young Stephen Younger Dihua Yu Lauren Zeise Dadi Zhou S. Lawrence Zipursky Bora Zivkovic Myrsini Zorba Ramunas Zydelis

Register now: www.aaas.org/meetings

www.sciencemag.org

EDITOR-IN-CHIEF Bruce Alberts EXECUTIVE EDITOR NEWS EDITOR

EXECUTIVE PUBLISHER Alan I. Leshner PUBLISHER Beth Rosner

Monica M. Bradford M. Sugden

Colin Norman
FULFILLMENT SYSTEMS

MANAGING EDITOR, RESEARCH JOURNALS Katrina L. Kelner DEPUTY EDITORS R. Brooks Hanson, Barbara R. Jasny,

Andrew

Waylon Butler;

SERVICE SUPERVISOR

1200 New York Avenue, NW Washington, DC 20005

Editorial: 202-326-6550, FAX 202-289-7562 News: 202-326-6581, FAX 202-371-9227

Bateman House, 82-88 Hills Road Cambridge, UK CB2 1LQ

+44 (0) 1223 326500, FAX +44 (0) 1223 326501
SUBSCRIPTION SERVICES For change of address, missing issues, new

orders and renewals, and payment questions: 866-434-AAAS (2227) or 202-326-6417, FAX 202-842-1065. Mailing addresses: AAAS, P.O. Box 96178, Washington, DC 20090-6178 or AAAS Member Services, 1200 New York Avenue, NW, Washington, DC 20005 I NSTITUTIONAL S ITE L ICENSES please call 202-326-6755 for any questions or information REPRINTS: Author Inquiries 800-635-7181 Commercial Inquiries 803-359-4578 PERMISSIONS 202-326-7074, FAX 202-682-0816 MEMBER BENEFITS AAAS/Barnes&Noble.com bookstore www.aaas.org/bn; AAAS Online Store www.apisource.com/aaas/ code MKB6; AAAS Travels: Betchart Expeditions 800-252-4910; Apple Store www. apple.com/eppstore/aaas; Bank of America MasterCard 1-800-833-6262 priority code FAA3YU; Cold Spring Harbor Laboratory Press Publications www.cshlpress.com/affiliates/aaas.htm; GEICO Auto Insurance www.geico.com/landingpage/go51.htm?logo=17624; Hertz 800-654-2200 CDP#343457; Office Depot https://bsd. ofcedepot.com/portalLogin.do; Seabury & Smith Life Insurance 800424-9883; Subaru VIP Program 202-326-6417; VIP Moving Services www.vipmayower.com/domestic/index.html; Other Benets: AAAS Member Services 202-326-6417 or www.aaasmember.org. [email protected] (for general editorial queries) [email protected] (for queries about letters) [email protected] (for returning manuscript reviews) [email protected] (for book review queries) Published by the American Association for the Advancement of Science (AAAS), Science serves its readers as a forum for the presentation and discussion of important issues related to the advancement of science, including the presentation of minority or conflicting points of view, rather than by publishing only material on which a consensus has been reached. Accordingly, all articles published in Scienceincluding editorials, news and comment, and book reviewsare signed and reect the individual views of the authors and not ofcial points of view adopted by AAAS or the institutions with which the authors are afliated. AAAS was founded in 1848 and incorporated in 1874. Its mission is to advance science, engineering, and innovation throughout the world for the benet of all people. The goals of the association are to: enhance communication among scientists, engineers, and the public; promote and defend the integrity of science and its use; strengthen support for the science and technology enterprise; provide a voice for science on societal issues; promote the responsible use of science in public policy; strengthen and diversify the science and technology workforce; foster education in science and technology for everyone; increase public engagement with science and technology; and advance international cooperation in science.

SENIOR EDITORS/COMMENTARY Lisa D. Chong, Brad Wible; SENIOR Gilbert J. Chin, Pamela J. Hines, Paula A. Kiberstis (Boston), Marc S. Lavine (Toronto), Beverly A. Purnell, L. Bryan Ray, Guy Riddihough, H. Jesse Smith, Phillip D. Szuromi (Tennessee), Valda Vinson, Jake S. Yeston; ASSOCIATE EDITORS Kristen L. Mueller, Jelena Stajic, Nicholas S. Wigginton, Laura M. Zahn; RESEARCH ASSOCIATE Alexis Wynne Mogul; ONLINE EDITOR Stewart Wills; ASSOCIATE ONLINE EDITORS Robert Frederick, Tara S. Marathe; WEB CONTENT DEVELOPER Martyn Green; BOOK REVIEW EDITOR Sherman J. Suter; ASSOCIATE LETTERS EDITOR Jennifer Sills; EDITORIAL MANAGER Cara Tate; SENIOR COPY EDITORS Jeffrey E. Cook, Cynthia Howe, Harry Jach, Barbara P. Ordway, Trista Wagoner; COPY EDITORS Chris Filiatreau, Lauren Kmec; EDITORIAL COORDINATORS Carolyn Kyle, Beverly Shields; PUBLICATIONS ASSISTANTS Ramatoulaye Diop, Joi S. Granger, Jeffrey Hearn, Lisa Johnson, Scott Miller, Jerry Richardson, Jennifer A. Seibert, Brian White, Anita Wynn; EDITORIAL ASSISTANTS Emily Guise, Michael Hicks, Patricia M. Moore, Miriam Weinberg; EXECUTIVE ASSISTANT Sylvia S. Kihara; ADMINISTRATIVE SUPPORT Maryrose Madrid; EDITORIAL FELLOW Melissa R. McCartney N EWS DEPUTY NEWS EDITORS Robert Coontz, Eliot Marshall, Jeffrey Mervis, Leslie Roberts; CONTRIBUTING EDITORS Elizabeth Culotta, Polly Shulman; NEWS WRITERS Yudhijit Bhattacharjee, Adrian Cho, Jennifer Couzin, David Grimm, Constance Holden, Jocelyn Kaiser, Sam Kean, Richard A. Kerr, Eli Kintisch, Andrew Lawler (New England), Greg Miller, Elizabeth Pennisi, Robert F. Service (Pacific NW), Erik Stokstad; INTERNS Michael Torrice, Jue Wang; CONTRIBUTING CORRESPONDENTS Jon Cohen (San Diego, CA), Daniel Ferber, Ann Gibbons, Robert Koenig, Mitch Leslie, Charles C. Mann, Virginia Morell, Gary Taubes; COPY EDITORS Linda B. Felaco, Melvin Gatling, Melissa Raimondi; ADMINISTRATIVE SUPPORT Scherraine Mack, Fannie Groom; BUREAUS New England: 207-549-7755; San Diego, CA: 760-942-3252, FAX 760-942-4979; Pacic Northwest: 503-963-1940 PRODUCTION DIRECTOR James Landry; SENIOR MANAGER Wendy K. Shank; ASSISTANT MANAGER Rebecca Doshi; SENIOR SPECIALISTS Steve Forrester, Chris Redwood; SPECIALIST Anthony Rosen; PREFLIGHT DIRECTOR David M. Tompkins; MANAGER Marcus Spiegler; SPECIALIST Jason Hillman ART DIRECTOR Yael Kats; ASSOCIATE ART DIRECTOR Laura Creveling; SENIOR ILLUSTRATORS Chris Bickel, Katharine Sutliff; ILLUSTRATOR Yana Greenman; SENIOR ART ASSOCIATES Holly Bishop, Preston Huey, Nayomi Kevitiyagala; ART ASSOCIATES Jessica Neweld, Matthew Twombly; PHOTO EDITOR Leslie Blizard

EDITORIAL

EDITORS

AND OPERATIONS ([email protected]); DIRECTOR Nomuna Nyamaa; CUSTOMER Pat Butler; SPECIALISTS Latoya Casteel, LaVonda Crawford, Vicki Linton, April Marshall; DATA ENTRY SUPERVISOR Cynthia Johnson; SPECIALISTS Shirlene Hall, Tarrika Hill, William Jones SENIOR SYSTEMS ANALYST

AND A DMINISTRATION DIRECTOR Deborah RiveraWienhold; ASSISTANT DIRECTOR, BUSINESS OPERATIONS Randy Yi; MANAGER, BUSINESS ANALYSIS Eric Knott; MANAGER , BUSINESS OPERATIONS Jessica Tierney; FINANCIAL ANALYSTS Priti Pamnani, Celeste Troxler; RIGHTS AND PERMISSIONS: ADMINISTRATOR Emilie David; ASSOCIATE Elizabeth Sandler; MARKETING DIRECTOR Ian King; MARKETING MANAGERS Allison Pritchard, Alison Chandler, Julianne Wielga; MARKETING ASSOCIATES Aimee Aponte, Mary Ellen Crowley, Adrian Parham, Wendy Wise; MARKETING EXECUTIVE Jennifer Reeves; DIRECTOR, SITE LICENSING Tom Ryan; DIRECTOR, CORPORATE RELATIONS Eileen Bernadette Moran; PUBLISHER RELATIONS, eRESOURCES SPECIALIST Kiki Forsythe; SENIOR PUBLISHER RELATIONS SPECIALIST Catherine Holland; PUBLISHER RELATIONS, EAST COAST Phillip Smith; PUBLISHER RELATIONS, WEST COAST Philip Tsolakidis; FULFILLMENT SUPERVISOR Iquo Edim; FULFILLMENT COORDINATOR Carrie MacDonald; MARKETING MANAGER Christina Schlecht; MARKETING ASSOCIATE Mary Lagnaoui; ELECTRONIC MEDIA : MANAGER Lizabeth Harman; PROJECT MANAGER Trista Snyder; ASSISTANT MANAGER Lisa Stanford; SENIOR PRODUCTION SPECIALISTS Ryan Atkins, Christopher Coleman, Walter Jones; PRODUCTION SPECIALISTS Nichele Johnston, Kimberly Oster; DIRECTOR, WEB AND NEW MEDIA Will Collins

B USINESS O PERATIONS

SCIENCE INTERNATIONAL EUROPE ([email protected]) EDITORIAL: INTERNATIONAL MANAGING EDITOR Andrew M. Sugden; SENIOR EDITOR/COMMENTARY Julia Fahrenkamp-

INFORMATION FOR AUTHORS

See pages 807 and 808 of the 6 February 2009 issue or access www.sciencemag.org/about/authors

Uppenbrink; SENIOR EDITORS Caroline Ash, Stella M. Hurtley, Ian S. Osborne, Peter Stern; ASSOCIATE EDITOR Maria Cruz; LOCUM EDITOR Helen Pickersgill; EDITORIAL SUPPORT Deborah Dennison, Rachel Roberts, Alice Whaley; ADMINISTRATIVE SUPPORT John Cannell, Janet Clements, Louise Moore; NEWS: EUROPE NEWS EDITOR John Travis; DEPUTY NEWS EDITOR Daniel Clery; CONTRIBUTING CORRESPONDENTS Michael Balter (Paris), John Bohannon (Vienna), Martin Enserink (Amsterdam and Paris), Gretchen Vogel (Berlin) LATIN AMERICA CONTRIBUTING CORRESPONDENT Antonio Regalado ASIA Japan Ofce: Asca Corporation, Tomoko Furusawa, Rustic Bldg. 7F, 77 Tenjin-cho, Shinjuku-ku, Tokyo 162-0808, Japan; +81 3 6802 4616, FAX +81 3 6802 4615, [email protected]; ASIA NEWS EDITOR Richard Stone (Beijing: [email protected]); CONTRIBUTING CORRESPONDENTS Dennis Normile [Japan: +81 (0) 3 3391 0630, FAX +81 (0) 3 5936 3531; [email protected]]; Hao Xin [China: +86 (0) 10 6307 4439 or 6307 3676, FAX +86 (0) 10 6307 4358; [email protected]]; Pallava Bagla [South Asia: +91 (0) 11 2271 2896; [email protected]]

Bill Moran Sean Sanders: 202-326-6430 PROJECT DIRECTOR, OUTREACH Brianna Blaser PRODUCT ([email protected]); MIDWEST/WEST COAST/W. CANADA Rick Bongiovanni: 330-405-7080, FAX 330-405-7081; EAST COAST/ E . CANADA Laurie Faraday: 508-747-9395, FAX 617-507-8189; UK / EUROPE / ASIA Roger Goncalves: TEL/FAX +41 43 243 1358; JAPAN ASCA Corporation, Nanako Ide +81 (0) 3 6802 4616, FAX +81 (0) 3 6802 4615; [email protected]; SENIOR TRAFFIC ASSOCIATE Deiandra Simms WORLDWIDE ASSOCIATE DIRECTOR OF SCIENCE CAREERS Tracy Holmes: +44 (0) 1223 326525, FAX +44 (0) 1223 326532 CLASSIFIED ([email protected]); U.S.: SALES MANAGER Daryl Anderson: 202-326-6543; MIDWEST Tina Burks: 202-326-6577; EAST COAST Alexis Fleming: 202-326-6578; WEST / SOUTH CENTRAL Nicholas Hintibidze: 202-326-6533; SALES COORDINATORS Rohan Edmonson, Shirley Young; SALES Susanne Kharraz, Dan Pennington, Alex Palmer; SALES ASSISTANT Lisa Patterson; JAPAN ASCA Corporation, Jie Chin +81 (0) 3 6802 4616, FAX +81 (0) 3 6802 4615; [email protected]; ADVERTISING SUPPORT MANAGER Karen Foote: 202-326-6740; ADVERTISING PRODUCTION OPERATIONS MANAGER Deborah Tompkins; SENIOR PRODUCTION SPECIALIST/GRAPHIC DESIGNER Amy Hardcastle; SENIOR PRODUCTION SPECIALIST Robert Buck; SENIOR TRAFFIC ASSOCIATE Christine Hall
ADVERTISING DIRECTOR, WORLDWIDE AD SALES COMMERCIAL EDITOR

AAAS BOARD
PRESIDENT

OF DIRECTORS RETIRING PRESIDENT, CHAIR James J. McCarthy; Peter C. Agre; PRESIDENT-ELECT Alice Huang; TREASURER David E. Shaw; CHIEF EXECUTIVE OFFICER Alan I. Leshner; BOARD Alice Gast, Linda P. B. Katehi, Nancy Knowlton, Cherry A. Murray, Julia M. Phillips, Thomas D. Pollard, David S. Sabatini, Thomas A. Woolsey

SENIOR EDITORIAL BOARD

John I. Brauman, Chair, Stanford Univ. Richard Losick, Harvard Univ. Linda Partridge, Univ. College London Michael S. Turner, University of Chicago

BOARD OF REVIEWING EDITORS

Adriano Aguzzi, Univ. Hospital Zrich Takuzo Aida, Univ. of Tokyo. Joanna Aizenberg, Harvard Univ. Sonia Altizer, Univ. of Georgia David Altshuler, Broad Institute Arturo Alvarez-Buylla, Univ. of California, San Francisco Richard Amasino, Univ. of Wisconsin, Madison Angelika Amon, MIT Meinrat O. Andreae, Max Planck Inst., Mainz Kristi S. Anseth, Univ. of Colorado John A. Bargh, Yale Univ. Cornelia I. Bargmann, Rockefeller Univ. Ben Barres, Stanford Medical School Marisa Bartolomei, Univ. of Penn. School of Med. Facundo Batista, London Research Inst. Ray H. Baughman, Univ. of Texas, Dallas Yasmine Belkaid, NIAID, NIH Stephen J. Benkovic, Penn State Univ. Ton Bisseling, Wageningen Univ. Mina Bissell, Lawrence Berkeley National Lab Peer Bork, EMBL Robert W. Boyd, Univ. of Rochester Paul M. Brakeeld, Leiden Univ. Joseph A. Burns, Cornell Univ. William P. Butz, Population Reference Bureau Mats Carlsson, Univ. of Oslo Peter Carmeliet, Univ. of Leuven, VIB Mildred Cho, Stanford Univ. David Clapham, Childrens Hospital, Boston David Clary, Oxford University J. M. Claverie, CNRS, Marseille Jonathan D. Cohen, Princeton Univ. Andrew Cossins, Univ. of Liverpool Robert H. Crabtree, Yale Univ. Wolfgang Cramer, Potsdam Inst. for Climate Impact Research

F. Fleming Crim, Univ. of Wisconsin William Cumberland, Univ. of California, Los Angeles Jeff L. Dangl, Univ. of North Carolina Stanislas Dehaene, Collge de France Edward DeLong, MIT Emmanouil T. Dermitzakis, Univ. of Geneva Medical School Robert Desimone, MIT Claude Desplan, New York Univ. Dennis Discher, Univ. of Pennsylvania Scott C. Doney, Woods Hole Oceanographic Inst. W. Ford Doolittle, Dalhousie Univ. Jennifer A. Doudna, Univ. of California, Berkeley Julian Downward, Cancer Research UK Denis Duboule, Univ. of Geneva/EPFL Lausanne Christopher Dye, WHO Michael B. Elowitz, Calif. Inst. of Technology Gerhard Ertl, Fritz-Haber-Institut, Berlin Mark Estelle, Indiana Univ. Barry Everitt, Univ. of Cambridge Paul G. Falkowski, Rutgers Univ. Ernst Fehr, Univ. of Zurich Tom Fenchel, Univ. of Copenhagen Alain Fischer, INSERM Scott E. Fraser, Cal Tech Chris D. Frith, Univ. College London Wulfram Gerstner, EPFL Lausanne Charles Godfray, Univ. of Oxford Diane Grifn, Johns Hopkins Bloomberg School of Public Health Christian Haass, Ludwig Maximilians Univ. Steven Hahn, Fred Hutchinson Cancer Research Center Gregory J. Hannon, Cold Spring Harbor Lab. Niels Hansen, Technical Univ. of Denmark Dennis L. Hartmann, Univ. of Washington Chris Hawkesworth, Univ. of St. Andrews Martin Heimann, Max Planck Inst., Jena James A. Hendler, Rensselaer Polytechnic Inst. Ray Hilborn, Univ. of Washington Michael E. Himmel, National Renewable Energy Lab. Kei Hirose, Tokyo Inst. of Technology Ove Hoegh-Guldberg, Univ. of Queensland Brigid L. M. Hogan, Duke Univ. Medical Center Ronald R. Hoy, Cornell Univ. Olli Ikkala, Helsinki Univ. of Technology Meyer B. Jackson, Univ. of Wisconsin Med. School

Stephen Jackson, Univ. of Cambridge Steven Jacobsen, Univ. of California, Los Angeles Peter Jonas, Universitt Freiburg Barbara B. Kahn, Harvard Medical School Daniel Kahne, Harvard Univ. Gerard Karsenty, Columbia Univ. College of P&S Bernhard Keimer, Max Planck Inst., Stuttgart Elizabeth A. Kellog, Univ. of Missouri, St. Louis Hanna Kokko, Univ. of Helsinki Lee Kump, Penn State Univ. Mitchell A. Lazar, Univ. of Pennsylvania David Lazer, Harvard Univ. Virginia Lee, Univ. of Pennsylvania Olle Lindvall, Univ. Hospital, Lund Marcia C. Linn, Univ. of California, Berkeley John Lis, Cornell Univ. Richard Losick, Harvard Univ. Ke Lu, Chinese Acad. of Sciences Laura Machesky, CRUK Beatson Inst. for Cancer Research Andrew P. MacKenzie, Univ. of St Andrews Raul Madariaga, cole Normale Suprieure, Paris Anne Magurran, Univ. of St Andrews Charles Marshall, Harvard Univ. Martin M. Matzuk, Baylor College of Medicine Virginia Miller, Washington Univ. Yasushi Miyashita, Univ. of Tokyo Richard Morris, Univ. of Edinburgh Edvard Moser, Norwegian Univ. of Science and Technology Sean Munro, MRC Lab. of Molecular Biology Naoto Nagaosa, Univ. of Tokyo James Nelson, Stanford Univ. School of Med. Timothy W. Nilsen, Case Western Reserve Univ. Helga Nowotny, European Research Advisory Board Stuart H. Orkin, Dana-Farber Cancer Inst. Elinor Ostrom, Indiana Univ. Jonathan T. Overpeck, Univ. of Arizona P. David Pearson, Univ. of California, Berkeley John Pendry, Imperial College Reginald M. Penner, Univ. of California, Irvine Simon Phillpot, Univ. of Florida Philippe Poulin, CNRS Colin Renfrew, Univ. of Cambridge Trevor Robbins, Univ. of Cambridge Barbara A. Romanowicz, Univ. of California, Berkeley Jens Rostrup-Nielsen, Haldor Topsoe

Edward M. Rubin, Lawrence Berkeley National Lab Shimon Sakaguchi, Kyoto Univ. Michael J. Sanderson, Univ. of Arizona Jrgen Sandkhler, Medical Univ. of Vienna David W. Schindler, Univ. of Alberta Paul Schulze-Lefert, Max Planck Inst., Cologne Christine Seidman, Harvard Medical School Terrence J. Sejnowski, The Salk Institute Richard J. Shavelson, Stanford Univ. David Sibley, Washington Univ. Joseph Silk, Univ. of Oxford Montgomery Slatkin, Univ. of California, Berkeley Davor Solter, Inst. of Medical Biology, Singapore Joan Steitz, Yale Univ. Elsbeth Stern, ETH Zrich Yoshiko Takahashi, Nara Inst. of Science and Technology Jurg Tschopp, Univ. of Lausanne Derek van der Kooy, Univ. of Toronto Bert Vogelstein, Johns Hopkins Univ. Ulrich H. von Andrian, Harvard Medical School Bruce D. Walker, Harvard Medical School Christopher A. Walsh, Harvard Medical School David A. Wardle, Swedish Univ. of Agric Sciences Graham Warren, Max F. Perutz Laboratories Colin Watts, Univ. of Dundee Detlef Weigel, Max Planck Inst., Tbingen Jonathan Weissman, Univ. of California, San Francisco Sue Wessler, Univ. of Georgia Ellen D. Williams, Univ. of Maryland Ian A. Wilson, The Scripps Res. Inst. Jerry Workman, Stowers Inst. for Medical Research Xiaoliang Sunney Xie, Harvard Univ. John R. Yates III, The Scripps Res. Inst. Jan Zaanen, Leiden Univ. Huda Zoghbi, Baylor College of Medicine Maria Zuber, MIT

BOOK REVIEW BOARD

John Aldrich, Duke Univ. David Bloom, Harvard Univ. Angela Creager, Princeton Univ. Richard Shweder, Univ. of Chicago Ed Wasserman, DuPont Lewis Wolpert, Univ. College London

16

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

RANDOMSAMPLES
E D I T E D B Y C O N S TA N C E H O L D E N

CEREBRAL CUT-UP AN INTERNET HIT

Before his death in 2008, the famous amnesic patient Henry Molaison (or H.M., as he is known to legions of Psychology 101 students) agreed to donate his brain for further research. Last month, scientists froze the brain and sliced it into 2401 paperthin sections, part of a project to create an open-access digital atlas (Science, 26 June, p. 1634). The 53-hour procedure was streamed live on the Internet. All that slicing turned out to be a major public attraction: Over the 3-day period, the Web site received 400,000 hits. Traffic to Molaisons Wikipedia entry also spiked, from about 400 to more than 40,000 hits daily. And hundreds of people around the world posted comments on Twitter and on the labs Facebook page. We had Twitter fans keeping us company all night, says Jacopo Annese (left), the neuroanatomist at the University of California, San Diego, who led the effort. He has posted a selection of comments (A sample: Live brain slicing! Hard to pull away but must go to 4 year olds birthday party.) on the labs Web site, thebrainobservatory.ucsd.edu.

Name That Author

If youre a casual reader who has trouble telling your Updike from your elbow, a team of physicists may be able to help you. Sebastian Bernhardsson and colleagues at Ume University in Sweden say they can distinguish one author from another by analyzing their writings statistically. They tracked how the number of different words in a sample of text grows 30 with the total number of 20 words in the Hardy sample for Melville 10 three authors: Lawrence Herman 0 0 40 80 120 Melville, D. H. Number of words in text (in tens of thousands) Lawrence, and Thomas Hardy. Each writer has his own distinctive curve describing that increase, the physicists recently reported in the New Journal of Physics. In a short sample, the number of different words increases almost as fast as the total does; it increases more gradually as the sample becomes longer. So the curves start out steep and eventually level out. Melville, who uses the biggest vocabulary, has the steepest curve. Hardy adds new words at a slower rate, followed by Lawrence. If the method works, that would be interesting because its such a simple statistic, says Daniel Rockmore, a mathematician at Dartmouth College. But he says the researchers would have to compare many more authors to prove it. R. Harald Baayen, a quantitative linguist at the University of Alberta in Edmonton, Canada, says others have been working on such statistical methods for decades and that the physicists method may be too simple. Variations among one authors books, he notes, often exceed the variations between authors.
Number of different words
(in thousands)

Myopia Out of Control

Americans are getting ever more nearsighted, according to scientists at the National Eye Institute in Bethesda, Maryland. In the early 1970s, about 25% of the population qualified as myopic. In the early 2000s, that proportion had leapt to almost 42%, says a team led by epidemiologist Susan Vitale. The report, in the Archives of Ophthalmology, created a buzz last week about the dangers of computers and texting. Similar increases have been happening around the world. Oddly, the condition is highly heritable yet

CREDITS (TOP TO BOTTOM): JACOPO ANNESE; DUANE FROESE, UNIVERSITY OF ALBERTA; S. BERNHARSSON ET AL., NEW JOURNAL OF PHYSICS 11 (2009)

malleable by the environment. But what environment? The suspicion has always been centered around near work, says vision scientist Donald Mutti of Ohio State University in Columbus. But recent studies of schoolchildren in both the United States and Singapore have failed to show an association between near work and an increase in myopia. Mutti says research instead is increasingly pointing to lack of outdoor exposure as the culprit. Were kind of a dim indoors people nowadays, he observes. If you ask me, I would say modern society is missing the protective effect of being outdoorsalthough whether its the light or the distance that does it is still not known.

Mammoths Last Stand

Most of North Americas large mammals are thought to have gone extinct some 13,000 to 15,000 years ago. But a new study of ancient DNA suggests that woolly mammoths and horses hung on until 10,000 years ago or later. A multinational team headed by geneticist Eske Willerslev of the University of Copenhagen reached that conclusion by analyzing soil samples for ancient DNA from animals urine and feces. The team picked a site on the banks of the Yukon River in central Alaska, a likely mammoth stomping ground where the permafrost has never melted and the stratigraphy is well dated. They took core samples from seven permafrost layers ranging from about 12,000 to about 7500 years old. The scientists reported last week in the Proceedings of the National Academy of Sciences that they found mitochondrial DNA from mammoth, bison, moose, horse, and snowshoe hare. Mammoth and horse DNA was dated to between 10,500 and 7600 years agopostdating the most recent fossils by at least 3000 years. The SCIENCE VOL 327

results show that some survived in the interior for several thousand years after the arrival of their human predators, the researchers say. Paul Koch, a paleoecologist at the University of California, Santa Cruz, says that obtaining sedimentary ancient DNA is relatively new and that the results put another nail in the coffin of the idea that large North American mammals went extinct suddenly.

www.sciencemag.org

1 JANUARY 2010

17

NEWS>>
U.S. SPACE POLICY

Turmoil at Australian synchrotron

Cuts in U.K. physical science

20

22
President George W. Bush proposed in 2004, this White House is more intrigued by human missions to asteroids, Phobos, and Deimos as a precursor to landing humans on the Red Planet. That option was given particular prominence by Augustine panel members when they testified this fall before congressional committees. To prepare for human visits, NASA may order additional robotic missions to the martian moons and asteroids. Before making his decision, Obama reviewed several options presented to him by NASA, the Office of Management and Budget, and the Office of Science and Technology Policy. The choices included keeping NASAs budget flat and delaying a new launcher, giving it an additional $1 billion to build a heavy-lift launcher, ramping up NASAs annual budget by $3 billion for an aggressive program, or reducing NASAs budget and abandoning space flight. The presidents decision to go with the second option is a major departure from his 2010 budget plan, which called for NASA to receive a 5% increase in 2010 followed by level funding through 2014. The Augustine panel concluded that NASA needed a $3 billion annual increase to move ahead with a robust spaceflight program. Last month, Congress belatedly completed action on NASAs 2010 budget, boosting it by $1 billion, to $18.7 billion. An additional $1 billion in 2011, combined with support from other countries, would put the agency close to the panels suggested level. There are a lot of different ways to reach that level, including help from abroad and increasing NASAs efficiency, Augustine added. Its not clear when the new policy will be formally announced. One White House source said it was imminent, while another hinted that it would wait until Obamas State of the Union address in late January. Another possibility is a 1 February release as part of the presidents 2011 budget request to Congress. Given the White Houses preoccupation with health care and climate change, however, NASA officials and their industry backers see the new policy as welcome proof that Obama also cares about space flight.
ANDREW LAWLER

Obama Backs New Launcher And Bigger NASA Budget

President Barack Obama plans to ask Congress to cancel work on a new rocket and instead fund a heavy-lift launcher to take humans to the moon, asteroids, and the moons of Mars. The president outlined the new direction for the U.S. human space flight program on 16 December at a meeting in the White House with NASA Administrator Charles Bolden, according to officials familiar with the discussion. NASA would see its 2011 budget grow by $1 billion both to get the new launcher on track and to bolster the agencys fleet of robotic Earth-monitoring spacecraft. The current NASA plan for human exploration is built around the $3.5 billion Constellation program, which was intended to provide a way to get humans to the space station and beyond after the space shuttle is retired this year. But its initial launcher, Ares 1, has faced a string of cost and technical problems, and an outside panel chaired by retired aerospace executive Norman Augustine was skeptical of that effort (Science, 25 September, p. 1606). Although NASA has done a good job confronting the rockets engineering challenges, says Augustine, the schedule has slipped so badly it doesnt fit into the future program well. But some lawmakers, such as Senator Richard Shelby (RAL), are sure to fight any changes to the program. According to knowledgeable sources, the White House has decided that scarce NASA funds would be better spent on a simpler heavy-lift vehicle that could be ready to fly as early as 2018. Meanwhile, Europe, Japan, and Canada would be asked to work on a lunar lander and modules for a moon base, a contribution that would save the United States several billion dollars. And commercial companies would take over the job of getting supplies and possibly humans to the international space station. The decision is not going to make anyone gasp, said one source in the White House, which hopes to ease congressional concerns about the impact of the new plan on existing aerospace jobs by transitioning workers from Ares 1 to the heavy-lift vehicle

Change in direction. President Obama has told NASA Administrator Charles Bolden (inset) that he wants to replace Ares 1 (above) with a heavy-lift rocket.

project. But Shelby and some of his colleagues fear that an Ares 1 cancellation will lead to mass layoffs in their states. Indeed, Shelby inserted language into the 2010 NASA spending bill that requires the agency to gain congressional approval before changing the existing rocket program. Last month, Shelby also wrote to NASAs inspector general asking his office to investigate alleged conflicts of interest on the Augustine panel. The legislator said that several panel members were registered lobbyists who took direct advantage of their temporary roles on the Commission to further their personal business. None of the panel members was actually a registered lobbyist, although Augustine says, Id be surprised if lobbyists had not provided the panel with their input. The form that the heavy-lift launcher would take has yet to be decided. But rather than pointing the rocket to the moon, as U.S.
VOL 327 SCIENCE

18

1 JANUARY 2010

www.sciencemag.org

CREDITS: NASA/SANDRA JOSEPH AND KEVIN O'CONNEL; (INSET) NASA/BILL INGALLS

Stem cell research in regulatory limbo in Ireland

The making of a Russian cosmologist

25
CLIMATE CHANGE

26

Senate Looms as Bigger Hurdle After Copenhagen

President Barack Obama knew the clock was ticking on reaching a climate-change agreement when he barged into a meeting with the leaders of China, Brazil, India, and South Africa on the last day of the deadlocked U.N. convention in Copenhagen. And while Obama called the 18 December deal he brokered an unprecedented breakthrough in getting developing countries to agree for the first time to voluntary reductions in greenhouse-gas emissions, he knew that it was missing major components that supporters had sought, ranging from mandatory mitigation targets to deforestation. The so-called Copenhagen Accord (see chart) didnt even contain a timetable for deciding when to decide. But as hard as it was to negotiate a deal in Denmark, Obama faces an even tougher time in Washington trying to convince the U.S. Senate to pass cap-and-trade legislation. And hes up against another ticking clock: the November 2010 elections that will decide whether his party retains control of Congress. The closer you get to the election around here, the less policy is going on and the more politics, says Senator Lisa Murkowski (RAK). Even so, Murkowski predicts that the Senate will need months, not weeks to debate the complex issue. One major obstacle is the Senates own rules. A supermajority is needed to pass nearly any controversial legislation, because opponents can launch a filibuster that requires 60 votes to bring to an end. Supporters can count on only about half that number, and getting the rest will require deals with some combination of coal-state Democrats, moderate Republicans like Murkowski, and liberals who want a stronger bill. And that means balancing concerns about trade, energy, and the economic effects of the bill. Its going to be a tough environment to find 60 votes, says Thomas Gibson of the American Iron and Steel Institute. For advocates of emissions reductions, the lessthan-satisfying outcome Dealmakers. President Obama engaged fellow world leaders on the last of the Copenhagen sum- day of the Copenhagen meeting. mit shows how difficult it will be to reach any agreement. In June, the and Public Works Committee has already House of Representatives narrowly passed attracted plenty of ill will. Republicans on landmark emissions reductions that would the committee boycotted the vote, and Boxer, bring U.S. emissions 20% below 2005 levels the chair, bucked precedent by proceeding by 2020. And such legislation hasnt fared without them. And hers is the only one of six well in the Senate. In 2003, cap-and-trade leg- committees with jurisdiction to have passed islation went down by a margin of 43 votes to the legislation. 55. Two years later, it secured only 38 votes. In Democrats from states where coal is king 2008, a procedural vote on a bill that con- will be central to any successful deal. Last tained more favorable language for nuclear month, 92-year-old Senator Robert Byrd power fell 12 votes short of the 60 required to (DWV), their unofficial leader, shocked the bring it to a votealthough six senators said coal industry when he declared that West Virthey would have voted for the bill had they ginians can choose to anticipate change and been present. adapt to it, or resist and be overrun by it. He This time around, legislation introduced also said he plans to stay at the table to make in October by Senators John Kerry (DMA) sure his state is part of any solution, increasand Barbara Boxer (DCA) and approved the ing the chances that the Senate bill would profollowing month by the Senate Environment vide more money for clean-coal projects

CREDITS (TOP TO BOTTOM): PETE SOUZA/WHITE HOUSE/SIPA PRESS/NEWSCOM; (SOURCE:) U.N. FRAMEWORK CONVENTION ON CLIMATE CHANGE AND SCIENCE

TWO WEEKS IN DENMARK

ISSUE Reducing Emissions Verification WHAT HAPPENED IN COPENHAGEN Agreed to hold temperature rise to 2C by 2100; each country will report this month on plans for voluntary cuts between 2010 and 2020. Agreed to have developed nations report their emissions cuts, subject to international verification. Developing countries will draw upon international consultations in reporting their performance. WHAT REMAINS UNRESOLVED Agreeing on an emissions goal for 2050 and legally binding cuts by both developing and developed nations. Agreeing on a mandatory process to check numbers from developing nations, with consequences if the numbers are wrong.

Financing

Developed nations agreed to provide roughly $30 billion between 2010 and Agreeing on who will pay what, when, and how the money will be distributed. 2012 for mitigation and adaptation, rising to $100 billion a year by 2020. Agreeing on the scalenational or regionalto be used in counting efforts to avoid deforestation.

Deforestation The financing would include $3.5 billion for developed nations to pay tropical countries to slow the destruction of forests.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

19

NEWS OF THE WEEK

and assistance to industry than the House version. Senator Tom Carper (DDE), who led a task force on coal for Boxer earlier this year, says he can already see momentum building. If the House members from a coal state voted for the House bill, its going to be hard for some senators to ignore what their House members have done, says Carper. Bridging the gap between coal-state lawmakers and environmentalists is not the only difficult task facing the White House. Getting a handful of Republicans to support the bill is another crucial one. One key step in that direction is an outline of cap-and-trade legislation released last month by senators Kerry, Joe Lieberman (IDCT), and Lindsay Graham (RSC). Last month, Graham said the legislation, if passed, would create jobs, protect our national security interests, and improve our environment.
RESEARCH MANAGEMENT

Winning Grahams vote James Inhofe (ROK), an outmight require more support for spoken climate skeptic, believes the nuclear industry than is in the sciencemag.org the scandal spells doom for cliFor interviews and House legislation, including mate legislation that hes already analyses, and to advanced research, regulatory labeled dead on arrival. But two changes, and some type of finan- learn where individual other Republicanssenators senators stand on cial incentives for new nuclear climate bills, go to Murkowski and Lamar Alexanpower plants. Grahams role is http://scienceinsider.org. der (RTN)told Science that the greatest indication that you the e-mails have not shaken can reach a 60-vote margin, says Lou their conf idence in the science behind Leonard of the World Wildlife Fund in Wash- global war ming. Indeed, Boxer told ington, D.C. On the other hand, such conces- reporters at a press conference this week sions could cause a mutiny among staunch that the new push by foes to challenge the liberals like Senator Bernie Sanders (IVT), fundamental facts of global warming will who says he would consider voting against backfire because it will put new attention the bill depending on how weak it is. on the scienceand thats good. One unknown may be fallout from the Its not clear how the watered-down leaked climate science e-mails (Science, Copenhagen pact might affect the Senate 4 December, p. 1329), which had virtually no vote. Before the summit, Boxer said that a effect on negotiations in Copenhagen. Senator political agreement there would help us

Online

Protests by Staff, Advisers Rattle Australian Synchrotron

and its chair were cordialright Amid media fanfare and the up to the day he was fired. Lamb smiles of politicians, the Aussays that Walter and Rod Hill, tralian Synchrotron (AS) here board deputy chair and pro viceunveiled a new tunnel in April chancellor for industry engagethat will eventually house a ment and commercialisation at medical beamline. Leaders Melbournes Monash University, said it would push the boundmet in Lambs office: I thought, aries of synchrotron science, amongst other things, I was going allowing clinicians to image a to get a pat on the back. Instead, single cancer cell in a womans they told me I was gone. The breast or plaques in the artery whole thing lasted about 15 minwall of a beating heart. In June, utes. He claims that he has the $330 million synchrotron received no official explanation won funding for a new science as to why. center and a hotel complex. In Happier days. Chemist Robert Lamb, former director of the Australian SynchroInitially, the board said little. September, it hosted a major tron, before his dismissal. But as concerns mounted among conference. Four months later, the staff and SAC, Walter, Hill, the entire facility was in turmoil. Director trons around a ring the diameter of a football and a third board member, Sean Gallagher Robert Lamb had been fired; a majority of the field, emitting intense x-rays that are filtered director of the University of Sydneys international scientific advisory committee through beamlines for customized applica- United States Studies Centremet and (SAC) had resigned; and staff began a work- tions. This third-generation AS was com- spoke with the staff on 9 December. to-rule slowdown, demanding that the board pleted in 2007; it has been a smashing suc- According to a text obtained by Science, chair, attorney Catherine Walter, be removed. cess, says Jeff Corbett, a synchrotron scien- Walter said that legal and confidentiality What went wrong? The crisis erupted tist from the SLAC National Accelerator constraints prevented discussion of Lambs shortly after the AS board removed Lamb Laboratory in Menlo Park, California, and an firing. She spoke of the boards deep confrom his job on 30 October. Lamb and board AS adviser. Last year, the AS ranked second cern about progress with the science and members have refused to discuss details. But in the world in reliability, with a beam avail- investment cases, which are crucial to many observers say that a gaping rift devel- ability of 98%. Nine beamlines have been obtaining funding for further development oped between the board of directors on one completed or are under way, but there are 29 and expansion after 2012. An audit by the hand and AS staff and SAC on the other over vacancies and no new lines since 2007. Victorian government last year praised AS the slow pace of developing long-term plans. In a recent interview, Lamb, chair of science but raised a red flag over the lack of The AS is one of 17 facilities of its kind in chemistry at the University of Melbourne, long-term science and business plans, the world. It accelerates a thin beam of elec- said his relations with the board of directors according to one insider. In an e-mail
MELBOURNE, AUSTRALIA

20

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

CREDIT: COURTESY OF ROBERT LAMB

NEWS OF THE WEEK

move a climate-change bill forward. But a central issue left unresolved in Denmark is Chinas refusal to agree to international verification of its reports on emissions reductions. Obama won the support of the China delegation for compromise language that embraces international consultations. But without a firmer Chinese commitment, it may be hard to convince leery lawmakers that China and other major emitters from developing nations are willing to carry their share of the burden. Theres nothing [in the text] to keep these countries from accelerating their emissions, says an Inhofe spokesperson, who says the talks were a complete disaster in terms of the pending legislation. The spokesperson also criticized the U.S. decision, announced at the talks, to participate in a $100-billion-ayear fund, starting in 2020, for helping the developing world cope with global warming: It probably made things worse, since the American people arent going to be happy seeing their tax dollars going abroad to support a global warming fund. Obama will also have to decide how much political capital to spend on an issue that is losing support among Americans, according to opinion polls, when one they care a great deal aboutunemploymentis at 10% and could get worse. Some critics note that Obama didnt say in Copenhagen that climate change would be a top priority in 2010. But Daniel Weiss of the liberal Center for American Progress in Washington, D.C., says that Obamas meetings with world leaders at the conference and with U.S. politicians at home show he is fully engaged. Hes succeeded in Copenhagen at pushing the process forward. Now he has to succeed at getting the Senate to ELI KINTISCH act, says Weiss.

ScienceNOW.org
From Sciences Online Daily News Site
Hand SizeNot Sex Determines Sense of Touch Sometimes a difference between the sexes is not based on sex at all. Women have a finer sense of touch than men do, but a new study shows that this is simply because their fingertips tend to be smaller. http://bit.ly/handsize Chimps Keep Their Cool With Fire When primatologist Jill Pruetz found herself threatened by wildfires in the savannas of Fongoli, Senegal, in 2006 she had two options: stay with the chimpanzees she was studying, or run. She chose the chimps. The primates were calm, andwith her in towthey carefully made their way around the blaze. The chimps actions, Pruetz would later report, set them apart from other nonhuman animalsand they may reveal the evolutionary origins of how we came to master fire. http://bit.ly/chimpfire Seasick? Try Controlling Your Breathing If you get seasick easily, you may prepare for boat rides with pressure-point bracelets, ginger, or a prescription skin patch. Now theres one more remedy: timing your breathing to counteract the nauseating motion. The technique presumably works because it helps control gravity sensors in the abdomena lesserknown input to our fine-tuned balance system. http://bit.ly/breathingtiming Bacteria Can Transform Minerals Electrically Got a messy cleanup problem that requires a molecule-by-molecule fix? Instead of nanotech, how about deploying an array of ready-made, versatile bacteria? Scientists studying a genus of the rock-dwelling bacteria called Shewanella have found out how the organisms can transform minerals by zapping them with tiny electrical currents. The discovery could lead to new types of fuel cells to generate electricity, to better environmental-cleanup techniques, and possibly even to a new generation of organically made materials. http://bit.ly/Shewanella Read the full postings, comments, and more on sciencenow.sciencemag.org.

response, Walter said that the review high- gan Medical School in Ann Arbor and a lighted a priority that was also a concern of continuing SAC member. the board, but she denied that it prompted Walter contests the view that the board the scapegoating of Lamb. did not act on SACs recommendations. In One of the boards critics, Frank an e-mail she wrote, All recommendations Larkins, chief scientist for energy of the from SAC were passed on to the AS manVictoria State Government and SAC chair agement. And Gallagher said in a teleuntil his resignation on 9 December, says a phone interview that paperwork is an clash of cultures separated inevitable result of the the board from the scientific I thought I was organizations complex staff and SAC. Our consisstructure. Stakeholders going to get a pat tent problem has been the include five state governsilence in response to our on the back. ments, the federal govern[SAC] recommendations to Instead, they told ment, New Zealand, 32 the board to get moving on universities, and 42 other long-term science and busi- me I was gone. organizations. ness plans, says Larkins. He Several SAC members ROBERT LAMB, has urged Victoria to interAUSTRALIAN have spoken about their conSYNCHROTRON cern for the synchrotrons vene and replace the chair of the board. future. Soichi Wakatsuki, a Michael Grunze, a professor of applied continuing member and director of the physical chemistry at the University of Photon Factory in Tsukuba, Japan, wrote Heidelberg in Germany, another SAC in a 12 December letter to the board that member who resigned on 9 December, also being a member of SAC had been the most blasted the board in a resignation letter. distressing period in my career, and that the The SACs advice, Grunze wrote, has status of AS as a world class synchrotron is been either consistently ignored or not now in danger. Ted Baker, a University of acted upon in a timely manner by the lead- Auckland professor of structural biology and ership of the AS Board. the new SAC chair, also warned the board, if Some critics of the board say that AS the level of disclosure and trust are not management started long-term planning at improved, radically, and fast, the AS will lose the beginning of the year but were held key staff and its use base will evaporate. back. They accuse the board of micromanThe staff slowdownwhich limits work agement and choking operations with to a standard 9-to-5 dayis ongoing at paperwork. The board is much more press time. The State Government of Victoinvolved in the administration than is nec- ria, which has the largest stake in the projessary or good, said Janet Smith, a protein ect, so far has declined to intervene. ELIZABETH FINKEL crystallographer at the University of Michi-

CREDIT: PHOTOS.COM

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

21

NEWS OF THE WEEK

RESEARCH FUNDING Darkening skies. STFC threatened to cut U.K. funding to the Gemini telescopes in 2007. Money will now be withdrawn in 2012.

U.K. Physicists Cry Foul At Major Budget Cuts

Its not been a festive time for many U.K. physicists following the mid-December announcement of a 5-year funding plan for the Science and Technology Facilities Council (STFC), the British body responsible for particle physics, astronomy, nuclear physics, and space science. Administrative changes and the ravages of the economy have left the council with a gaping hole in its finances, so researchers braced for cuts and even offered advice on how to make them. Although the STFC tried to spread the losses fairly, nuclear physics was pared to the bone, with just 30 million for the next 5 years. This amounts to a cut of 29% over that period, according to nuclear physicist William Gelletly of the University of Surrey. If the U.K. is serious about nuclear new build and maintaining high standards in nuclear medicine, then it should take a very hard look at how much it should be spending on nuclear physics, he says. Although astronomy was cut less severely, by 10%, the U.K. government will
SCIENTIFIC PUBLISHING

withdraw support from a number of projects, including the Pier re Auger Cosmic Ray Observatory, the Liverpool Telescope, and the U.K. Infra-Red Telescope. And in 2012 it will withdraw from the Gemini Telescopes, the James Clerk Maxwell Telescope, and the Isaac Newton Group of Telescopes. After that, U.K. astronomers will no longer have access to a major observatory in the Northern Hemisphere, their own sky. We are now seriously concerned at the effect the loss of so many smaller projects will have on the health and morale of physics groups in British universities, says Andrew Fabian, president of the Royal Astronomical Society. The STFCs problems date back to 2007, when it was created from the merger of two existing research councils, and nuclear physics moved into it from a third (Science,

Science field Astronomy Nuclear physics Particle physics Space science

Total 5-year budget Cut 10% 267 million 29% 30 million 4% 690 million 6% 639 million

21 December 2007, p. 1851). Both of the merged councils had overcommitted themselves to costly international projects, and STFC soon found a 40 million hole in its budget. The situation has since gotten worse as the weakening pound makes subscriptions to international facilities more expensive. And with the poor economy, the STFC expects to get no budget increase in the

Errors in Chemistry Claims Cast Doubt on Reactome Paper

A newly developed research tool called a reactome array, which has attracted widespread interest from biologists, has come under intense f ire from scientists who say the description of the device in the 9 October issue of Science (p. 252) includes impossible chemical reactions and makes little sense. The publication drew immediate attention because the array promises to establish the functions of a myriad of enzymes and probe the metabolism of bacteria and other kinds of cells. Last week, Science acknowledged the furor, publishing online an Editorial Expression of Concern in which the journals editorin-chief, Bruce Alberts, notes that serious questions have been raised about the methods and data presented. It was just so obvious the chemistry was flawed, says biochemist Laura Kiessling of the University of Wisconsin, Madison, editor-in-chief of ACS Chemical Biology. While admitting to serious errors in describing the methods used to create the reactome arraywhich required the synthesis of thousands of compounds representing metabolites and other enzyme substrates, linking them to a fluorescent dye, and fixing them on a glass slidethe studys corresponding authors stand behind the device. Were confident in the results and technology. Many researchers have used the array without problems, says Manuel Ferrer of the Spanish National Research Councils (CSICs) Institute of Catalysis in Madrid. According to Ferrer, at the request of Science, CSIC will conduct an investigation of his lab, and related inquiries will be held at collaborators labs in Germany and the United Kingdom. He says that Nobel laureate Richard Roberts of New England Biolabs and other researchers have also agreed to conduct blinded tests to verify that the reactome array works. Roberts confirmed that fact, noting that he recently visited the Madrid lab and came away impressed after initially thinking the reactome array was too good to be true. In private chats and online postings, chemists began expressing skepticism about the reactome array as soon as the article describing it was published, noting several significant errors in an initial figure. Some also questioned how a relatively unknown group could have synthesized so many complex compounds. The dismay grew when supplementary online material providing further information on the synthesized compounds wasnt available as soon as promised. We failed to put it in on time. The data is quite voluminous, says co-corresponding author Peter Golyshin of Bangor University in Wales, a microbiologist whose team provided bacterial samples analyzed by Ferrers lab. Science is also coming under fire. It was stunning no reviewer caught [the errors], says Kiessling. Ferrer says the papers peer reviewers did not raise major questions about the chemical synthesis methods described; the journals executive editor, Monica Bradford, acknowledged that none of the papers primary reviewers was a synthetic organic chemist. We do not have evidence of fraud or fabrication. We do have

STFC

22

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

CREDIT: (PHOTO) GEMINI OBSERVATORY PUBLIC INFORMATION & OUTREACH

NEWS OF THE WEEK

United Kingdoms next 3-yearly comprehensive spending review. Stung by criticism that its 2007 cuts involved no consultation, the STFC carried out an extensive prioritization exercise with the research community before announcing its latest funding plan. For 5 years from fiscal year 201112, the STFC will save 115 million by withdrawing from 24 major national and international projects and squeezing the budgets of 38 others. It will also require a 10% cut in grants to university researchers involved in those projects and a 25% cut in new postgraduate studentships and postdoctoral fellowships. Particle physics, now mostly focused on exploiting CERNs new Large Hadron Collider, got the largest slice of the 5-year pie, 690 million, and had the smallest cut, 4%. In contrast, nuclear physicists, as STFCs newcomers, feel theyve been sidelined by the much bigger communities already in place. The cuts in nuclear physics will be devastating for the field. The number of academic nuclear physics researchers will drop by at least a factor of 2 from [currently around] 55, says Paul Nolan of the University of Liverpool. At this level, they will not be able to make an international impact. The STFC says that the decisions were made by peer-review panels, reflecting the quality of the science. But some nuclear physicists argue that it is hard for such panels to be fair when some fields are poorly represented. When the going gets rough, most people will tend to mark up the things they know and love. So it is no surprise that the big battalions, who make up 90% of the membership of these committees, fared better, says Gelletly. Others point out that at the end of the 5 years, the STFC has chosen to give its in-house facilitiesthe DIAMOND x-ray source and the ISIS neutron sourcea 25 million increase, while cutting funds to external centers. There is a major conflict of interest here, says Nolan. Researchers are now talking about whether the STFC should be reorganized, separating facility funding from research grants, and whether it should oversee studentships and fellowships. These are generally considered essential for the health of the science, and cuts make little sense, says Robert Cywinski of the University of Huddersfield. Perhaps it is time that the whole concept of funding research studentships across all research councils should be reviewed and possibly taken out of the hands of the individual councils.
DANIEL CLERY

From the Science Policy Blog

Merck has hired the former director of the U.S. Centers for Disease Control and Prevention, Julie Gerberding, as its head of vaccine development. The company says the infectious-disease expert will expand the companys vaccine offerings internationally. http://bit.ly/4R7nOL Henrik Thomsen, a Danish clinician who alerted patients and regulators to potential risks of Omniscan, a drug used to improve MRI scans, is facing a libel suit. Manufacturer GE Healthcare says a 2007 presentation by Thomsen was defamatory. http://bit.ly/4Mb1qp The National Football League has changed its stance on the issue of concussions and long-term brain-damage risk. Doctors and experts working for the league had dismissed the link before, but now the NFL is exploring a partnership with scientists at Boston University who have been among its harshest critics. The partnership could include funding for the schools research. http://bit.ly/6cps5f A new report details alleged errors associated with the Propatria study, a research trial into the use of probiotic therapies in which live microorganisms are administered to patients to treat disease. The Dutch Health Care Inspectorate found that scientific rules were not followed in the study, in which 24 patients receiving the treatment died. By comparison, there were nine deaths among those receiving the placebo. http://bit.ly/4NbieG In an interview with ScienceInsider, retiring science committee chair Representative Bart Gordon (DTN) said that science education was among his top priorities for his last year in office. Hell be tackling that as part of a reauthorization of the America COMPETES Act. One of the things weve been doing is inventorying all the current STEM education programs, and were finding tens of millions of additional dollars being spent on programs that nobody knew about, said the lawmaker. http://bit.ly/5Bz1sw For more science news and analysis, visit http://scienceinsider.org

CREDIT: A. BELOQUI ET AL., SCIENCE 326, 5950 (9 OCTOBER 2009)

Figured out. After spotting errors in this figure and being unsatisfied by supplementary online material, chemists challenged whether the so-called reactome array could work as claimed.

concerns about the inconsistencies and have asked the authors institutions to try to sort all of this out by examining the original data and lab notes, she says. Ferrer says he takes responsibility for all the mistakes and apologizes: I understand the disappointment of Science and Sciences readers. Yet some chemists, including those who have sought clarifications from the authors, remain unconvinced

by supplementary data that has since been posted and the explanations offered so far. Many colleagues think it must be fraud. Im trying to keep an open mind, says chemical biologist Ben Davis of the University of Oxford in the United Kingdom, who wrote a skeptical review of the reactome array article on the Faculty of 1000 Web site. But clearly there are a lot of unexJOHN TRAVIS plained elements.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

23

NEWS OF THE WEEK

HIGHER EDUCATION

Recession Hits Some Sciences Hard At Florida State University

Philip Froelich, 63, is the tenured Francis Eppes Professor of Oceanography at Florida State University (FSU) in Tallahassee. He wont be much longer. Despite a distinguished 31-year career as a researcher and administrator, he will be laid off next May. And despite a positive external evaluation within the past year, his departmentmuch diminished by layoffswill be no more, folded along with the geological sciences department into a new department dominated by meteorology. Why would you cut Flip Froelich? It doesnt make any sense, says geologist Michael Perfit of the University of Florida (UF), Gainesville. Its all about money, of course. When the cash-strapped Florida state legislature recently slashed funds for higher education for the third straight year, big across-theboard cuts spilled down through individual state university budgets. But at FSUone of the big four Florida state schoolsthe fiscal crisis has turned into a ravaging torrent for a few departments, most of them in the sciences. In the end, unlike at other universities, FSU administrators balanced their budget by firing many faculty members, including many tenured professors like Froelich. That decimated the geological sciences, oceanography, and anthropology departments. The layoffs at FSU have truly devastated faculty morale across the campus, says anthropologist Cheryl Ward of Coastal Carolina University in Conway, South Carolina, who left FSU before the cuts. They caused lasting harm to science programs. But to the FSU administration, slashing small departments that were far from supporting themselves was the only way to avoid permanently undermining education across the university. The deep, targeted cuts were unfortunately the best option, says FSU Provost Lawrence Abele. I dont believe in cutting across the board; that weakens everything, he says. Theres no disagreement on the 39,000student FSU campus that the budget situation has gone from serious to dire. The big, tax-generating housing bubble burst early in Florida, which has no state income tax to cushion the loss. The Florida legislature cut the state university systems annual $380 million budget by $82 million over

year of budget cutting, the subsidy gets harder to find, says Travis. Targeted departments and programs within departments took heavy hits. Geological Sciences and Oceanography will be merged with Meteorology at the end of this 2007 to 2010$55 million of it in this aca- academic year to form a new department of demic year. That budget, plus tuition, pays Earth, Ocean, and Atmospheric Sciences. all salaries in the state system. And at Froelich thinks the merger is a reasonable slightly over $3100 last year (up 15% this idea but says the university should have year), tuition at Florida state universities done it 2 to 5 years ago, when more favorwas the lowest in the nation, notes Joseph able economic conditions would have given Travis, FSUs dean of the College of Arts any merged department better prospects. and Sciences. Were trying to run a Major Anthropology barely survived elimination League Baseball operation on a AAA or but will be diminished and restructured. AA [minor league] budget, he says. The numerous and often focused faculty firings have been much more controCUTS IN FLORIDA STATE UNIVERSITY SYSTEM versial. Of the faculty Tenured and tenure-track laid off from 2007 to Faculty laid off (103) faculty laid off (43) 2010 in the 10,700Others USF faculty Florida State USF University System, UF 60% were laid off from UF FSU FSUs approximately TENURE-TRACK 1750 faculty alone, accordFSU UCF ing to data collected by FSU FSU TENURED faculty members and provided by Froelich. Approximately 43 tenured or tenureFSU - Florida State University UF - University of Florida track faculty were laid off UCF - University of Central Florida USF - University of South Florida across the system. But Heavy FSU layoffs. The majority of faculty firings in the Florida State according to FSU English University System came at Florida State University (left), where about professor and Faculty Senate 80% of tenure-line layoffs occurred (right). President Eric Walker, 35 of those 43 tenure-line facAcross-the-board cuts could cripple ulty were lost at FSU. the institutional missions, starving everyOf about 21 tenured faculty let go across body, says Travis, so you do elaborate the system, all or all but one were at FSU. cost-benef it analyses. Which of the pro- The College of Education was hit hard, but grams are the weakest? FSU President T. K. 10 of that 21 came out of the College of Arts Wetherell made the criteria for judging the and Sciences, all of them scientists in the strength of a department or program relatively small departments of geological explicit: student credit hours generated, sciences, oceanography, and anthropology. degrees awarded, contract and grant expen- Oceanography, at least, had just this spring ditures, and tuition collected, all on a per- received a glowing evaluation from the faculty as well as an absolute basis. university that included an external The state legislature funds FSU based reviewer, according to Froelich. on enrollment, Travis notes. The departElsewhere in the Florida state system, ments of geological sciences, meteorology, faculty fared better. At UF, we did not cut a and oceanography came in at the bottom of lot of existing people, says Provost Joseph 15 Arts and Sciences departments with Glover. We did cut a lot of vacant and newly about 6000 student credit hours per year vacant positions. And weve spread [the each, according to Travis. Anthropology was cuts] over a period of time, [so] this year we fourth from the bottom with 11,000 hours; only have left a small amount to do. The English, for example, had 50,000. Sciences geological sciences department was on the never pay for themselves, says Travis. block for a while, notes Perfit, its chair. The Theres always a subsidy arrangement in problem, as he sees it, was that we were just which larger departments in effect help sup- small. Some of the best [geoscience] schools port smaller ones. But in the third straight in the nation are being cut just because
VOL 327 SCIENCE www.sciencemag.org

24

1 JANUARY 2010

SOURCE: P. FROELICH/FSU; E. WALKER/FSU

NEWS OF THE WEEK

theyre small and would therefore seem to create only small losses to the university. His department survived, though for now the university is using stimulus money to pay everyone in the department. FSU was the only university in the system to lean so heavily on faculty layoffs, but FSUs Travis still sees no way around that.
IRELAND

As to Oceanography in particular, the discussion was never about value of research, he writes in an e-mail. The discussion was always about whether we could continue to afford to subsidize a department generating so few undergraduate credit hours and relatively few graduate degrees. And in 40 years the legislature has never

come back and restored funds it cut from the budget, he says. Other Florida universities like UF may be using stimulus money to avoid extensive faculty firings until budget cuts are restored or tuition increases accumulate, he says, but we at FSU chose not to take this kind of chance.
RICHARD A. KERR

Embryo Ruling Keeps Stem Cell Research Legal

A ruling from the Irish Supreme Court has reignited that countrys debate over the legal status of human embryos, confirming the legality of research with human embryonic stem cells (hESCs) but leaving such work in a regulatory limbo that may not be resolved soon. On 15 December, the court ruled that human embryos outside the womb are not unborn and therefore are not protected under the countrys constitution. The case before the court, in which a woman wanted to implant frozen embryos against the wishes of her estranged husband, does not directly involve stem cell research, but an opposite ruling could have made such work unconstitutional. Its not a green light for hESC research, says Siobhn OSullivan, director of the Irish Council for Bioethics. However, the ruling means that certainly hES cell research is not banned in Ireland. Ireland, a largely Catholic country that has experienced a growth in biomedical research over the past decade, has no laws governing human embryos outside the womb. Abortion is illegal, but assisted reproduction and research with hESCs, which are derived from lab-grown embryos, are both unregulated. Scientists have been uncertain, however, whether the Irish Constitution, which acknowledges the right to life of the unborn, prohibits derivation of hESCs or even work with ones derived elsewhere. Public funding agencies have been similarly perplexed about whether they could fund hESC research. No such work has so far been funded. OSullivan says that because of the legal vacuum, it is difficult to say if any hESC research is going on in Ireland. If you were using them, you wouldnt want to publicize the fact, she says. But several scientists in Ireland have said they would like to work with the cells, and Science Foundation Ireland acknowledges that a few, whose names have not been revealed, have applied for public funding to do so. Frank Barry, head of the Regenerative Medicine Institute at the National University of Ireland, Galway, says he has not submitted a grant involving hESCs, but he would welcome the chance to work with the cells. The ruling is a major milestone for Ireland in terms of research with human ES cells, he says. However, I think it also puts Ireland in a place where it does not want to be, where hESC research is both legal and unregulated. In 2008, University College Cork (UCC) became the first major university in Ireland to explicitly allow work with hESCs. The univerSo far, other universities have not followed UCCs lead. The Irish legislature, the Oireachtas, has also avoided dealing with the controversial issue, although ethical and law experts as well as scientists have been urging lawmakers to take action for years. A government Commission on Assisted Human Reproduction in 2005 recommended legislation to regulate fertility treatments. And in 2008, OSullivans Council for Bioethics said in a paper that the governments failure to act undermines the moral value of the embryo, and it recommended that a national regulatory body oversee embryo research. That could be the councils final word on the subject. On 16 December, the government announced that the body would no longer be fundeda victim of severe budget cuts in Ireland. There is now a vacuum in terms of ethical oversight, Barry says. But OSullivan says that the expertise that went into the councils report will still be available to politicians as they wrestle with the issue. In the Supreme Court ruling, the judges urged the countrys lawmakers to address the legal status of in vitro human embryos, calling Irelands lack of regulation of fertility treatments disturbing and undesirable. One judge cited hESC research as an example of how science has gotten ahead of the law. Health Minister Mary Harney responded by promising that the government would propose legislation in 2010 to regulate assisted reproduction, but she failed to reference hESC research. OSullivan says speedy action in any case is unlikely, given that a lengthy public consultation period will be necessary. Id be very surprised if we have regulation in 2010, she says. But OSullivan hopes the process moves forward as quickly as possible. The fact that the government has not yet regulated this area is absolutely incredible and very unfortunate indeed, she says. It doesnt actually matter which side of the debate you sit on; what we have right now is GRETCHEN VOGEL cowboy territory.

CREDITS (TOP TO BOTTOM): WIKIMEDIA; ADAM DINAN

Controversial move. University College Cork (above) issued regulations on research with hESCs, prompting a billboard campaign by opponents of such work.

sitys governors voted 16 to 15 to require scientists who want to work with hESCs to get approval from the universitys ethics board. UCC also established a subcommittee to examine the proposed source of the cells, the goals of the intended research, and the applicants expertise in the relevant fields. The move sparked widespread debate in Ireland, and no UCC researcher has publicly acknowledged seeking approval for hESC work.
SCIENCE VOL 327

www.sciencemag.org

1 JANUARY 2010

25

NEWSFOCUS
Cosmic quest. Sunyaevs contributions have revolutionized cosmology.

Robert Kirshner, an astronomer at Harvard University. Sunyaevs scientific journey from Tashkent to the astrophysical hall of fame is one that most researchers in the West would find difficult to imagine. Although he wasnt involved in military projects, Sunyaevs early career as a researcher was constrained by the culture of secrecy and paranoia that prevailed in the Soviet Union during the Cold War. Like most of his contemporaries, he was practically forbidden from traveling outside the U.S.S.R. and didnt have access to English-language journals until a year after they were published. In compensation, he had the mentorship of Zeldovich, regarded as one of the most brilliant physicists of the 20th century. A father figure to his juniors, Zeldovich inspired Sunyaev with his extraordinary analytical abilities and a dose of tough love, at times refusing to acknowledge him in the hallways if his protg had no new results to report. I was so lucky to meet him in my life, Sunyaev says of the legend, who died in 1987. The other good fortune for Sunyaevand for astrophysicswas that the Iron Curtain did not block scientific exchanges entirely. Sunyaevs contact with a visiting scholar from Toiling behind the Iron Curtain under a tough mentor, a Russian astrophysicist the University of Cambridge helped uncovered secrets of the universe that have led to discoveries 4 decades later spark the Wests interest in the SZ effect, initiating a decades-long effort In 1960, at the height of the Cold War, Sunyaevs best-known work dates back by experimental physicists in the United Rashid Sunyaev left his home in Tashkent, to the late 1960s, when he and his legendary Kingdom and the United States to validate the capital of Soviet Uzbekistan, to study mentorYakov Zeldovich, one of the and then exploit the phenomenon. In 2008, physics in Moscow. He was then 17 years fathers of the Soviet hydrogen bombpre- when a team of U.S. astrophysicists reported old, with exceptional mathematical talent dicted a phenomenon that causes massive the first galaxy clusters discovered using the the kind of student the Soviet government clusters of galaxies to make an imprint upon SZ effect, the result represented the bridging would have liked to groom into a weapons the cosmic microwave background (CMB). of two divides: between theory and applicascientist. With genuine apprehension, But the impact of the so-called Sunyaev- tion, and between two scientific cultures that Sunyaevs grandmother asked him to make a Zeldovich (SZ) effect is only now being until 2 decades ago had been isolated from promise: Could young Rashid stay away fully realized. In the past decade, the effect each other by secrecy and mutual distrust. from work that might help in the building of has enabled cosmologists to measure astromissiles and bombs? nomical distances precisely and determine Useless science Half a century later, she would have been the expansion rate of the universe, independ- Once restricted from making trips overseas, proud of her grandson, who now directs the ent of other techniques such as the observa- Sunyaev is now a globetrotter, flying freMax Planck Institute for Astrophysics in tion of Type 1A supernovae. And in the past quently between Garching and Moscow and Garching, Germany, and is a chief scientist at 2 years, thanks to advances in detection speaking at conferences around the world. the Space Research Institute in Moscow. Not technology, astronomers have begun using Last October, he flew to Austin for an invited only did Sunyaev manage to keep his word the phenomenon to discover distant clusters talk at the University of Texas (UT). He got about avoiding secret military programs, but of galaxies. The SZ effect has gone from there 2 days later than planned because of a he also helped unlock secrets of the universe being Gee, good for you, Sunyaev to a last-minute snafu: He kept the passport bearthat are now pillars of modern cosmology. powerful tool for probing the universe, says ing his multiple-entry U.S. visa in a bank

In the Afterglow Of the Big Bang

26

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

CREDIT: HANS HERMANN HEYER/ESO

NEWSFOCUS
locker and forgot to collect it before the bank closed for the weekend. The incident seems consistent with Sunyaevs image as a man totally absorbed in his work, immersed in science 24/7 except when hes tending to his garden and unable to pull himself away from the computer even at mealtimes without some prodding from his wife. Sunyaev is stocky, with a square face and silver hair, and a shy manner that manifests an affable, teddy-bearish personality. One of his default expressions is of childlike wonderment, which he sported at the Texas conference while picking out greens from the salad bar and later as he listened to a talk in the auditorium, chewing the end of a pen. Sunyaev might never have ended up as an astrophysicist if it hadnt been for a meeting with Zeldovich in March 1965, when Sunyaev was a graduate student at the Moscow Insti- Cold skies. Using the South Pole Telescope tute of Physics and Technology planning a (above), John Carlstrom (right) and his colPh.D. in elementary particle theory. Snowed leagues have discovered new galaxy clusters under with coursework and itching to do real revealed through the SZ effect. research, Sunyaev went to see Zeldovich, who was looking for students to join his research suggesting ways in which this fossil group at the Institute of Applied Mathematics. radiation could be used as a marker and He told Sunyaev that he would have to work a backdrop for studying the evolution on astronomy. Sunyaev, who had heard his of the universe. The consequences of department chair say that astronomy was an the CMB were immediately obvious to absolutely useless science, told Zeldovich him, says Sunyaev. that hed prefer to study elementary particles. As his first assignment, Sunyaev Zeldovich was a little amused, Sunyaev began to work out the physics of the says. He said, Please help me solve one or first few seconds after the big bang. If two problems in astrophysics, and after that the universe began as a fireball, as the we will work on what you want. CMB suggested, it was a soup of phoZeldovichs after that really meant tons, protons, and electrons sharing the never, but it didnt matter. Sunyaev was same temperature at the moment of immediately taken with astrobirth. As the universe physics, in no small part because cooled down, some of the of Zeldovichs brilliance and protons and electrons came charm. He was like mercury, sciencemag.org together to form hydrogen. Podcast interview very fast, Sunyaev says. I Working with Zeldovich and with Yudhijit would spend 5 days solving an Dima Kurt, Sunyaev calculated Bhattacharjee. equation; he would take 10 minthe rate at which this process had utes to understand it. occurred. It was groundbreaking It was a grueling apprenticeship. One in work, says Eiichiro Komatsu, a researcher at 10 students survived. There were 4-5 of us UT Austin. But the best was yet to come. living in a small room in the dormitory. We came only to sleep for a few hours, Sunyaev Heavy artillery says. You had to work in the night because After Sunyaev completed his Ph.D. in 1966, the next day you had to show that you were his future was uncertain. Under government solving problems. rules, he needed a special residence permit and In late 1965, the group received word of a a place to live in order to work in Moscow. discovery that would revolutionize cosmol- Moscow was considered paradise, so you ogy. Researchers at Bell Labs had detected the could get permission to live there only if you CMB, the leftover radiation from the big bang agreed to do work that nobody else agreed to, that confirmed that the cosmos had originated he says. Or you had to be very accomplished, as a fireball. Zeldovich, a proponent of an like being the best ballet dancer or something. alternative cold modelin which the starting Zeldovich, who had been decorated with point was a very low temperatureaccepted three stars from the government for his conright away that he had been wrong and began tributions to bombmaking projects, was able

Online

to get Sunyaev a small room in an apartment shared with another family. It was tiny compared to our house in Tashkent, but I called my mother to tell her; I was so proud, Sunyaev says. But there was another obstacle. The chief of the KGB at the institute called Sunyaev to his office and told him to look for work elsewhere. He said, Dont tell Zeldovich why you are leaving. Just tell him you dont want to work with him, Sunyaev says. The reason was that his family, which had been prominent under the Tsars, was viewed as a political threat, and intelligence officials were uneasy about letting Sunyaev work at an institution where many people were working on military programs. I said I cannot lie to Zeldovich, he says. The KGB finally left him alone. Between 1966 and 1970, Sunyaev worked on two fundamental ideas related to

CREDITS (TOP TO BOTTOM): KEITH VANDERLINDE, 2008; ERIK LEITCH

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

27

NEWSFOCUS
the CMB. The first was an analytical deter- it were a shadow of the galaxy cluster, and mination of how sound waves rippled would be brighter at submillimeter wavethrough the fireball of radiation and matter lengths. Only objects as massive as clusters in the moments following the big bang, of galaxies could cause such a distortion in shaping cosmic structure on large scales. the CMB. The most exciting thing about the These so-called baryonic acoustic oscilla- effect was that it was independent of redshift: tions, Sunyaev predicted, must have left a The distortion in the CMB would be just as signature in the CMB. (James great whether a galaxy were Peebles, an astronomer at Everybody began to close or far away. Princeton University, reached say, Yes, yes, yes Most Soviet astrophysithe same conclusion indecists were not convinced, pendently around the same everything is however. At one talk Sunyaev time.) The balloon-borne correct. I realized gave in Moscow, he recalls, a BOOMERANG and MAXwhite-haired professor told IMA experiments f inally what it meant to him the theory was nonsensimeasured the oscillations have such heavy cal. He said, I have been some 35 years later (Science, teaching thermodynamics for 28 April 2000, p. 595; 4 May artillery behind me. 35 years, and Ive never heard RASHID SUNYAEV of radiation decreasing in 2001, p. 823). The second idea is what led brightness that way. I became to the discovery of the SZ effect. In 1967 or very upset, and I was trying to explain that I 1968, Sunyaev heard about an astronomical have three techniques to prove this and so mystery that had puzzled physicists since the on, but nobody was listening to me. 1930s: why the mass of galaxy clusters, as The reaction changed when Zeldovich, estimated from their light, seemed lower than late for the talk, walked in. He said, the estimate from a measurement of the Rashids right and everything is correct, motion of galaxies within them. Although Sunyaev says with a chuckle. And everythis missing matter is now believed to be body began to say Yes, yes, yes everydark matter, which puts out no light but exerts thing is correct. I realized what it meant to a gravitational effect on stellar motion, in the have such heavy artillery behind me. 60s, researchers wondered whether the missWorlds apart ing matter could in fact be intracluster gas. The question led astrophysicists to focus In 1968, a 27-year-old Scottish astronomer attention on the physics of this gas, and named Malcolm Longair ar rived in Sunyaev decided to analyze how it might Moscow to spend a year with Soviet scieninteract with the CMB radiation. Every so tists as part of an exchange program often, he found, an energetic electron bounc- between the Royal Society and the U.S.S.R. ing around inside the hot gas would slam into Academy of Sciences. The Royal Society had entered. Western journals like Nature and The Astrophysical Journal routinely arrived more than a year late, with all advertisements clipped out by censors. Rarely, a recent issue would find its way to scientists with friends in the West. But the quality of the research blew Longairs mind. Rashid and the others were fantastically well trained, and they were under tremendous pressure to solve problems very fast, he says. Another Western visitor, cosmologist R. Bruce Partridgethen at Princeton witnessed the intensity on a visit to Zeldovichs house in August 1968. There was a large room with a table in the middle and blackboards at either end, he says. Zeldovich himself, Partridge says, was a Santa Claus without the beard; almost entirely bald, utterly vibrant, with shining eyes. After a midmorning snack, the researchers grilled Partridge for 2 hours about what experimental and theoretical physicists in the United States were up to. It was the single most draining and exhilarating experience I have ever had, Partridge says. It made my Ph.D. look like a walk in the park. For Sunyaev, Longair was an ambassador from another world with a style of discussing science and writing papers that was much freer than anything he had seen. Sunyaevs first awareness of the world of science outside the Soviet Union had come a year before, when Zeldovich took him and some of his colleagues to an International Astronomical Union meeting in Prague. The impact on me was unbelievable; it was the same as when a child goes to kindergarten for the first time and sees other children who have their own fathers and mothers, he says. I realized how much we were losing by being cut off. The isolation didnt just prevent Sunyaev and his colleagues from learning about developments overseas; it also created hurdles in publishing both in Soviet and Western journals. All research papers authored were reviewed by special committees that would take months to approve them and would edit out words such as nuclear. Partridge recalls that one preprint from Zeldovich arrived in the mail long after it had been written, with a coffee stain and crumbs on it: Somebody had literally eaten their lunch on the paper. Ironically, it was the challenge of getting published in a timely way that in part prompted Zeldovich to drive Sunyaev and the others so hard. He told me several times that I had to identify a problem and solve it at least a year prior to my Western colleagues, Sunyaev says.

Shadow in the CMB. A new galaxy cluster imaged by the South Pole Telescope as a distortion in the cosmic microwave background (left). The telescope has 966 energy detectors (center) arrayed on six wedges (right).

a CMB photon, boosting its energy and raising its frequency. As a result, Sunyaev f igured, astronomers observing a cluster would see a greater number of CMB photons above a certain frequencyaround 218 GHzand a smaller number below it. In other words, the CMB would appear fainter than normal at centimeter and millimeter wavelengths, as if

had warned him to mind his words and actions during his visit because the KGB would be watching him at every step. A lover of Russian music and theater, Longair hit it off with Sunyaev. He met with Zeldovichs group often, but never at the Institute of Applied Mathematics, which was closed to outsiders. He quickly discovered what a different scientific environment he
VOL 327 SCIENCE

28

1 JANUARY 2010

www.sciencemag.org

CREDITS (LEFT TO RIGHT): SPT TEAM; UC BERKELEY SPT DETECTOR GROUP (2)

NEWSFOCUS
Sunyaev was keen to have radio astronomers in Europe and the United States begin efforts to detect the SZ effect, which he knew would be a technological challenge unlikely to be met in the Soviet Union. After returning to the United Kingdom, Longair gave talks about the phenomenon at different universities and urged colleagues to try to detect it. Meanwhile, Zeldovich heard a rumor that a Russian astronomer named Yuri Pariiski had detected the effect. Sunyaev says Zeldovich quickly scolded him for not A rare mingling. Sunyaev and Zeldovich having written up a sepa- (right) had little contact with Western scienrate paper about the topic. tists. One exception was a 1968 meeting in They co-authored the first Tbilisi, where R. Bruce Partridge, Sunyaev, one in 1972. Vladimir Dashevsky, and Malcolm Longair
(left to right) posed for a photo.

The hunt is on In 1976, a graduate student at Cambridge named Mark Birkinshaw decided to make the detection of the SZ effect his Ph.D. project. It was a risky topic for a dissertation; everybody knew that the effect would be difficult to detect with instruments available at the time. I took it up because it was a brandnew thing, says Birkinshaw, now an astronomer at the University of Bristol in the United Kingdom. Building on efforts by his thesis adviser, Birkinshaw made an attempt using the 25meter radio telescope at Chilbolton Observatory in Hampshire. After 2 years of work on minimizing statistical noise from clouds, I was able to detect the effect in two or three clusters, but it needed further confirmation, he says. Birkinshaw and his colleagues reported a cast-iron detection in 1984 with the help of the 40-meter dish at the Owens Valley Radio Observatory (OVRO) near Bishop, California. Over the next decade, researchers got better SZ images of clusters by combining signals from multiple radio telescopes. In 1995, John Carlstrom and Marshall Joy at the California Institute of Technology made another significant improvement by mounting centimeter-wavelength receivers on the OVRO millimeter array. Soon, Carlstroms group and a group led by Lyman Page at Princeton were hatching plans to survey the entire sky for the SZ effect. We decided we wanted to do SZ science big-time, says Carlstrom, who is now at the University of Chicago in Illinois.

CREDITS (LEFT TO RIGHT): COURTESY OF MALCOLM LONGAIR; COURTESY OF RASHID SUNYAEV

Because millimeterand centimeter-wavelength radiation can be absorbed by water va p o r, b o t h t e a m s sought to find a cold, dry place with calm, clear skies. Carlstrom c h o s e Antarctica, where his g roup built the South Pole Telescope (SPT) a 10-meter instrument with an array of 966 energy detectors designed at the University of California, Berkeley, that make it many times more sensitive to the SZ effect. It saw first light in February 2007. In its first year of operation, the SPT helped spot four galaxy clusters, three of them new discoveries. Since the October 2008 results, the survey has found many more clusters, says Carlstrom. Still more are expected from Pages group, working at the 6-meter Atacama Cosmology Telescope on Cerro Toco in northern Chile. How these clusters are distributed will help cosmologists get a better picture of large-scale cosmic evolution and answer questions about dark energy and the expansion of the universe. After the fall In the years following the Soviet Unions collapse in 1991, Sunyaev received job offers from institutes around the world. It
SCIENCE VOL 327

would have made perfect sense to quit, as many Russian scientists were doing, but he was reluctant to leave the Space Research Institute. Many of my friends asked me why I would not move to the West, he says. But I was very afraid to leave because I knew if I left, my research group in Moscow would die. Sunyaevs acquaintances say he went to great lengths to keep his group together. In 1992, while having lunch with Partridge at a food court in Washington, D.C., Sunyaev reflected guiltily on how much the meal had cost. He said, What Ive just eaten here would pay a Russian scientist for a month, recalls Partridge, now a professor at Haverford College in Pennsylvania. On more than one occasion, he says, Sunyaev used money from his prizes to pay salaries for some of his colleagues. Since taking his appointment at Max Planck, Sunyaev has promoted collaborations between his Moscow lab and his new institute, as well as the rest of the world. He is still just enormously productive and motivational, says Carlstrom. One product of those collaborations is eROSITAan x-ray telescope being built in Germany, which will be dispatched into space on board a Russian spacecraft named Spectrum-X. The telescopes primary goal is to investigate dark energy by surveying up to 100,000 galaxy clusters. For Sunyaev, eROSITA and other international scientific projects represent a quest for knowledge that binds humanity together. I think astronomy is very useful, he says, contradicting his department chairs comment from 47 years ago. What we are studying has great importance for understanding our place in the universe regardless of what ones place may be on the geopolitical map.
YUDHIJIT BHATTACHARJEE

www.sciencemag.org

1 JANUARY 2010

29

COMMENTARY
Island lizards Research opportunities and NIH

LETTERS I BOOKS I POLICY FORUM I EDUCATION FORUM I PERSPECTIVES

33

36

LETTERS
edited by Jennifer Sills

Bushmeat Hunting and Climate: An Indirect Link

J. F. BRODIE AND H. K. GIBBS (BUSHMEAT HUNTING AS CLIMATE THREAT, Letters, 16 October 2009, p. 364) argue that bushmeat extraction threatens the carbon stocks of tropical forests because (i) bushmeat hunting reduces abundances of large-bodied vertebrates; (ii) tree species with large seeds reproduce poorly without large-bodied vertebrates on which they depend for seed dispersal; (iii) large seed size is correlated with high wood density in tropical trees; and (iv) trees with high wood density contribute disproportionately to the carbon stock. Their first point is well-established, but evidence regarding the others is mixed. Killing animals reduces seed dispersal of vertebrate-dispersed trees (14) but does not necessarily reduce the reproduction of large-seeded trees (5), perhaps because large-bodied animals also function as seed predators and herbivores (2, 6). Likewise, the correlation between seed size and wood density in tropical trees is at best weak (7). Finally, plots with trees of higher wood density do not necessarily have higher total tree carbon stocks; depending on the site, carbon stocks may be positively related, negatively related, or unrelated to mean wood denLianas climbing a tropical canopy tree.

sity, because of the usually countervailing effects of tree volume (8). Lianas (woody vines that climb into the tree canopy) provide an alternative possible link between bushmeat hunting and carbon storage. Hunting is a disadvantage for species with seeds dispersed by animals, and therefore gives a comparative advantage to species with seeds dispersed by wind (5, 9). This strategy is much more common among liana species than trees (60 versus 20%). Liana leaves displace an equal mass of tree leaves (10), and lianas store much less carbon per leaf area than trees (11). Thus, hunting may favor lianas, and an increase in lianas is likely to reduce carbon storage. Whatever its effect on forest carbon stores, the bushmeat crisis is unarguably a major threat to tropical biodiversity (2, 12, 13). This by itself is reason to fight it.
PATRICK A. JANSEN,1,2* HELENE C. MULLER-LANDAU,3 S. JOSEPH WRIGHT3
and Conservation Ecology Group, University of Groningen, Haren, Netherlands. 2Forest Ecology and Forest Management Group, Wageningen University, Wageningen, Netherlands. 3Smithsonian Tropical Research Institute, Balboa, Ancon, Republic of Panama. *To whom correspondence should be addressed. E-mail: [email protected]
1Community

References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. P. M. Forget, P. A. Jansen, Conserv. Biol. 21, 106 (2007). S. J. Wright et al., Conserv. Biol. 14, 227 (2000). N. J. Cordeiro, H. F. Howe, Conserv. Biol. 15, 1733 (2001). K. M. Holbrook, B. A. Loiselle, Ecology 90, 1449 (2009). S. J. Wright, A. Hernandez, R. Condit, Biotropica 39, 363 (2007). E. Mendoza, R. Dirzo, Oikos 116, 1841 (2007). I. J. Wright et al., Ann. Bot. 99, 1003 (2007). J. C. Stegen, N. G. Swenson, R. Valencia, B. J. Enquist, J. Thompson, Glob. Ecol. Biogeogr. 18, 617 (2009). G. Nunez-Iturri, O. Olsson, H. F. Howe, Biol. Conserv. 141, 1536 (2008). T. Kira, H. Ogawa, in Productivity of Forest Ecosystems, P. Duvigneaud, Ed. (UNESCO, Paris, 1971), pp. 309321. F. E. Putz, Biotropica 15, 185 (1983). K. H. Redford, Bioscience 42, 412 (1992). H. C. Muller-Landau, Biotropica 39, 372 (2007).

Gray Wolves Not Out of the Woods Yet

IN APRIL 2009, THE U.S. FISH AND WILDLIFE Service (FWS) removed the northern Rocky Mountain population of gray wolves (Canis lupus) from all protections under the Endangered Species Act (ESA). Following the ESAs mandate to base listing determinations solely on thebest scientific and commercial data available, FWS conducted an extensive analysis of regional threats to

wolves. They concluded that while [p]ublic hostility toward wolves led to excessive human-caused mortality that extirpated the species, subsequent improvement in attitudes toward wolves ensured the long-term viability of the species. We agree that human behaviors (and the attitudes and values underlying them) ultimately caused the extirpation of wolves in the northern Rockies, but we find little support for FWSs conclusion that attitudes toward wolves have improved, or are improving. Indeed, the
VOL 327 SCIENCE

larger body of research points to the opposite conclusion (15). Although FWS provided more than 200 citations in their analysis, they cited just one empirical study that examined attitudes toward wolves (4). [This cannot be explained by a lack of published literature; a recent review identified 50 publications that specifically addressed the topic (6).] Thus, it appears FWS was either unaware of the extensive body of research on attitudes toward wolves, or chose to ignore this research. In fact, the only empirical article cited by FWS

30

1 JANUARY 2010

www.sciencemag.org

CREDIT: P. A. JANSEN

Capture of a transient guest

Protein shell designs

38
a meta-analysiscomes to a very different conclusion: Across the 37 attitude surveys we studied, the reported statistics were stable over the last 30 years[t]his contradicts a recent perception among some ecologists that wolf support has recently grown (4). The FWSs analysis of the threat posed by negative attitudes toward wolves is wholly inadequate. When threats to a species continued survival are primarily social in nature, FWS must use the same standard that goes into analyzing biological and ecological threats. It is time for FWS to expand its view of what constitutes science and fully incorporate the social sciences into listing determinations.
JEREMY T. BRUSKOTTER,1* ERIC TOMAN,1 SHERRY A. ENZLER,2 ROBERT H. SCHMIDT3
1School

42
State University, Columbus, OH 43210, USA. 2Institute on the Environment, University of Minnesota, St. Paul, MN 55108, USA. 3Department of Environment and Society, Utah State University, College of Natural Resources, Logan, UT 84322, USA. *To whom correspondence should be addressed. E-mail: [email protected]

Patents: A Threat to Innovation?

IN THE POLICY FORUM BALANCING INNOVAtion and access: Patent challenges tip the scales (16 October 2009, p. 370), M. J. Higgins and S. J. H. Grahams claim that Paragraph IV patent challenges are increasingly stifling new drug innovation is misleading. Economists have repeatedly cautioned that correlation is not causation. The increasing number of Paragraph IV challenges, coupled with the decreasing number of FDA-approved new compounds is an interesting, but not causal, relationship. Declines in approvals could be due to a range of factors, including decreasing research productivity. Reasons for the decline in productivity include the increasing difficulty of understanding the science of more complex diseases and the focus of pharmaceutical companies on low-risk me too drug development (1). Not all Paragraph IV challenges lead to early generic entry. In research documenting Paragraph IV challenges between 2004 and 2006, I found that only 13 (11%) of the 115 lawsuits resulted in a generic win (2). When

References
1. J. T. Bruskotter, R. H. Schmidt, T. L. Teel, Biol. Conserv. 139, 211 (2007). 2. M. D. Duda, S. J. Bissell, K. C. Young, Wildlife and the American Mind (Response Management, Harrisonburg, VA, 1998). 3. J. W. Enck, T. L. Brown, Wildl. Soc. Bull. 30, 16 (2002). 4. C. Williams, G. Ericsson, T. A. Heberlein, Wildl. Soc. Bull. 30, 575 (2002). 5. G. Ericsson, T. A. Heberlein, Biol. Conserv. 111, 149 (2003). 6. C. Browne-Nunez, J. G. Taylor, Americans attitudes toward wolves and wolf reintroduction: An annotated bibliography, Tech. Report No. 2002-0002 (U.S. Geological Survey, 2002).

of Environment and Natural Resources, The Ohio

BREAKTHROUGH IN RNA ISOLATION

The single step method without phase separation

RNAzol RTmRNA isolates total RNA, with

*

TOTAL RNA

and small RNA (200 - 10 bases) in separate fractions.

Higher RNA yield and quality than with previous single-step reagents. No chloroform-induced phase separation. Just add water. RNA is ready for RT-PCR, microarrays, poly A+ selection, northern blotting and RNase protection. No DNase treatment necessary. No need for a refrigerated centrifuge. All steps performed at room temperature.
SMALL RNA

MOLECULAR RESE ARCH CENTER, INC.

5645 Montgomery Road, Cincinnati, Ohio 45212
*

www.mrcgene.com
Phone: (888) 841-0900

Piotr Chomczynski, patent pending

RNAzol is a trademark of Molecular Research Center, Inc.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

31

LETTERS
the branded company prevails, there is neither early generic entry nor revenue loss. Perhaps more important, 73% ended in settlements, of which a subsetabout 60% (3)did not appear to result in early generic entry (2). Hence, a count of patent challenges is an unreliable indicator of losses for branded companies. The fact that so many of the lawsuits end in settlements precludes information on the patent strength and boundaries that is revealed through judicial determination. In addition to these factors, branded pharmaceuticals are increasingly retaliating by producing or licensing authorized generics that dampen their revenue loss and the incentives for future patent challenges. For instance, in 2003, Apotex was granted the 180-day exclusivity period for its generic version of Paxil, an anti-depressant marketed by GlaxoSmithKline (GSK). Apotex had expected to generate sales of approximately $575 million during its 6 month exclusivity, but the introduction of GSKs authorized generic lowered the actual sales to around $200 million (4). Paragraph IV challenges are an important mechanism for identifying patents that should not have been granted in the first place. They should be allowed to continue unless there is much more compelling evidence that they are in some way slowing innovation.
MERLINA MANOCARAN
Mirzayan Science and Technology Policy Fellow, The National Academies, Washington, DC 20001, USA. E-mail: [email protected]

References and Notes

1. GAO, New drug development: Science, business, regulatory, and intellectual property issues cited as hampering drug development efforts (U.S. Government Accountability Office, GAO-07-49, November 2006); www.gao.gov/new.items/d0749.pdf. 2. M. Manocaran, dissertation, Harvard Business School, Boston, MA (2009). 3. Details of settlements between branded and generic companies are not always made public, so it is not possible to confirm the true number of settlements that include provisions for early entry. This estimate may be overstated due to this lack of data. 4. T. Gilbert, Comment of Apotex Corp. in Support of Citizen Petition Docket No. 2004P-0075/CP1 (Food and Drug Administration, 24 March 2004); www.fda.gov/ohrms/dockets/dailys/04/apr04/040204/ 04P-0075-emc00001.pdf.

This fact increases uncertainty for innovators, and the U.S. federal courtsweakening of patent rights over the past several years has made eventual payoffs from innovation even less likely. Added incentives offered to generic entrants in the Hatch-Waxman Act to launch Paragraph IV patent challenges, and the strategic use of this regime by generic firms, are exacerbating this problem by further undermining the market incentives to do pharmaceutical research. We are unconvinced by Manocarans assertion that the use of authorized generics will dampen incentives to engage in Paragraph IV challenges; preliminary evidence actually suggests otherwise (4, 5). Regarding the outcomes of Paragraph IV patent suits, the most credible evidence to our knowledge comes from a 2002 Federal Trade Commission (FTC) report cited in our article (6). Although we cannot definitively refute Manocarans claim that 73% of a sample of cases she studied ended in settlement, we note that another researcher, using a larger sample that includes more recent cases, finds settlement rates at trial were as low as 20% and no higher than 39% during the 20062009 period (7). This same research shows that generics won 25% of all cases at trial (and 41% of those that did not settle), a share substantially different than the 11% that Manocaran reports (7). Moreover, we note that the 41% figure is virtually identical to the 42% generic win rate reported by the Federal Trade Commission for earlier years (6). These inconsistencies notwithstanding, Manocaran correctly raises the issue of collusive settlements, an issue with which the competition authorities are rightfully dealing (8). However, her general observations about settlement do not affect our main thesis: The Hatch-Waxman Act is creating incentives to challenge patents and, with shifting treatments of patent law by the courts, industry revenues are increasingly threatened. In a market system, the predictable outcome without intervention will be less innovation, to the detriment of public health in the long run.
STUART J. H. GRAHAM AND MATTHEW J. HIGGINS*
College of Management, Georgia Institute of Technology, Atlanta, GA 30308, USA. *To whom correspondence should be addressed. E-mail: [email protected]

7. G. Glass, The Paragraph Four report: The May 2009 case outcomes data (Parry Ashford, 2009); www.paragraphfour.com. 8. L. Layton, FTC sues in pay-for-delay pact: Drugmaker paid rivals to withhold generic, agency says, Washington Post, 3 February 2009.

Let Top Students Go Forth and Prosper

THE NEWS OF THE WEEK STORY STUDY FINDS science pipeline strong, but losing top students (Y. Bhattacharjee, 30 October 2009, p. 654) decried the steep drop in the percentage of the highest performing students taking science and engineering jobs. But why not let these talented, scientifically trained human catalysts shift gears and move into areas such as public policy, legislation, law, finance, economics, public relations, and yes, even entertainmentthat seemingly silly place where ideas and visions are formed? It would help to have scientifically trained policy makers and legislators who truly understand the scientific and technologic issues they are voting on, with enough clout to get others on board. It would also help to have management consultants and financial analysts who avoid entrenched mindsets and realize that some visionary business approaches are de facto Ponzi schemes. A protectionist attitude that expects the best students to stay within their formative disciplines has pernicious consequences. Top students in science and engineering form a gift to societyand to the scientific enterprisewhen they fly forth to pollinate areas of vital importance to the public discourse. Cross-disciplinary ambassadors should be encouraged, not discouraged, if we are to build a bright new, sustainable future.
GURUPRASAD MADHAVAN1* AND BARBARA ANN OAKLEY2
and Global Affairs, National Academy of Sciences, Washington, DC 20001, USA. 2Department of Industrial and Systems Engineering, Oakland University, Rochester, MI 48309, USA. *To whom correspondence should be addressed. E-mail: [email protected]
1Policy

Response
WE AGREE WITH MANOCARAN THAT CORRElation is not causation and say as much in our Policy Forums first paragraph. But Manocarans suggestion that the pharmaceutical industry may not be facing an innovation crisis is not supported by the weight of evidence (13). Companies rely critically on patents to recoup their R&D expenditures, and patent rights are only as good as the human agents (such as judges and juries) who review them.

Letters to the Editor

Letters (~300 words) discuss material published in Science in the previous 3 months or issues of general interest. They can be submitted through the Web (www.submit2science.org) or by regular mail (1200 New York Ave., NW, Washington, DC 20005, USA). Letters are not acknowledged upon receipt, nor are authors generally consulted before publication. Whether published in full or in part, letters are subject to editing for clarity and space.

References
1. 2. 3. 4. 5. H. Grabowski, PharmacoEconomics 22, 15 (2004). H. Moses et al., New Engl. J. Med. 294, 1333 (2005). M. J. Higgins, D. Rodriguez, J. Fin. Econ. 80, 351 (2006). E. Berndt et al., Health Aff. 26, 790 (2007). E. Berndt et al., Do authorized generic drugs deter Paragraph IV certifications? Recent evidence. (Analysis Group Inc., Working Paper, 2007); www.analysisgroup.com/uploadedFiles/Publishing/ Articles/PhRMA_Authorized_Generic_Entry.pdf. 6. Federal Trade Commission, Generic Entry Prior to Patent Expiration: An FTC Study (2002).

32

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

BOOKS ET AL.
ECOLOGY AND EVOLUTION

Leaping Lizards
Rosemary G. Gillespie

ropical islands, tion can run counter to natuLizards in an Evolutionary Tree vibrancy, color, ral selectionbut it doesnt Ecology and Adaptive Radiation and sex make always. Indeed, anoles display of Anoles for an intoxicating key innovations in the form com bination. For a of toe pads that allow them to by Jonathan B. Losos biologist, if you add in explore a new ecological arena University of California Press, high diversity, endethrough natural selection and Berkeley, 2009. 527 pp. $75, 52. mism, and a strikingly dewlaps that may enhance the ISBN 9780520255913. Organisms charismatic group of rate of speciation through sexand Environments. animals, it is difcult ual selection. Perhaps the most to imagine a more intriguing element in the syscaptivating system. In Lizards in an Evolu- tem is that, despite the apparent diverse selectionary Tree: Ecology and Adaptive Radia- tive pressures acting on these lizards, there is tion of Anoles, Jonathan Losos describes such remarkable predictability in the repeated evoa system in ardent detail. The book repre- lution of ecomorphs on the Greater Antilles. sents a rich compendium of information by Except for a brief discussion in the nal an extraordinarily insightful biologist with a chapter of parallels in other systems, Losos deep and broad understanding of the diver- quite wisely makes little attempt to analyze the sity of Anolis lizards in the Caribbean. Losos lizards in the context of other studies: Had the (an evolutionary biologist and herpetologist book taken this avenue, at Harvard University) indicates two audi- it would have grown ences for the book: those deeply interested to encyclopedic proin anoles and those interested in general ques- portions [like those of tions of biodiversity, evolutionary biology and (1)]. However, readers ecology. I fall in the latter category, although always draw parallels the book certainly enhanced my appreciation to systems with which of the former. The solid foundation in natural they are most familiar. history makes this a compelling read even for So I will mention some biologists with a marginal interest in lizards of these in the context of or evolutionary biology. As Losos comments, radiations in the Hawaionly by having a rich and deep understand- ian Islands, because ing of the organisms we study can we have such comparisons allow insights into how and why they vary and how us to evaluate common they have evolved. It is this foundation of themes. For example, understanding that has made books by such in the anoles, body natural history giants as Jane Goodall, Ger- size diverged early ald Durrell, George Schaller, and Bernhard in the radiation withGrzimek so influential. Although Lososs out much subsequent book is aimed at a somewhat higher level, it change, but habitat Uniquely variable. Anolis distichus, a widespread and highly polymorphic offers the same inspiration. use has been diverging trunk anole, exhibits substantial intra- and interpopulational variation in One of the most intriguing aspects of the throughout the radia- dewlap color (here A. d. vinosus, from the Tiburon Peninsula, Haiti). Anolis system is that it offers so many direc- tion. In the same way, tions for research: Whereas some species Hawaiian sap-feeding planthoppers in the of the web rather than the body phenotype. have bizarre adaptations for crypticity, oth- genus Nesosydne (Delphacidae) appear to Third, anole ecomorphs have arisen almost ers appear to run, using speed and alacrity have undergone extensive ecological shifts entirely through convergent evolution, whereas to elude predators and catch prey. Likewise, early in their radiation and relatively minor in Hawaiian Tetragnatha spiders, communities although much of their behavior and morphol- changes subsequently (2). have been lled by a combination of colonizaogy provides mechanisms for escaping detecIn the context of community assembly, the tion and evolution. The domination of evolution, the lizards display colorful dewlaps and very clear pattern of repeated evolution of eco- tion over colonization in the anoles (perhaps perform staccato head bobbing to signal to morphs in the Caribbean anoles suggests that also in Hawaiian Ariamnes spiders) indicates mates and competitors. Clearly, sexual selec- evolution has allowed species to occupy the that movement in the anoles must be severely ecological space more rapidly than has colo- curtailed, an attribute that Losos also discusses The reviewer is at the Department of Environmental Scinization. In contrast, many Hawaiian plants in the context of population structure. ence, Policy, and Management, University of Califorunderwent major ecological changes early in Lizards in an Evolutionary Tree offers a nia, Berkeley, CA 947203114, USA. E-mail: gillespie@ berkeley.edu their radiation, with communities on younger winning combination of enchanting animals
www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

islands being lled simply by colonization of ecological equivalents from older islands. For these plants, it appears that colonization occurs more readily than evolutionary shifts within an island. Yet, in other lineages (both plants and animals), there are varying levels of independent ecological radiation into the diverse habitats on each island (3). Repeated evolution of ecomorphs with discrete sets in any one habitat, as in the anoles, is most pronounced in spiders (genera Tetragnatha and Ariamnes), and comparisons are intriguing. First, as in anoles, some habitats may be missing a member (usually the same one) of the ecomorph set. Losos considers the most likely explanation for this to be island size. The geochronology of the Hawaiian Islands may allow the phenomenon to be scrutinized in some detail. Second, the anole radiation is characterized by several taxa with unique ecological attributes. In Hawaiian Tetragnatha, taxa outside the spiny leg clade have several unique representatives, and convergence in the more-encompassing lineage seemsif anythingto be limited to the form

CREDIT: RICHARD E. GLOR (HTTP://LACERTILIA.COM/)/UNIVERSITY OF ROCHESTER

33

BOOKS ET AL.
and an idyllic setting that addresses some key biological questions. Losos documents an extraordinary history of research on almost every imaginable attribute of Anolis lizards. He frequently stops to take stock before presenting the hypotheses and asking how these could be tested, and he whets our appetites by presenting avenues for future study. What an exciting time it is for evolutionary biology, and anoles provide one of the most compelling systems to further our understanding of the eld.
References
1. R. G. Gillespie, D. A. Clague, Eds., Encyclopedia of Islands (Univ. California Press, Berkeley, CA, 2009). 2. G. K. Roderick, D. M. Percy, in Specialization, Speciation, and Radiation, K. J. Tilman, Ed. (Univ. California Press, Berkeley, CA, 2007). 3. R. G. Gillespie, B. G. Baldwin, in The Theory of Island Biogeography Revisited, J. B. Losos, R. Ricklefs, Eds. (Princeton Univ. Press, Princeton, NJ, 2009). 10.1126/science.1182503

BROWSINGS
The Red Book of C. G. Jung. Rubin Museum of Art, New York, through 15 February 2010. On the lower level of the Rubin Museum of Art in New York, there is a small exhibition, The Red Book of C. G. Jung: Creation of a New Cosmology. Jung, who was a highly inuential gure in the history of psychology and psychoanalysis, spent a period of time during World War I in selfinvestigation and waking visions. He created phantasmagorical, multicolored, and detailed images that illuminate his description of this exploration in the recently released Liber Novus (commonly known as The Red Book). The beautiful images, as well as a video describing him by guest curator Sonu Shamdasani, are shown in the exhibit. During the exhibit, the museum has been sponsoring a series of dialogues between notable individuals (including Twitter co-founder Jack Dorsey and philosopher Cornel West) and psychoanalysts; audio podcasts of many of these dialogues will be available at www.wnyc.org. The museum is also showing a lm series based on Jungian themes. Barbara Jasny

COMMUNICATING SCIENCE

If Our Messages Are To Be Heard

Peter Kareiva

The reviewer is at The Nature Conservancy, 4722 Latona Avenue NE, Seattle, WA 98105, USA. E-mail: pkareiva@ tnc.org

34

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

CREDIT: FROM C. G. JUNG, THE RED BOOK FOUNDATION OF THE WORKS OF C. G. JUNG/COURTESY W. W. NORTON

cientists know important things. They know about the role of greenhouse gases in global warming. They know how genes are inherited. They know how the body ghts off infections. They know that the worlds ecosystems are being needlessly degraded. But most scientists do not know how to talk to anyone other than scientists. As a consequence, political leaders and the public at large either ignore or, perhaps more accurately, are bored by whatever it is that scientists are trying to tell them. The general populations attitude toward climate change has become the iconic story of a public that pays no heed to the message of scientists. This inability of scientists to connect with the nonscientists has far-reaching consequences well beyond any single issue such as global warming. Randy Olson and Cornelia Dean have written two very different books with the same goal: to school scientists on how to communicate with and reach the public. Dean, formerly a science editor for the New York Times, knows well how caveats kill the message. And she has seen rsthand the freezing out that instantly accompanies even

a hint of patronizing utterances. As a journal- creating their own news, and building a comist who was in at the founding of the Tuesday munity of like-minded activists. However, as Science Times, Dean saw thoughtful media Dean cautions, the work required for maincoverage of science initially grow but then taining an effective blog is enormous, and the dwindle under the scal pressures of failing return on investment from a scientists pernewspapers. Am I Making Myself Clear? is as spective may be too low. The solution may well much about why scientists need to talk to the be science collectives that maintain blogs and public as it is about how we should talk science can respond instantly to the latest story about a to the public. She argues that scientists need to child dying from a u vaccine or some article develop a civic persona that nds some way to that purportedly overturns 30 years of global communicate science. circulation models. But Deans wisdom, especially before we give ourselves over Dont Be Such a Scientist for engaging in the political to the Internet, Dean reminds Talking Substance in arena, is delivered with a mix us what we all knowthere an Age of Style of authority and charm, as is is too much information out by Randy Olson evident in her advice on how to there, so the key is to master respond to questions from a conthe arts of standing out above Island Press, Washington, gressional committee or staffer: the confusion and deliverDC, 2009. 215 pp., illus. Paper, $19.95. ISBN 9781597265638. Say I dont know when approing a message that is heard, priate and offer to provide the understood, and rememneeded information later. But bered. This is hard enough Am I Making as the old saying goes, dont let for a captive audience in Myself Clear? your mouth write checks your a classroom and orders of A Scientists Guide to ass cant cash. If you promise to magnitude harder when tryTalking to the Public provide additional information, ing to reach a public audience by Cornelia Dean memos, or the like, be prepared that has many vibrant options Harvard University Press, to produce them, and fast. for reading, viewing, and lisCambridge, MA, 2009. Blogs and e-mail campaigns tening. Yet parents with teen284 pp. $19.95, 14.95, 18. have become hugely influenagers in their household will ISBN 9780674036352. tialfor spreading information, have some idea of the power

BOOKS ET AL.
of YouTube postings that go viral and suddenly become talked about in every high school in the country, having been viewed by millions. The Internet can be a powerful means for communication, and scientists need to better tap into it. Olsons Dont Be Such a Scientist is also about reaching the public in fresh ways, in particular through movies and the entertainment industry. Although his writing style is irreverent and much more raw than Deans elegant prose, Olsons insights are equally valuable. They come partly from his having lived in the academic world for much of his life. He was a marine biology professor who gave up his tenured university position to go to Hollywood and learn how to make movies. Olsons latest lm, Sizzle: A Global Warming Comedy, uses goofy humor to inform nontechnical audiences about global warming. Olsons shtick is that science must join the 21st century and reach people where they livein a world of celebrities, videos, and movies. Olson advocates using entertainment to convey scientic content, and he emphasizes the need to reach people in their hearts and guts (and maybe even their groins). Some readers will nd Olsons autobiographical treatise offputting and a bit narcissistic. But to be turned off by Olsons style only proves his point. Get with it. Film and visual images have enormous capacity to tell stories and change thinking. The traditional mode of communicating science is not working; surveys that probe the publics mastery of basic scientic issues consistently document that scientists are failing to reach the public (1). Stuffy and dry science is a losing proposition. Olson recommends that researchers experiment with new approaches, take risks, develop their own voices, and above all recognize the power of storytelling. Whereas social scientists, linguists, and political scientists might advise us how to better frame the issues, these ists are not where Olson turns for inspiration. His book is a plea for indulging ones artistic nature in pursuit of more heartfelt connections to the public. That message will make many scientists squirm, especially those who take refuge in the caricature of science as objective, fact-based, and free from personal values. If scientists were seen as adventurers and explorers instead of as fact-mongers and talking encyclopedias, people might stay awake long enough to learn their science lessons.
an exceedingly clever vehicle for making science engaging Variety

Olson is at his best while recounting how unlikeable scientists can be with their relentless critical thinking, negativity, and smarterthan-thou condescension. A particularly telling anecdote concerns a public debate in New York City between two teams arguing whether or not global warming is a crisis. When the moderator asked the global warming is a crisis team why it thought the other side was misrepresenting the issues, one scientist responded, I dont think they [the global warming is not a crisis team] are completely doing this on a level playing eld that the people here will understand. With that statement, the researcher insulted and instantly alienated his highly educated Manhattan audience. Before-and-after polling revealed that, as a result of watching the debate, the audience (which, admittedly, had been stacked by the organizers) had shifted its position by 16 percentage points against the global warming is a crisis view. It is not hard to gure out why Olson, Dean, and others (1) are in 2009 tackling the cultural and communication divide between science and the rest of humanity. Scientists everywhere are bemoaning popular misunderstandings regarding global warming, stem cell research, and childhood vaccination programs, to name just a few topics where science intersects public policy. Fifty years ago, C. P. Snow gave a famous warning

CREDIT: COURTESY SHIFTING BASELINES/PRAIRIE STARFISH PRODUCTIONS

about the dangerous divide between science and the humanities, a divide that he thought put human destiny at risk (2). Today Snows warning is even more pertinent, and yet scientists continue to be resoundingly inept at reaching the public. Both Dean and Olson mention that Carl Sagan was spurned by the National Academy of Sciences, purportedly because he was too successful a communicator. The professional reward system in science routinely belittles the media scientist or the advocate scientist. One senses that this is beginning to change, but scientists still have a great deal to learn about effective communication. Dean and Olson both underemphasize the single biggest reason why scientists are often such ineffective communicators. The failure of scientists as communicators is that they do not know how to listen, especially when it comes to the uneducated public. Brilliant scientists can be stunningly dumb when it comes to dealing with people. I recall one world-renowned ecologist who nearly caused a brawl in a Pacific Northwest tavern by preaching to the bartender about the extinction crisis and self-righteously scolding the tavern for advertising fried spotted owl on the bar menu. Instead of trying to understand the values and thinking behind attacks on the Endangered Species Act, global warming, or the theory of evolution, scientists too often deride what they see as an ignorant public, with potentially devastating consequences (3). The foundation of successful communication is listening to and respecting your audience. Dont Be Such a Scientist and Am I Making Myself Clear? ought to be required reading in all science graduate programs, but they should be supplemented with the wisdom of Nelson Mandela, who knew how to reach a public that initially vilied him (4). Scientists could learn from Mandela that to win peoples minds you must rst get them to listen, and people will listen only if they feel that they are respected.
References
1. C. Mooney, S. Kirshenbaum, Unscientic America: How Scientic Illiteracy Threatens Our Future (Basic, New York, 2009); reviewed in (5). 2. C. P. Snow, The Two Cultures and the Scientic Revolution (Cambridge Univ. Press, Cambridge, 1959). 3. A. C. Revkin, New York Times, 21 November 2009, p. A1; www.nytimes.com/2009/11/21/science/earth/21climate. html?_r=1. 4. J. Carlin, Playing the Enemy: Nelson Mandela and the Game That Made a Nation (Penguin, New York, 2008). 5. J. Coyne, Science 325, 678 (2009). 10.1126/science.1183465

www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

35

POLICYFORUM
RESEARCH AGENDA

Opportunities for Research and NIH

Francis S. Collins

The promise of fundamental advances in diagnosis, prevention, and treatment of disease has never been greater.

High-Throughput Technologies

In the past, most biomedical basic science projects required investigators to limit their scope to a single aspect of cell biology or physiology. The revolution now sweeping the eld is the ability to be comprehensivefor example, to dene all of the genes of the human or a model organism, all of the human proteins and their structures, all of the common variations in the genome, all of the major pathways for signal transduction in the cell, all of the patterns of gene expression in the brain, all of the steps involved in early development, or all of the components of the immune system. Further development of technologies in areas such as DNA sequencing, imaging, nanotechnology, proteomics, metabolomics, small-molecule screening, and RNA interference are ripe for aggressive investment. Furthermore, these technologies will spur the production of massive and complex data sets and will require major investments in computational biology. As one example, the Cancer Genome Atlas (1) is now poised to derive comprehenNational Institutes of Health, Bethesda, MD 20892, USA. E-mail: [email protected]

sive information about the genetic underpinnings of 20 major tumor types. This information will likely force a complete revision of diagnostic categories in cancer and will usher in an era where abnormal pathways in specific tumors will be matched with the known targets of existing therapeutics. Another example is the opportunity to understand how interactions between ourselves and the microbes that live on us and in us (the microbiome) can inuence health and disease (2).
Translational Medicine

Critics have complained in the past that NIH is too slow to translate basic discoveries into new diagnostic and treatment advances in the clinic. Some of that criticism may have been deserved, but often the pathway from molecular insight to therapeutic benet was just not discernible. For many disorders, that is now changing. Three major factors have contributed to this: (i) the discovery of the fundamental basis of hundreds of diseases has advanced dramatically; (ii) with support from the NIH Roadmap, academic investigators supported by NIH now have access to resources to enable them to convert fundamental observations into assays that can be used to screen hundreds of thousands of candidates for drug development; (iii) public-private partnerships are being more widely embraced in the drugdevelopment pipeline to enable biotech and pharmaceutical companies to pick up promising compounds that have been effectively de-risked by academic investigators and to

bring them to clinical trials and U.S. Food and Drug Administration (FDA) approval. As one example, the NIH Therapeutics for Rare and Neglected Diseases (TRND) (3) program will allow certain promising compounds to be taken through the preclinical phase by NIH, in an open environment where the worlds experts on the disease can be involved. Furthermore, as information about common diseases increases, many are being resolved into distinct molecular subsets, and so the TRND model will be even more widely applicable. The rst human protocol (for spinal cord injury) involving human embryonic stem cells (hESCs) was approved by the FDA in 2009, and the opening up of federal support for hESC research will bring many investigators into this field. The capability of transforming human skin broblasts and other cells into induced pluripotent stem cells (iPSCs) opens up a powerful strategy for therapeutic replacement of damaged or abnormal tissues without the risk of transplant rejection (46). Although much work remains to be done to investigate possible risks, the iPSC approach stands as one of the most breathtaking advances of the last several years, and every effort should be made to pursue the basic and therapeutic implications with maximum speed.
Beneting Health Care Reform

U.S. expenditures on health care now represent 17% of our Gross Domestic Product, are continuing to grow, and are excessive as a percentage of per capita gross income com-

36

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

CREDITS: (LEFT TO RIGHT) ODOM AND OHALLORAN/NORTHWESTERN UNIVERSITY; LINDA BARTLETT/NCI, NIH; BILL BRANSON/NCI,NIH; NASA; BILL BRANSON/NIDCR, NIH

he mission of the National Institutes of Health (NIH) is science in pursuit of fundamental knowledge about the nature and behavior of living systems and the application of that knowledge to extend healthy life and to reduce the burdens of illness and disability. The power of the molecular approach to health and disease has steadily gained momentum over the past several decades and is now poised to catalyze a revolution in medicine. The foundation of success in biomedical research has always been, and no doubt will continue to be, the creative insights of individual investigators. But increasingly those investigators are working in teams, accelerated by interdisciplinary approaches and empowered by open access to tools, databases, and technologies, so a careful balance is needed between investigator-initiated projects and large-scale community resource programs. For both individual and large-scale efforts, it is appropriate to identify areas of particular promise. Here are ve such areas that are ripe for major advances that could reap substantial downstream benets.

POLICYFORUM
pared with other developed countries. Yet few would argue that the quality of care is what it should be. Reinventing health care is thus an urgent national priority, and NIH can make substantial contributions. Among projects that must be pursued are the following. Comparative effectiveness research. NIH has supported clinical studies for many years that evaluate outcomes of medical treatment options. For example, the Diabetes Prevention Program (7) demonstrated substantially better benets of exercise and life-style changes over medication in preventing the onset of diabetes. The Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study (8) compared older, cheaper antipsychotic drugs with newer ones, demonstrating that the older drugs worked just as well and had a better sideeffect prole. With support from the American Recovery and Reinvestment Act (ARRA), NIH is investing $400 million in such studies in scal year (FY) 2009FY 2010 and expects to continue high levels of support. Prevention and personalized medicine. Advances in pinpointing individual genetic and environmental risk factors for disease now make it possible to focus prevention strategies more effectively. For example, research to establish the utility of information about individual genetic risks associated with breast cancer, colon cancer, or prostate cancer may help inform the timing of mammography, colonoscopy, or prostatespecic antigen screening. Also, both retrospective and prospective analyses of how individual information about disease risk actually alters health behaviors and clinical outcomes will be critical. Health disparities research. The health of racial and ethnic minorities, people living in poverty, and other disadvantaged groups in the United States is substantially worse than the health of the overall population (9). Any successful reform of the health-care system will require attention to these groups. Using new and powerful tools to disaggregate environmental and genetic contributions, NIH will seek to pinpoint causes of health disparities and to point the way toward solutions. Pharmacogenomics. There is compelling evidence of a correlation between genotype and drug response for more than a dozen drugs (10), and that number is growing. Prospective studies will be needed for many of these applications, if FDA is to be convinced that genotyping should be required on the label and if insurance companies are to be persuaded to reimburse for the cost of genotyping. Health research economics. Although the major justication for biomedical research will always be the relief of human suffering and the prolongation of life, further precision is needed in assessing the economic value of research initiatives, especially those that are large and expensive. Models that attempt to capture this cost-benet balance in Disability Adjusted Life Years (DALYs), Quality Adjusted Life Years (QALYs), Value of Investment approaches, or other metrics are only partially successful in providing the kind of information that policymakers need. NIH plans to initiate a grants program to encourage development and application of more rigorous models.
Focusing More on Global Health

Much of recent global health research has justiably been focused on AIDS, tuberculosis, and malaria (11). It is also critical to go beyond the focus on the big three diseases to neglected tropical diseases of low-income countries that contribute to staggering levels of morbidity and mortality. In collaboration with other sources of support such as the Bill and Melinda Gates Foundation, NIH can play a major role in ramping up the discovery of novel targets in both pathogen and host and work to facilitate advances in prevention, diagnostics, and therapeutics. Helping to build capacity and training opportunities in the developing world will be a critical component of such progress. Additional resources will also be needed to respond to the growing challenge of chronic noncommunicable diseases and injuries.
Reinvigorating and Empowering the Biomedical Research Community

The U.S. biomedical research community has been under stress since the flattening of the NIH budget in 2003 and may potentially face even more severe disruptions at the end of ARRA funding in FY 2011. Looking toward the future, a critical feature must be an emphasis on innovation. Although the two-level NIH peer-review process is much admired and much copied around the world, its potential tendency toward conservatism is a chronic concern and invariably worsens when funding is very tight. Recognizing these problems, NIH announced a series of concrete steps in June 2008 to enhance the peer-review process (12). Effects of these new steps will be closely monitored, and additional reforms to encourage innovation will be undertaken as needed. Meanwhile, it will be critical to resist political attacks on certain areas of sensitive research (such as drug abuse and sexually transmitted diseases); peer review should remain the appropriate standard for making funding decisions. The success of biomedical research rests squarely on the robustness of NIH training

programs for the next generation of basic and clinical scientists. These training programs face many challenges: (i) the number of supported positions is insufcient to support all of the best applicants; (ii) stipends for graduate students have failed to keep up with ination; (iii) the relative paucity of new faculty positions over the last few years has forced many talented scientists to remain for long periods in postdoctoral positions; (iv) the typical age at which an investigator obtains his or her rst independent NIH grant support has risen to 40 or older; (v) training programs to encourage underrepresented minority participation have thus far generally failed to generate a scientic workforce that resembles the rest of the nation. Solutions in all these areas are badly needed. One initiative that could encourage earlier independence of the most talented young scientists would be a program modeled after the Whitehead Institute Fellows program, where carefully chosen scientists who have just obtained Ph.D., M.D., or M.D.-Ph.D. degrees are provided with laboratory space, technical support, nancial resources, and senior mentorship, but are allowed to pursue independent projects, effectively skipping over 5 years or more of postdoctoral training. Finally, it is time for NIH to develop better models to guide decisions about the optimum size and nature of the U.S. workforce for biomedical research. A related issue that needs attention, though it will be controversial, is whether institutional incentives in the current system that encourage faculty to obtain up to 100% of their salary from grants are the best way to encourage productivity. Recruiting, retaining, and empowering scientists from many disciplines to work together, supported by a stable trajectory for biomedical research support, are critical to realize the unprecedented opportunities that lie in front of us. It is time to be bold.
References and Notes
1. TCGA, http://cancergenome.nih.gov. 2. Human Microbiome Project, http://nihroadmap.nih.gov/ hmp/. 3. TRND, www.rarediseases.info.nih.gov/. 4. K. Takahashi, S. Yamanaka, Cell 126, 663 (2006). 5. K. Takahashi et al., Cell 131, 861 (2007). 6. J. Yu et al., Science 318, 1917 (2007). 7. Diabetes Prevention Program Research Group, N. Engl. J. Med. 346, 393 (2002). 8. J. A. Lieberman et al., N. Engl. J. Med. 353, 1209 (2005). 9. Healthy People, www.healthypeople.gov/. 10. D. A. Flockhart et al., Clin. Pharmacol. Ther. 86, 109 (2009). 11. Committee on the U.S. Commitment to Global Health, Board on Global Health, The U.S. Commitment to Global Health: Recommendations for the New Administration (National Academies Press, Washington, DC, 2009). 12. Enhancing Peer Review, http://enhancing-peer-review. nih.gov/. 10.1126/science.1185055

www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

37

PERSPECTIVES
CHEMISTRY

Molecular Donuts and Donut Holes

Kenton H. Whitmire

A transient template steers the self-assembly of a giant metal oxide wheel.

he molecular host-guest complex reported by Miras et al. on page 72 of this issue (1) provides insight into a remarkable system in which a template molecule forms, builds a host around itself, and then exits once its job is done. By using a ow reactor system, the authors capture an early stage in this process, where a molecular donut has its hole still inside (see the gure, panel A). In this complex, the two metal oxide anions are held together by a ring of Na+ ions and hydrogen bonds. The success of this approach suggests possibilities for nding other transient host-guest complexes and designing new open framework structures. At a donut shop, you might have the opportunity to buy not only donuts but some donut holes as well. So if you encounter a molecule or ion with a large void within it, you might ask yourself what happened to the molecular donut hole around which it formed. This is exactly the situation encountered by Miras et al. In a synthetic tour de force, the authors isolated an elliptical, attened donut-shaped ion containing 150 metal atoms, with the template that led to its formation still present (see the figure, panel A). Wheel-like structures have been known for some time, but understanding how they formed has been left to speculation (2, 3). The idea of creating molecules that selectively encapsulate other molecules or ions (4, 5) is now known as supramolecular chemistry. Widely used in chemistry, chemical engineering, biology, and materials science, it exploits secondary intermolecular interactions such as hydrogen bonding to direct molecular structure and reactivity. Not only are molecules designed to encapsulate specific molecules or ions, but molecules or ions also serve as templates for creating intricate structures around themselves. Complex molecular architectures with tailored physical and chemical properties have been synthesized in this way. For example, the size and shape of the cavities in zeolites (microporous silicon and aluminum oxide materials) can be determined by the presence of a templating molecule or ion that can then be removed, leaving the porous architecture intact (6). Similarly, tissue engineering makes use of molecular scaffolds to template the growth
Department of Chemistry, Rice University, Houston, TX 77005, USA. E-mail: [email protected]

Getting ready to leave. The host-guest complex reported by Miras et al. (1) (A) is stabilized by sodium atoms (gray) on the surface of the guest molecule. Addition of more electrons leads to elimination of the guest molecule. A similar situation is found in mica (B), a layered mineral that cleaves easily along the planes of the intercalating ions. For color coding in panel A, see (1).

of tissues and organs. Supramolecular assemblies determine the structures and functions of proteins, DNA, and cell membranes, and the principles that make them work so effectively can be employed in vitro to solve nonbiological problems in materials design or catalysis. For example, a designed molecule containing four iron atoms connected by organic molecules into a tetrahedral cage can reversibly encapsulate highly reactive P4 molecules, effectively producing a molecular bottle that isolates the molecules from the atmosphere, in which they would burn spontaneously (7, 8). Miras et al. describe the structure of Na22[MoVI36O112(H2O)16] [MoVI130MoV20 O442(OH)10(H2O)60]180H2O, abbreviated as Na22{Mo36}{Mo150}180H2O. They obtained the molecule by carefully controlling the degree

of reduction of a sodium molybdate solution in a ow system. The {Mo36} guest ion is known to form spontaneously in acidied solutions of Na2MoO42H2O. The oxidation state of Mo does not change, and consequently no reducing agent is needed to produce it. Under such conditions the {Mo150} host cannot form, because it requires the addition of 20 electrons, which are provided by introducing the reducing agent sodium dithionite. Given that both host and guest are negatively charged ions, one might think that electrostatic repulsions would prevent the system from being stable. However, this problem is circumvented by the binding of Na+ ions between the {Mo36}8 and {Mo150}14 ions. The host can thus effectively be viewed as forming around a positively charged [Nax{Mo36}](x8)+ rather than a negatively charged ion. This is not unlike the structure of mica, where slabs of aluminum and silicon oxide are separated by layers of metal cations such as sodium, magnesium, or calcium (see the gure, panel B). Some uses of mica come from the ready cleavage of the layers along the planes of the intercalating ions. A similar situation occurs in the host-guest complex described by Miras et al., where cutting the ions apart is initiated by addition of eight more electrons to {Mo150}. As a result, the electrostatic interactions become overwhelming, and the templating guest is eliminated. Both the {Mo36} template/guest and the {Mo150} host can be isolated independently after the reduction process is complete. It is not necessary to have a template involved in the formation of a hollow structure. In some cases, the bond lengths and angles for different metal ions in an oxide structure will lead to spontaneous curvature that can cause the structure to close on itself, giving a ring, sphere, or tube. Aluminosilicates accomplish this in nature by bringing together the appropriate ratio and arrangement of Si4+ and Al3+ ions; the slightly different bonding requirements of the two ions cause the array to curve spontaneously into allophane (9) or imogolite (10, 11)spherical and tubular metal-oxide analogs to fullerenes and carbon nanotubes, respectively. In the present case, not only is the curvature of the {Mo150} guest facilitated by the presence of the template, but it is also stabilized by the molybdenum ions adopting two different oxidation states, Mo6+ and Mo5+.

38

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

PERSPECTIVES
The discovery of an ephemeral guest template that forms its host and then leaves will no doubt prompt others to look for molecular donut holes. The ndings suggest new strategies for creating other open structures by judicious combinations of ions of different metals and/or the same metals in differing oxidation states.
References and Notes
1. H. N. Miras et al., Science 327, 72 (2010). 2. K. L. Taft, S. J. Lippard, J. Am. Chem. Soc. 112, 9629 (1990). 3. A. Muller, S. Roy, Coord. Chem. Rev. 245, 153 (2003). 4. D. J. Cram, Nature 356, 29 (1992). 5. J. M. Lehn, Science 227, 849 (1985). 6. P. Wagner, M. E. Davis, in Supramol. Organization and Materials Design, 83 (Cambridge Univ. Press, Cambridge 2002), pp. 83102. 7. P. Ma et al., Science 324, 1697 (2009). 8. 9. 10. 11. K. N. Raymond, Nature 460, 585 (2009). B. Cretonet al., J. Phys. Chem. C 112, 358 (2008). L. A. Bursill et al., Philos. Mag. A 80, 105 (2000). P. D. G. Cradwick et al., Nat. Phys. Sci. (Lond.) 240, 187 (1972). 12. Support for this work comes from the Robert A. Welch Foundation (C-0976) and the NSF (CHE-0719396). We are grateful to Adam Colson for help in preparation of the gure. 10.1126/science.1184533

ASTRONOMY

Serendipitous Astronomy
Kenneth R. Lang

Many of the seminal discoveries in astronomy have been unanticipated.

CREDIT: COURTESY OF THE CHARLES HAYDEN PLANETARIUM, MUSEUM OF SCIENCE, BOSTON

our hundred years ago, Galileo Galilei turned his newly constructed spyglass to the skies, and thus began astronomers use of novel telescopes to explore a universe that is invisible to the unaided eye. The search for the unseen has resulted in many important unexpected discoveries, including Jupiters four large moons, the planet Uranus, the rst asteroid Ceres, the large recession velocities of spiral nebulae, radio emission from the Milky Way, cosmic x-ray sources, gamma-ray bursts, radio pulsars, the binary pulsar with its signature of gravitational radiation, and the cosmic microwave background radiation. The observable universe is a modest part of a much vaster, undiscovered one that remains to be found, often in the least expected ways (1, 2). A few months after presenting the Venetian Doge a spyglass, or telescope, as a valuable tool of war, Galileo turned a telescope of his own making to the nearly full Moon, and on 7 January 1610, Jupiter appeared to him (3, 4). As luck would have it, the planet was then just above the Moon and the second brightest object in the sky (see the gure). Three, then four, Medicean stars were found accompanying Jupiter, orbiting the planet at speeds that decreased with distance from it. No one had predicted the possible existence of moons orbiting Jupiter. In the next century, two other unanticipated discoveries occurred when astronomers were observing stars with telescopes of thenunsurpassed quality. William Herschel was using his reecting telescope to survey bright stars, locating nearby faint stars that might help determine the distances of the bright ones. Giuseppe Piazzi was using a nely calibrated instrument to compile a catalog of accurate star positions. Herschel discovered
Department of Physics and Astronomy, Tufts University, Medford, MA 02155, USA. E-mail: [email protected]

Uranus on 13 March 1781 (5), and Piazzi found the first asteroid, Ceres, on 1 January 1801 (6); both objects moved against the background stars and were initially thought to be comets. In the early 20th century, Vesto Slipher unexpectedly helped us move beyond the stars into the expanding universe. Working at the Lowell Observatory in Flagstaff, Arizona, he was measuring the rotations of spiral nebulae, whose bright centers were then thought to be newborn starsthe surrounding spiral arms had been interpreted as nascent planetary systems. Using a spectrographic camera with a modest 24-inch (0.61 m) refractor telescope, he found, in 1917,

that the outward velocities of 25 spiral nebulae were well in excess of the velocity of any known cosmic object. Almost all of the spiral nebulae were moving away from the Earth, at astonishingly high velocities, up to 1100 km s1. This suggested to Slipher that the spiral nebulae were stellar systems, or island universes. By 1929, Edwin Hubble showed that the measured distances, established by him using the superb light-gathering power of the 100-inch (2.5 m) Hooker telescope on Mount Wilson, were roughly correlated with Sliphers velocities. This relationship is now attributed to the expanding universe, which no one had anticipated at the time Slipher made his measurements.

Looking up. A rendering of how the night sky of Padova, Italy, would have looked on 7 January 1610, 6:30 p.m., looking east.

www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

39

PERSPECTIVES
Other serendipitous discoveries revealed an entirely different, explosive universe, which was unveiled when new technologies let us capture invisible forms of radiation, rst as radio waves and then as x-rays and gamma rays. Radio noise of cosmic origin was, for example, inadvertently found in 1933 by Karl Jansky, working for the Bell Telephone Laboratory to determine the sources of interference with terrestrial radio communications. He built a rotating radio antenna that pointed sideways at the horizon instead of up into the sky. Fortunately, the broad eld of view, of 25, occasionally included the Milky Way, which produced an interstellar interference comparable in intensity to lightning discharges in distant thunderstorms. The new eld of radio astronomy nevertheless had to await the development of radio and radar technology during World War II and the subsequent construction of radio interferometers to provide angular resolving power. The radio galaxies that were then discovered emit unanticipated radio power, equivalent to the visible light of a million million (1012) stars which is attributed to electrons moving at nearly the speed of light. Enormous amounts of energy are also released from cosmic x-ray sources, which were discovered during a 5-min rocket ight primarily designed to detect x-rays from the Moon. As x-rays are absorbed in our atmosphere, cosmic x-ray sources must be observed with instruments launched above the obscuring air, in rockets or satellites. By the mid-20th century, such brief rocket ights had shown that the Sun radiates detectable x-rays, and it was thought that lunar material might also emit them when illuminated by solar x-rays. Riccardo Giacconis group at the American Science and Engineering Company (AS&E) tried to detect the Moons x-rays in 1962. Instead, they found evidence for the rst known, discrete x-ray source outside the solar system. The x-ray emission of any Sun-like star would be far too faint to be detected with existing equipment, due to the stars greater distances. Nevertheless, the AS&E group intended to search for other unknown cosmic x-ray sources. The consequences of their resulting discovery were enormous, leading to the development of focusing x-ray telescopes placed aboard a host of orbiting x-ray observatories and to the discovery of new cosmic objects, such as million-degree gas spiraling into stellar black holes. The most energetic sources found in the universe, at least so far, are the gamma-ray bursts whose duration is measured in seconds and which never reappear in exactly the same part of the sky. Their discovery was the unexpected result of defense satellite observations designed to detect clandestine nuclear bomb explosions in Earths atmosphere, on the Moon, or in outer space. As reported by R. Klebesadel, I. Strong, and R. Olson in 1973 (7), brief, intense gamma-ray bursts were instead found to be coming from the distant cosmos. The long-lasting afterglow of the cosmic gamma-ray bursts was eventually detected at visible wavelengths, enabling the distances to be determined by spectroscopy. We now know that some of these bursts radiate, for a few seconds, gamma-ray energy equivalent to the visible-light energy emitted by all the galaxies in the observable universe. Pulsars were discovered during a survey of the scintillations, or twinkling, caused when radiation from cosmic radio sources passes through the Suns winds. To study this effect, Antony Hewish and his colleagues at Cambridge University built a large array of 2048 dipole antennas connected to a radio receiver and chart recorder with a time constant of 0.1 s, the time scale of the scintillations. Graduate student Jocelyn Bell found a source of strong radio scintillation uctuating in the middle of the night, when the array was pointed away from the Sun and the effects of the solar wind should have been small. Further investigations led to the detection of periodic radio pulses, repeating with an exceedingly precise repetition period of 1.3372795 s. Within months, other large radio telescopes were used with rapid time sampling, rather than the long integration times formerly used to detect faint cosmic radio signals, resulting in the discovery of many more pulsars. No one had foreseen either the existence of pulsars or their subsequent interpretation as rotating neutron stars with intense, beaming magnetic elds. The first binary radio pulsar, designated PSR 1913+16, was found in 1975 by Russell Hulse and Joseph Taylor as the result of a highsensitivity, computerized search for new radio pulsars using the Arecibo Observatory. The orbital period of the two neutron stars was found to decrease with time, indicating that they are slowly approaching each other as they expend gravitational radiation, a scenario they had not anticipated but which has since provided tests of gravitation and general relativity (8). The Bell Telephone Laboratory provided the setting for yet another accidental discovery, involving a horn-reector antenna that had been used in the rst tests of a communication satellite. Arno Penzias and Robert Wilson were measuring the temperatures of all sources of noise in the horn-antenna system to make accurate measurements of the intensity of several extragalactic radio sources. But a persistent, ubiquitous, and unvarying noise source was detected, contributing an antenna temperature of 3.5 1.0 K, equally strong in all directions, wherever the antenna was pointed, independent of the time of the day and of the year and with no dependence on the location of any known cosmic radio source. Penzias and Wilson did not know what they had found and avoided any mention of the cosmological implications in their publication. But a group at Princeton University, which was attempting to make a similar measurement at the time, drew attention to the implications in a companion paper. So this particular discovery was not entirely unanticipated. In the late 1940s and early 1950s, George Gamow, Ralph A. Alpher, and Robert C. Herman had previously speculated that the hot, 109 K radiation of the early universe would still be around, cooled to about 5 K over the past 14 billion years of expansion. Penzias and Wilson were nevertheless unaware of this work until after their discovery of what is now known as the cosmic microwave background radiation. So our celestial science seems to be primarily instrument-driven, guided by unanticipated discoveries with unique telescopes and novel detection equipment. With our current knowledge, we can be certain that the observed universe is just a modest fraction of what remains to be discovered. Recent evidence for dark, invisible matter and mysterious dark energy indicate that the main ingredients of the universe remain largely unknown, awaiting future, serendipitous discoveries.
References and Notes
1. An account of our gradual awareness of a vast, previously concealed universe, primarily as the result of unanticipated discoveries, is given in (9). 2. Many of the seminal papers mentioned in this Perspective have been reproduced with commentary in (10). 3. G. Galilei, Sidereus Nuncius (Ventis, Apud Thoman Baglionum, 1610). 4. A. van Helden, The Sidereal Messenger (Univ. Chicago Press, Chicago, 1989). 5. W. Herschel, Phil. Trans. R. Soc. 71, 497 (1781). 6. G. Piazzi, Risultati delle Osservazioni della Nuova Stella scoperta il d 1. Gennajo allOsservatorio Reale di Palermo (Nella Reale Stamperia, Palermo, 1801). 7. R. W. Klebesadel, I. B. Strong, R. A. Olson, Astrophys. J. 182, L85 (1973). 8. J. A. Taylor, J. M. Weisberg, Astrophys. J. 253, 908 (1982). 9. K. R. Lang, Parting the Cosmic Veil (Springer, New York, 2006). 10. K. R. Lang, O. Gingerich, A Source Book in Astronomy and Astrophysics 19001975 (Harvard Univ. Press, Cambridge, MA, 1979). 11. This Perspective is an abbreviated version of an invited lecture, The Serendipitous Nature of Astronomical Discovery, given at the Joint International Astronomical UnionINAF Astronomical Observatory of Padova Symposium Astronomy and its Instruments Before and After Galileo, held in Venice-San Servolo Isle, 28 September to 3 October 2009. 12. Special thanks to L. Pigatto for informed discussions about Galileo, and to J. Bredekamp for support from the NASA AISR program. 10.1126/science.1183653

40

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

PERSPECTIVES
CHEMISTRY

Janus Catalysts Direct Nanoparticle Reactivity

David J. Cole-Hamilton

Metal oxide nanoparticles decorated with carbon nanotubes can be turned into readily recovered catalysts that function at the interface between oil and water phases.

m going to set you a challenge. Achieving high reaction rates Go and make a cup of tea. Add depends on creating as much interfaWAT E R OIL milk and sugar, and stir well. Now, cial area between the phases as possiplease get just the sugar back out for ble for the catalyst to do its job, which me. Difcult, isnt it? The same probis accomplished by vigorous stirring lem faces chemists who want to make to create an emulsion. The silica form synthetic products more environmenof these nanoparticles can stabilize O tally friendly. Soluble compounds either water-in-oil or oil-in-water O O +O that are used to speed up desired reacemulsions (6) and accumulate at high OH tionshomogeneous catalystscan concentrations at the interfaces. end up in nal products, where they Crossley et al. added a catalytipose a nightmare of a separation probcally active metalpalladium (Pd) lem. Ideally, if these catalysts could to these nanoparticles to create cata+ H2 be completely recovered, they could lysts that achieve high reaction rates be recycled and kept out of the prodin biphasic reactions through the ucts, in which they could be toxic even combination of high interface conat trace levels. One general approach centration and high interfacial surO O OH OH to recovering such catalysts is phase face area. The announcement of rate transfer, which takes advantage of the enhancements alone is an important different solubility of compounds in development, but the authors furwater versus organic solvents. On page ther advance the eld by using their 68 of this issue, Crossley et al. (1) have catalysts in reactions to form fuels converted solid nanoparticles that About face for catalysts. Reactions at the interface between organic from bio-oils. Naturally derived have solubility in both water and oils and aqueous phases (blue and gray areas, respectively) can be catalyzed molecules are upgraded for use as into catalysts that can operate in both by different parts of the same nanoparticle. Crossley et al. grew carbon fuels by coupling small molecules phases. These catalysts can be recov- nanotubes (shown in white) on metal oxide nanoparticles (shown in together, removing oxygenated ered even from complex mixtures, orange) that, like the Roman god Janus, present two faces. The oxide groups, or both. such as those that result when biomass surface prefers to be in water and the carbon nanotubes prefer to be in The removal of oxygenated groups oil, so these particles seek out water-oil interfaces. The addition of palproducts are upgraded into fuels. illustrates the power of this method. ladium (shown in yellow) to these nanoparticles creates catalysts that Phase separation between organic Acid and alcohol groups increase the can work in both phases. In the system depicted here, basic magnesium and aqueous phases has been oxide catalyzes a coupling reaction of 5-methylfurfural and acetone water solubility of molecules, but if exploited in commercial methods for that is useful in biofuel production. The product transfers to the oil the molecules are coupled together in catalyst recovery (2), but the success phase, where a palladium catalyst attached to the carbon nanotubes a way that removes these groups, then of these approaches depends on the catalyzes a subsequent hydrogenation reaction. The nanoparticle cata- the desired product will be more solparticular system. For example, if lyst can be separated after reaction via ltration. uble in the organic phase. The catathe reactants have some solubility in lysts not only accelerate the coupling water, and the products have less solubility, Crossley et al. have developed nanopar- reactions, but the biphasic system also allows then separation can be achieved by modify- ticles that selectively locate at the interface the desired product to move into the organic ing the catalyst so that it dissolves in water. between the aqueous and organic phases. phase and leave everything else, including the The catalyst can be recovered at the end of They deposited carbon nanotubes on metal recoverable catalyst, in the aqueous phase. In the reaction by simply decanting the oil from oxide nanoparticles, such as silicon oxide systems with a single liquid phaseeven the water (3). However, the reaction takes (silica) or magnesium oxide, with diameters ones that use insoluble heterogeneous cataplace in the aqueous phase, so if the water of 50 nm or less. The oxides are hydrophilic lysts on solid supportsall of the products solubility of the reactants is too low, the and attracted to the water, while the carbon would remain together in solution. Separation reaction may be unacceptably slow. Some nanotubes are hydrophobic and prefer the would require complex distillation steps that recent approaches have tackled this prob- organic layer (hence they are described as might decompose or rearrange the thermally lem through the use of additives (4, 5) that Janus particles, like the two-faced character sensitive products. improve catalyst solubility. of mythology). Like a large surfactant molCrossley et al. also exploit another asymecule, the nanoparticles compromise by sit- metry of these Janus particles: The carbon ting at the interface. Unlike surfactants, the nanotubes grown on silica are more perfect Eastchem, School of Chemsitry, University of St. Andrews, nanoparticles are solids that can be separated than those grown on magnesium oxide. When St. Andrews, Fife KY16 9ST, Scotland, UK. E-mail: djc@ st-and.ac.uk with methods such as ltration. they deposit the Pd catalyst in the silica syswww.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

CREDIT: P. HUEY/SCIENCE

41

PERSPECTIVES
tem, almost all of it binds on the silica side. Because the catalyst orients itself at the interface with the silica in contact with the water, and the carbon nanotubes in the organic phase, the catalyst is predominantly in contact with the water. Reactants that are water-soluble react more readily in the emulsion than those in the organic phase. Thus, the catalyst will select among reactants in a complex mixture only on the basis of water solubility. When the Pd catalyst is deposited in the magnesium oxide system, some of it now binds to the carbon nanotubes at defect sites. Hydrogenation reactions can occur in the organic phase, and the magnesium oxide, which is basic, catalyzes coupling reactions in the aqueous phase. The result is that different reactions can be run in the oil and in the water phases at the same time (see the gure). One intriguing extension would be to rethink conventional organic synthesis of drug molecules, which depends on several cycles of reaction and separation. A complex molecule could undergo reaction on just one of its functional groups in water. The product could rearrange and become organic-soluble, and then transfer to the other phase, where a different reaction could occur. In this way, a series of reactions could create a desired molecule in one pot, without needing to protect vulnerable functional groups in the reactants. Simple ltration followed by phase separation and solvent removal would then give the desired product in high yield and containing no catalyst residuesa pharmaceutical chemists dream come true.
References
1. S. Crossley, J. Faria, M. Shen, D. E. Resasco, Science 327, 68 (2010). 2. D. J. Cole-Hamilton, Science 299, 1702 (2003). 3. E. Wiebus, B. Cornils, in Catalyst Separation, Recovery and Recycling: Chemistry and Process Design, D. J. ColeHamilton, R. P. Tooze, Eds. (Springer, Dordrecht, Netherlands, 2006), pp. 105143. 4. S. L. Desset, S. W. Reader, D. J. Cole-Hamilton, Green Chem. 11, 630 (2009). 5. C. Machut et al., Angew. Chem. Int. Ed. 46, 3040 (2007). 6. M. Shen, D. E. Resasco, Langmuir 25, 10843 (2009). 10.1126/science.1184556

BIOCHEMISTRY

Some Enzymes Just Need a Space of Their Own

Sebyung Kang and Trevor Douglas

Protein shells that sequester enzymatic reactions are found in diverse organisms and may provide blueprints for nanoparticle design.

n many ways biology is dened by the idea (and reality) of containers and welldened barriers. These enable the distinction of self from the rest of the universe, separation of cells from each other, and the definition of organelles within a cell. Although many biological barriers and compartmental boundaries are lipid-based membranes, there is a growing awareness, brought about by some spectacular structural biology, of protein-based compartments that act as isolated environments within the cell. On page 81 of this issue, Tanaka et al. (1) add to the growing number of examples of protein-based microcompartments, reporting the structure of a microcompartment that sequesters ethanolamine metabolism in the bacterium Escherichia coli. These protein-based containers challenge a long-standing assumption that bacteria and archaea, which lack membrane-enclosed organelles, are devoid of internal compartmentalization. Protein-based compartments often exhibit highly symmetric structures assembled from a limited number of subunit building blocks. These structures are reminiscent of viruses and share properties (2) that make them more than just curiositiesthey incorporate catalytic activities and sequester reactions from the cellular environment, and control smallDepartment of Chemistry and Biochemistry and Center for Bio-Inspired Nanomaterials, Montana State University, Bozeman, MT 59717, USA. E-mail: tdouglas@chemistry. montana.edu

molecule access across the protein barrier through structural dynamics. Tanaka et al. report the three-dimensional crystal structures of the major shell constituents (EutS, EutL, EutK, EutM) of an ethanolamine utilization (Eut) microcompartA C

ment, where ethanolamine metabolism is isolated in E. coli (see the gure). The Eut microcompartment prevents the release of acetaldehyde into the cytosol, mitigating the potentially toxic effects of excess aldehyde and limiting the loss of carbon by diffusion
D

F
100 A 1000 A

Microcompartment architectures. (A) Model of the ethanolamine utilization (Eut) microcompartment. (B) Model of the carboxysome microcompartment. (C) Shell structure of T. maritima encapsulin (PDB: 3DKT). (D) Shell structure of B. subtilis lumazine synthase (PDB: 1RVV). (E) Three-dimensional reconstruction of bovine pyruvate dehydrogenase complex [derived from (6) with permission from the National Academy of Sciences, USA]. (F) Structure of human ferritin (PDB: 2HFA). Structures in (C), (D), and (F) were generated with the molecular graphics program Chimera, with the indicated PDB les.

42

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

PERSPECTIVES
of acetaldehyde across the cell membrane. Although the four Eut shell proteins have similar protein folds, the individual proteins have distinguishing structural features that suit them to specialized architectural and biochemical roles within the microcompartment. EutM and EutS share a conserved / core structure that serves as the basic assembly component of the shell. EutL has distinct open and closed crystal forms, which are potentially involved in gated molecular transport through pores in the protein shell. The crystal structure of the carboxyl-terminal domain of EutK suggests that it is a nucleic acid binding protein. Overall, the unique functions of this microcompartment arise from well-dened combinations of its protein components. Among the examples of protein-based microcompartments found in nature (see the gure) is the carboxysome, which enhances CO2 fixation inside many photosynthetic and chemoautotrophic bacterial cells by encapsulating the key enzymes ribulose-1,5bisphosphate carboxylase-oxygenase and carbonic anhydrase. It has a roughly polyhedral proteinaceous outer shell with a diameter of 800 to 1400 (3). Encapsulin, isolated from Thermotoga maritima, consists of a thin icosahedral shell with a diameter of 240 , formed from 60 copies of a single monomer (4). Its interior surface is lined with conserved sites that bind to the carboxyl termini of enzymes involved in oxidative stress response, such as peroxidase. Lumazine synthase, from Bacillus subtilis, similarly forms an icosahedral shell and catalyzes the formation of 6,7-dimethyl-8-ribityllumazine in the penultimate step of riboavin biosynthesis (5). This icosahedral shell encloses a trimer of a complex called the riboavin synthase forming bifunctional enzyme complex (formerly called heavy riboavin synthase). Similarly, the pyruvate dehydrogenase complex serves as the link between glycolysis and the tricarboxylic acid cycle in many organisms. This complex assembles from multiple copies of three enzymes (a decarboxylase, a dihydrogenase, and a avoenzyme) to form the icosahedral shell (6). Some protein cage architectures even serve as reaction chambers for inorganic chemistry. Ferritin, for example, is an octahedral protein cage with 24 structurally identical subunits (7). It is found in all domains of life and forms an iron oxide nanoparticle in its cavity as a storage mechanism for maintaining iron homeostasis. There is a growing interest in mimicking protein compartments for synthetic applications involving the encapsulation and sequestration of catalyst species. To this end, the icosahedral capsid of the cowpea chlorotic mottle virus has been used as a molecular container to encapsulate an individual enzyme of horseradish peroxidase (8), ferritin has been chemically derivatized to incorporate a catalytically active organometallic moiety (7), and a smaller cage constructed from a Listeria innocua DNA binding protein has been modied with catalytically active platinum clusters for light-driven H2 production (9). The advantage offered by these encapsulated catalytic systems lies in the ability to sequester and protect active sites of enzymes, control substrate access, and maintain separation between competing reactions. The potential for artificially incorporating multiple catalytic centersperhaps part of a pathway or cascade where high local substrate concentrations can be maintainedhas yet to be realized, but the large protein microcompartments, such as the Eut system and large virus-like particles, have the capacity for this level of molecular engineering. The protein shells of the microcompartments reported by Tanaka et al. are not merely static walls or barriers between the interior and exterior environments; they are dynamic skins. Conformational flexibility encoded within protein shells allows selective transportation of components across the barrier and prevents leakage of toxic materials while retaining reaction intermediates. A deep understanding of such fine-tuned dynamics, superimposed on the static architectural features of these microcompartments, may enable the development of bioinspired multifunctional and dynamic nanomaterials as well as fundamental new insights into complex reactions within the cell.
References and Notes
1. S. Tanaka, M. R. Sawaya, T. O. Yeates, Science 327, 81 (2010). 2. T. Douglas, M. Young, Science 312, 873 (2006). 3. S. Tanaka et al., Science 319, 1083 (2008). 4. M. Sutter et al., Nat. Struct. Mol. Biol. 15, 939 (2008). 5. K. Ritsert et al., J. Mol. Biol. 253, 151 (1995). 6. Z. H. Zhou, D. B. McCarthy, C. M. OConnor, L. J. Reed, J. K. Stoops, Proc. Natl. Acad. Sci. U.S.A. 98, 14802 (2001). 7. S. Abe et al., J. Am. Chem. Soc. 130, 10512 (2008). 8. M. Comellas-Aragones et al., Nat. Nanotechnol. 2, 635 (2007). 9. S. Kang et al., Angew. Chem. Int. Ed. 47, 7845 (2008). 10. T.D. is a founder of SpeciGen, a biotechnology com-

pany with a focus on protein cages for drug delivery, and has consulted with them in the past 3 years.
10.1126/science.1184318

NEUROSCIENCE

Brain Activity to Rely On?

D. Sam Schwarzkopf 1 and Geraint Rees 2

The characteristics of neuronal activity that mark whether consciousness arises include how reproducible neuronal response patterns are to a sensory stimulus.

T
1

he human brain is a noisy place. The responses of single neurons to sensory stimuli are highly variable. Yet our conscious experience of the environment is stable and consistent. How can a stable conscious representation of the environment arise from noisy individual neuronal responses? Not all activity in the brain reaches consciousness, so one way to address this question is to exam-

UCL Institute of Cognitive Neuroscience, University College London, 17 Queen Square, London WC1N 3AR, UK. 2 Wellcome Trust Centre for Neuroimaging, University College London, 12 Queen Square, London WC1N 3BG, UK. E-mail: [email protected]

ine how neurons respond when we are aware of sensory stimuli relative to when stimuli are unnoticed and invisible. On page 97 of this issue, Schurger et al. (1) use functional magnetic resonance imaging (fMRI) to show that in the human brain, visible stimuli that enter awareness elicit spatial patterns of neuronal activity that are more reproducible than for invisible stimuli. This difference may be useful in studying brain function in conditions such as coma or schizophrenia. fMRI indirectly measures the activity of large populations of neurons simultaneously at many points in the brain, and has provided

much information about the functional organization of visual processing in the human brain. Even the local spatial pattern of fMRI signals over a few centimeters carries information about underlying neuronal processing (2). Schurger et al. have discovered that the consistency of the local spatial patterns of activity in sensory cortex elicited by stimuli changes, depending on whether the stimuli reach awareness or not. The authors examined brain responses while observers viewed images of faces or houses. Images were presented to each eye simultaneously. If the same image is shown

www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

43

PERSPECTIVES
to both eyes, it is clearly visible. single stimulus enters consciousActivity patterns of neuronal populations However, when images of opponess or not, but must indicate site color contrast are presented to something about the nature of High reproducibility Low reproducibility each eye, the visual system fuses the underlying mechanisms. One (visible stimulus) (invisible stimulus) the two images, and the resultant attractive possibility is that recurperception is of a uniformly colrent connections between differTrial 1 ored yellow eld; thus, the indient brain areas reduce the varividual images are rendered invisiability of responses for a stimuble. Schurger et al. examined neulus that is consciously perceived. ronal populations in the human Both the strength (9) and direction ventral visual cortex whose activ[feedforward or feedback (10)] ity discriminated whether a face or of connections may be impora house was being presented. They tant determinants of conscious Trial 2 found that the consistency of the processing. Reduced variability spatial patterns of activity in these resulting from recurrent processlocations differed depending on ing might align the responses of whether the stimulus was visible stimulus-selective neurons across or invisible. If invisible, the patdifferent brain regions. This might terns were more variable. Thus, explain why Schurger et al. were invisibility of a stimulus is assounable to discriminate whether a Trial 3 ciated with lower reproducibility visible stimulus was a face or a of responses, and suggests that house from the responses of those reproducibility may be an imporbrain regions that encoded invistant difference between conscious ible stimuli. Time and unconscious sensory repreEven in the absence of any sentations (see the gure). stimulation, the brain is active The main challenge for any and exhibits highly correlated study that seeks to compare con- Reproducibility. These simulated data illustrate hypothetical activity patterns of spontaneous ongoing response scious and unconscious stimulus neuronal populations in response to three presentations of a stimulus, measured patterns (11). Although this activwith fMRI in small areas of the human visual cortex. Although each response patprocessing is that changes in brain ity has sometimes been regarded tern in the left column carries some noise, the reliability of the pattern across activity may be simply due to three presentations is apparent. The right column illustrates different response as noise, it has been suggested changes in the stimulus (3). Thus, patterns, but of lower reliability. The difference in reliability between the left and that incoming sensory stimulamany studies compare physically right columns is analogous to the difference in response patterns in the human tion sculpts the activity and temidentical stimuli that differ only temporal lobe for visual stimuli that either entered awareness and were visible, poral dynamics of these spontain whether they reach awareness. or remained outside awareness and invisible (1). neous networks (12). An alternaFor example, if two deliberately tive explanation for the ndings incompatible images are presented to each very ne eye movements (microsaccades). of Schurger et al. is that conscious perception eye (binocular rivalry), conscious perception Such differences will affect the responses of of a stimulus has a direct effect on these sponalternates spontaneously between each mon- visual neurons and hence the reliability of taneous network dynamics, reducing the variocular view (4). As stimulation remains con- responses in the visual cortex. ability of their uctuations. The explanation stant, changes in conscious contents become What implications do these ndings have for the reduced variability of local spatial patdissociated from changes in the physical for our understanding of how consciousness terns of responses associated with conscious stimulus. Schurger et al. instead used binocu- arises in the brain? Several characteristic sig- perception awaits further empirical work. lar fusion of images presented to both eyes, natures of neural activity that are correlated References which allows direct experimental control of with consciousness have been proposed (7). 1. A. Schurger, F. Pereira, A. Treisman, J. D. Cohen, Science whether a stimulus enters awareness. How- One suggests that the intensity or the dura327, 97 (2010); published online 12 November 2009 ever, in this approach, there is physical vari- tion of activity in sensory neurons must pass a (10.1126/science.1180029). 2. J. D. Haynes, G. Rees, Nat. Rev. Neurosci. 7, 523 (2006). ability between stimuli when opposite color threshold to reach awareness. Visible images 3. C. Frith, R. Perry, E. Lumer, Trends Cogn. Sci. 3, 105 contrast images are viewed (making the face can elicit greater activity in some regions of (1999). or house stimuli invisible), compared to when visual cortex than invisible stimuli (8). Alter4. P. Sterzer, A. Kleinschmidt, G. Rees, Trends Cogn. Sci. 13, 310 (2009). identical images are viewed (visible stimuli). natively, consciousness might be correlated 5. I. Ohzawa, R. D. Freeman, J. Neurophysiol. 56, 221 (1986). Many neurons in the primary visual cortex, with the synchrony of responses in stimulus6. J.-D. Haynes, R. Deichmann, G. Rees, Nature 438, 496 even if they respond selectively to one eye selective neuronal assemblies. Schurger et al. (2005). 7. G. Rees, Philos. Trans. R. Soc. Lond. Ser. B 362, 877 (5), modulate their responses when dissimilar now add that the reproducibility of responses (2007). images are presented to each eye (6). Thus, is a marker of conscious processing. 8. K. Moutoussis, S. Zeki, Proc. Natl. Acad. Sci. U.S.A. 99, response variability at this early stage of Reproducibility of response patterns to 9527 (2002). 9. J.-D. Haynes, J. Driver, G. Rees, Neuron 46, 811 (2005). visual processing might be transmitted on to a sensory stimulus can by denition only be the higher visual areas explored by Schurger assessed across multiple presentations of that 10. V. A. F. Lamme, Trends Cogn. Sci. 10, 494 (2006). 11. B. Biswal et al., Magn. Reson. Med. 34, 537 (1995). et al. In addition, physically different visible stimulus. But we do not have any difculty 12. T. Kenet et al., Nature 425, 954 (2003). and invisible stimuli may elicit differences in seeing a single stimulus on its own. So repro10.1126/science.1184242 involuntary behavior such as the pattern of ducibility per se cannot determine whether a

44

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org

PERSPECTIVES
RETROSPECTIVE

Rossiter H. Crozier (19432009)

Jacobus J. Boomsma and Pekka Pamilo

A geneticists fascination with ants and other social insects yielded deep insights into the evolution of animal societies.

CREDIT: A. WILD/UNIVERSITY OF ILLINOIS

n 12 November, Ross H. Crozier died unexpectedly from cardiac arrest while at work in his laboratory. He pioneered the use of genetic markers for the study of social insects, made crucial contributions to the theory of social evolution, and was a unique mentor for many who continued in his footsteps. About 40 years ago, the study of social evolution shifted gears after seminal papers by William D. Hamilton made it clear that the evolution of reproductive altruism could be explained by considering inclusive tness (the number of gene copies in future generations as they are passed on both via own offspring and the offspring of relatives). This crucial insight allowed Robert L. Trivers to overturn the notion that insect societies are harmonious by default. But these advances only accelerated developments in social insect evolutionary biology after Ross Crozier established the rst genetic and statistical techniques to estimate relatedness in groups where pedigrees are unknown. Throughout his career, he bridged the gap between inclusive tness theory and empirical genetics and mentored two academic generations who applied and elaborated his pioneering initiatives. Ross Crozier was born in an Australian family living in India during World War II. Already during his school years he was a keen ant observer, an interest that would later shape his academic career. He graduated from the University of Melbourne, where he was inuenced by the famous evolutionary cytogeneticist Michael J. D. White, himself a student of J. B. S. Haldane. Crozier went on to do his Ph.D. at Cornell University under the renowned ant biologist William L. Brown. His rst scientic papers in the late 1960s mapped chromosomal variation in a large number of ant species. During his early research, Crozier quickly grasped the potential of using molecular markers to infer family relationships and the importance of inclusive tness modeling. In the early 1990s, he became the rst to take the social insects into the genomics era by sequencing the honeybee mitochondrion. At
1

Department of Biology, University of Copenhagen, Universitetsparken 15, 2100 Copenhagen, Denmark. 2Department of Biosciences, Box 65, 00014 University of Helsinki, Finland. E-mail: [email protected]; pekka.pamilo@ helsinki.

this time, sequencing was a highly demanding endeavor because it had to be done manually. In 1993, he published the honeybee mitochondrial genome, with his wife Ching as coauthor (1). It was the second insect mitochondrion to be sequenced, preceded only by Drosophila, and it paved the way for the later honeybee genome project. Soon after receiving his Ph.D. degree in 1969, Crozier became an assistant professor at the University of Georgia. By the time he returned to Australia in 1975, he had initiated the lines of research that were to establish his lasting inuence. He showed that the coeff icients of relatedness, which are fundamental in Hamiltons inclusive tness theory, must be formulated differently for ants, bees, and wasps, where males develop from unfertilized eggs. This meant that the entire richness of relatedness differences in the colonies of these insects could be used to test predictions from inclusive tness theory. He retained his focus on ants, in which differences in relatedness within and among colonies should hold many of the keys for explaining the expression of reproductive conict from rst principles. Ross Crozier was also the rst to realize that Hamiltons concept of animals being able to discriminate between kin, non-kin, and different degree of kin needed genetic underpinning. He formalized the challenges and paradoxes involved, and identied several ways by which kin recognition can evolve. This started a literature in which his original models continue to be cornerstones. He later summarized the state of the eld in a monograph with one of us (2). The advances in DNA technology also gave him effective tools to make pioneering contributions to the reconstruction of ant phylogenetic trees and for applying molecular phylogenetics to conservation biology. Crozier had long-lasting research interests in social insect immunity and the genetics of sex and caste determination, and he kept returning to these topics with new

molecular tools. There are very few areas of social insect evolutionary biology in which he was not involved as a pioneer or later innovator. He aptly demonstrated his scholarship in recent essays on the root of the ant phylogeny (3) and the origin of eusociality (4), in which he opined on recent controversies in his typically modest but insightful manner. In recognition of his contributions to the eld, Crozier became the rst recipient of the Hamilton Award of the International Union for the Study of Social Insects in 2006. His characteristic combination of conceptual synthesis and innovative tool development made his laboratories at the University of New South Wales in Sydney, La Trobe University in Melbourne, and James Cook University in Townsville ideal training sites for young researchers and sabbatical havens for senior researchers. Visitors were met with great hospitality and became quickly immersed in the friendly innovative atmosphere that surrounded him. Croziers inuence has been much more substantial than the list of his coauthors suggests: He created a eld and remained an active participant and inspiring tutor through a vast e-mail correspondence across the globe, the messages leaving his home computer often in the evenings or during the weekends. The thoughtful and scholarly constructive way in which Ross Crozier interacted with his peers and students was highly appreciated in the learned societies in which he was active and by the editors of the scientic journals that he served. He will be sadly missed for the new ideas he continued to generate and for his insightful counseling.
References
1. R. H. Crozier, Y. C. Crozier, Genetics 133, 97 (1993). 2. R. H. Crozier, P. Pamilo, Evolution of Social Insect Colonies (Oxford Univ. Press, Oxford, 1996). 3. R. H. Crozier, Proc. Natl. Acad. Sci. U.S.A. 103, 18029 (2006). 4. R. H. Crozier, Aust. J. Entomol. 47, 2 (2008). 10.1126/science.1185061

www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

45

REVIEW Lipid Rafts As a MembraneOrganizing Principle

Daniel Lingwood and Kai Simons* Cell membranes display a tremendous complexity of lipids and proteins designed to perform the functions cells require. To coordinate these functions, the membrane is able to laterally segregate its constituents. This capability is based on dynamic liquid-liquid immiscibility and underlies the raft concept of membrane subcompartmentalization. Lipid rafts are fluctuating nanoscale assemblies of sphingolipid, cholesterol, and proteins that can be stabilized to coalesce, forming platforms that function in membrane signaling and trafficking. Here we review the evidence for how this principle combines the potential for sphingolipid-cholesterol self-assembly with protein specificity to selectively focus membrane bioactivity. he lipid raft hypothesis proposes that the lipid bilayer is not a structurally passive solvent, but that the preferential association between sphingolipids, sterols, and specific proteins bestows cell membranes with lateral segregation potential. The concept has long suffered assessment by indirect means, leading to questions of fact or artifact (1). The resistance of sphingolipid, cholesterol, and a subclass of membrane proteins to cold detergent extraction (2) or mechanical disruption (3) has been widely used as an index for raft association with little or no regard for the artifacts induced by these methods. Though the acquisition of resistance to disruption may point to physiologically relevant biases in lateral composition (4), this disruptive measure tells us little about native membrane organization. Support from light microscopy was also missing because, with the exception of organization into specialized membrane domains such as caveolae or microvilli, putative raft componentsspecifically glycosylphosphatidylinositol (GPI)anchored proteins, fluorescent lipid analogs, raft transmembrane (TM) domains, and acylated proteinsoften show a homogeneous distribution at the cell surface (5). Moreover, early investigations into submicron membrane organization often yielded conflicting evidence regarding the distribution or motion of these constituents in the living cell (1). Today, however, the advancement of technology has produced compelling data that self-organization of lipids and proteins can induce subcompartmentalization to organize bioactivity of cell membranes. pids, the region of chemical linkage between the head group and sphingosine base contains both acceptors and donors of hydrogen bonds, thus increasing associative potential, both with cholesterol and other sphingolipids (11). Other explanations for cholesterol selectivity include the proposed umbrella effect, in which cholesterol hydrophobicity is preferentially shielded by the strongly hydrated head groups of sphingolipid (15) or stoichiometric, but reversible, complex formation between cholesterol and sphingolipid or saturated glycerophospholipid (16). Immiscible liquid phase coexistence in vitro has been suggested as the physical principle underlying rafts in vivo (17). Of central importance is the demonstration of selectivity in association between certain lipids. However, phase separation in simple systems at thermodynamic equilibrium in vitro cannot be translated into proof for membrane domain formation in living cells (1). Instead, model-membrane work emphasizes the fact that certain lipids exhibit preferential association and provides a framework for understanding how heterogeneity in cell membranes may arise (18). In this respect, the terms Lo and Ld should not be applied to the living cell, as they refer only to the liquid-ordered and liquid-disordered phases of model-membrane systems where parameters relating to translational order (lateral diffusion) and conformational order (trans/gauche ratio in the acyl chains) can be accurately measured (11). Glimpses of Nano-Assemblies in Living Cells Currently, lipid rafts are viewed as dynamic nanoscale assemblies enriched in sphingolipid, cholesterol, and GPI-anchored proteins (19) (Fig. 2A). To reach this viewpoint, membrane research has had to contend with the observers effect, akin to Heisenbergs uncertainty principle: We can change and/or induce heterogeneity in membranes simply by trying to observe it. Initially, this required moving away from detergent extraction as a means to infer native organization. In a first step, detergentfree, chemical cross-linking of GPI-anchored proteins at the plasma membrane suggested that the intrinsic heterogeneity by rafts was present in nanoscale complexes below the optical resolution limit set by the diffraction of light (19). This nanometer-size scale was later supported by viscous drag measures of antibody-bound raft proteins (21) and electron microscopic observation of immunogold-labeled raft antigens (20). Indeed, recent near-field scanning optical microscopy has confirmed a nanoscale bias in the distribution of raft-associating proteins in fixed cells (22). Less perturbing measures of spatial and temporal dynamics in living cells have also provided correlating data. For example, single-particle tracking of colloidal goldlabeled GPI-anchored receptors reveals stimulation-induced temporary arrest of lateral diffusion, or STALL, in short-lived (~0.5-s) 50-nm areas as a bioactive feature of receptor function (23). Parallel advances in microscopy and spectroscopy have revealed similar heterogeneity

planation: Self-associative properties unique to sphingolipid and cholesterol in vitro could facilitate selective lateral segregation in the membrane plane and serve as a basis for lipid sorting in vivo (7). This proposal for compartmentalization by lipid rafts suggested a nonrandom membrane architecture specifically geared to organize functionality within the bilayer. This function was initially thought to be membrane trafficking; however, rafts could influence organization of any membrane bioactivity (Fig. 1). Here, we highlight advances in technology that point to the existence of raft-based membrane heterogeneity in living cells and discuss the levels of preferential association underlying dynamic domain structure and biological function(s). Lipid Interactions in Model Membranes Assembly into two-dimensional liquid crystalline biomembranes is a fascinating property characteristic of lipids. Long thought to be incapable of coherent lateral structure (8), it is now apparent that principles of lipid self-association can also confer organization beyond nonspecific measures of fluidity. An important advance in modelmembrane systems was the discovery of phase separation in wholly liquid bilayers (9, 10). It is a cholesterol-dependent lateral segregation, wherein the planarity (molecular flatness) of the rigid sterol ring favors interaction with straighter, stiffer hydrocarbon chains of saturated lipids and disfavors interaction with the more bulky unsaturated lipid species (11). Interaction with cholesterol also forces neighboring hydrocarbon chains into more extended conformations, increasing membrane thickness and promoting segregation further through hydrophobic mismatch (12). In purified lipid systems, the combined effect is a physical segregation in the membrane plane: A thicker, liquid-ordered, Lo phase coexists with a thinner, liquid-disordered, Ld phase (13). Sphingolipids tend to display longer and more saturated hydrocarbon chains, thus potentiating interdigitation between leaflets (14) and favoring interaction with cholesterol. Moreover, unlike glycerophospholiVOL 327 SCIENCE

Origins of the Lipid Raft Concept Biochemically, it is clear that lipids are sorted within the cell (6). This is particularly notable in polarized epithelia where glycosphingolipids (GSLs) are enriched at the apical surface (7). Lipid rafts were originally proposed as an exMax Planck Institute for Molecular Cell Biology and Genetics, Pfotenhauerstrasse 108, 01307 Dresden, Germany. *To whom correspondence should be addressed. E-mail: [email protected]

46

1 JANUARY 2010

www.sciencemag.org

REVIEW

1970s

R O O

R O

C HC HC
O

Glycerolipids

Sphingolipids

Influenza

Apical membranes and budded viral envelopes reveal selective sorting of sphingolipid in polarized epithelia

VSV

The steroldependent Lo phase is discovered in model membranes

1990s
Detergent-resistant membranes suggest sterol-dependent sphingolipid and protein association in cell membranes

DRMs

Functionalization of Nanoscale Heterogeneity Antibody cross-linking at the cell surface causes raft proteins and lipids to co-patch and exclude non-raft proteins (31). This selec2000s tivity in patching is cholesterol90 80 dependent and can be transmitted 70 Macroscopic phases 60 across the plasma-membrane leafseparate in model 50 lets (32). The nonrandom coalesSTALL 40 membranes and cell 30 cence behavior observed in these membranes 20 artifactual cross-linking studies 100 nm 10 0 suggests how raft organizing po775 785 785 790 795 800 805 810 815 820 875 m/z tential may be functionalized to Advances in microscopy and larger, more physiologically releLipidomics reveals that spectroscopy (e.g. SPT, FCS, sphingolipids and sterol are FRET, STED, FPALM) reveal vant temporal and spatial scales sorted in the TGN during dynamic nanoassemblies of sterol, (Fig. 2B). A contention of the Plasma GUV membrane transport to the plasma sphingolipid, and protein in living lipid raft hypothesis is that dymembrane cells namic nanoscale heterogeneity can be stabilized to coalesce into larger Fig. 1. Evolution of the raft concept for subcompartmentalization in cell membranes. A bold H indicates hydrogen bonding. raft domains by specific lipid-lipid, VSV, vesicular stomatitus virus; DRMs, detergent-resistant membranes; GUV, giant unilamellar vesicle; m/z, mass/charge protein-lipid, and protein-protein ratio; SPT, single-particle tracking; FCS, fluorescence correlation spectroscopy; FRET, fluorescence resonance energy interactions (20). In this sense, cell transfer; STED, stimulated emission depletion; FPALM, fluorescence photoactivation localization microscopy. membranes would possess an underlying sphingolipid/cholesterolfor raft molecules in uncross-linked, resting con- been seen (23, 25). However, other techniques based connectivity that can be activated to cluster ditions. For GPI-anchored proteins, variable waist have indicated that nanoheterogeneity is actin- membrane bioactivity with little energy cost. Indeed, fluorescence correlation spectroscopy points to independent (26). The situation for TM proteins multimerization promotes the sorting of GPI<120-nm assemblages that fluctuate on a sub- is not yet clear. However, fluorescence photo- anchored proteins into sphingolipid/cholesterolsecond time scale (24). High spatial and temporal activation localization microscopy has revealed a enriched carriers during clathrin-independent resolution fluorescence resonance energy transfer dynamically clustered nanoscale distribution of endocytosis (33). Along similar lines, cluster(25) has generated a more conservative size es- hemagglutinin (27), a TM protein previously de- ing of cell surface Gb3 or GM1 (both GSLs) by timate with GPI-anchored receptors residing in scribed as raft-associating (21). The role of the their respective ligands Shiga toxin and cholera more temporally stable clusters of ~10 nm. Assem- association between cholesterol and sphingolip- toxin induces energy-independent tubular inbly formation is always cholesterol-dependent, ids in assembly formation has been analyzed vaginations of sphingolipid-biased membrane and, in some cases, an actin requirement has also recently by stimulated emission depletion mi- composition (34, 35). Similar behavior has
Rel. Int. (%)

www.sciencemag.org

SCIENCE

HC

Antibody-patched proteins display selective steroldependent coalescence at the cell surface

H O

C H

C CH 2 H

CH2
O

CH2

O C

Sphingolipids self-associate by hydrogen bonding

1980s

Ld

Lo

croscopy. This study revealed that, unlike glycerophospholipids, plasma-membrane sphingolipids display transient cholesteroldependent confinement in areas of <20 nm (28). In this case, differences in diffusion were attributed to differential hydrogenbonding capacities of glycerolversus sphingosine-based lipids. However, spin-labeled lipid probes at the cell surface have also revealed heterogeneity in membrane order on an electron spin resonance time scale (29). Different techniques are yielding a range of values for different molecular constituents in diverse cell types. However, these methods all point to the existence of small, dynamic and selective cholesterol-related heterogeneity in the plasma membranes of living cells. Recent data point to critical behavior as a potential physical basis for the existence of fluctuating nanoscale assemblies in plasma membranes (30).

VOL 327

1 JANUARY 2010

47

REVIEW
also been reported during the multivalent binding of SV40 virus to its GM1 receptor (35). Invagination from the plasma membrane was dependent on having longer receptor hydrocarbon chains, which are common to sphingolipid, and suggests that the effect is mediated by line tension arising between membrane domains of different compositional order (35). Coalescence of dynamic heterogeneity also occurs during signaling, for example, during the formation of B cell receptor (BCR) or T cell receptor (TCR) foci. Antigen binding induces the dynamic association of BCR to its signaling effector Lyn kinase and leads to the formation of an immune synpase. The interaction is dependent on the nature of Lyn lipid anchorage, with membrane orderdisrupting bulky hydrocarbon chains preventing association with the BCR (36). During TCR activation, raft components of this receptor complex (e.g., GPI- anchored proteins) become selectively immobilized in nanoscale clusters (37), seeding the accumulation of cholesterol, sphingomyelin, and saturated and long-chain phosphatidyl choline into the synapse, effectively sorting proteins according to their affinity for raft domains (38). Rafts in this activated or coalesced condition constitute a more ordered assemblage: a fluid membrane environment in which proteins can be modulated specifically (39), yet that exists separately from the surrounding membrane rich in unsaturated glycerophospholipid. Raft activation is often stabilized or nucleated by scaffolding elements such as cortical actin (40) and may become dominating when the mole fraction of sphingolipids and cholesterol increases, as is the case in the apical membrane of epithelial cells (41).

A
GPL

GSL / SM

Raft dissociating

Raft associating

Cholesterol

B
GPI-anchored protein

Acylated protein

Raft platform

Raft platform

Non-raft TM protein

Raft TM protein

Raft phase

Actin

Phase Separation in Cell Membranes Despite their selective co-patching with raft markers at the cell surface, raft TM proteins are depleted from the tightly packed Lo phase when reconstituted in model systems (42, 43). Thus, the Lo phase as it exists in simple model membranes is unlikely to be identical to raft-based heterogeneity in plasma membranes that selectively includes TM proteins (44, 45). Giant plasma-membrane vesicles isolated by a chemical membrane blebbing procedure can be cooled to phase separate into Lo- and Ld-like phases (46), and here also, raft TM proteins are typically excluded from the ordered membrane

Fig. 2. Hierarchy of raft-based heterogeneity in cell membranes. (A) Fluctuating nanoscale assemblies of sterol- and sphingolipid-related biases in lateral composition. This sphingolipid/sterol assemblage potential can be accessed and/or modulated by GPI-anchored proteins, certain TM proteins, acylated cytosolic effectors, and cortical actin. Gray proteins do not possess the chemical or physical specificity to associate with this membrane connectivity and are considered non-raft. GPL, glycerophospholipid; SM, sphingomyelin. (B) Nanoscale heterogeneity is functionalized to larger levels by lipid- and/or protein-mediated activation events (e.g., multivalent ligand binding, synapse formation, protein oligomerization) that trigger the coalescence of membrane orderforming lipids with their accompanying selective chemical and physical specificities for protein. This level of lateral sorting can also be buttressed by cortical actin. (C) The membrane basis for heterogeneity as revealed by the activation of raft phase coalescence at equilibrium in plasma-membrane spheres. Separated from the influence of cortical actin and in the absence of membrane traffic, multivalent clustering of raft lipids can amplify the functional level to a microscopic membrane phase. Membrane constituents are laterally sorted according to preferences for membrane order and chemical interactions. phase (47). Remarkably, this phase coexistence indicates that after chemical modification of protein (e.g., formaldehyde cross-linking, thiol treatment), the capacity for physical or lipidbased liquid-liquid phase separation can be manifested by the plasma membrane, despite its compositional complexity. Now the question is, how might phase lengthscale separation take place in plasma membranes at physiological temperatures? Some insight has come from a cell-swelling procedure to separate plasma-membrane spheres VOL 327 SCIENCE from the influence of cytoskeletal, endocytic, or exocytic processes in a cell line enriched in the raft ganglioside GM1. Pentavalent clustering by cholera toxin resulted in sterol-dependent coalescence of a micron-scale raft phase at 37C, selectively reorganizing the lateral distribution of proteins and lipids according to their predicted affinity for raft domains (44). In this case, selective incorporation of TM proteins was achieved at a lipid-ordering level far below that observed in model-membrane Lo phases (45). Thus, whereas preferential lipid-lipid associations do under-

48

1 JANUARY 2010

www.sciencemag.org

REVIEW
lie raft clustering, at physiological temperature they do not form a Lo phase when raft proteins become included, and specific lipid-protein interactions must come into play to modify organization (Fig. 2C). Along these lines, a comparison of lipid-packing in vesicles formed from lipids of the plasma membrane versus the plasma membrane itself reveals that lipid domainforming order is tightly regulated by the presence of protein (48). Rafts as Entities of Physical and Chemical Specificities Selective coordination of TM protein organization suggests that cells functionalize lipid order based sorting by including another specificity, most likely interactions involving proteins (45). Cell membranes are crowded with membrane proteins and their associated biases in regional composition (49). Proteins can specifically organize the distribution of lipids, a property that combines with sphingolipid-cholesterol assemblage potential to produce raft-based membrane heterogeneity. During vertical distortion of the bilayer, certain lipids of varying chain length are perturbed by the protein surface to different extents via the hydrophobic matching condition. Generally, it is assumed that the hydrophobic membranespanning part of the protein is stiff with no appreciable internal flexibility (50). However, by distorting lipids in the vertical direction, it would be possible to counter mismatch. The lipid species best adapted to the matching condition will have an increased probability of being close to the proteinlipid interface (50). Defined as wetting (51), the membrane protein surface is proposed to stabilize a sterically favored lipid environment. Electron spin resonance has identified a highly dynamic selection of boundary lipids for a number of proteins (52). However, some membrane proteins retain tightly bound lipids, even in the detergent solution present after purification (53). In such cases, lipids may have defined binding sites, where specific intercalation into protein structure is achieved (54). Because lipids must vertically complement the rigid hydrophobic surface of the membrane domain of integral membrane proteins, variation in the protein boundary also has direct consequences for the thickness and conformational order of the bilayer (50). In model membranes, long amphiphilic peptides order and thicken the bilayer in the absence of cholesterol, whereas shorter peptides offset the membrane order and thickness induced by the presence of cholesterol (55). It has previously been argued that cholesterolbased increases in membrane thickness influence the subcellular distribution of membrane proteins in accordance with the length of their TM domain (56). Conversely, changes in protein TM length itself have been argued to be the thicknessdetermining factor (57). Heterogeneity at the protein boundary is intensified by the fact that most membrane proteins are oligomeric, acting in specific macromolecular complexes to organize function (49). Superficially, these complexes are a source of steric restrictions and molecular crowding (49), but they can also incorporate specific lipids as integral features of their quaternary structure, thus functionally uniting protein-protein and lipid-protein interactions (54). Lipid incorporation is a function of specific polarhead group interactions and hydrocarbonchain space-filling functions within the oligomeric complex (54). Many of these oligomeric structures are also formed by strong associations resistant to detergents, with the binding of cholesterol to oligomers of caveolin-1 being a prominent example (58). In the raft field, we should be asking what it means for proteins to be wetted or, as we define the term in this context, lubricated (59), by specific lipids or lipid environments, particularly when it involves constituents that are important components of heterogeneity by rafts. A number has crafted additional specificity to a lipid-based connectivity, effectively reducing lateral space with minimal energy input. A cholesterol-binding pocket, as well as six palmitate residues, has recently been identified in the crystal structure of the b2-adrenergic receptor dimer interface (61). Palmitoylation of some membrane proteins has been shown to enhance association with raft nano-assemblies (62). Perhaps in the context of forming functional protein oligomers, the propensity of palmitate for raft association is augmented by combination with cholesterol. However, whether the b2-adrenergic receptor harnesses this lipid to connect with raft-based heterogeneity is not yet clear. Given the contribution of both physical and chemical specificities to lateral selectivity in the bilayer, lipid rafts are probably functionalized by both lipid-lipid and specific protein-lipid interactions. These lateral associations are governed by both physical and chemical specificity. Lipid-

Fig. 3. The lubrication of a raft TM protein by lipid. Membrane proteins bind and/or enrich certain lipids through chemical and physical specificities. These lipids may themselves exhibit sphingolipid/ sterol assemblage potential. In this scheme, a TM raft protein (light blue) specifically interacts with sterol and GSL, an interaction that lubricates its inclusion to and the assembly of functionalized (coalesced) raft membrane. of sphingolipid binding motifs have been described for membrane proteins (60). We propose that specific protein interaction with membraneordering raft lipids provides a functionalizing connection to the sphingolipid-cholesterol basis for raft assembly (Fig. 3). Interestingly, cholera toxincross-linking of GM1 was found to increase the partitioning of the raft enzyme betasecretase to the Lo phase in giant unilamellar vesicles (43). As pointed out earlier, the Lo phase does not reproduce the modestly ordered, TM proteininclusive raft structure of cell membranes. Thus, the fact that an undefined specificity for GSLs overcomes this stringent lateral sorting condition suggests that the specific lubrication of protein by lipid is likely to couple proteins to raftbased heterogeneity. Under this scheme, the functionalization of this heterogeneity depends on both lipid physical parameters and specific interactions that may include or even require proteins. For example, the TM protein LAT is an obligate component of raft-based accumulation of membraneordering lipids during the formation of the immunological synapse (38). As previously mentioned, membrane proteins work in functional complexes, so it is not surprising that evolution SCIENCE VOL 327 protein interactions alone cannot describe lipid rafts (63), because these do not account for the preferentially connecting lipid-lipid interactions that have so convincingly been demonstrated in model lipid membranes. Rather, we assert that sphingolipid-cholesterol assemblage potential forms a core raft connectivity that can be precisely modulated by protein specificity. In this view, raft-based membrane heterogeneity couples specific chemistries of association to the physical order preferences of lipids and proteins. Moreover, the assembly of proteins into rafts may be accompanied by conformational changes that modify protein activity. Rafts Inside the Cell The propensity to form heterogeneity by rafts is positively correlated with sterol content, which is maximized in the plasma membrane (6, 64), where the actin cortex also plays a central role in influencing or organizing sphingolipid-cholesterol assemblage potential (23, 25). However, for intracellular membranes the situation is less clear. But the emerging field of lipidomics is proving an important tool in evaluating both surface and intracellular membrane heterogeneity (38, 65).

www.sciencemag.org

1 JANUARY 2010

49

REVIEW
The concentration of sphingolipids and sterols increases along the biosynthetic pathway from the endoplasmic reticulum to the trans-Golgi network (TGN) (65). Recently, the comprehensive lipidome of immunoisolated post-Golgi carriers has revealed that order-forming sphingolipids, long and saturated glycerophospholipids, and sterols are selectively enriched in raft protein carriers bound from the TGN to the plasma membrane (66). Thus, functional raft clustering probably underlies the lateral sorting of cell surfacedestined constituents within the TGN, in keeping with the hypothesis of raft phase segregation principles as a means to selective carrier formation. Compositional Evolution of the Cell Membrane The cellular lipodome is theoretically made up of 9600 species of glycerophospholipids; more than 100,000 species of sphingolipids; thousands of mono-, di-, or triglycerides variants; and numerous fatty-acid and sterol-based structures (67). This amounts to an abundance of composition that might seem geared to dampen collective behavior in the membrane. However, this is not the case. Activating the sphingolipid-cholesterol assemblage potential in thermodynamically equilibrated plasmamembrane spheres leads to demixing of only two phases (P) as presently observed (44). The Gibbs phase rule states that the number of de-mixed entities (P) for a system at equilibrium is strictly correlated with the number of chemically independent components (C) by the expression P = C F + 2, where F is the number of independently variable intensive properties (i.e., pressure, temperature, and mole fractions of phase components). Thus, one could venture to postulate that these physical segregation principles have guided the coevolution of both membrane lipid and protein species, such that instead of having the vast P complexity possible from the phase rule, very few different P have survived. This could be explained by the fact that many components of the plasma membrane are not chemically independent, often forming specific complexes to reduce the lateral dimensionality of function. How then can long-range collective behavior arise from a chemically cross-talking plasma membrane? The answer is physicochemical teamwork. Activating the sphingolipid-cholesterol assemblage potential does not involve a purely physical phase transition with defined melting points and the like, but rather the coalescence of raft membrane arises through the functionally relevant cooperation of physical order (from lipid hydrocarbon chains, sterols, and the protein boundary) with specific chemical interactions (between proteins and lipids). In this sense, the cell appears to have designed a membrane composition that manipulates the physically selective behavior of lipids in a chemically specific manner, enabling organized heterogeneity to occur in the living condition. The introduction of membrane-organizing cholesterol seems to have coincided with the evolution of multicellular complexity after the oxidation of our atmosphere (68). This correlation may imply that, in the pre-sterol era, other chemical means of reducing lateral dimensionality could have evolved. Interestingly, Caenorhabditis elegans does not use sterol as a structural element in its membranes (69). Principles of organized heterogeneity in such organisms are unknown but when revealed will potentially unravel a new chemical toolkit for membrane subcompartmentalization. Conclusions Cell membranes are complicated in composition but precise in purpose: to selectively compartmentalize the constituents of life away from environmental lifelessness. Thus, it is not surprising that membranes have innovated a means to laterally organize gatekeepers of this task. In living cells, there is strong evidence for dynamic raftbased membrane heterogeneity at the nanoscale, which can be functionally coalesced to more stable membrane-ordered assemblies. At its core, sphingolipid-cholesterol assemblage potential supplies membranes with a subcompartmentalization propensity that can be accessed or organized by proteinaceous input at little energetic cost. Raft proteins are envisioned as being equipped with a dynamic sterol-sphingolipiddependent bias in composition at the nanoscale, allowing for the partitioning to and assembly of more stable raft platforms in the functionalized state. During raft activation, protein-lipid interactions are coupled to lipid-orderbased sorting, generating heterogeneity serving to functionalize, focus, and coordinate the bioactivity of membrane constituents.
References and Notes
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. S. Munro, Cell 115, 377 (2003). D. A. Brown, K. J. Rose, Cell 68, 533 (1992). K. S. Song et al., J. Biol. Chem. 271, 9690 (1996). D. Lingwood, K. Simons, Nat. Protoc. 2, 2159 (2007). K. Jacobson, O. G. Mouritsen, R. G. W. Anderson, Nat. Cell Biol. 9, 7 (2007). G. van Meer, D. R. Voelker, G. W. Feigenson, Nat. Rev. Mol. Cell Biol. 9, 112 (2008). K. Simons, E. Ikonen, Nature 387, 569 (1997). S. J. Singer, G. L. Nicolson, Science 175, 720 (1972). B. R. Lentz, D. A. Barrow, M. Hoechli, Biochemistry 19, 1943 (1980). J. H. Ipsen et al., Biochim. Biophys. Acta 905, 162 (1987). K. Simons, W. L. C. Vaz, Annu. Rev. Biophys. Biomol. Struct. 33, 269 (2004). A. J. Garca-Sez, S. Chiantia, P. Schwille, J. Biol. Chem. 282, 33537 (2007). S. L. Veatch, S. L. Keller, Phys. Rev. Lett. 94, 148101 (2005). P. S. Niemel, M. T. Hyvnen, I. Vattulainen, Biophys. J. 90, 851 (2006). J. Huang, G. W. Feigenson, Biophys. J. 76, 2142 (1999). H. M. McConnell, M. Vrljic, Annu. Rev. Biophys. Biomol. Struct. 32, 469 (2003). R. Schroeder, E. London, D. Brown, Proc. Natl. Acad. Sci. U.S.A. 91, 12130 (1994). G. W. Feigenson, Biochim. Biophys. Acta 1788, 47 (2009). J. F. Hancock, Nat. Rev. Mol. Cell Biol. 7, 456 (2006). T. Friedrichson, T. V. Kurzchalia, Nature 394, 802 (1998). A. Pralle, P. Keller, E. L. Florin, K. Simons, J. K. Hrber, J. Cell Biol. 148, 997 (2000). T. van Zanten et al., Proc. Natl. Acad. Sci. U.S.A. 106, 18557 (2009). 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. K. G. Suzuki et al., J. Cell Biol. 177, 717 (2007). P. F. Lenne et al., EMBO J. 25, 3245 (2006). D. Goswami et al., Cell 135, 1085 (2008). R. Lasserre et al., Nat. Chem. Biol. 4, 538 (2008). S. T. Hess, Proc. Natl. Acad. Sci. U.S.A. 104, 17370 (2007). C. Eggeling et al., Nature 457, 1159 (2009). M. J. Swamy et al., Biophys. J. 90, 4452 (2006). S. L. Veatch et al., ACS Chem. Biol. 3, 287 (2008). T. Harder, P. Scheiffele, P. Verkade, K. Simons, J. Cell Biol. 141, 929 (1998). G. Gri, B. Molon, S. Manes, T. Pozzan, A. Viola, Immunol. Lett. 94, 247 (2004). S. Mayor, R. E. Pagano, Nat. Rev. Mol. Cell Biol. 8, 603 (2007). W. Rmer et al., Nature 450, 670 (2007). H. Ewers et al., Nat. Cell Biol.; published online 20 December 2009 (10.1038/ncb1999). H. W. Sohn, P. Tolar, S. K. Pierce, J. Cell Biol. 182, 367 (2008). K. Drbal et al., Int. Immunol. 19, 675 (2007). T. Zech et al., EMBO J. 28, 466 (2009). Y. Kaizuka, A. D. Douglass, S. Vardhana, M. L. Dustin, R. D. Vale, J. Cell Biol. 185, 521 (2009). A. Viola, N. Gupta, Nat. Rev. Immunol. 7, 889 (2007). D. Meder, M. J. Moreno, P. Verkade, W. L. Vaz, K. Simons, Proc. Natl. Acad. Sci. U.S.A. 103, 329 (2006). A. T. Hammond et al., Proc. Natl. Acad. Sci. U.S.A. 102, 6320 (2005). L. Kalvodova et al., J. Biol. Chem. 280, 36815 (2005). D. Lingwood, J. Ries, P. Schwille, K. Simons, Proc. Natl. Acad. Sci. U.S.A. 105, 10005 (2008). H.-J. Kaiser et al., Proc. Natl. Acad. Sci. U.S.A. 106, 16645 (2009). T. Baumgart et al., Proc. Natl. Acad. Sci. U.S.A. 104, 3165 (2007). P. Sengupta, A. T. Hammond, D. Holowka, B. Baird, Biochim. Biophys. Acta 1778, 20 (2008). M. Gie et al., Biophys. J. 85, 1278 (2003). D. M. Engelman, Nature 438, 578 (2005). O. G. Mouritsen, M. Bloom, Annu. Rev. Biophys. Biomol. Struct. 22, 145 (1993). S. A. Akimov et al., Phys. Rev. 77, 051901 (2008). D. Marsh, L. I. Horvth, Biochim. Biophys. Acta 1376, 267 (1998). R. Callaghan, G. Berridge, D. R. Ferry, C. F. Higgins, Biochim. Biophys. Acta 1328, 109 (1997). A. G. Lee, Biochim. Biophys. Acta 1612, 1 (2003). F. A. Nezil, M. Bloom, Biophys. J. 61, 1176 (1992). M. S. Bretscher, S. Munro, Science 261, 1280 (1993). K. Mitra, I. Ubarretxena-Belandia, T. Taguchi, G. Warren, D. Engelman, Proc. Natl. Acad. Sci. U.S.A. 101, 4083 (2004). M. Murata, Proc. Natl. Acad. Sci. U.S.A. 92, 10339 (1995). M. L. Allende, R. L. Proia, Curr. Opin. Cell Biol. 12, 587 (2002). J. Fantini, Cell. Mol. Life Sci. 60, 1027 (2003). V. Cherezov et al., Science 318, 1258 (2007); published online 24 October 2007 (10.1126/science.1150577). K. Chakrabandhu et al., EMBO J. 26, 209 (2007). R. G. W. Anderson, K. Jacobson, Science 296, 1821 (2002). B. Mesmin, F. R. Maxfield, Biochim. Biophys. Acta 1791, 636 (2009). B. Brgger et al., J. Cell Biol. 151, 507 (2000). R. W. Klemm et al., J. Cell Biol. 185, 601 (2009). L. Yetukuri, K. Ekroos, A. Vidal-Puig, M. Oresuc, Mol. Biosyst. 4, 121 (2008). O. G. Mouritsen, M. J. Zuckermann, Lipids 39, 1101 (2004). T. V. Kurzchalia, S. Ward, Nat. Cell Biol. 5, 684 (2003). We thank F. Jlicher, G. Feigenson, J. Howard, P. Schwille, C. Brangwynne, H.-T. He, G. van Meer, and members of the Simons lab for critical reading of this paper. A special thank you goes to M. Surma, M. Gerl, and I. Levental for their construction of and contribution to the figures. This work was supported by European Union FP6 Lipid PRISM grant no. LSHB-CT2007-037740, Deutsche Forschungsgemeinschaft Schwerpunktprogramm 1175, and Bundesministerium fr Bildung und Forschung BioChance Plus grant no. 0313827.

10.1126/science.1174621

50

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

BREVIA
can provide an entirely additional benefit to crypsis (difficulty in visually detecting prey) (2) and functions even in the absence of crypsis, when the prey is in clear view. Although the role of predator cognition in the evolution of warning coloration is widely studied (3), researchers studying the evolution of John Skelhorn,1 Hannah M. Rowland,2 Michael P. Speed,2 Graeme D. Ruxton1* other aspects of adaptive coloration have concenrganisms are under strong selection to Selenia dentaria) as prey. Before testing, chicks trated on understanding how predators sensory sysavoid detection by predators and to cap- were divided into nine groups, each containing tems have influenced the evolution of prey coloration ture prey, and understanding how animals eight individuals. Birds in all groups received four (2). Our results show that predators cognitive stratvisual appearances are adapted for these purposes 2-min trials in which they were placed in the ex- egies (recognition and identification), rather than continues to pose interesting questions for evolu- perimental arena individually. The items placed in their sensory capabilities, are the selective force tionary theory (1). Although the function of cryp- the experimental arena with them during these trials driving the evolution of masquerade and raise the sis (avoiding detection) (2), aposematism (warning differed among groups. Three groups encountered possibility that predator cognition may be a more important selective agent than previously realized. Masquerade is more widespread than betterstudied aspects of adaptive coloration such as aposematism and is found in a large number of species from a wide range of taxa. Plants from the genus Lithops look remarkably like stones; stick insects Fig. 1. (A) The time in seconds (mean T SE) taken to make a first peck at the test stimulus (Brim indicates brimstone; Thorn, early resemble twigs; the Amthorn;andTwig,hawthorntwig).Thelatencytoattacktheteststimulidifferedamongourexperimentalgroups(Kruskal-Wallistest;c2 =56.10,P< azon fish Monocirrhus 0.001, df = 8). Specifically, birds trained with branches (Br) took significantly longer to attack the test stimuli than did birds in the control groups polyacanthus is visually trainedonmanipulatedbranches(Man)ornothing(No)(Kruskal-Wallistest;c2 =47.39,P<0.001,df=1).SamplesizeisN=8ineachgroup.(B) almost indistinguishable Thetimeinseconds(meanT SE)thatbirdshandledpreyfor inthetesttrial.Handlingtimes (4)differedamongourexperimentalgroups(Kruskal- from leaves, and birds Wallis test;c2 = 35.34, P < 0.001, df = 5). Birds trained with branches took significantly longer to handle caterpillarsthan didbirds in the control from the family Nyctibiidae bear an uncanny groups trained on manipulated branches or nothing (Kruskal-Wallis test; c2 = 31.16, P < 0.001, df = 1). Sample size is N = 8 in each group. likeness to tree stumps. coloration) (3), and mimicry (resembling a defended a hawthorn branch complete with leaves (hawthorn Therefore, masquerade appears to have evolved organism) (4) are intensively studied, one aspect of Crataegus spp. being a common host plant for the on multiple occasions, and its ecological imporadaptive coloration has been almost completely two caterpillar species). Three groups encountered tance requires further investigation in light of our ignored: masquerade. Masquerading organisms a manipulated hawthorn branch that had been experiments. appear to closely resemble inedible and generally bound in purple cotton thread to change its visual References and Notes inanimate objects such as twigs, leaves, stones, and appearance without influencing its physical struc1. G. D. Ruxton, T. N. Sherratt, M. P. Speed, Avoiding bird droppings. Individuals using this defensive ture or odor. The final three groups experienced an Attack: The Evolutionary Ecology of Crypsis, Warning strategy are assumed to avoid predation or gain empty arena. The test stimulus differed among Signals and Mimicry (Oxford Univ. Press, New York, groups given the same previous experience: one access to prey by being misidentified as either 2004), pp. 2325. 2. M. Stevens, S. Merilaita, Philos. Trans. R. Soc. London Ser. B inedible objects by their predators or as innocu- group received a single Brimstone larvae, another 364, 423 (2009). ous objects by their prey. There is currently no a single Early thorn larvae, and the remaining 3. J. Mappes, N. Marples, J. A. Endler, Trends Ecol. Evol. 20, group received a single hawthorn twig (7). empirical evidence to support this theory (5, 6). 598 (2005). Birds with prior experience of twigs took Demonstrating that organisms benefit from 4. M. P. Speed, Anim. Behav. 46, 1246 (1993). 5. J. A. Endler, Biol. J. Linn. Soc. Lond. 16, 25 (1981). masquerade is methodologically challenging: It is longer to attack both species of twig-resembling 6. J. A. Allen, J. M. Cooper, J. Biol. Educ. 19, 268 (1985). difficult to determine whether a predator has de- caterpillars, and handled them more cautiously, 7. Materials and methods are available as supporting tected and misidentified an individual (that is, mas- compared with birds that had either no experimaterial on Science Online. querade) (5, 6) or whether it has simply failed to ence of twigs or experience only of twigs whose detect the prey item (which would be crypsis) (2). visual appearance had been manipulated by bind- Supporting Online Material www.sciencemag.org/cgi/content/full/327/5961/51/DC1 For predators to misidentify a masquerading prey ing them in colored thread (Fig. 1). Our results show that masquerade functions Materials and Methods item as the object that it closely resembles, the Figs. S1 to S3 predator must have previous experience of that to promote misidentification of the masqueradobject. By manipulating predators previous expe- ing organism. The caterpillars in our experiment 14 September 2009; accepted 27 October 2009 rience of the putative model but keeping their ex- were treated differently by predators with previ- 10.1126/science.1181931 posure to the masquerader the same, it is possible ous experience of twigs in ways that are likely to 1 Division of Ecology and Evolutionary Biology, Faculty of to determine whether predators are misidentifying lead to antipredatory protection: increased latency Biomedical and Life Sciences, University of Glasgow, Glasgow masquerading prey as their models or simply failing to attack and more cautious handling. Most powerG12 8QQ, UK. 2Division of Population and Evolutionary to detect them. We used domestic chicks (Gallus fully, this effect occurred at close range, entirely out Biology, School of Biological Sciences, University of Liverpool, gallus domesticus) as predators and putative twig- of context, on a visually contrasting substrate and Liverpool L69 7ZB, UK. resembling caterpillars (the Brimstone moth, Opis- in an empty arena with no other prey or objects *To whom correspondence should be addressed. E-mail: thograptis luteolata, and the Early thorn moth, present. We therefore can conclude that masquerade [email protected]

Masquerade: Camouflage Without Crypsis

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

51

RESEARCH ARTICLE Dendritic Mechanisms Underlying Rapid Synaptic Activation of Fast-Spiking Hippocampal Interneurons
Hua Hu,1,2 Marco Martina,1,3 Peter Jonas1,4* Fast-spiking, parvalbumin-expressing basket cells (BCs) are important for feedforward and feedback inhibition. During network activity, BCs respond with short latency and high temporal precision. It is thought that the specific properties of input synapses are responsible for rapid recruitment. However, a potential contribution of active dendritic conductances has not been addressed. We combined confocal imaging and patch-clamp techniques to obtain simultaneous somatodendritic recordings from BCs. Action potentials were initiated in the BC axon and backpropagated into the dendrites with reduced amplitude and little activity dependence. These properties were explained by a high K+ to Na+ conductance ratio in BC dendrites. Computational analysis indicated that dendritic K+ channels convey unique integration properties to BCs, leading to the rapid and temporally precise activation by excitatory inputs. feedforward and feedback inhibition, contribute to the generation of network oscillations, and are thought to be involved in higher brain function and dysfunction (17). After stimulation of excitatory input synapses in vitro, BCs respond with remarkable speed, exquisite temporal precision, and preferential activity in the onset phase of a stimulus train (8, 9). Similarly, during network activity in vivo, such as sharp-wave ripple or theta rhythms, BCs are activated by input from principal neurons with short latency and minimal jitter (1013). The mechanisms underlying this rapid and precise activation are unclear. It is generally thought that synaptic factors, such as the time course of the postsynaptic conductance and the extent of depression or facilitation, play an im1 Institute of Physiology I, Universitt Freiburg, Engesserstrae 4, D-79108 Freiburg, Germany. 2Department of Physiology at Institute of Basic Medical Sciences and Centre for Molecular Biology and Neuroscience (CMBN), University of Oslo, Blindern, NO-0317 Oslo, Norway. 3Department of Physiology, The Feinberg School of Medicine of Northwestern University, 303 East Chicago Avenue, Chicago, IL 60611, USA. 4Freiburg Institute for Advanced Studies, Albertstrae 19, D-79104 Freiburg, Germany.

ast-spiking, parvalbumin-expressing, gaminobutyric acid (GABA)releasing (GABAergic) interneurons (BCs) play a

key role in the function of neuronal networks. These neurons set a narrow time window for temporal summation in principal neurons by fast

*To whom correspondence should be addressed. E-mail: [email protected]

Fig. 1. Confocally targeted recording from BC dendrites. (A) Confocal image (pseudocolor representation) of a BC in the dentate gyrus filled with Alexa Fluor 488 taken during the experiment. (B) Infrared differential interference contrast videoimage of the apical dendrite of the same cell as in (A). (C) Light micrograph of a BC filled with biocytin during recording and labeled by using 3,3-diaminobenzidine. 10-mm stack projection. Arrows indicate the axonal arbor, forming baskets around granule cell somata. (D) Train of APs

evoked by a 1-s, 0.75-nA current pulse applied at the soma (top) and AP frequency ( f ) current (I) relation (bottom). Same cell as in (C). Bottom right graph shows mean maximal AP frequency (bar) and data from individual cells (points). (E) (Left) Confocal micrograph of a BC filled with biocytin and stained with fluorescein isothiocyanate (FITC)conjugated avidin; (center) parvalbumin immunoreactivity of the same BC; (right) overlay. 40-mm stack projection. SCIENCE www.sciencemag.org

52

1 JANUARY 2010

VOL 327

RESEARCH ARTICLE
portant role (8, 9, 1416). Alternatively or additionally, the electrical properties of interneuron dendrites may contribute. Whereas the dendrites of pyramidal neurons were extensively characterized (1722), those of fast-spiking, parvalbuminexpressing BCs have not been directly examined. However, Ca2+ imaging experiments suggest that Na+, K+, and Ca2+ channels may be present in the dendrites of neocortical fast-spiking interneurons (23). To study the dendrites of BCs directly, we used confocally targeted patch-clamp recording in hippocampal slices (Fig. 1) (20). After filling BCs in the dentate gyrus with Alexa Fluor 488 via somatic recording, we traced dendrites into molecular layer or hilus by using confocal imaging (Fig. 1A) and obtained dendritic recordings by using correlated infrared-differential interference contrast videomicroscopy (Fig. 1B). BCs were identified on the basis of the location of the axon in the granule cell layer (24) (Fig. 1C), the fastspiking AP phenotype (25) (Fig. 1D; mean maximal frequency 104.2 T 2.2 Hz; n = 98 dual recordings), and the immunoreactivity to the Ca2+binding protein parvalbumin tested in a subset of neurons (Fig. 1E; 23 of 25 cells). Detailed analysis of the axonal arbor revealed that our sample was mainly composed of classical basket cells with tangential collaterals (78 of 83 recovered cells) but also included a subpopulation of cells with radial collaterals, suggestive of axo-axonic cells (5 of 83 cells) (fig. S3) (26). APs propagate into BC dendrites with attenuated amplitude. We made simultaneous recordings from somata and apical dendrites of dentate gyrus BCs at distances up to 300 mm from the soma, close to the physical dendritic length (Fig. 2). Current injection at both the soma and the apical dendrite evoked high-frequency trains of APs (Fig. 2A). Although the AP frequency was identical at the soma and the dendrite, single APs at the two locations differed substantially (Fig. 2B). First, APs in the apical dendrite showed markedly attenuated amplitude in comparison with somatic APs. For the first AP in the train, the peak amplitude measured from threshold was 80.8 T 2.0 mVat the soma and 23.4 T 2.1 mV at the dendrite (at distances >100 mm from soma; P < 0.01; Fig. 2, B and D). Second, the maximal rate of rise declined as a function of distance (444.4 T 29.5 mV ms1 at the soma versus 81.3 T 10.8 mV ms1 at the dendrite at >100 mm; P < 0.01; Fig. 2, B and E). Lastly, the duration at half-maximal amplitude was slightly prolonged (0.50 T 0.02 ms at the soma versus 0.84 T 0.08 ms at the dendrite at >100 mm; P < 0.01; Fig. 2, B and F). To test whether dendritic AP propagation was activity-dependent, we compared the properties of the first and the last AP at distal dendrites in the high-frequency train (Fig. 2, B and C). The amplitude of the dendritic AP at the end of the train was similar to that at the onset (22.0 T 1.8 mV versus 23.4 T 2.1 mV; P > 0.5 for distal recordings >100 mm). Likewise, the maximal rate of rise and the duration at half-maximal amplitude were comparable between the first and the last AP (Fig. 2, D to F). To examine whether the properties of the dendritic AP depended on the site of current injection, we compared APs evoked by somatic and dendritic current injection in the same cell (Fig. 2, B to C). Dendritic APs evoked by somatic and dendritic current injection had similar amplitudes (22.7 T 2.8 mV versus 24.8 T 3.4 mV for the first AP; P > 0.4 for distal recordings). Likewise, the maximal rate of rise and the duration at halfmaximal amplitude were independent of the site of current injection (Fig. 2, D to F).

Fig. 2. APs in BC dendrites show marked amplitude attenuation, moderate broadening, and little activity dependence. (A) Train of APs evoked by somatic (left) and dendritic (right) current injection. Black traces, somatic voltage and corresponding current; red traces, dendritic voltage and corresponding current. (B and C) First AP (B) and last AP (C) in the 1-s train shown at expanded time scale. Dendritic recording site on apical dendrite 124 mm from soma. (D to F) Summary plot of AP peak amplitude measured from threshold (D), maximal slope of the rise (E), and duration at half-maximal amplitude (F) plotted against distance (positive distance, www.sciencemag.org SCIENCE

apical dendrite; negative distance, basal dendrite; both measured from the center of the soma). Data from 42 simultaneous somatodendritic and 8 somatic recordings (distance 0). Recording temperature ~23C. Solid symbols, somatic current injection; open symbols, dendritic current injection. Red, first AP; blue, last AP in a 1-s train. For the half-duration of the first AP, only 39 out of 42 recordings could be analyzed in which voltage crossed the half-maximal amplitude level during repolarization. Lines represent exponential functions fit to the data points. For length constant values, see table S1. VOL 327 1 JANUARY 2010

53

RESEARCH ARTICLE
Analysis of APs at near-physiological temperature gave comparable results (fig. S1). Furthermore, APs recorded in basal dendrites [receiving granule cell input (14)] were similar to those in apical dendrites (receiving mossy cell and entorhinal cortex input) (fig. S2). Lastly, similar results were obtained for neurons with radial axon collaterals, representing putative axo-axonic cells (fig. S3). APs are initiated in the axon. To determine the site of AP initiation, we measured latency differences of APs at different subcellular locations (Fig. 3). In the majority of apical dendritic recordings (19 of 28 BCs), somatic APs preceded dendritic APs during long current pulses, independently of the site of current injection (Fig. 3, A and B). When measured at the half-maximal amplitude, the somatic AP preceded the dendritic AP by 0.21 T 0.05 ms (14 apical dendritic recordings at distances of >100 mm). Likewise, in the majority of basal dendritic recordings (12 of 14 BCs), the somatic AP preceded the dendritic AP (dendritic recordings 17 to 62 mm from the soma). In pyramidal neurons, brief high-intensity dendritic current pulses or distributed activation of excitatory synapses trigger dendritic spikes more effectively than long current pulses (2729). We therefore tested these stimulation paradigms in BCs. However, 0.5-ms current pulses applied to the dendrite (137 T 17 mm) failed to initiate dendritic spikes in six out of six BCs and even failed to trigger somatic APs in four out of six recordings, despite the injection of large current amplitudes of up to 3.5 nA (Fig. 3C). Failure of AP initiation was presumably related to the marked rectification of the voltage-current relation, showing that it becomes increasingly difficult to depolarize the dendritic membrane potential beyond 30 mV (Fig. 3C). Activation of distal synapses by stimulation of axons of the lateral perforant path initiated APs (Fig. 3D). However, as with the other stimulation paradigms tested, the somatic AP constantly preceded the dendritic AP. Similarly, stimulation of two synaptic inputs with stimulation electrodes ~30 mm apart initiated APs in which the somatic consistently preceded the dendritic response (four BCs; fig. S4). In a subset of BCs, either the dendritic APs preceded the somatic APs (10 of 42 BCs) or dendritic and somatic APs occurred simultaneously (1 of 42 BCs; Fig. 3, E and F). This could be due to either dendritic AP initiation (2729) or dendritic origin of the axon (3032). We therefore performed a correlated analysis of the morphological properties of BCs by using post-hoc biocytin labeling. In all cases in which the AP occurred first in the dendrite, the axon originated from the dendrite and was closer to the dendritic than the somatic recording electrode (Fig. 3G). To test whether the results from all morphologically recovered BCs were quantitatively consistent with axonal AP initiation, we plotted AP latency against the difference of distances of the recording locations from the axon origin (Fig. 3H). The linear regression line intersected the abscissa near 0, indicating AP initiation in the axon (correlation coefficient r = 0.90; P < 0.001). High K+ to Na+ conductance ratio in BC dendrites. Both the robust axonal AP initiation and the marked dendritic AP attenuation could suggest that BC dendrites behave as passive cables. We therefore determined the density of voltage-gated Na+ and K+ channels in outsideout patches isolated at various locations (Fig. 4). In somatic outside-out patches obtained with Cs+internal solution, voltage pulses from 120 mV to 0 mV evoked inward currents in the majority of patches (Fig. 4A). These currents showed fast activation and inactivation and were blocked by 1 mM tetrodotoxin in the external solution, demonstrating that they were mediated by voltage-gated Na+ channels (33) (Fig. 4B). In contrast, in dendritic outside-out patches, the amplitude of inward currents was substantially smaller (Fig. 4, A and C). Quantitative analysis revealed that the Na+ current density was 13.3 T 2.1 pA mm2 at the soma but steeply declined as a

Fig. 3. The axon is the exclusive site of AP initiation in BCs under a variety of conditions. (A and B) Recording from a BC in which the somatic preceded the dendritic AP (31 of 42 BCs). First somatic (black) and dendritic (red) APs evoked by 1-s somatic (A) and dendritic (B) current pulses. Left graphs, somatic and dendritic APs at absolute voltage scale; right graphs, APs normalized to same peak amplitude. Dendritic recording site on apical dendrite 176 mm from soma. (C) Responses to brief depolarizing current injection in the dendrite. (Left) Dendritic responses and (right) voltage-current relation for the voltage immediately after the pulse (arrow) in a subset of BCs in which no somatic spikes could be elicited (four out of six cells). Line represents linear regression of data points for voltages 30 mV. (D) Somatic (black) and dendritic (red) APs initiated by evoked postsynaptic potentials (PSPs). Axons of the lateral perforant path were stimulated in the outer molecular layer. Left traces, somatic and dendritic APs at absolute voltage scale; right traces, APs normalized to same peak amplitude. (E and F) Recording from a BC in which the dendritic preceded the somatic AP (10 of 42 BCs). First somatic (black) and dendritic (red) APs evoked by 1-s somatic (E) and dendritic current pulses (F). (G) Correlated light microscopic morphological analysis of the same cell shown in (E and F). The axon (arrows) originated near the dendritic recording site. (H) Plot of latency between somatic and dendritic APs against the corresponding difference of distances, using the axon origin as reference point. Solid circles, somatic current injection; open circles, dendritic current injection. Line represents the result of linear regression; average dendritic AP propagation velocity was 0.53 m s1. Recording temperature ~23C except for synaptic experiments in (D) (~31C). VOL 327 SCIENCE www.sciencemag.org

54

1 JANUARY 2010

RESEARCH ARTICLE
function of distance, with estimated length constants of 87 mm in apical dendrites and 25 mm in basal dendrites (Fig. 4D and table S1). Expression of Kv3-type K+ channels is a hallmark property of fast-spiking GABAergic interneurons (34, 35). We therefore explored the possible presence of Kv3 or other types of K+ channels (19, 36) in BC dendrites. In somatic outside-out patches isolated with K+-internal solution, voltage pulses from 120 mV to 70 mVevoked large voltage-dependent outward currents (Fig. 4E). Quantitative analysis revealed a K+ current density of 91.5 T 21.1 pA mm2 at the soma. In apical dendrites, the K+ current density decayed moderately as a function of distance, with an estimated length constant of 763 mm (Fig. 4F). In contrast, in basal dendrites, the decay was steeper, with a length constant of 57 mm. To determine the molecular identity of the K+ channels expressed in BC dendrites, we measured functional parameters that discriminate among K+ channel subtypes (25, 34, 35) (fig. S5). Dendritic K+ channels in BCs showed several characteristic properties. First, they had a high activation threshold of ~40 mV (fig. S5, a and b). Second, their activation time course was fast and highly voltage-dependent (fig. S5c). Third, they were blocked by low concentrations of extracellular tetraethylammonium (TEA), with only minimal distance dependence (fig. S5, d and e). Application of 1 mM TEA reduced the K+ current amplitude in dendritic patches to 62.7 T 5.7%, similar to somatic patches (P > 0.05; fig. S5e), whereas 200 nM a-dendrotoxin (a-DTX) had almost no effect (amplitude 97 T 0.6% of control; n = 3). Lastly, K+ channels showed only minimal inactivation, again with little distance dependence (fig. S5f). Because only Kv3 channels combine the properties of high activation threshold, rapid activation, high TEA sensitivity, and insensitivity to a-DTX, our results indicate that Kv3-type channels prevail in BC dendrites, consistent with immunocytochemical data [(34), but see (23)]. Dendritic K+ channels shape excitatory postsynaptic potential (EPSP) time course and coincidence detection in BCs. One implication of our results is that dendritic K+ channels may be efficiently activated by excitatory synaptic input (8, 9, 1416). We therefore injected artificial excitatory postsynapticcurrentlike (EPSC-like) waveforms (aEPSCs) into the dendrite (231 T 12 mm) while recording the corresponding artificial EPSPs (aEPSPs) at the soma (Fig. 5). The parameters of the aEPSCs were chosen to mimic the amplitude and time course of unitary EPSCs in BCs (14). Bath application of 5 mM TEA reversibly prolonged the half-duration of single aEPSPs from 12.5 T 0.7 ms to 14.8 T 0.9 ms (P < 0.01; Fig. 5A). Because TEA had no significant effect on the membrane time constant (t = 10.8 T 0.8 ms in control versus 12.0 T 1.8 ms in TEA; P > 0.2), these results indicate that voltage-gated K+ channels accelerate the decay of EPSPs (37). To corroborate that the K+ channels activated by EPSPs were located dendritically, we simulated EPSPs in a previously established passive BC cable model (Fig. 5B) (38) with and without dendritic voltage-gated K+ channels. Insertion of K+ channels into the apical dendrites near the activated synapses accelerated EPSP kinetics (Fig. 5, C and D), reproducing the experimental observations (Fig. 5A). In contrast, insertion of K+ channels into the basal dendrites remote from the synapses had only minimal effects on EPSP kinetics. Activation of BCs requires the coincident activation of multiple excitatory synaptic inputs (8, 14). To test how the K+ channelmediated acceleration of the EPSP decay time course affects coincidence detection in BCs, we simulated pairs of EPSPs separated by variable time intervals (Fig. 5, E and F). If the same synapse was activated repetitively at variable time intervals, dendritic K+ channels reduced the extent and shortened the time window of temporal summation (Fig. 5E). In contrast, if two spatially separated synapses were activated at different times, the maximal extent of summation was unchanged, whereas the summation window was shortened by dendritic K+ channels (Fig. 5F). Thus, dendritic K+ channels enable BCs to selectively detect nearly coincident, spatially separated activity.

Fig. 4. High K+ to Na+ conductance density ratio in BC dendrites. (A) Na+ currents evoked in outside-out patches from soma, apical dendrite, and basal dendrite. Test pulse potential was 0 mV. Na+ currents were recorded with Cs+-internal solution. (B) Tetrodotoxin (TTX) sensitivity of Na+ channels. Top graphs, Na+ currents in control conditions and in the presence of 1 mM TTX (somatic outside-out patch). Bottom, plot of Na+ peak current against time during TTX application (horizontal bar); each point represents the mean of 10 consecutive measurements. (C) Difference between somatic and dendritic Na+ channel density in the same cell. (Left) Location of somatic and dendritic recording pipette, superimposed with morphological reconstruction of the somatodendritic domain of the BC. (Right) Na+ currents recorded in outside-out patches isolated from these locations by using two patch pipettes pulled from same glass capillary. (D) Na+ current density as a function of distance from the soma. Data from 17 somatic (black), 15 apical dendritic (red, positive distances), and 9 basal dendritic (red, negative distances) outside-out patches. (E) K+ currents evoked in outside-out patches from soma, apical dendrites, and basal dendrites. Test pulse potential was 70 mV. K+ currents were recorded with K+-internal solution. (F) K+ current density as a function of distance from the soma. Data from 7 somatic (black), 17 apical dendritic (red, positive distances), and 14 basal dendritic (red, negative distances) outside-out patches. Red lines represent exponential functions fit to the data points. For length constant values, see table S1. Na+ currents are the average of 38 to 50 sweeps; K+ currents are either single traces or the average of three sweeps. Leakage and capacitive currents were digitally subtracted. www.sciencemag.org SCIENCE VOL 327

1 JANUARY 2010

55

RESEARCH ARTICLE
Dendritic K+ channels control EPSP-AP coupling and AP phenotype of BCs. To test how dendritic K+ channels influence EPSP-AP coupling, we inserted Na+ and K+ channels in densities consistent with our experimental results and compared scenarios with passive and active dendrites (Fig. 6). In the BC model with passive dendrites, APs evoked by both brief current pulses and synaptic activation were followed by a marked afterdepolarization. In contrast, in the model with active dendrites, the afterdepolarization was largely suppressed (Fig. 6B). Analysis of the underlying mechanisms revealed that the activation of dendritic K+ channels, which was particularly efficient for APs evoked by EPSPs, reduced the dendrosomatic current flow after APs (Fig. 6B). Correlated with this inhibition of the afterdepolarization, dendritic K+ channels changed the input-output characteristics of BCs. In the model with passive dendrites, strong synaptic stimuli (activation of >25 synapses) triggered bursts of APs. In contrast, in the model with active dendrites, single APs were generated over a wide range of synaptic stimulus intensities (Fig. 6C). Furthermore, with a repetitive synaptic stimulation paradigm the model with passive dendrites showed a greatly reduced AP initiation threshold for the second synaptic stimulus, whereas the model with active dendrites had an almost constant AP initiation threshold (Fig. 6D). Thus, dendritic K+ channels generate strength- and timing-independent activation properties of BCs. Lastly, dendritic K+ channels contribute to the classical fast-spiking AP phenotype of BCs (25). In the model with passive dendrites, long depolarizing current pulses evoked trains of APs with markedly declining amplitude, and APs were occasionally initiated after the termination of the pulse (Fig. 6, E and F). In contrast, in the model with active dendrites, the AP amplitude was relatively constant during the train. Discussion. The dendrites of BCs differ from those of pyramidal neurons in several ways. First, in BCs, the axon is the invariant AP initiation site. In contrast, in pyramidal neurons APs are preferentially triggered in the axon (17, 18) but can be also initiated in dendrites under specific conditions (2729). Second, in BCs, APs backpropagate into dendrites with marked amplitude attenuation, but little AP broadening and minimal activity dependence. In contrast, in pyramidal neurons backpropagation is less decremental and more activity-dependent (17, 18). Third, BC dendrites are endowed with a low Na+ channel density but a high Kv3 channel density, whereas pyramidal neuron dendrites show a high density of both Na+ channels and A-type (Kv4) K+ channels (19, 36). Thus, Kv3 channels are expressed in dendrites, somata, and axons of fastspiking interneurons (35, 39). Lastly, K+ channels in BC dendrites shorten the EPSP time course, whereas Na+ channels in pyramidal neuron dendrites boost the amplitude and prolong EPSPs (37, 40). Dendritic K+ channels shape the electrical properties of fast-spiking GABAergic interneurons at the level of input, EPSP-AP conversion, and output. At the input level, dendritic K+ channels accelerate the decay time course of unitary EPSPs, reduce the time window for temporal summation, and help BCs to detect the synchronous activity of converging, spatially separated excitatory inputs. Dendritic location puts K+ channels into a strategic position to be efficiently activated by EPSPs and enables them to implement synapse-specific processing rules. At the level of EPSP-AP conversion, dendritic K+ channels ensure precise 1:1 coupling between EPSP and AP, suppressing the generation of AP bursts by large synaptic inputs. Furthermore, dendritic K+ channels ensure a constant AP ini-

Fig. 5. Dendritic K+ channels shape EPSPs and coincidence detection properties of BCs. (A) (Left) Artificial EPSPs (aEPSPs) evoked by injection of an EPSC-like current into the dendrite and recorded at the soma in control conditions (black) and in the presence of 5 mM TEA in the extracellular solution (blue). Peak amplitude of the aEPSC was 0.5 nA. (Right) Summary graph of half-duration of the aEPSP. Data from the same experiment are connected by lines. Peak amplitude of aEPSC was 0.5 3 nA; 10 mM 6-cyano-7nitroquinoxaline-2,3-dione (CNQX) and 2 mM SR95531 were added in four out of nine experiments. (B) Simulation of K+ channel activation in a BC 2 ms after onset of an EPSP in the distal apical dendrite. Color code (right) shows density of activated K+ conductance. (C) Dendritic K+ channels lead to synapse-specific acceleration of the EPSP decay time course. EPSPs with passive dendrites (blue), K+ channels in the apical dendrites (black), and K+ channels in the basal dendrites (gray). Lower right graph shows normalized superposition. Synapse was located on the distal apical dendrite in all cases. Synaptic peak conductance was 1 to 10 nS. (D) Corresponding plot of EPSP half-duration against synaptic peak conductance. (E and F) Dendritic K+ channels enable synapse-specific coincidence detector properties. Plot of paired pulse summation ratio (EPSPmax/EPSP1) for dual activation of the same synapse [(E), continuous lines, distal apical dendrite; and dashed lines, basal dendrite] and activation of different spatially separated synapses (F). Corresponding traces are shown on top. Blue, passive dendrites; black, dendrites enriched with K+ channels.

56

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

RESEARCH ARTICLE
ing on the same BC during movement (42, 43). At the output level, dendritic K+ channels contribute to the fast-spiking AP phenotype (25), enhancing AP repolarization and thereby promoting recovery of Na+ channels from inactivation. Dendritic channels represent an almost infinite pool, which can be efficiently recruited during physiological and pathophysiological AP activity. After APs, dendritic channels will be activated with longer delays than somatic channels with identical gating, enhancing their delayed rectifier properties. Whether and how dendritic K+ channels also contribute to the complex (sometimes anti-Hebbian) induction rules for long-term potentiation at glutamatergic principal neuroninterneuron synapses remains to be determined (16, 44). Because the dendrites of BCs differ significantly from those of somatostatin-expressing interneurons (30, 31), our results demonstrate that the diversity of GABAergic interneurons extends to the dendritic level. Previous results suggested that synaptic facilitation or depression and passive cable properties underlie the routing that leads to a switch from perisomatic to dendritic inhibition during repetitive activity (9). Our findings suggest that dendritic properties may participate in this dynamic switch. In fast-spiking, parvalbumin-expressing BCs, limited synaptic dynamics (14, 16) and dendritic K+ channel activation will facilitate stimulus-locked AP generation in the early phase of repetitive input synapse stimulation. In contrast, in somatostatinpositive interneurons, synaptic facilitation (4547) and a high Na+ channel density (30) will promote asynchronous AP generation late in the train. Thus, dendritic properties may contribute to setting the rules for routing of activity in inhibitory microcircuits.
References and Notes

Fig. 6. Dendritic K+ channels control EPSP-AP coupling and AP phenotype of BCs. (A) Simulation of K+ channel activation in a BC 2 ms after the onset of a somatic current stimulus triggering an AP. Color code (right) shows density of activated K+ conductance. (B) APs (bottom left), total activated dendritic K+ conductance (top right), and axial current in all primary dendrites (bottom right; negative peak truncated) versus time. APs were initiated either by brief somatic current pulses (2 ms, 1.2 nA, 1) or by simultaneous activation of 10 synapses randomly placed on apical dendrites (2). Note that dendritic K+ channels abolished the afterdepolarization by reducing the dendrosomatic current flow (positive values of axial current, Iax). (C) Dendritic K+ channels ensure 1:1 EPSP-AP coupling over a wide range of synaptic strength. (Top) APs evoked by simultaneous activation of 50 synapses. (Bottom) Plots of number of evoked APs against number of activated synapses. (D) Dendritic K+ channels normalize the AP threshold for paired synaptic activation. (Top) APs evoked by consecutive activation of two sets of synapses on different dendrites (10 and 1 to 3 active synapses, respectively). (Bottom) Plots of threshold for AP initiation (in units of number of synapses) against time interval between EPSPs. Continuous lines, dual activation of the same synapse; dashed lines, activation of synapses on different dendrites. (E) Trains of APs evoked by 250-ms 2-nA depolarizing current pulses for passive and active dendrites. (F) Plot of AP amplitude, measured from the peak of AP to the trough of the subsequent afterpotential, versus time for different stimulus intensities (0.5 nA to 2.5 nA in 0.5-nA increments). Blue, passive dendrites; black, active dendrites. In all simulations, voltage given refers to the soma.

tiation threshold during repetitive stimulation. These properties are relevant during in vivo network activity, for example, during sequential ac-

tivation of feedforward and feedback inputs on BCs (41) or sequential activation of presynaptic granule cells with adjacent place fields convergSCIENCE VOL 327

1. G. Buzski, E. Eidelberg, Brain Res. 230, 346 (1981). 2. P. Andersen, J. C. Eccles, Y. Lyning, Nature 198, 540 (1963). 3. B. E. Alger, R. A. Nicoll, J. Physiol. 328, 105 (1982). 4. F. Pouille, M. Scanziani, Science 293, 1159 (2001). 5. M. Bartos, I. Vida, P. Jonas, Nat. Rev. Neurosci. 8, 45 (2007). 6. J. A. Cardin et al., Nature 459, 663 (2009). 7. D. A. Lewis, T. Hashimoto, D. W. Volk, Nat. Rev. Neurosci. 6, 312 (2005). 8. M. Galarreta, S. Hestrin, Science 292, 2295 (2001). 9. F. Pouille, M. Scanziani, Nature 429, 717 (2004). 10. J. Csicsvari, H. Hirase, A. Czurko, G. Buzski, Neuron 21, 179 (1998). 11. J. Csicsvari, H. Hirase, A. Czurk, A. Mamiya, G. Buzski, J. Neurosci. 19, 274 (1999). 12. T. Klausberger et al., Nature 421, 844 (2003). 13. T. Klausberger et al., Nat. Neurosci. 7, 41 (2004). 14. J. R. P. Geiger, J. Lbke, A. Roth, M. Frotscher, P. Jonas, Neuron 18, 1009 (1997). 15. R. Miles, J. Physiol. 428, 61 (1990). 16. H. Alle, P. Jonas, J. R. P. Geiger, Proc. Natl. Acad. Sci. U.S.A. 98, 14708 (2001). 17. G. J. Stuart, B. Sakmann, Nature 367, 69 (1994). 18. N. Spruston, Y. Schiller, G. Stuart, B. Sakmann, Science 268, 297 (1995). 19. D. A. Hoffman, J. C. Magee, C. M. Colbert, D. Johnston, Nature 387, 869 (1997).

www.sciencemag.org

1 JANUARY 2010

57

20. T. Nevian, M. E. Larkum, A. Polsky, J. Schiller, Nat. Neurosci. 10, 206 (2007). 21. G. Stuart, N. Spruston, M. Husser, Eds., Dendrites (Oxford Univ. Press, Oxford, ed. 2, 2007). 22. D. Johnston, J. C. Magee, C. M. Colbert, B. R. Cristie, Annu. Rev. Neurosci. 19, 165 (1996). 23. J. H. Goldberg, G. Tamas, R. Yuste, J. Physiol. 551, 49 (2003). 24. E. H. Buhl, K. Halasy, P. Somogyi, Nature 368, 823 (1994). 25. B. Rudy, C. J. McBain, Trends Neurosci. 24, 517 (2001). 26. Materials and methods are available as supporting material on Science Online. 27. G. Stuart, J. Schiller, B. Sakmann, J. Physiol. 505, 617 (1997). 28. N. L. Golding, N. Spruston, Neuron 21, 1189 (1998). 29. A. Losonczy, J. C. Magee, Neuron 50, 291 (2006). 30. M. Martina, I. Vida, P. Jonas, Science 287, 295 (2000). 31. K. M. M. Kaiser, Y. Zilberter, B. Sakmann, J. Physiol. 535, 17 (2001). 32. M. Husser, G. Stuart, C. Racca, B. Sakmann, Neuron 15, 637 (1995).

33. B. Hille, Ion Channels of Excitable Membranes (Sinauer, Sunderland, MA, ed. 3, 2001). 34. J. Du, L. Zhang, M. Weiser, B. Rudy, C. J. McBain, J. Neurosci. 16, 506 (1996). 35. M. Martina, J. H. Schultz, H. Ehmke, H. Monyer, P. Jonas, J. Neurosci. 18, 8111 (1998). 36. A. Korngreen, B. Sakmann, J. Physiol. 525, 621 (2000). 37. D. Fricker, R. Miles, Neuron 28, 559 (2000). 38. A. Nrenberg, H. Hu, I. Vida, M. Bartos, P. Jonas, Proc. Natl. Acad. Sci. U.S.A. 10.1013/pnas.0910716107 (2010). 39. E. M. Goldberg et al., J. Neurosci. 25, 5230 (2005). 40. G. Stuart, B. Sakmann, Neuron 15, 1065 (1995). 41. A. Sik, M. Penttonen, A. Ylinen, G. Buzski, J. Neurosci. 15, 6651 (1995). 42. M. W. Jung, B. L. McNaughton, Hippocampus 3, 165 (1993). 43. J. K. Leutgeb, S. Leutgeb, M.-B. Moser, E. I. Moser, Science 315, 961 (2007). 44. K. P. Lamsa, J. H. Heeroma, P. Somogyi, D. A. Rusakov, D. M. Kullmann, Science 315, 1262 (2007). 45. A. B. Ali, A. M. Thomson, J. Physiol. 507, 185 (1998). 46. C. Kapfer, L. L. Glickfeld, B. V. Atallah, M. Scanziani, Nat. Neurosci. 10, 743 (2007).

47. G. Silberberg, H. Markram, Neuron 53, 735 (2007). 48. We thank G. Buzski and A. Roth for critically reading the manuscript; P. Somogyi for help with cell identification; A. Nrenberg for providing the passive cable model; and S. Becherer, I. Koeva, M. Northemann, U. Thirimanna, and K. Winterhalter for technical assistance. Supported by the Deutsche Forschungsgemeinschaft (SFB 780/A5, SFB-TR 3/B10, and Leibniz program), the Bundesministerium fr Bildung und Forschung (01 GQ 0420), the Norwegian Research Council (178670/V40), and the Epilepsy Foundation.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1177876/DC1 Materials and Methods Figs S1 to S5 Tables S1 and S2 References 17 June 2009; accepted 19 October 2009 Published online 3 December 2009; 10.1126/science.1177876 Include this information when citing this paper.

REPORTS An Unusually Fast-Evolving Supernova

Dovi Poznanski,1,2* Ryan Chornock,1 Peter E. Nugent,2 Joshua S. Bloom,1 Alexei V. Filippenko,1 Mohan Ganeshalingam,1 Douglas C. Leonard,3 Weidong Li,1 Rollin C. Thomas2 Analyses of supernovae (SNe) have revealed two main types of progenitors: exploding white dwarfs and collapsing massive stars. Here we describe SN 2002bj, which stands out as different from any SN reported to date. Its light curve rose and declined very rapidly, yet reached a peak intrinsic brightness greater than 18 magnitude. A spectrum obtained 7 days after discovery shows the presence of helium and intermediate-mass elements, yet no clear hydrogen or iron-peak elements. The spectrum only barely resembles that of a type Ia SN, with added carbon and helium. Its properties suggest that SN 2002bj may be representative of a class of progenitors that previously has been only hypothesized: a helium detonation on a white dwarf, ejecting a small envelope of material. New surveys should find many such objects, despite their scarcity. The spectrum we obtained a week after detection is extremely blue, with weak yet remarkable features (3). Using a c2 fit, we have digitally compared our (continuum-removed) spectrum with about 4000 spectra of nearly 1400 SNe, allowing for velocity offsets. Not a single spectrum fits well. The closest matches were SNe Ia, mostly due to the absorption feature near 6150 (rest frame), usually attributed to Si II (Fig. 2). The best of those (SN 2009dc) is a superluminous, slowly declining, C-rich, possibly super Chandrasekhar-mass SN Ia (46). These few very luminous SNe reported so far evolve slowly and eject substantial amounts of unburned material, suggesting massive white dwarf progenitors. That is, the closest spectroscopic match has one of the most substantially different light curves. Although the spectra are broadly similar, SN 2002bj has prominent He I lines, which are not expected in a SN Ia. In addition, the spectrum of SN 2009dc had to be artificially redshifted by 3000 km s1 in order to match that of SN 2002bj, implying that SN 2002bj had slower ejecta at the time when the spectrum was taken. Using the code SYNOW (7), we produced synthetic spectra and identified most of the features as coming from helium and intermediatemass elements such as C, Si, and S, but no H (SOM). Although this empirical fit does not produce meaningful abundances, the lack of Fe or other Fe-peak elements in the fit is peculiar. As exceptional is the considerable S II contribution, when compared to Ca II (a ratio never seen before in other SNe). We also report a tentative identification of V II. Although based on only a single line, the relevant spectral region (~3950 ) would have emission from Ca II without it. The spectrum was taken in spectropolarimetry mode, yet there are no polarization line features down to 0.1 to 0.2%. The continuum is consistent with

upernovae (SNe) are usually classified on the basis of telltale lines in their spectra (1). Those empirical types are routinely associated with progenitor systems according to the current understanding of their explosion mechanisms. Type Ia SNe are interpreted as the thermonuclear disruption of a white dwarf, and the other types are interpreted as the core collapse of a massive star. SN 2002bj, which we describe here, would formally belong, according to that classification, to the type Ib class because of the lack of H and the presence of He in the optical spectra we have obtained. However, the overall observed properties of this SN are unprecedented, and the taxonomic classification is misleading.

Department of Astronomy, University of California, Berkeley, CA 947203411, USA. 2Computational Cosmology Center, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA. 3Department of Astronomy, San Diego State University, Mail Code 1221, San Diego, CA 921821221, USA. *To whom correspondence should be addressed. E-mail: [email protected]

SN 2002bj was discovered independently at magnitude (mag) 14.7 by the Lick Observatory SN Search (LOSS) and by amateur astronomers (2) on 28.2 February 2002 (universal time dates are used throughout this paper) in the galaxy NGC 1821. The distance, corrected for local bulk flows (assuming a Hubble constant of 73 km s1 Mpc1), is 50 T 5 Mpc [see supporting online material (SOM) for a discussion of the host galaxy properties and distance]. A prediscovery LOSS image with limiting magnitude 18.4 (Galactic extinction corrected) on 21.2 February 2002 (a week before discovery) shows nothing at that position. As part of our SN followup program, we obtained optical broadband photometry of SN 2002bj in the B, V, R, and I bands for nine epochs over 20 days until it faded below the detection threshold (SOM). Our photometry does not show a rising phase, but the nondetection constrains the rise to be less than 7 days long. The decline was almost as fast, dropping by 4.5 mag (in the B band) in 18 days. SN 2002bj evolves on unprecedented time scales (Fig. 1). VOL 327 SCIENCE

58

1 JANUARY 2010

www.sciencemag.org

REPORTS
polarization by dust in the Milky Way (see SOM for details). Using our photometry, we determined the bolometric evolution of SN 2002bj (SOM). The total radiated energy in optical bandpasses was on the order of 1049 erg, starting at a peak of 1043 erg s1. Assuming blackbody emission, we derived the temporal evolution of the effective temperature, radius, and photospheric velocity. The temperature and velocity declined very rapidly, indicating rapid recession of the photosphere in a low-mass envelope. We estimated the mass of the ejecta using the scaling relation that ties it to the photospheric velocity and rise time v Mej,1 tt1 2 v1 Mej,2 (8). This scaling assumes 2 2 that the opacity is similar to that of a SN Ia. SN 2002bj rose at least three times faster than a normal SN Ia (depending on the assumed explosion date); thus, although its velocity at peak is uncertain (see discussion in SOM), the ejected mass has to be smaller than ~0.15 solar mass (M), about 10% that of a SN Ia. The luminosity and short rise time of SN 2002bj translate to 0.15 to 0.25 M of 56Ni when using Arnett's law (8, 9), if the light curve is solely powered by radioactive 56Ni and its decay product 56Co. Under this same assumption, the rapid decline we measured requires a sharp drop of the gamma-ray deposition efficiency of an order of magnitude in less than 3 weeks. The small ejected mass we derived, the lack of Fe-peak elements in the spectrum, and the relatively large amount of 56Ni required to explain the high luminosity are difficult to reconcile without assuming an additional energy source. SN 2002bj looks spectroscopically somewhat like a SN Ia but with He, C, and an exceptionally fast light curve. Recently, a mechanism has been proposed (10) by which binary white dwarfs of the AM CVn class may undergo a thermonuclear explosion of the He accreted on the primary star. Such a scenario will produce roughly 10% of the luminosity of a SN Ia, for about 10% of the typical time; hence, these objects were dubbed W.IaW SNe. These SNe are expected to be faint (between 15 and 18 mag at peak in the V band) and rapidly evolving (1 to 6 days of rise time, with the brighter objects usually rising more slowly). The decline was not explicitly discussed by Bildsten et al. (10), but the low ejected mass implies a rapid decline. The short time scales of these events may allow the detection of the shortlived radioactive nuclei 52Fe or 48Cr, in addition to the standard 56Ni that drives SN Ia light curves. 48 Cr decays to 48V within a day and then to 48Ti in a week. The decay of these nuclei may (partially) power the optical light curve. The rate of .Ia events is predicted to be roughly a few percent of the SN Ia rate per unit of local volume. The spectral signature was not predicted, but some properties seem to result naturally in that scenario. Because this is a thermonuclear He detonation on a white dwarf, we do not expect any H, but He does seem reasonable, as well as intermediate-mass elements that either survive the convective burning phase and detonation (11) or are produced in the explosion. Although other recent SNe have been proposed to be related to He detonations on a white dwarf [SN 2005E (12) and perhaps also SN 2008ha (1214)], these events have more massive ejecta (0.2 to 0.3 M) and much slower light curves, and thus do not fit the current predictions of .Ia models, though they may be explained with related phenomena involving much more massive He shells. The light curve of SN 2002bj is as fast as predicted in this model but slightly more luminous than expected (15). The high luminosity yet small ejecta mass may be reconciled if some short-lived 48 Cr or 52Fe are synthesized, as their yield per unit of mass is higher than that of 56Ni on short time scales. Our tentative identification of V II in the spectrum supports this hypothesis (48V is the daughter of 48Cr), and the peculiar composition of the spectrum may support it, as well. SNe Ia usually display a prominent secondary peak in their infrared light curve, attributed to line-blanketing by singly ionized Fe-peak elements (16). The lack of a secondary peak in the light curve of SN 2002bj is consistent, within that picture, with our nondetection of Fe-peak elements in the spectrum. Bildsten et al. (10) assumed that the ejecta will have velocities of ~15,000 km s1, which is

Fig. 1. Comparison of the light curve of SN 2002bj to those of SNe of various types (gray dashed lines; R-band magnitudes offset to the same B-band maximum date). SN 2002bj is quite luminous at peak for a core-collapse event, yet faint compared with typical SNe Ia. SN 1994I (21) is often cited as a WfastW SN Ic; SN 2003gs (22) was recently presented as one of the fastest SNe Ia; and SN 2008ha (14) is a faint, peculiar, and fast SN of debated breed. SN 2002bj is significantly faster than any of these. SN 1998S (23), SN 2005cf (24), and SN 2008D (25) are standard representatives of type IIn, Ia, and Ib SNe, respectfully; they are shown for reference. The dashed red line shows the slowest rise slope of SN 2002bj allowed by the data.

Fig. 2. The unique spectral features of SN 2002bj (shown in red; continuum removed) are difficult to identify a priori. Fl , flux per unit of wavelength. This spectrum, taken 7 March 2002 (7 days after discovery), is reminiscent of SNe Ia, with the notable exception of the prominent He and C lines, never seen before in such SNe. We show (in black) a typical SN Ia spectrum near maximum light (SN 2001bf), redshifted by 10,000 km s1 in order to match ejecta velocities. The spectral features identified in black are present in both objects, and the ones in red are seen only in SN 2002bj. www.sciencemag.org SCIENCE VOL 327

1 JANUARY 2010

59

REPORTS
equivalent to all of the binding energy released from fusing He converted into kinetic energy (because the star is assumed to be left bound). We infer for SN 2002bj photospheric velocities that drop rapidly from about 8400 km s1 at detection to 2000 km s1 3 weeks later. Extrapolating to an explosion date 7 days before detection, the initial velocity could have been between 14,000 km s1 (linear extrapolation) and ~25,000 km s1 [exponential extrapolation as often seen in SNe (17, 18)]. A rise time that is faster by a factor of 2 would imply velocities twice as high. The only direct measurement we have is from the spectrum 7 days after detection: about 4000 km s1. This is consistent with the derived photospheric velocity at that time if the rise time was about 7 days. Out to a distance of 60 Mpc, the LOSS survey is complete (99%) for SNe Ia, and 31 have been found. Because SN 2002bj is quite luminous, the incompleteness correction for it is almost as small (94%), resulting in a relative rate of 3.4% of the SN Ia rate for SN 2002bjlike SNe (19). This is in good agreement with the predictions for SNe .Ia. The SN .Ia model, still in its infancy, lacks more stringent predictions such as detailed light curves and spectral composition and evolution. Nevertheless, all the diagnostics we could apply seem consistent with this interpretation. The evidence here is tentative, but the existence of V, if seen in future discoveries of objects of this class, points to a different nucleosynthetic chain and therefore may serve as a smoking gun for a truly different SN explosion channel. Regardless of the interpretation, current and future surveys should focus on short cadencesrepeat visits on daily rather than weekly time scalesin order to find many more SNe resembling SN 2002bj.
References and Notes
1. A. V. Filippenko, Annu. Rev. Astron. Astrophys. 35, 309 (1997). 2. T. Puckett, J. Newton, M. Papenkova, W. D. Li, IAU Circ. 7839, 1 (2002). 3. The blue continuum, in combination with residual host galaxy lines, explains the original erroneous classification of this object as a SN IIn on the basis of a noisier spectrum (20). 4. D. A. Howell et al., Nature 443, 308 (2006). 5. M. Hicken et al., Astrophys. J. Lett. 669, L17 (2007). 6. M. Yamanaka et al., http://arxiv.org/abs/0908.2059 (2009). 7. A. Fisher, D. Branch, P. Nugent, E. Baron, Astrophys. J. Lett. 481, L89 (1997). 8. W. D. Arnett, Astrophys. J. 253, 785 (1982). 9. P. G. Sutherland, J. C. Wheeler, Astrophys. J. 280, 282 (1984). 10. L. Bildsten, K. J. Shen, N. N. Weinberg, G. Nelemans, Astrophys. J. Lett. 662, L95 (2007). 11. K. J. Shen, L. Bildsten, Astrophys. J. 699, 1365 (2009). 12. H. B. Perets et al., http://arxiv.org/abs/0906.2003 (2009). 13. S. Valenti et al., Nature 459, 674 (2009). 14. R. J. Foley, et al., Astron. J. 138, 376 (2009). 15. If our interpretation is correct, it would not be too surprising that the first SN .Ia found would be unusually luminous, because such an object would be easier to find and to recognize as such. 16. D. Kasen, Astrophys. J. 649, 939 (2006). 17. P. Nugent et al., Astrophys. J. 645, 841 (2006). 18. X. Wang et al., Astrophys. J. Lett. 699, L139 (2009). 19. Poisson statistics allow fractions in the range from 0.7 to 11% (1s). 20. 21. 22. 23. 24. 25. 26. T. Matheson, P. Berlind, IAU Circ. 7844, 5 (2002). M. W. Richmond, et al., Astron. J. 111, 327 (1996). K. Krisciunas, et al., http://arxiv.org/abs/0908.1918 (2009). A. Fassia et al., Mon. Not. R. Astron. Soc. 318, 1093 (2000). X. Wang et al., Astrophys. J. 697, 380 (2009). M. Modjaz et al., Astrophys. J. 702, 226 (2009). We thank L. Bildsten for valuable insights into the SN .Ia model; A. Gal-Yam, D. Kasen, D. Maoz, T. Matheson, P. Mazzali, E. Ofek, E. Quataert, K. Shen, and N. Smith for useful discussions; R. Foley for reducing the Lick 3-m spectrum of SN 2002bj; and A. A. Miller and A. Merritt for the DeepSky analysis. A.V.F.'s group has been supported by NSF grants AST-0607485 and AST0908886, by U.S. Department of Energy grants DE-FC0206ER41453 (SciDAC) and DE-FG02-08ER41563, and by the TABASGO Foundation. The Katzman Automated Imaging Telescope and its ongoing operation were made possible by donations from Sun Microsystems, the Hewlett-Packard Company, AutoScope Corporation, the Lick Observatory, NSF, the University of California, the Sylvia & Jim Katzman Foundation, and the TABASGO Foundation. Some of the data presented here were obtained at the W. M. Keck Observatory, which is operated as a scientific partnership among the California Institute of Technology, the University of California, and NASA; the observatory was made possible by the generous financial support of the W. M. Keck Foundation. We thank the staffs at the Lick and Keck observatories for their assistance.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1181709/DC1 SOM Text Figs. S1 to S4 Tables S1 to S3 References 9 September 2009; accepted 29 October 2009 Published online 5 November 2009; 10.1126/science.1181709 Include this information when citing this paper.

Polarization-Induced Hole Doping in WideBand-Gap Uniaxial Semiconductor Heterostructures

John Simon, Vladimir Protasenko, Chuanxin Lian, Huili Xing, Debdeep Jena* Impurity-based p-type doping in wideband-gap semiconductors is inefficient at room temperature for applications such as lasers because the positive-charge carriers (holes) have a large thermal activation energy. We demonstrate high-efficiency p-type doping by ionizing acceptor dopants using the built-in electronic polarization in bulk uniaxial semiconductor crystals. Because the mobile hole gases are field-ionized, they are robust to thermal freezeout effects and lead to major improvements in p-type electrical conductivity. The new doping technique results in improved optical emission efficiency in prototype ultraviolet light-emittingdiode structures. Polarization-induced doping provides an attractive solution to both p- and n-type doping problems in wideband-gap semiconductors and offers an unconventional path for the development of solid-state deep-ultraviolet optoelectronic devices and wideband-gap bipolar electronic devices of the future. he direct-gap III-V nitride semiconductor family and its alloys span the widest spectral range of band gaps (Eg) among all semiconductors, ranging from the infrared (InN, Eg = 0.7 eV) through the visible and the ultraviolet (UV) (GaN, Eg = 3.4 eV) to the deep UV range (AlN, Eg = 6.2 eV). This property is the basis for its applications in short-wavelength

lasers (1, 2) and in light-emitting diodes (LEDs) for solid-state lighting applications (3, 4). In addition, the wide band gaps, availability of heterojunctions, high electron-saturation velocities, and high breakdown fields enable high-speed and high-power electronic devices. Compact shortwavelength, solid-state light sources will enable a wide range of applications such as high-density VOL 327 SCIENCE

optical data storage, water treatment, sterilization of medical equipment, UV-enabled security marks on credit cards and currency bills, and biological and cellular imaging. Currently, the III-V nitride semiconductors offer the most viable approach toward the realization of high-efficiency, deep-UV optical emitters based on semiconductors (2). A problem that has persisted since the early 1990s and is becoming increasingly troublesome is the high resistivity of p-type GaN and AlGaN layers. The activation energy EA of the most commonly used acceptor dopant (Mg) in GaN is ~200 meV (57), several times the thermal energy kBT at room temperature (where kB is the Boltzmann constant, and T is temperature). The activation energy of acceptors increases with the band gap, reaching EA ~ 630 meV in AlN (1). For comparison, the donor (Si) activation energies are ED ~ 15 meV for GaN and ED ~ 282 meV for AlN (1). Thus, the thermal activation of holes is highly inefficient at room temperature for GaN and becomes increasingly problematic for higherband-gap AlGaN and AlN layers. As a result, injection of holes is a severe impediment for light-emitting devices in
Department of Electrical Engineering, University of Notre Dame, 275 Fitzpatrick Hall, Notre Dame, IN 46556, USA. *To whom correspondence should be addressed. E-mail: [email protected]

60

1 JANUARY 2010

www.sciencemag.org

REPORTS
the UV and deep-UV spectral windows. High p-type resistance leads to excessive Joule heating of p-doped AlGaN layers for Al composition xAl 20%. Instead, p-GaN layers must be used and absorption losses incurred in the narrower band-gap region. Furthermore, hole reflection and trapping at heterojunction valence-band offsets block hole injection into optically active AlGaN regions (2) and reduce the efficiency of such devices. An alternative strategy for efficient p-type doping and hole injection in widebandgap semiconductors is therefore highly desirable at this time. The large ionic component of the Ga(Al)-N bonds, combined with the deviation of their equilibrium lattice structure from ideal wurtzite crystals, give rise to giant spontaneous polarization fields in III-V nitride semiconductors (8, 9). In addition, the strain-induced piezoelectric component of the fixed charge in the nitrides is the highest among all III-V semiconductors. At abrupt Al(Ga)N/GaN heterojunctions, the sharp discontinuity in the polarization field leads to the formation of a bound sheet charge sp at the heterointerface, captured by the Gauss law boundary condition sp (P1 P2 ) n, where % n is the unit vector normal to the heterointerface, % and (P1 , P2 ) are the polarization fields across the heterojunction. When wurtzite nitride crystals are grown along the [0001] orientation (metal or Fig. 1. Schematic illustration of polarization-induced p-type doping in graded polar heterostructures. (A) Sheets of charge dipoles in every unit cell of the crystal. The net unbalanced polarization charge is shown in (B), which leads to the electric field in (C), and the energyband bending in the valence band in (D) if holes are not ionized. Field ionization of holes results in a steady-state energy-band diagram shown in (E), which highlights the smooth valence-band edge without any potential barriers for hole flow. Ef, is the Fermi level; Ec and Ev are the conduction and valence-band edges, respectively; and Eg is the band gap. Ga-face), a positive bound polarization charge creates a high electric field and energy-band bending, such that a mobile two-dimensional electron gas (2DEG) forms at AlGaN/GaN heterojunctions without the need for intentionally introduced impurity dopants. The bound sheet-charge density can be as high as sp ~ 6 1013 cm2 at pseudomorphic AlN/GaN heterojunctions, facilitating mobile 2DEGs with a very high charge carrier density. For example, in AlN/GaN semiconductor heterostructures, the mobile 2DEG concentrations are 4 1013 cm2 (10). Such polarization-induced 2DEGs form the basis of nitride high-electron mobility transistors that have surpassed transistors made from any other semiconductor family in RF power performance (11). The polarization fields have also been exploited to create parallel sheets of 2D hole gases in Mg-doped AlGaN/GaN multiplequantumwell structures (12, 13). Although such parallel 2D hole sheets have high conductivity in the plane of the heterojunctions, they suffer from low conductivity perpendicular to the interfaces because of potential barriers in the valence band that require transport to occur through tunneling or thermionic emission processes. Even in shortperiod superlattice structures, the large effective mass of holes in minibands results in low mobility and high resistance (1). An alternate strategy for hole doping without potential barriers will facilitate higher conductivities. If instead of sharp heterojunctions we grew a compositionally graded crystal, the bound polarization-induced sheet charge spreads to a bound 3D form in accordance with rp (z) P(z), where rp(z) is the volume charge density in the polar (z) direction, and is the divergence operator. For [0001]-oriented Ga-face crystals graded from GaN to AlGaN, the polarization bound charge is positive and induces the formation of a mobile 3D electron gas. These 3D electron slabs are quite distinct from those formed by donor impurity doping: Because the carriers are created by effective electrostatic field ionization, they require no impurity incorporation, and thus exhibit virtually no freezeout at cryogenic temperatures as opposed to thermally ionized carriers in shallow, donor-doped layers (14). The resulting 3D electron gases have higher n-type conductivity than impurity-doped layers of comparable carrier concentration, because ionized impurity scattering is absent. The absence of freezeout and high mobilities made it possible to observe Shubnikovde Haas oscillations (15). Polarization-induced field-effect transistors have also been demonstrated recently with this technique (16). By the same measure, flipping the polarity of the crystal (growing along the N-face, which is

polarization charge dipoles

0001 (N-Face)

unbalanced charges graded AlGaN

charge neutrality lines

unit cells GaN

Charge

B
Net polarization-induced bound unbalanced charges depth from the surface

Field

C
Electric field due to unbalanced polarization charges band gap

Deep acceptor levels

Field-induced hole ionization

Ev

E
Energy Eg Ec Ef Field-ionized 3D hole gas Smooth valence band for enhanced vertical transport Energy-band Diagram Ev

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

61

REPORTS
the 0001 direction) and compositional grading from GaN to AlGaN should result in mobile 3D hole slabs. The ability to do so without the introduction of Mg-acceptor dopants hinges on the propensity of the surface to act as a remote acceptor state. The surface of III-V nitride semiconductors freely provides mobile electrons, but not holes, and this difference has been attributed to the presence of deep-level traps that localize holes (17). Lowering of defect and trap densities may enable dopant-free p-type carriers, but intentionally introducing Mg-acceptor dopant atoms in the N-face graded AlGaN layer serves as the necessary source of holes. This work demonstrates the ability to use the polarization charges Fig. 2. Structural and transport properties of p-type samples (A) Concentration of Al and Ga atoms in a compositionally graded AlGaN sample (sample d), with the measured concentration of Mg dopant atoms. The thickness of the graded layer is d ~ 85 nm, capped with a thin, heavily doped p++ layer for ohmic contacts. SIMS, secondary ion mass spectrometry. (B) Measured temperature-dependent resistivity for samples a to c, highlighting the polarization boost in p-type conductivity. in N-face 0001 layers to generate polarizationinduced, p-type graded AlGaN slabs that are highly conductive. The mechanism of polarization-induced hole formation is illustrated schematically in Fig. 1. The total polarization (spontaneous plus piezoelectric) can be pictured as charge dipoles in every unit cell of the crystal (Fig. 1A). Because AlxGa1xN (where x is the Al mole fraction) has higher polarization than GaN, the sheet-charge dipoles in unit cells of the AlGaN layer are of a higher magnitude than in GaN, so the dipole strength increases linearly with the composition. When the composition of the layer is graded with increasing Al mole fraction, the net unbalanced bound polarization charge is negative (Fig. 1B), given by rp(z) = P(z) ~ 5 1013 (x2 x1)/d cm3, where x1 and x2 are the Al compositions at the ends of the graded layer of thickness d (in centimeters). This bound charge creates a built-in electric field (Fig. 1C) and energy-band bending that would be greater than the band gap of the semiconductor layer if left uncompensated (Fig. 1D). To neutralize the bound, negative polarization charge, holes are consequently fieldionized from the deep Mg-acceptor atoms and form a high-density mobile 3D hole gas. The concentration of the 3D hole gas should then exhibit a weak temperature dependence and resist freezeout at low temperatures. In addition, the

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 50 100 150 200 10 10 10 10 10 250
20 19 18 17 16

B
100

10

0.1 100

150

200

250

300

300

200

100

B
50

C
70 10
20

60 10
18

40 50 30 10
19

10

17

40

20

30

~0.17 eV

0.16
20 10

10

18

10

16

0.30
10 1017 0 100 200 300

EA Ev
10
15

10

0 100

0 150 200 250 300

Fig. 3. Hall-effect temperature-dependent (A) hole concentration, (B) hole mobilities, and (C) hole concentration and mobility measured down to T = 4 K. The polarization-doped graded AlGaN p-type layers show higher hole

concentrations and conductivities. Holes in polarization-doped layers are resistant to freezeout at low temperatures, and their mobility and concentration can be measured down to cryogenic temperatures. SCIENCE www.sciencemag.org

62

1 JANUARY 2010

VOL 327

REPORTS
smooth spatial variation of the valence-band edge (Fig. 1E) should further facilitate highconductivity p-type transport in both lateral and vertical directions. To test this concept of polarization-induced p-type doping, Mg-doped graded AlGaN layers were grown on top of semi-insulating N-face 0001 GaN substrates by plasma-assisted molecular beam epitaxy. A Mg-doped GaN sample (Na ~ 1019/cm3, sample a; here, Na is the acceptor concentration) was used as a control sample. Graded AlGaN samples doped with the same Mg concentration but linearly graded from x = 0 to 0.16 (sample b) and x = 0 to 0.3 (sample c) over d ~ 85 nm were grown. The sample structures and compositions were characterized by x-ray diffraction, in situ reflection, highenergy electron diffraction patterns and atomic force microscopy [see supporting online material (SOM) for a description of crystal growth and characterization (18)]. Secondary ion mass spectrometry measurements were performed on a separate graded AlGaN sample (x = 0 to 0.4, sample d) as well as the control Mg-doped GaN sample to verify the incorporation of Mg atoms into the crystal and the linear grading of Al composition in the polarization-doped AlGaN layers (Fig. 2A). Samples a to c were subsequently processed for Hall-effect measurements, as described in the SOM (18). The measured resistivities of the bulk p-GaN and polarization-doped AlGaN layers from T = 300 to 100 K are shown in Fig. 2B. The roomtemperature resistivity of both polarization-doped samples b and c (rb, rc ~ 0.6 ohmcm) is lowered by more than a factor of 2 compared with that of the control sample a (ra ~ 1.22 ohmcm). The resistivity of the control sample a increased monotonically by more than two orders of magnitude as the temperature was lowered from 300 to 100 K (Fig. 2B); this increase is expected from the freezeout of thermally activated holes. In comparison, the resistivities of the polarization-doped samples b and c actually decreased with temperature, which is indicative of metallic behavior. This decrease in resistivity can occur if (i) polarization-induced holes do not freeze out at low temperatures and (ii) the mobilities of polarization-induced holes increase when temperature is lowered from 300 to 100 K. Temperature-dependent Hall-effect measurements performed at a magnetic field of 0.5 T confirmed the two hypotheses. The measured hole concentrations and mobilities are shown in Fig. 3, A and B. Compared with the exponential freezeout (activation energy EA ~ 170 meV) of mobile holes for the Mg-doped GaN control sample (a), the hole densities in the polarization-induced graded AlGaN samples (b and c) remain essentially independent of temperature, and are near the theoretical prediction [rp ~ 5 1013 (x2 x1)/d cm3], as indicated by the thick gray lines in Fig. 3A. Polarization charges are atomic in origin and do not require thermal energy to be activated, so they enhance the hole concentration independent of temperature. In addition, because polarization charges are spatially distributed, the band-edge potential variations are smooth, and no abrupt potential barriers exist for the flow of holes along any direction. These properties are a major advantage and novelty of this method of p-type doping. Polarization enhancement of hole densities are 2 and 6 for samples b and c at room temperature and many orders of magnitude at lower temperatures. The measured hole mobilities as a function of temperature in samples a to c are shown in Fig. 3B. Sample c has lower hole mobility because of increased alloy scattering. Although it was not possible to perform Hall-effect measurements for control sample a below T ~ 150 K because of carrier freezeout, we measured the polarizationenhanced hole concentration and mobility down to T = 4 K for sample b. As shown in Fig. 3C, the hole concentration showed a very small decrease with temperature, whereas the hole mobility increased to mp ~ 65 cm2/Vs at 30 K before decreasing, indicating competition between phonon and impurity scattering. To test the effectiveness of such polarizationenhanced p-type layers as hole injectors in optical devices, a Mg-doped graded AlGaN layer (x = 0 to 30%, identical to sample c) was grown on a n-type doped N-face 0001-oriented GaN substrate. A control p-n junction with a Mg-doped GaN p-type layer was also grown, and these structures were fabricated into light-emitting diode structures [see SOM for the fabrication procedure (18)]. These junctions serve as prototype LEDs, requiring electrical injection of holes and electrons into the depletion region where they recombine radiatively to emit photons. Under forward bias at room temperature, both devices exhibit electroluminescence in the UV spectral range (Fig. 4A). We observed characteristic subband-gap emission attributed to deep acceptor levels in nitrides (19). Furthermore, we note that the graded AlGaN p-layer structure showed much brighter optical emission (Fig. 4B), which we attribute to two factors: (i) improved p-type conductivity in the vertical direction due to polarization-induced hole doping and (ii) the existence of a built-in quasi-

A
140 120 100 80 60 40 20 0

3.4

3.2

3.0

2.8

2.6

B
35 30 25

Electroluminescence @40 mA

electron blocking layer also blocks holes p-GaN Conventional LED structure p-AlGaN p-GaN n-GaN
EC EF

Graded AlGaN p-n junction

visible blue emission Graded AlGaN p-n junction x10 Bulk GaN p-n Junction

20 15 10 5
polarization boost
electron blocking

active region
e-

EV

Polarization-doped LED structure

EC EF

360

380

400

420

440

460

480

500

0 0.0 0.2 0.3 0.5 0.7 0.8

p-graded AlGaN

EV

Fig. 4. (A) Room temperature electroluminescence of the graded p-type AlGaN junction and the control GaN p-n junction at 40-mA drive current. Both samples have an area of 80 mm by 150 mm. (Inset) Optical microscope micrograph displaying the blue part of the emission of the graded AlGaN junction. a.u., arbitrary units. (B) Relative output intensity with increasing drive current for the graded AlGaN p-n junction and the control bulk-doped p-n junction. The www.sciencemag.org SCIENCE

polarization-doped diode shows much brighter emission than the bulk-doped p-n junction. Schematic energy-band diagrams of a conventional (C) LED device and (D) a polarization-doped device. The graded AlGaN p-n junction uses the entire band offset DEg in the conduction band as an electron-blocking layer, resulting in enhanced electroluminescence. In comparison, a traditional electron-blocking layer (C) also blocks holes through a valence-band offset DEv. VOL 327 1 JANUARY 2010

n-GaN

Bulk GaN p-n Junction

no hole blocking

63

REPORTS
electric field imposed on minority electrons injected into the p-type layer of the graded AlGaN. The compositional grading in the p-type AlGaN layer causes the increase in the band gap DEg to appear entirely in the conduction band, which acts as a natural electron-blocking layer. This feature is shown in the energy-band diagram in Fig. 4D. Electron-blocking layers have been shown to improve the efficiency of emission by preventing the spillover of higher mobility electrons from the optically active regions of nitride LEDs (20). Conventional electron-blocking layers implemented in nitride LEDs and lasers consist of a thin AlGaN layer placed on the p-doped side [schematic band diagram shown in Fig. 4C]. In addition to blocking electron overflow through a conduction band barrier DEC, such layers also prevent efficient hole injection because of the unavoidable valence-band offset DEV (21). Polarization-doped graded layers provide a solution to this design bottleneck. In addition to improving the p-type conductivity, the polarization-induced graded p-type AlGaN layer facilitates electron blocking without adding barriers to hole injection and offers an added degree of freedom in gradedrefractive-index design, all of which are useful for UV laser applications. The polarization-doped layer is also of a larger band gap than the active region of the p-n junction and serves as a natural optically transparent layer with minimal absorption losses. This method of polarization doping should prove particularly useful for deep-UV optoelectronic applications where both p- and n-type doping of high Al composition AlGaN is a major challenge. The technique presented here could be applied to produce highly conductive p-type regions in wideband-gap nitrides composed of highAl composition AlGaN and in the more general AlInGaN material system with proper choice of the crystal direction of growth and management of strain within allowable limits. The doping scheme can be used to obtain desired hole or electron concentrations in spite of poor ionization efficiencies of deep-level dopants in any semiconductor crystals that possess sufficiently strong spontaneous and piezoelectric polarization (for example, in the ZnO material family).
References and Notes
1. Y. Taniyasu, M. Kasu, T. Makimoto, Nature 441, 325 (2006). 2. A. Khan, K. Balakrishnan, T. Katona, Nat. Photonics 2, 77 (2008). 3. D. A. Steigerwald et al., IEEE J. Sel. Top. Quantum Electron. 8, 310 (2002). 4. O. Brandt, K. H. Ploog, Nat. Mater. 5, 769 (2006). 5. P. Kozodoy et al., J. Appl. Phys. 87, 1832 (2000). 6. W. Gtz, N. M. Johnson, J. Walker, D. P. Bour, R. A. Street, Appl. Phys. Lett. 68, 667 (1996). 7. C. J. Eiting, P. A. Grudowski, R. D. Dupuis, Electron. Lett. 33, 1987 (1997). 8. F. Bernardini, V. Fiorentini, D. Vanderbilt, Phys. Rev. B 56, R10024 (1997). 9. C. Wood, D. Jena, Polarization Effects in Semiconductors: From Ab-Initio Theory to Device Applications (Springer, New York, 2007). 10. Y. Cao, D. Jena, Appl. Phys. Lett. 90, 182112 (2007). 11. Y.-F. Wu et al., IEEE Electron Device Lett. 25, 117 (2004). 12. I. D. Goepfert, E. F. Schubert, A. Osinsky, P. E. Norris, Electron. Lett. 35, 1109 (1999). 13. P. Kozodoy, M. Hansen, S. P. DenBaars, U. K. Mishra, Appl. Phys. Lett. 74, 3681 (1999). 14. J. Simon, A. K. Wang, H. Xing, S. Rajan, D. Jena, Appl. Phys. Lett. 88, 042109 (2006). 15. D. Jena et al., Phys. Rev. B 67, 153306 (2003). 16. S. Rajan, H. Xing, S. DenBaars, U. K. Mishra, D. Jena, Appl. Phys. Lett. 84, 1591 (2004). 17. S. Rajan et al., Appl. Phys. Lett. 102, 044501 (2007). 18. Details of the growth and fabrication procedure are described in the supporting online material. 19. F. Calle et al., MRS Internet J. Nitride Semicond. Res. 3, article 24 (1998). 20. R.-C. Tu et al., IEEE Photon. Technol. Lett. 15, 1342 (2003). 21. S.-H. Han et al., Appl. Phys. Lett. 94, 231123 (2009). 22. We thank the U.S. Office of Naval Research and the NSF (award no. 0907583) for financial support and C. Wood for discussions. N-face semi-insulating substrates were obtained from D. Hanser, Kyma Technologies, Raleigh, North Carolina.

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/60/DC1 Materials and Methods Figs. S1 to S3 References 12 October 2009; accepted 30 October 2009 10.1126/science.1183226

Translocation of Single-Stranded DNA Through Single-Walled Carbon Nanotubes

Haitao Liu,1* Jin He,2* Jinyao Tang,1 Hao Liu,2,3 Pei Pang,2,4 Di Cao,2,4 Predrag Krstic,5 Sony Joseph,5 Stuart Lindsay,2,3,4 Colin Nuckolls1 We report the fabrication of devices in which one single-walled carbon nanotube spans a barrier between two fluid reservoirs, enabling direct electrical measurement of ion transport through the tube. A fraction of the tubes pass anomalously high ionic currents. Electrophoretic transport of small single-stranded DNA oligomers through these tubes is marked by large transient increases in ion current and was confirmed by polymerase chain reaction analysis. Each current pulse contains about 107 charges, an enormous amplification of the translocated charge. Carbon nanotubes simplify the construction of nanopores, permit new types of electrical measurements, and may open avenues for control of DNA translocation. e report the use of single-walled carbon nanotubes (SWCNTs) as nanopores for analyzing molecular transport properties. Nanopores are orifices of molecular diameter that connect two fluid reservoirs. At this length scale, the passage of even a single molecule generates a detectable change in the flow of ionic current through the pore (1, 2). They can be used as single-molecule Coulter counters and form the basis of proposed new approaches to DNA sequencing (3). The first nanopore devices were based on pore proteins (47), but more

recently pores have been fabricated by drilling (and sometimes partially refilling) solid-state materials (812). Nanochannels have been formed by etching silicon nanowires (13), and channels with one nanoscale dimension have been etched into glass (14) or quartz (15). Carbon nanotubes are obvious candidates for the fabrication of nanopore structures. Pressuredriven gas, water, and ion transport has been recorded through membranes composed of many multiwalled carbon nanotubes (16) or doublewalled carbon nanotubes (17). These experiVOL 327 SCIENCE

ments showed that the water flow rate is greatly enhanced inside the tube, an effect confirmed by molecular dynamics simulations (18). DNA has been passed through a 100-nm diameter carbon nanotube (19) and 50-nm-wide hydrophilic channels (13). It seems counterintuitive that hydrophilic DNA would enter the hydrophobic interior of a SWCNT, but simulations show that both RNA (20) and DNA (21) will translocate through 1.5- to 2-nm diameter tubes. The simulations were carried out using very large electric fields (tenths of a volt per nm) to generate observable motion on the simulation time scale. This result leaves open the possibility that some measurable translocation might occur at the much smaller fields that could be implemented in the laboratory. Here, we report direct measurement of this translocation. We have made a device in which one SWCNT spans a barrier between two fluid reservoirs [see Fig. 1 and Supporting Online Material (22)]. Relative to CNT membranes (16, 17), this arrangement makes it possible to detect signals from the translocation of a single molecule and to correlate
1 Department of Chemistry, Columbia University, New York, NY 10027, USA. 2Biodesign Institute, Tempe, AZ 85287, USA. 3 Department of Chemistry and Biochemistry, Tempe, AZ 85287, USA. 4Department of Physics, Arizona State University, Tempe, AZ 85287, USA. 5Physics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA.

*These authors contributed equally to this work. To whom correspondence should be addressed. E-mail: [email protected] (S.L.); [email protected] (C.N.)

64

1 JANUARY 2010

www.sciencemag.org

REPORTS
A B
Input channel IR AM OR Output channel

C
20

10

DNA Translocates Normal

0 0 0.01 0.1 1 10 100 IONIC CONDUCTANCE IN 1M KCl (nS)

Fig. 1. (A) A single nanopore device was fabricated by growing SWCNTs at low densities on an oxidized Si wafer. We used cobalt catalyst particles with ethanol vapor as the carbon source in conditions most likely to produce high-quality SWCNTs with an outside diameter of 1 to 2 nm (30). A 700-nm layer of PMMA resist is spun on and reservoirs opened over selected tubes with electronbeam lithography. The exposed regions of SWCNTs were removed by O2 plasma etch. The SEM image of the device shows a 2-mm barrier before removal of the exposed SWCNT (arrows). Pillars in the reservoir support the PDMS cover. (B) Optical micrograph taken through a PDMS cover. The reservoirs (IR, input; OR, output) span the barrier between PDMS channels at an angle of about 60. AM marks the location of one set of alignment markers. (C) Current flows through the single SWCNTs and not through a leakage path. With the SWCNT bridging the gap and opened, most tubes pass currents in the expected range (Normal), but 20% pass unexpectedly large currents. Some of these (marked in red) also passed DNA oligomers. These data are limited to the subset of devices exposed to short plasma etches for which control experiments show no leakage (22). Table 1. Relation of ionic conductance with electrical properties (VT is the threshold voltage for semiconducting tubes). These measurements do not discriminate between metallic SWCNTs and bundles containing a metallic tube, but most of the tubes are single-walled (30). Raman scattering was used to determine diameters marked * and confirm electronic properties marked . The tube marked translocated DNA.
FET device ID HL_4_1_41 HL_4_1_10 HL_4_1_39 HL_4_1_41 HL_4_1_41 AP3 AB6 P6 AZ3 N2 Ionic current 10.7 T 0.05 nA/0.4 V 3.4 T 0.04 nA/0.5 V 2.5 T 0.07 nA/0.4 V 1.91 T 0.05 nA/0.4 V 0.98 T 0.04 nA/0.4 V (VT ~ 10 V) 0.46 T 0.03 nA/0.5 V (VT ~ 10 V) 0.07 T 0.02 nA/0.4 V (VT ~ 25 V) 0.10 T 0.03 nA/0.5 V (VT ~ 10 V) <10 pA/0.4 V(VT ~ 25V) Diameter (nm) 2.0 1.7 (1.5*) 4.2 0.9 1.3* 1.8 1.1* 3.4 Electrical property Metallic Metallic Metallic Metallic Semiconducting Semiconducting Semiconducting Semiconducting Semiconducting

HL_4_1_37 AB20 HL_4_1_41 AS3 HL_4_1_37 Z22 HL_4_1_41 M8

Ion conductance (nS)

12 10 8 6 4 2 0

Ion current (nA)

100 80 60 40 20

Ion conductance (nS)

14

120

25 20 15 10 5 0

1M

-5 -0.4 -0.2 0 0.2 0.4

0.5

1.5

0 2.5

10

Bias (V)

KCl concentration (M)

External diameter (nm)

Fig. 2. Ion transport in the subset of SWCNTS with high ionic conductance. (A) Current versus voltage applied to Ag/AgCl reference electrodes for a 2-mm long, 1.7-nm diameter SWCNT for various concentrations of KCl electrolyte as marked. The solid lines are simulated as described in the text. (B) Ionic conductance as a function of salt concentration. The red line is a fit to the cm dependence suggested by molecular dynamics simulations. We found 0.33 < m < 0.4 in three different tubes. The blue line shows the salt dependence of conductance measured in a planar nanopore (27). (C) In this subset of tubes, current at 1 M KCl is better related to diameter (green squares). The red dashed lines show simulations for excess charge densities of 2, 3, and 4 M. www.sciencemag.org SCIENCE VOL 327

transport with the properties of individual SWCNTs. We grew well-separated SWCNTs on the surface of oxidized silicon wafers and formed fluid reservoirs along the path of chosen tubes with electron-beam lithography. A scanning electron microscopy (SEM) image of a device at this stage is shown in Fig. 1A, where the SWCNT is just visible on each side of the barrier. An oxygen plasma was used to remove the exposed parts of the SWCNT, leaving the SWCNT under the barrier intact (fig. S4I) (22, 23). The fluidic pathway was completed by placing a poly(dimethylsiloxane) (PDMS) cover on top of the chip (Fig. 1B). Each chip also contained control devices lacking the bridging SWCNT (22) to check the integrity of the barrier, including devices with unopened SWCNTs. We used a mild plasma treatment such that 100% of the devices lacking CNTs did not leak (fig. S5), although this approach resulted in a large fraction of tubes that were not opened (20%), as determined by SEM imaging (fig. S4F). The fluid reservoirs were filled with 1 M KCl, and Ag/AgCl electrodes (BASI MF-2078) were used to measure the conductance across the reservoirs connected by the SWCNT. The devices passed current if, and only if, they were spanned by a SWCNT that was opened (Fig. 1C), so the interface between the tube and the poly(methyl methacrylate) (PMMA) does not appear to provide a leakage path. This conclusion was verified by chemically tethering poly(ethylene glycol) molecules to one or both ends of the CNTs. The current was reduced in one direction of bias when the tube was modified at one end, and in both directions of bias when the tube was modified at both ends (fig. S6). The ionic conductance of a tube of electrolyte should be given by G = 6.02 1026 (mK + mCl)cKClepD2 / 4L, where mK = 7.62 108 m2/ V s, mCl = 7.91 108 m2/ V s, cKCl is the KCl concentration in mole/l, e the electronic charge, D the tube diameter, and L the tube length. Table 1 shows that there is no correlation between the tube diameter and ionic conductance. The ionic conductance spans nearly four orders of magnitude (Fig. 1C), with only the lowest conductances (the range marked normal in Fig. 1C) being consistent with the classical formula for cKCl = 1 M, 1 nm < D < 5 nm (fig. S7) and L = 2 mm. We also measured the electronic properties of some of the tubes (Table 1) using both their response as field-effect transistors and Raman scattering (figs. S8 to S10). The SWCNTs with the highest ionic conductance are all metallic. We considered whether the excess current could be accounted for by electrochemical currents stemming from reduction and oxidation reactions at the end of metallic tubes. A conducting tube suspended in a potential gradient in an electrolyte acts as a bipolar electrode (24), but enormous fields are required to drive electrochemical processes at the ends of a bipolar carbon nanotube electrode (25). Measurements with

DEVICE COUNT

1 JANUARY 2010

65

REPORTS
an electrode contacting the SWCNT directly revealed that electrochemical currents were negligible for the potentials used here (fig. S11). To look for clues to a mechanism for the large ionic currents, we used molecular dynamics simulations coupled with solutions of the Poisson-Nernst-Planck equation for transport in the SWCNT and the outside reservoirs (22). The flow rate of water is greatly enhanced inside SWCNTs (17), but the molecular dynamics simulations showed that the electrophoretic mobility of ions is similar to that in the bulk electrolyte. However, the selective filtering of anions or cations owing to charged end groups (26) can result in a net excess concentration, n, of one charge inside the tube. This charge will, in turn, drive an electroosmotic current. Molecular dynamics simulations further showed that both water and ions flow with an electroosmotic velocity, v, given by v n0.74 for a (10,10) SWCNT. Both anions and cations are driven in the same direction by an extremely large electroosmotic flow, but only the charge imbalance inside the tube results in a net ionic current proportional to nv, that is, n1.74. The mechanism of charge accumulation is complex and involves both charged end groups and the electronic properties of the SWCNT, and we have not yet developed a quantitative model for it (see fig. S12 for further evidence of the role of charged end groups). However, current-voltage curves obtained at different salt concentrations in the reservoirs, c, can be fitted if n = 3.31 c0.22 M (Fig. 2A). This result is equivalent to an ionic conductance that varies as c0.39, shown by the red curve passing through the measured data in Fig. 2B. This dependence on concentration is quite different from the linear dependence expected for a tube of electrolyte or the saturation at low salt observed for a planar nanopore carrying a surface charge (27). In contrast to the full set of devices, the subset with anomalously high conductance does show some relation between conductance and tube diameter (Fig. 2C, green squares). The red dashed lines show simulated values of ionic conductance as a function of diameter for n = 2, 3, and 4 M. The measured data can be accounted for by assuming that variability in the charge of end groups leads to some variability in n. Although it might be instructive to study the translocation of simpler polyelectrolytes as a prelude to the study of DNA, methods such as dye-labeling are much less sensitive than polymerase chain reaction (PCR) for detecting and counting small numbers of molecules. To test for DNA translocation of SWCNTs, we used 60-nt and 120-nt DNA oligomers with sequences that were predicted to be relatively free of secondary structure, with forward and reverse primers chosen to have high melting temperatures to minimize primer dimers and false priming (22). Devices were characterized by measuring current flow with 1 or 2 M KCl alone, and then a DNA solution [1 or 2 M KCl, 1 mM phosphate-buffered saline (PBS), pH 7] was flowed into the input reservoir side. A control sample was collected from the output reservoir to check for DNA contamination, and a positive bias was then applied to the output side of the device. In the subset of high current tubes, we first observed a slow increase in the background current (Fig. 3, A and B; data are for 0.1 nM DNA). After a time, which varied from a few to tens of minutes, depending on the DNA concentration in the input reservoir, large transient increases in current were observed. These spikes were accompanied by large fluctuations in the background current (Fig. 3C). The spikes disappeared when the polarity of the bias across the tube was reversed and reappeared when the original bias (positive on the output side) was restored. Quantitative PCR (QPCR) (22) showed that DNA was translocated in devices manifesting these large spikes. Translocation occurred only in tubes with conductances (before DNA addition) of >2 nS (Fig. 1C). Some devices that showed instabilities in the background but no large current spikes (Fig. 3F) gave negative PCR results. We also tested for transloca-

Fig. 3. Ion current signals of DNA translocation. (A) Current (2 M KCl, 1 mM PBS, pH 7) before DNA addition. (B) After DNA addition, current slowly increases. (C) 5 min after addition of 0.1-nM 60-nt DNA, large positive current spikes appear. These spikes are followed by a drop in baseline over a period of a second or so and then by a gradual rise leading to the next spike. (D) Representative data from another tube (also 60-nt DNA), with the distribution of currents shown in (E). The DNA causes large changes in baseline in addition to the spikes. (F) Data from a tube that showed both a current increase on DNA addition and baseline fluctuations but no spikes. No translocation was detected by PCR. The insets in (C) and (F) show the fluorescence signal from double-stranded DNA dye labels as a function of the PCR cycle number for samples collected from these particular runs. Table 2. Results of QPCR tests for translocation in tubes with conductance >2 nS that gave uncontaminated control signals (data from four other devices that showed contamination in the control sample were rejected). Errors in spike count reflect the consequences of different cut-off criteria for selecting spikes. Errors in the molecule count were dominated by uncertainties in the filter recovery efficiency, except for the data marked *, which were calibrated with a second oligomer.
Tube AD1 AD2 AA New1 AA New2 HL-4-1-36 A136 HL-4-1-41 HL-4-1-40 O8 Tube conductance nS (1 M KCl) 9.7 9.5 19.6 2.7 9.6 1.6 4.8 2.7 DNA sample (nt) 60 60 120 120 60 60 60 60 Number of spikes 350 30 64 1500 36 46 T T T T T T 0 0 50 10 10 200 4 5 Number of molecules 8000 T 2000 400 T 200 8500 T 3100 24,400 T 5700 1224 T 774* 1900 T 200* 0 0 Molecules per spike 23 T 10 13 (+17, 13) 88 (+126, 88) 16 T 7 34 T 21* 41 T 10*

66

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

REPORTS
for production of devices with particular pore sizes.
References and Notes
1. M. Rhee, M. A. Burns, Trends Biotechnol. 25, 174 (2007). 2. C. Dekker, Nat. Nanotechnol. 2, 209 (2007). 3. D. Branton et al., Nat. Biotechnol. 26, 1146 (2008). 4. J. J. Kasianowicz, E. Brandin, D. Branton, D. W. Deamer, Proc. Natl. Acad. Sci. U.S.A. 93, 13770 (1996). 5. M. Akeson, D. Branton, J. J. Kasianowicz, E. Brandin, D. W. Deamer, Biophys. J. 77, 3227 (1999). 6. J. Kasianowicz, S. Henrickson, H. Weetall, B. Robertson, Anal. Chem. 73, 2268 (2001). 7. A. Meller, D. Branton, Electrophoresis 23, 2583 (2002). 8. J. Li et al., Nature 412, 166 (2001). 9. A. Storm, J. Chen, X. Ling, H. Zandbergen, C. Dekker, Nat. Mater. 2, 537 (2003). 10. H. Chang et al., Appl. Phys. Lett. 88, 103109 (2006). 11. M. J. Kim, M. Wanunu, D. C. Bell, A. Meller, Adv. Mater. 18, 3149 (2006). 12. P. Chen et al., Nano Lett. 4, 1333 (2004). 13. R. Fan et al., Nano Lett. 5, 1633 (2005). 14. D. Stein, M. Kruithof, C. Dekker, Phys. Rev. Lett. 93, 035901 (2004). 15. X. Liang, S. Y. Chou, Nano Lett. 8, 1472 (2008). 16. B. J. Hinds et al., Science 303, 62 (2004). 17. J. K. Holt et al., Science 312, 1034 (2006). 18. S. Joseph, N. R. Aluru, Nano Lett. 8, 452 (2008). 19. T. Ito, L. Sun, R. M. Crooks, Chem. Commun. 2003, 1482 (2003). 20. I.-C. Yeh, G. Hummer, Proc. Natl. Acad. Sci. U.S.A. 101, 12177 (2004). 21. Y. Xie, Y. Kong, A. K. Soh, H. Gao, J. Chem. Phys. 127, 225101 (2007). 22. Materials and methods are available as supporting material on Science Online. 23. We also made some devices using multiwalled tubes but found that these were much more difficult to open (fig. S3) (22). 24. K.-F. Chow, F. Mavre, R. M. Crooks, J. Am. Chem. Soc. 130, 7544 (2008). 25. C. Warakulwit et al., Nano Lett. 8, 500 (2008). 26. S. Joseph, R. J. Mashl, E. Jakkobsson, N. R. Aluru, Nano Lett. 3, 1399 (2003). 27. R. M. M. Smeets et al., Nano Lett. 6, 89 (2006). 28. X. Jin, S. Joseph, E. Gatimu, P. Bohn, N. R. Aluru, Langmuir 23, 13209 (2007). 29. X. Tu, S. Manohar, A. Jagota, M. Zheng, Nature 460, 250 (2009). 30. X. Guo et al., Science 311, 356 (2006). 31. We acknowledge valuable discussions with G. Zhang, O. Sankey, D. Crooks, C. Yamashiro, and M. Muthukumar. X. Cui provided us with samples of larger-diameter SWCNTs, and T. Liu assisted us with atomic force microscopy measurements. This work was supported by grants from the DNA sequencing technology program of the National Human Genome Research Institute, Arizona Technology Enterprises, and the Biodesign Institute (S.L.), an NSF NIRT Award, Nanoscale Science and Engineering Initiative of the NSF, and NYSTAR (C.N.), the OFES, U.S. DOE, and NSF through the National Institute for Computational Sciences of the University of Tennessee (P.K.). A patent has been filed jointly by Arizona State University and Columbia University naming J. He, S. Lindsay, and C. Nuckolls as inventors in the area of DNA sequencing.

Fig. 4. Characteristics of the translocation signals for 60-nt DNA. (A) Spike rate increases with bias after a threshold that depends on the particular CNT; the two devices here show spike signals above 0.1 (squares) and 0.4 V (circles). (B) Spike amplitude (squares) increases linearly with bias and is about 5% of the background current (circles). (C) Distribution of the charge in each spike for the SWCNT in units of the electronic charge, e. tion in failed control devices (i.e., lacking the CNT and deliberately overetched) that displayed leakage current. A few with very large leakage current showed evidence of DNA in the output well, but none displayed spikes, regardless of the magnitude of the leakage current. Thus, the spikes signal translocation of DNA through the SWCNTs. QPCR also provides a measure of the number of molecules collected. We collected small samples of fluid from the output reservoir by flushing the system through with excess buffer and concentrated the solution using a Microcon YM-10 centrifugal filter so that we could redilute with PCR buffer. The filter losses were found to be highly variable, more so at low DNA concentrations, and account for much of the stated uncertainty in our results. We calibrated the PCR reaction with known amounts of DNA and, for two data points, calibrated filter losses by adding a known amount of a second sequence (with orthogonal primers) and carrying out a PCR analysis of both the target and calibration samples. The final molecule count was corrected for filter losses and dilution during the sample collection. The various errors in these steps tend to underestimate the amount of DNA that translocated, so the final results are probably lower limits. PCR was limited to the first use of a device, and we rejected samples from chips that showed contamination in the control samples collected. We were able to carry out PCR on samples collected from 12 devices that had a conductance >2 nS. Of these, four had DNA contamination in control samples, leaving the eight devices listed in Table 2. Two of these showed no spikes and yielded no PCR signal. The remaining six all appeared to pass more than one molecule per spike. In particular, tubes HL-4-136 and A136, for which the filter recovery was directly measured with a control sample, passed at least 30 to 40 molecules for each spike. It is possible that the tube fills entirely with DNA, the spike signaling the cooperative emptying (or possibly filling) of the tube. The uncertainties in the PCR measurement are too large to reveal any significant difference between the number of molecules per spike for the 60-nt sample (23, 13, 34, and 41) and the 120-nt sample (88 and 16), although the spike frequency was much lower in the two 120-nt runs, and the spike duration significantly longer (figs. S13 and S14). Figure 4A shows data for the spiking rate as a function of bias for two different tubes passing 60-nt DNA. The spike rate increased with applied bias, and the two tubes showed different threshold biases for the onset of spikes (and hence translocation). For the 60-nt DNA, the spike amplitudes are about 5% of the baseline current (Fig. 4B), and their duration is between 3 and 100 ms, independent of applied bias, as long as it is above the threshold for translocation. The product of the spike duration and amplitude yields the charge contained in each spike (Fig. 4C). This is remarkably large, at about 1 pC or 107 electrons in each spike. Fan et al. explained positive charge spikes observed in nanochannels as a consequence of additional mobile ions brought into the channel by DNA molecules (13). Filling the tubes (2 mm long) with 100 (20 nm long) 60-nt DNA oligomers, each carrying 60 excess electronic charges would account for only 1 part in 10,000 of the observed charge in each spike. The spikes must originate with large changes in the polarization outside the tubes, much as observed in junctions between micro- and nanochannels (28). The charge accumulation caused by the asymmetrical current in the SWCNT might be the source of this polarization, but further modeling is required to shed light on this unusual signal. The excess ionic conductance appears to be a characteristic of metallic tubes, and we have proposed a mechanism based on electroosmotic flow resulting from trapped charge. Tubes with high ionic conductance will transport DNA molecules, giving a distinctive and unexpectedly large electrical signal of translocation. This kind of nanopore combines a long channel (in which translocation speed might be slowed) with an integrated electrode that might prove useful in new schemes for sequencing DNA by tunneling (3). The ability to select metallic SWCNTs of a desired diameter (29) may open the way SCIENCE VOL 327

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/64/DC1 Materials and Methods Figs. S1 to S14 Tables S1 to S3 References 10 September 2009; accepted 28 October 2009 10.1126/science.1181799

www.sciencemag.org

1 JANUARY 2010

67

REPORTS

Solid Nanoparticles that Catalyze Biofuel Upgrade Reactions at the Water/Oil Interface
Steven Crossley, Jimmy Faria, Min Shen, Daniel E. Resasco* A recoverable catalyst that simultaneously stabilizes emulsions would be highly advantageous in streamlining processes such as biomass refining, in which the immiscibility and thermal instability of crude products greatly complicates purification procedures. Here, we report a family of solid catalysts that can stabilize water-oil emulsions and catalyze reactions at the liquid/liquid interface. By depositing palladium onto carbon nanotubeinorganic oxide hybrid nanoparticles, we demonstrate biphasic hydrodeoxygenation and condensation catalysis in three substrate classes of interest in biomass refining. Microscopic characterization of the emulsions supports localization of the hybrid particles at the interface. n phase-transfer catalysis, reactions are carried out in a biphasic mixture of two immiscible solvents, often water and a hydrophobic organic liquid; added surfactants such as quaternary ammonium salts enhance interfacial surface area (through emulsification) and facilitate transfers between the phases (1). The technique is especially useful in cases in which a product might otherwise be unstable under the reaction conditions in one phase but can partition into the other phase after rapid formation. More broadly, ongoing partitioning of by-products on the basis of their relative solubility can result in substantial simplifications at the isolation and purification stages, obviating the need for procedures such as distillation that might damage heat-sensitive compounds. Such process improvements could have a major impact in the field of biomass conversion to fuels. For example, bio-oil obtained from pyrolysis of biomass is a complex liquid that is only partially soluble in either water or hydrocarbon solvents (2). Rather than carrying out multiple consecutive purification steps during refining to separate out the hydrophilic by-products incompatible with fuel applications (3), it would be desirable to perform sequential reactions under phase-transfer conditions in a single reactor medium. A serious drawback in such systems, however, is that the surfactants can be difficult to separate from final product mixtures. Solid particles are more easily recoverable and have also been shown in many cases to stabilize aqueous-organic emulsions (47), but these solid-stabilized emulsions have not been widely used in catalytic contexts. Moreover, in cases such as the refining of bio-oils in which the system is biphasic and contains up to 30% water, the most efficient way of catalyzing reactions is to place the solid catalyst at the liquid/liquid interface and to maximize the extent of interface by creating an emulsion. Otherwise,

the catalyst particles will preferentially remain in the heavier phase, such as water. In that case, only the water-soluble molecules will be converted. If further conversion of water-insoluble molecules is wanted, one would need to remove them from the top of the reactor and send them to another reactor with a catalyst operating in the organic phase. Therefore, the concept of solid particles that can simultaneously stabilize an emulsion and catalyze reactions in both phases becomes an attractive proposition. In particular, oxide nanoparticles have previously been used to stabilize oil-in-water emulsions (810) because their hydrophilicity preferentially orients them toward the aqueous phase at the interFig. 1. (A) Optical microscopy image of a waterin-oil emulsion formed by sonicating a 1:1 mixture of decalin and water in the presence of 5 weight percent (wt %) Pd/SWNTSiO2 nanohybrids. (B) The same mixture as in (A) before sonication. It is seen that the nanohybrids preferentially migrate to the interface. (C) TEM image of the 5 wt % Pd/SWNTSiO2 nanohybrids. It is apparent that Pd clusters are preferentially located over the silica surface rather than over the nanotubes.

face. Carbon nanotubes have also been shown to produce emulsions, but of the water-in-oil variety because they are hydrophobic (11). We recently prepared hybrid nanoparticles by fusing carbon nanotubes to silica (12). By tuning their composition, we could modify the hydrophilic-hydrophobic balance and assemble water-in-oil or oil-in-water emulsions systematically and reproducibly (13). The emulsion volume fraction in the three-layer system formed (oil-emulsion-water) and the droplet size were greatly influenced by the oil-to-water ratio in the mixture and the degree of oxidative functionalization of the nanotubes. In this contribution, we have undertaken a more detailed characterization of the structure of these emulsions using optical and electron microscopy [figs. S14 to S16 and supporting online material (SOM) text] (14). One of the main objectives of the present study was to extend the utility of these nanohybrids by incorporating a transition metal, rendering them catalytically active for hydrogenation. The second objective was to add a solid base function to catalyze condensation reactions. As shown in Fig. 1, after the addition of Pd metal particles these particles still straddle the organic/aqueous interface, and the appearance and stability of the emulsions formed are unaffected. We envisioned that their selective metal functionalization would be a powerful strategy for directing reactivity in a specific phase: Depositing a metal such as Pd on the hydrophilic face would catalyze aqueous reactions, whereas deposition on the hydrophobic face would favor chemistry in the organic solvent. In fact, established catalyst

School of Chemical, Biological, and Materials Engineering, University of Oklahoma, Norman, OK 73019, USA. *To whom correspondence should be addressed. E-mail: [email protected]

68

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

REPORTS
preparation methods exist that can be used to selectively deposit metal precursors from solution on either silica or nanotubes, depending on the pH of the solution and the point of zero charge of the substrate (1517). In this work, we have explored two preparations with nanotubes of different type. This difference affects the deposition of Pd. As shown in Fig. 1C, in the first preparation the incipient wetness impregnation method used to deposit Pd on the single-walled carbon nanotube (SWNT) silica nanohybrids results in preferential deposition onto the silica side. This preferential deposition is due to the very low density of defects on the SWNT walls, as demonstrated by means of transmission electron microscopy (TEM) and Raman spectroscopy (fig. S1). Without defects, the SWNT cannot effectively anchor the Pd nanoparticles, which during the heat treatments end up on the silica surface. In contrast, on the second preparation type, which involves MgO as a support instead of SiO2, the resulting nanotubes are more defective than those in the first preparation, as also shown by means of TEM and Raman spectroscopy (fig. S2). Consequently, they are more effective in retaining Pd particles. Therefore, we have been able to demonstrate different degrees of hydrogenation activity in the organic phase by simply using different formulations of the nanohybrids. We present here the results obtained for several reactions of relevance to biomass-refining chemistry. Two of the major goals in the refining of liquids derived from lignocellulosic sources are the elimination of oxygen and the condensation of small molecules. The former is needed to improve the low stability caused by the high reactivity of the oxygenated functional groups in molecules such as the phenolic compounds de-

Fig. 2. (A) Schematic illustration of the reactions taking place at the water/oil interface in the solid-stabilized emulsions. Depending on the reaction temperature, the prevailing reactions are hydrogenation, hydrogenolysis, or decarbonylation, and depending on the relative solubilities, the products remain in the aqueous phase or migrate to the oil phase. (B) Total weight fraction of the various products as a function of temperature after 30-min reaction in a batch reactor, from gas chromatographic analysis of each phase (combined). (C) Partition of the various products as in (B) between the individual aqueous and organic phases. www.sciencemag.org SCIENCE VOL 327 1 JANUARY 2010

69

REPORTS
rived from lignin (18, 19). The latter is particularly important to increase the molecular weight of those light fragments derived from the less refractory parts of the biomass (cellulose and hemicellulose). To illustrate the application of the catalytic nanohybrids in emulsions, we chose the hydrodeoxygenation of a phenolic compound and the hydrogenation and etherification of an aldehyde on Pd as model reactions that occur in the aqueous side of the interface. Therefore, we have used the nanohybrids that preferentially have Pd on the hydrophilic side. In addition, we have been able to replace silica as the hydrophilic component in the nanohybrids with a basic oxide, magnesia, which not only is effective in stabilizing emulsions but also provides catalytic activity for base-catalyzed reactions, such as the Aldol condensation. In this case, as mentioned above, the hydrophobic side contains nanotubes that are more defective, which allow stabilization of Pd. Therefore, hydrogenation on the oil side is readily observed. To explore the catalytic application of Pdcontaining nanohybrids to phenolic hydrodeoxygenation in a water-in-decalin emulsion, we chose vanillin (4-hydroxy-3-methoxybenzaldehyde) as a test substrate because it is a common component of pyrolysis oil derived from the lignin fraction (20, 21). This compound was appealing for study because of its three different types of oxygenated functional groups (aldehyde, ether, and hydroxyl) and its partial solubility in both the organic and aqueous phases. The reaction was carried out in a batch reactor at three different temperatures (100, 200, and 250C) for different reaction periods. In each run, the molar ratio of vanillin to Pd in the reactor was about 100. After each reaction period, the emulsion was broken by filtering out the nanohybrid particles. The filtering was conducted in two steps. In the first one, a common paper filter was used. This coarse filter (8-mm pores) was able to trap a large fraction of nanohybrids because they quickly agglomerate when they contact the filter. In the second step, a polytetrafluoroethylene (0.2-mm pores) filter was used to separate the small fraction of nanohybrids that passed the first filter. The two liquid phases were separated and analyzed individually by means of gas chromatography with flame ionization (FID) and mass spectrometry (MS) detectors. It was then possible to monitor the migration that the different products underwent from the aqueous to the organic phase, as summarized in Fig. 2. Whereas vanillin and vanillin alcohol were highly soluble in the aqueous phase, the products resulting from both hydrogenolysis and decarbonylation were much more soluble in the decalin phase. As a result, the water-insoluble compounds, which are more valuable as fuel components, can be readily incorporated in the product stream. The turnover number measured at the lowest temperature tested (100C) was about 0.2 s1, which is of similar magnitude to that observed by others in the hydrogenation of guaiacol catalyzed by Pd/C in a monophasic system but at higher temperatures (150C) (19). The higher activity observed in the biphasic system could be ascribed to (i) a better state of particle dispersion at the interface as compared with that in the monophasic system (fig. S13) and/or (ii) enhanced hydrogen concentration at the interface (hydrogen has higher solubility in organic phase than in water). As illustrated schematically in Fig. 2A, we observed a range of different products and phasemigration processes that were due to the varying extents of hydrogenation, hydrogenolysis, and decarbonylation reactions catalyzed by Pd as the reaction conditions were modified. As shown in Fig. 2B, the chemoselectivity changes significantly with increasing temperature. At 100C, initially only hydrogenation of the aldehyde to the vanillin alcohol is observed, whereas at longer reaction times the alcohol is further converted into to 2-methoxy-4-methylphenol (p-creosol) via hydrogenolysis. At 200C, hydrogenolysis becomes the dominant path even at short reaction times. At 250C, the dominant reaction is the decarbonylation of the aldehyde group, leading primarily to o-methoxyphenol (guaiacol). The variation of product distribution with reaction time in the batch reactor at 100C is shown in fig. S5. The vanillin alcohol is the primary product, but at longer times it is consumed by hydrogenolysis to form p-creosol, which migrates to the organic phase upon formation, preventing further conversion. In contrast, the alcohol remains in the aqueous phase and continues reacting. This result illustrates the concept of simultaneous reaction and separation of intermediate products. We next explored the reactivity of molecules that were exclusively soluble in either the organic or aqueous phase. Octanal and glutaraldehyde were chosen as the model compounds soluble in the decalin and water phases, respectively. Three reaction runs were compared at the same conditions (14). In one of them, octanal and glutaraldehyde were partitioned in an equimolar mixture of water and decalin. In the other two, octanal and glutaraldehyde were separately dissolved in pure decalin and pure water, respectively. These molecules were readily hydrogenated over Pd to the corresponding alcohols. However, the nanohybrids that were used in this case had the metal preferentially deposited on the hydrophilic side. Therefore, we anticipated to see a greater effect on the conversion of glutaraldehyde. Indeed, although only 58% conversion was obtained in the single aqueous phase after 3 hours at 100C, practically all (98%) of the glutaraldehyde was converted in the emulsion under the same conditions. This enhancement can be ascribed to a combination of selective deposition of Pd on the polar silica as well as the increased exposure of the catalyst at the interface, as compared with the highly aggregated state of the solid catalyst suspended in the one-phase system. Although 5-hydroxypentanal was the expected primary product from the initial hydrogenation of one of the carbonyl groups in glutaraldehyde, this product was not observed. Instead, the cyclic hemiacetal, valerolactol, emerged as a major product. We suspect that the cyclization of the hydroxyaldehyde occurs via attack by the nucleophilic oxygen of the alcohol at the carbonyl carbon, which is analogous to the well-known cyclization of glucose (22). The time evolution of the products (fig. S6) indicates that the reaction sequence is glutaraldehyde d-valerolactol 1,5 pentanediol, with the diol dominating at long reaction times. An additional product, 5-hydroxy-1-pentyl tetrahydropyranyl ether, was also observed, but in low yield (<2%). Whereas glutaraldehyde, octa-

Fig. 3. Possible reaction paths and products arising from the condensation and direct hydrogenation of 5-methylfurfural and acetone over 5 wt % Pd/CNT/MgO. VOL 327 SCIENCE www.sciencemag.org

70

1 JANUARY 2010

REPORTS
nediol, and valerolactol remained in the aqueous phase, the ether migrated to the decalin phase. This is an example that illustrates the potential of the method by which one can easily separate an intermediate product from the reaction mixture even when it is obtained at low yieldbefore it continues reacting. At higher temperature, the ether underwent decarbonylation or hydrogenolysis before leaving the aqueous phase, producing 2-butoxytetrahydro2H-pyran or 2-pentyloxytetrahydro-2H-pyran, respectively. In good agreement with the very high log P values of these two products, they were observed exclusively in the oil phase, once again demonstrating the concept of separation after reaction. This result shows that one could arbitrarily modify selectivity by controlling the relative rates of reaction and migration out of the phase in which the catalyst is located. Analogous to shapeselectivity in zeolites (23) and selectivity affected by mass transfer phenomena (24), we can introduce the concept of a phase-transfer selectivity, by which the product distribution resulting from reactions catalyzed at the liquid/liquid interface is modified by the transport to and through the interface, where the catalyst resides. One could design emulsion systems with varying characteristics (such as droplet size, emulsion type, and relative solubilities) that may affect the residence time of intermediate products in one phase before it migrates to the other phase. This controlled migration may in turn affect the extent of reaction of one specific intermediate. Because very little Pd was present on the hydrophobic side, in the case of the octanal reaction we observed the opposing result: an activity enhancement in the pure organic phase relative to the emulsion. The octanol yield on the single phase was 9.1%, whereas that in the emulsion was only 2.3%. We ascribe this outcome to the larger proportion of Pd deposited on the hydrophilic oxide than on the hydrophobic nanotubes, as shown in Fig. 1C by means of TEM. The direct hydrogenation of octanal in the organic phase yielded 1-octanol as the (expected) dominant product, together with a small quantity of the symmetric dioctyl ether derived from alcohol coupling. Finally, we explored a tandem reaction sequence in which Pd-catalyzed hydrogenation was paired with a preceding Aldol condensation of 5-methylfurfural and acetone. This reaction was catalyzed by MgO, which was incorporated in the nanohybrids instead of silica. When using MgO during the nanotube synthesis, the SWNT product contains about 20% multiwalled carbon nanotubes (MWNTs) as an impurity; however, the use of MgO imparts basicity to the nanohybrids, which is crucial for the Aldol condensation. Therefore, because the MgO-based nanohydrids are still effective in stabilizing emulsions, we have been able to conduct reactions at the liquid/liquid interface by using a bifunctional catalyst that contains both metal and basic sites. Moreover, as mentioned above, this nanohybrid contains nanotubes that are more defective and is able to stabilize Pd particles not only on the hydrophilic side but also on the hydrophobic side. Therefore, hydrogenation can occur on both sides of the emulsion. As represented in Fig. 3, the expected products from the combination of these reactions are 4-(5methylfuran-2-yl)buten-2-one, 4-(5-methylfuran2-yl)buten-2-ol, 5-methylfuran-2-yl methanol, and 2- propanol. To better determine how these products evolve, we conducted this reaction in tandem. That is, we first ran the reaction under N2 at 80C for 3 hours and analyzed the products, which is indicated as reaction 1 in Table 1. In this case, no hydrogenation took place. Self-condensation of acetone was not observed. Ketone-ketone condensation reactions are thermodynamically less favorable and much slower than ketone-aldehyde condensation, as pointed out by West et al. (25). As a result, the major product was 4-(5-methylfuran2-yl)buten-2-one, which in line with its high log P value (~1.5) migrated almost completely to the organic phase. In the second experiment, which is indicated as reaction 2 in Table 1, an additional 1-hour reaction step at 100C under H2 flow was added to the initial 3 hours at 80C under N2. The 4-(5-methylfuran-2-yl)buten-2-one is totally hydrogenated in the second step, indicating that hydrogenation has occurred in the organic phase as well as in the aqueous phase. The shorter oxygenates, in this case propanol, remain in the aqueous phase, thus enhancing the probability of realizing further condensation reactions that would be advantageous in the production of fuel components. Similar reactions have been previously described by Dumesic and coworkers (26, 27), operating with a monophasic system. The advantage of operating in a biphasic system, with the catalyst at the liquid/liquid interface, is the possibility of conducting the sequential reactions in a single reactor instead of two. The carbon chains migrate to the organic phase after growing long enough to become hydrophobic, facilitating their isolation as desirable products, whereas the shorter chains remained in the aqueous phase to undergo further growth. The use of a biphasic system provides the possibility of doing the next reaction (for example, hydrogenationhydrogenolysis exclusively in the oil phase) but in the same reactor. In previous studies conducted in a monphasic aqueous system (26), once the water solubility of the condensed product becomes low enough, it leaves this phase, stopping the conversion. In many cases, these intermediate products

Table 1. Product distribution of the Aldol-condensationhydrogenation reaction of 5-methylfurfural and acetone over 5 wt% Pd/(CNT/MgO). Reaction conditions were 3 hours at 80C in 250 pounds per square inch (psi) of N2 and then 1 hour at 100C in 250 psi of H2. Aq., aqueous phase; Org., organic phase.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

71

REPORTS
still contain oxygen, and the only way to completely deoxygenate them is by using a second reactor operating in the organic phase or a hightemperature vapor phase hydrotreatment. With solid-stabilized emulsions, a continuous process could be designed in which the two homogeneous phases coexist with the emulsion in a layered configuration: oil/emulsion/water. One can achieve full conversion on both sides of the emulsion followed by constant removal of oilsoluble products from the top layer and watersoluble products from the bottom layer while the reaction keeps occurring in the emulsion. Our results highlight the preliminary applications of solid catalysts localized at the interface between two liquid phases. We anticipate that tailoring such emulsion-stabilizing solids with additional catalytic functional groups will facilitate a broad range of reactions.
References and Notes
1. C. M. Starks, J. Am. Chem. Soc. 93, 195 (1971). 2. G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 106, 4044 (2006). 3. G. W. Huber, J. A. Dumesic, Catal. Today 111, 119 (2006). 4. L. L. Dai, R. Sharma, C.-Y. Wu, Langmuir 21, 2641 (2005). 5. G. M. Whitesides, B. Grzybowski, Science 295, 2418 (2002). 6. A. D. Dinsmore et al., Science 298, 1006 (2002). 7. B. P. Binks, S. O. Lumsdon, Langmuir 16, 8622 (2000). 8. B. P. Binks, Curr. Op. in Coll. and Interf. Sci. 7, 21 (2002). 9. B. P. Binks, C. P. Whitby, Langmuir 20, 1130 (2004). 10. B. P. Binks, J. Philip, J. A. Rodrigues, Langmuir 21, 3296 (2005). 11. R. K. Wang et al., J. Am. Chem. Soc. 130, 14721 (2008). 12. D. E. Resasco et al., J. Nanopart. Res. 4, 131 (2002). 13. M. Shen, D. E. Resasco, Langmuir 25, 10843 (2009). 14. Materials and methods are available as supporting material on Science Online. 15. J. P. Brunelle, Pure Appl. Chem. 50, 1211 (1978). 16. J. R. Regalbuto et al., J. Catal. 184, 335 (1999). 17. S. Lambert et al., J. Catal. 261, 23 (2009). 18. D. C. Elliott, Energy Fuels 21, 1792 (2007). 19. C. Zhao, Y. Kou, A. A. Lemonidou, X. Li, J. A. Lercher, Angew. Chem. Int. Ed. 48, 3987 (2009). 20. C. Saizjimenez, J. W. Deleeuw, Org. Geochem. 10, 869 (1986). 21. D. Mohan, C. U. Pittman Jr., P. H. Steele, Energy Fuels 20, 848 (2006). 22. R. Brckner, in Advanced Organic Chemistry: Reaction Mechanisms (Academic Press, New York, 2002), pp. 283285. 23. N. Y. Chen, W. W. Kaeding, F. G. Dwyer, J. Am. Chem. Soc. 101, 6783 (1979). 24. E. Iglesia, Appl. Catal. Gen. 161, 59 (1997). 25. R. M. West, Z. Y. Liu, M. Peter, J. A. Dumesic, Chem. Sus. Chem. 1, 417 (2008). 26. D. A. Simonetti, J. A. Dumesic, Chem. Sus. Chem. 1, 725 (2008). 27. G. W. Huber, J. N. Chheda, C. J. Barrett, J. A. Dumesic, Science 308, 1446 (2005). 28. This research was supported by the Oklahoma Secretary of Energy and the Oklahoma Bioenergy Center. Partial support from the U.S. Department of Energy is gratefully acknowledged.

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/68/DC1 Materials and Methods SOM Text Figs. S1 to S16 References 19 August 2009; accepted 6 November 2009 10.1126/science.1180769

Unveiling the Transient Template in the Self-Assembly of a Molecular Oxide Nanowheel

Haralampos N. Miras,1 Geoffrey J. T. Cooper,1 De-Liang Long,1 Hartmut Bgge,2 Achim Mller,2 Carsten Streb,1 Leroy Cronin1* Self-assembly has proven a powerful means of preparing structurally intricate nanomaterials, but the mechanism is often masked by the common one-pot mixing procedure. We employed a flow system to study the steps underlying assembly of a previously characterized molybdenum oxide wheel 3.6 nanometers in diameter. We observed crystallization of an intermediate structure in which a central {Mo36} cluster appears to template the assembly of the surrounding {Mo150} wheel. The transient nature of the template is demonstrated by its ejection after the wheel is reduced to its final electronic state. The templates role in the self-assembly mechanism is further confirmed by the deliberate addition of the template to the reaction mixture, which greatly accelerates the assembly time of the {Mo150} wheel and increases the yield. he bottom-up self-assembly of large inorganic architectures is a key synthetic route for the preparation of a whole range of systems from cages (13) to metal organic frameworks (4, 5) and the formation of macrocycles (6) from comparatively simple small molecule building blocks. In these systems the building blocks, and the underlying selfassembly processes, are understood to such a high degree that many complex and intricate structures can be designed from first principles, and the resulting architectures can even be

postsynthetically modified (7). However, when nanostructures in the 2- to 10-nm range are targeted (e.g., metallic nanoparticles and quantum Fig. 1. A photograph of the flow-reactor system showing the blue reduction wavefront gradient formed within the vessel during the assembly of compound 1 (the reduced region is blue, and the more oxidized region is pale to clear). The structure of the {Mo150} wheel present in compound 1 is shown in space-filling ball-and-stick mode. Mo ions are green spheres; O ligands are red spheres. VOL 327 SCIENCE

dots), the self-assembly can be critically limited by the high number of degrees of freedom, and it is not trivial to target narrow size distributions. In contrast, in supramolecular chemistry, molecular templates have been successfully employed as external directors in the design of receptors (3, 810) and can facilitate the assembly of molecular nanostructures that are intrinsically monodisperse. Spectacular assembly control has been demonstrated by deliberate targeting and synthesis of templates that by design enable the formation of a predetermined structure, yet it is difficult to design large structures (2). The discovery of a similar templating strategy for the reliable fabrication of 2- to 10-nm molecular nanoparticles would revolutionize the synthesis and applications of molecular materials in the same way that templated synthesis has revolutionized the field of organic macrocyclic synthesis over the past 40 years. In recent years, Mller and co-workers reported the solution-phase assembly of a family

WestCHEM, Department of Chemistry, The University of Glasgow, Glasgow, G12 8QQ, UK. 2Fakultt fr Chemie, Universitt Bielefeld, Postfach 100131, 33501 Bielefeld, Germany. *To whom correspondence should be addressed. E-mail: [email protected]

72

1 JANUARY 2010

www.sciencemag.org

REPORTS
of nanoscale metal-oxide rings composed of 140 to 154 molybdenum centers by reduction of an acidic (pH = 1 to 3) molybdate solution, which have a range of interesting physical and chemical properties arising from their molecular nature, nanoscale size, and electronic distribution (1115). These rings crystallized either as ordered chain/layer assemblies or discrete molecules. Diffraction analysis revealed that each individual ring appeared to have assembled from several classes of discrete [MoxOy] building blocks, most commonly {Mo2}, {(Mo)Mo5}, and {Mo8} bonded together through oxo-bridges to form {Mo154-x} (x = 0 to 14) rings 3.6 nm in diameter. We hypothesized that some sort of internal templating must have played a role in such an intricate assembly process, but the deceptively simple reaction conditions have so far effectively concealed the enormous complexity involved in this self-assembly process. We probed the self-assembly of the molybdenum blue (MB) nanoparticles using a dynamic synthetic procedure in a flow system that enabled real-time adjustment of the three input variables (pH, concentration of molybdate, and reducing agent) controlling the synthesis of the molecular nanosized-wheels (Fig. 1). By using a flow system, rather than combining all reagents at once in a single flask (one pot), we were able to maintain an off-equilibrium reaction system in which we carefully controlled the degree of reduction of the polyoxomolybdate clusters. To achieve this reaction state, screening of the synthetic parametersfor example, the concentration of molybdate, reducing agent, pH, and ionic strengthwas required to determine the correct flow rate. If the flow rate was too low, then the reduced molybdate would be produced in such low concentration that crystallization would not occur and the reduced molybdate within the system would be reoxidized by oxygen over a period of days (i.e., the solution would become colorless). If the flow rate was too high, then the whole system would become over-reduced and no gradient between the reduced and oxidized regions would be set up on a time scale that would allow crystallization (i.e., the solution would become uniformly dark blue). However, after optimizing the flow system, we were also able to isolate and trap, by crystallization, a key intermediate in the assembly of wheel-type MB nanocluster whereby single crystals were formed in the flow reactor and isolated by halting the flow and filtering the product after a given run time of 2 to 3 days. Specifically, we characterized a hollow {Mo150} wheel that hosts a {Mo36} cluster that is bound to the central cavity of the ring species by chargebalancing sodium cations, and in the solid state the wheel itself is weakly covalently linked through 5 oxo-bridges to chains (16). This hostguest complex shows features indicative of an intermediate electronic and structural state, and we postulate that the {Mo36} cluster acts as the key template in the formation of the MB ring. We isolated the host-guest complex as the crystalline compound 1, Na22[MoVI36O112(H2O)16] [MoVI130MoV20O442(OH)10(H2O)61]180H2O Na221a180H2O in a gram yield of 4.5 g,17.4% by the reduction of an aqueous acidic solution of Na2MoO42H2O with Na2S2O4 under continuous addition of HNO3, and we were able to determine the formula of the host-guest compound unambiguously using many lines of investigation (16). The key to this discovery was the use of nitric acid in the flow system in a dual role as a proton source and an oxidant leading to incomplete reduction of the wheel. The archetypal wheel {Mo154} has 14 two-electron reduced compartments (i.e., a total of 28 4d electrons), but in the present case, only 10 of the 14 compartments are two-electron reduced, rendering the overall intermediate wheel 20-electron reduced (as confirmed by a combination of structural studies, chemical analysis, redox titration, and solution UV-VIS spectroscopy). Compound 1 crystallizes in the space group C2/m and has a large unit cell with a volume of 41,734 3. The corresponding formula was determined by elemental analyses, single-crystal x-ray structure analysis, bond valence sum calculations, redox titrations, and thermogravimetry (16). Structural analysis revealed that the {Mo150} wheel features an ellipsoidal central cavity of ~2.6 1.8 nm in which a {Mo36} unit resides. The principal axis of the {Mo36} template is tilted by 38.5 with respect to the main wheel plane and is linked to the {Mo150} wheel through sodium cation bridges that reduce the electrostatic repulsion between the two negatively charged constituents (Fig. 2), and the binding of the guest is also stabilized by hydrogenbonded interactions to the host. The {Mo150} wheel in 1a consists of 14 {Mo8}-type pentagonal building units (blue, Fig. 2) which are linked along the inner rim of the wheel by 12 {Mo2} linker units (red, Fig. 2), as opposed to the 14 {Mo2} units present in the archetypal wheel (4, 15), and connected by 14 {Mo1} groups along the central equatorial region of the structure (yellow, Fig. 2), so that the wheeltype cluster in 1a can be formulated as {Mo150} = ({Mo1}14{Mo2}12{Mo8}14); the decrease in the number of the {Mo2} building blocks in MB

Fig. 2. A polyhedral representation of the nanoscale {Mo36}{Mo150} wheel as part of the chain 1a [for linking details, see fig. S3 and the chains with pure wheels in (19)]. Sodium cations have been omitted for clarity, although there are ~22 positions on the inner side of the ring where sodium resides, of which 12 directly bridge the template to the outer ring. The host and the guest are also connected by a series of hydrogen bonds. The polyhedral building blocks are colored as follows: {Mo1}, yellow; {Mo2}, red; {Mo8}, blue with a light blue pentagonal central group.

Fig. 3. Representation of the expulsion process. The addition of eight additional electrons increases the repulsion between the ring and the template, leading to the expulsion of the {Mo36} unit, that is, a separation into a pure {Mo150} phase and a pure {Mo36} phase (both isolated as crystalline solids). Color scheme as in Fig. 2. SCIENCE VOL 327 1 JANUARY 2010

www.sciencemag.org

73

REPORTS
neously, versus flow conditions in which {Mo36} was added to a reduced molybdate solution. In the latter case, crystallization started after only 6 to 8 hours, yielding gram quantities of the wheel-based MB nanoparticles within 1 day. The static system required between 3 and 5 days (after crystallization commenced) to produce the same amount of material as isolated from the flow system over the period of 1 day. Our results illustrate how a bottom-up assembly process can be used to rapidly obtain gram quantities of a nanomaterial with welldefined size, shape, and composition. Furthermore, the use of a flow reactor proved to be a powerful tool in unveiling the mechanism of assembly of the {Mo154-x} nanowheel family, and we envision the technique emerging as a versatile means of generating off-equilibrium reaction conditions for mechanistic studies of self-assembly processes.

Fig. 4. Conceptual representation of the MB assembly showing the building block synthons (which are assigned on the basis of structural considerations) that form the template complex. The {Mo36}, {Mo150}, and {Mo36Mo150} complexes have each been isolated separately. Continuous flow-reaction conditions, along with a finely tuned reducing environment, are required to trap the template complex. Color scheme as in Fig. 2. wheels has been previously documented as the formation of defect sites (1719). The {Mo150} wheel in 1a can be geometrically related to the ring-shaped{Mo154} [Mo154O462H14(H2O)70]14 cluster archetype (11,14). According to the classification devised to describe these MB ring type architectures (14, 15, 17), the formula of the {Mo150} wheel in 1a can be approximately expressed in terms of the building blocks [{Mo2}12{Mo8}14{Mo1}14]14, and the overall formula is [MoVI130MoV20O442(OH)10(H2O)61]14 (20). The formula was deduced with the help of bond valence sum analysis on the structural data (distinguishing between the reduced and nonreduced Mo centers and between protonated and nonprotonated O atoms), and this was confirmed chemically using redox titrations and chemical analysis [see (16) for a detailed analysis of the wheel structure]. The transient nature of the wheel-template complex is supported by comparative reactivty studies and is reflected in its geometric and electronic structure. In contrast to the archetypal, highly symmetrical D7d {Mo154} MB wheel, the {Mo150} wheel in 1a is elliptical with maximum outer and inner ring diameters of ~3.6 and 2.6 nm, and minimum outer and inner ring diameters of ~3.5 and 1.8 nm (Fig. 3). This ellipsoidal structure appears to be a result of the central {Mo36} template: Comparison of the ratio between the maximum and minimum cluster dimensions for both the wheel and the template (Rl/s) shows a close match between {Mo36} (Rl/s = 1.42) and the inner ring of {Mo150} (Rl/s = 1.44). Another striking feature of the {Mo36}{Mo150} template-wheel assembly is the nonuniform delocalization of 20 4d electrons over the molybdenum centers. Bond valence sum calculations (16) for the Mo centers of the {Mo150} wheel show that the Mo centers close to the two defect sites at the most compressed sections of the wheel are fully oxidized (+6), whereas those in the least compressed region are reduced to +5. This electronic anisotropy stands in contrast to the 28-electron reduced {Mo154-X} rings, where the electrons are delocalized more symmetrically over the cluster surface (14, 15, 18).The intermediate 20-electron reduced MB wheel clearly favors the inclusion of the anionic {Mo36} template more than the 28-electron reduced {Mo154-X} systems. This observation is confirmed by its further reduction in solution upon replacing nitric acid with HCl in the flow reactor, which results in the expulsion of the template and the crystallization of two separate crystalline phases, including the empty, fully symmetric, and 28-fold reduced {Mo150} wheel and the {Mo36} template (Fig. 3). On the basis of these data, we postulate that the overall mechanism underpinning the formation of the {Mo154-x} family involves the {Mo36} cluster as a structure-directing template. In keeping with this hypothesis, the {Mo36} cluster is well known to form spontaneously in acidified molybdate solutions in the absence of reducing agent. As a result, we can formulate the mechanism (Fig. 4). To test this hypothesis, we compared the time necessary to synthesize the wheel nanoparticles under static conditions, where the molybdate, reducing agent, and acid were added simultaVOL 327 SCIENCE

References and Notes

1. S. Sato et al., Science 313, 1273 (2006). 2. P. Mal, B. Breiner, K. Rissanen, J. R. Nitschke, Science 324, 1697 (2009). 3. P. N. W. Baxter, in Comprehensive Supramolecular Chemistry, Vol. 9, J. L. Atwood, J. E. D. Davies, D. D. MacNicol, F. Vgtle, Eds. (Pergamon/Elsevier, New York, 1996), pp. 165211. 4. K. Chae et al., Nature 427, 523 (2004). 5. R. Matsuda et al., Nature 436, 238 (2005). 6. B. J. Holliday, C. A. Mirkin, Angew. Chem. Int. Ed. 40, 2022 (2001). 7. K. L. Mulfort, O. K. Farha, C. Stern, A. A. Sarjeaut, J. T. Hupp, J. Am. Chem. Soc. 131, 3866 (2009). 8. J. D. Badjic, V. Balzani, A. Credi, S. Silvi, J. F. Stoddart, Science 303, 1845 (2004). 9. D. J. Cram, Nature 356, 29 (1992). 10. J.-M. Lehn, Science 227, 849 (1985). 11. A. Mller et al., Angew. Chem. Int. Ed. Engl. 34, 2122 (1995). 12. A. Mller, S. Roy, Coord. Chem. Rev. 245, 153 (2003). 13. D.-L. Long, E. Burkholder, L. Cronin, Chem. Soc. Rev. 36, 105 (2007). 14. A. Mller, C. Serain, Acc. Chem. Res. 33, 2 (2000). 15. A. Mller et al., Chem. Eur. J. 5, 1496 (1999). 16. Materials and methods are available as supporting material on Science Online. 17. A. Mller et al., Z. Anorg. Allg. Chem. 625, 1187 (1999). 18. S. Shishido, T. Ozeki, J. Am. Chem. Soc. 130, 10588 (2008). 19. A. Mller et al., Angew. Chem. Int. Ed. Engl. 36, 484 (1997). 20. Crystal structure parameters as a CIF file are available from [email protected] (CSD reference 380343). We thank the UK Engineering and Physical Sciences Research Council (EPSRC), The University of Glasgow, and WestCHEM for funding.

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/72/DC1 Materials and Methods Figs. S1 to S5 References Movie S1 9 September 2009; accepted 2 November 2009 10.1126/science.1181735

74

1 JANUARY 2010

www.sciencemag.org

REPORTS

Synchronous Deglacial Overturning and Water Mass Source Changes

Natalie L. Roberts,1* Alexander M. Piotrowski,1 Jerry F. McManus,2 Lloyd D. Keigwin3 Understanding changes in ocean circulation during the last deglaciation is crucial to unraveling the dynamics of glacial-interglacial and millennial climate shifts. We used neodymium isotope measurements on postdepositional iron-manganese oxide coatings precipitated on planktonic foraminifera to reconstruct changes in the bottom water source of the deep western North Atlantic at the Bermuda Rise. Comparison of our deep water source record with overturning strength proxies shows that both the deep water mass source and the overturning rate shifted rapidly and synchronously during the last deglacial transition. In contrast, any freshwater perturbation caused by Heinrich event 1 could have only affected shallow overturning. These findings show how changes in upper-ocean overturning associated with millennial-scale events differ from those associated with whole-ocean deglacial climate events.

C
1

hanges in ocean overturning affect regional climate through the redistribution of heat energy to the high latitudes (1). Reconstructions of past ocean circulation changes suggest a strong link between the strength of Atlantic Meridional Overturning Circulation (AMOC) and circumNorth Atlantic temper-

Godwin Laboratory for Palaeoclimate Research, Department of Earth Sciences, University of Cambridge, Cambridge CB2 3EQ, UK. 2Department of Earth and Environmental Science, LamontDoherty Earth Observatory of Columbia University, Palisades, NY 10964, USA. 3Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA. *To whom correspondence should be addressed. E-mail: [email protected]

atures, as evidenced by the close match between rapid shifts in nutrient proxy records (2) and Greenland ice core paleotemperature records (3). Models of thermohaline circulation have suggested rapid changes in overturning strength and water mass stratification during meltwater-forced events, but also suggest that rate of overturning can be decoupled from stratification changes. However, changes in these primary physical variables cannot be unambiguously reconstructed through time by means of nutrient proxies of ocean circulation, because their sensitivity to source, circulation, and biological change complicates interpretation. This study directly compares water mass source changes, reconstructed

using Nd isotopes, with 231Pa/230Th0 ratios recording overturning rates at the same site. Nd isotopes can be used as a water mass source proxy because (i) the 143Nd/144Nd ratio (eNd) is a tracer of deep water masses originating in different ocean basins (4, 5); (ii) the 500- to 1000-year residence time of Nd in the ocean (6) is shorter than the time required for whole-ocean mixing; and (iii) Nd isotopes are not fractionated by biological or low-temperature processes (7). The eNd value for North Atlantic Deep Water (NADW) is ~ 14, and for Antarctic Bottom Water (AABW) it is ~ 7 to 9 (4, 5, 7), at modern deep water formation sites. Nd isotopes can be used to track these water masses over long path lengths (7). In marginal settings with high suspended particle concentrations, the water mass eNd appears to be altered by exchange with sedimentary Nd (8). Our site on the northeast Bermuda Rise (Fig. 1) is in the open ocean, oceanographically distant from major continental boundaries, and not located downstream of Bermuda. Although sediment deposition rates are high, the particles are primarily transported by recirculating gyres and any leachable material is predominantly authigenic (9). Application of eNd together with other oceanographic proxies provides accurate reconstructions of ocean circulation and changes in climate. Recent millennial-scale Nd isotope studies have used Fe-Mn oxides leached from bulk detrital sediment (10). However, horizontal advection of fine sediments (11) can transport distal Nd

Fig. 1. Core locations (white stars) and seawater profile locations (black triangles) are indicated on the map. Seawater eNd values (5) (white lines) are plotted on the phosphate concentration cross section along the transect A-B, where the core locations (white stars) are not plotted in relation to the eNd axes. Both profiles show similar water mass structure at the Bermuda Rise and Blake Ridge.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

75

REPORTS
isotopic signatures, complicating reconstruction of deep water composition. This is particularly problematic for sediment drift deposits. The Bermuda Rise has deposition rates reaching 100 cm per 1000 years, with >63 mmsized fraction as low as 0.25%, allowing submillennial resolution, but also indicating sediment transport (11, 12). Alternative phases such as fish teeth (13) or benthic foraminiferal calcite (14) are typically in low abundance, precluding high-resolution records. In this study we developed an earlier method established by Palmer and Elderfield (15) using Fe-Mn oxide coatings on planktonic foraminifera, which are underpinned by fish teeth measurements where possible, from core OCE326-GGC6 (GGC6; 3341.443N, 5734.559W, 4541 m) (9). Previous studies have shown that Nd is incorporated into foraminiferal calcite at very low concentrations (16, 17); for example, off the coast of Somalia, plankton tow Nd concentrations average 0.11 ppm (17), relative to core top measurements in this study of 2.63 ppm. Thus, high Nd concentrations exhibited by planktonic foraminifera from sediment cores likely result from metal oxide coatings on the shell (17). The precipitation of Fe-Mn oxide coatings onto sediment particles has been demonstrated to take place at the sediment-water interface (10, 15). Dissolved Nd concentrations increase with depth in the water column (18), and because sediment particles are in contact with Nd-rich bottom waters far longer than with other water masses, they retain a bottom water signature. All reductively cleaned and uncleaned foraminifera [calcite plus coating, with all clay and silicates removed (9)] at three North Atlantic core tops are within error of each other and the bottom water (NADW) composition (Fig. 2). The reductively cleaned foraminifera fall along mixing curves predicted by plankton tow end members, in the direction of surface water concentrations and eNd values. We calculate that in order to obtain surface water values, between 90 and >98% of all coatings must be removed; any remaining coatings are likely due to inefficient cleaning or readsorption of Nd onto clean calcite during cleaning (17). Because planktonic foraminifera Fe-Mn oxide coatings accurately archive bottom water eNd at the Bermuda Rise, this may be a promising approach in regions of high sediment accumulation (9). Downcore Nd isotope measurements were made on bulk sediment leachate (10), reductively cleaned and uncleaned planktonic foraminifera, and reductively cleaned fish debris from Bermuda Rise core GGC6 (Fig. 3). Although core top bulk sediment leachate eNd values are within error of modern-day NADW (4, 5) (Fig. 2), downcore measurements are always more positive than the foraminiferal and fish debris eNd and do not record the same changes as benthic d13C (Cibicidoides wuellerstorfi, epibenthic) from nearby Bermuda Rise core EN120-GGC1 (3340N; 5737W, 4450 m) (19). This suggests variable addition of a radiogenic Nd contribution to the Fig. 2. Core top measurements of reductively cleaned (open symbols) and unclean planktonic foraminifera (solid symbols) from cores ODP172 1061 (circles), OCE326-GGC6 (triangles), and ODP172 1063 (squares). Mixing curves are plotted on the basis of surface and deep water eNd measurements (5) and average plankton tow Nd concentrations (0.11064 ppm; center dashed line) and minimum and maximum envelopes (0.01224 ppm and 0.34992 ppm, respectively) (17), and the maximum measured Fe-Mn coating Nd concentration. Surface water (light gray) and bottom water (dark gray) eNd values are indicated by bands including 2s errors (5). Numbers denote the fraction of Fe-Mn oxide coatings contributing to the eNd signal. Fig. 3. GGC6 sediment leachates (purple triangles), unclean foraminifera (black triangles), and reductively cleaned fish debris (red triangles), plotted with GGC5 231Pa/230Th0 ratios (blue line) (22), GGC1 benthic d13C (green circles) [(19); for age model see (9)], and 14C paired benthic and planktonic apparent ventilation ages (brown squares) (25, 26). The Holocene, Younger Dryas (YD), Blling-Allerd (BA), Heinrich event 1 (H1), and 14,600 years ago (gray dashed line) are indicated; ka, thousands of years ago. Modern NADW eNd ~ 14 and AABW eNd ~ 7 to 9 (4, 5, 7).

bulk sediment leachate record. One plausible source is volcanic ash transported southward by bottom currents (11), providing a source of more positive Nd isotopic composition mobilized during leaching. Because foraminifera are not transported horizontally by currents (11), the observation that the uncleaned foraminifera and fish debris have no significant deviation from a 1:1 relationship throughout the record (fig. S2) confirms that we are measuring Fe-Mn oxide coatings precipitated on the planktonic foraminifera shells in bottom waters at the Bermuda Rise, as does the finding that core top samples match local deep water. The overall trend of the uncleaned foraminifera Nd isotopes (eNd-UF), from more radiogenic southern source deep water VOL 327 SCIENCE

(SSW) values in the glacial to less radiogenic northern source values in the Holocene, is consistent with benthic d13C at this site and elsewhere (20). Changes in the strength of AMOC are believed to cause variations in the 231Pa/230Th0 ratio (21, 22) because faster overturning rates will result in a higher flux of Pa to the Southern Ocean relative to more particle-reactive Th, lowering the sedimentary 231Pa/230Th0 ratio in the North Atlantic (23). Measurements of planktonic and benthic 12C/14C (14CB-P) provide another means of estimating AMOC rate, because the age difference between the two samples indicates the time since ventilation. However, these proxies are not without ambiguities. Sedimentary 231Pa/230Th0

76

1 JANUARY 2010

www.sciencemag.org

REPORTS
However, the early Holocene eNd value of 16 recorded by foraminifera and fish teeth indicates that NSW was 2 e units more negative than modern NADW. This suggests an increased proportion of the Labrador Sea component of NADW, which is thought to have begun advecting at ~8000 years ago (33). Fe-Mn oxide coatings that are precipitated onto planktonic foraminifera reliably record bottom water eNd at the Bermuda Rise today, and provide an accurate archive of past water mass structure changes in areas of sediment redistribution. If meltwater at Heinrich event 1 did affect ocean overturning, it was most likely in the upper part of the water column, enhancing the effects of vertical mixing and resulting in unchanging chemical signatures in the deep. This calls for modeling studies that distinguish between intermediate and deep overturning changes and do not invoke total shutdown of AMOC. After 14,600 years ago we observe coupled North Atlantic water mass source and overturning strength shifts through deglacial climate events linking the relative proportion of NSW and SSW with rates of overturning. There is invariably a strong oceanclimate link over the last deglacial, but the sensitivity of this link must be tested during more stable times (for example, the Holocene) in order to determine future climate change implications.
References and Notes
1. W. S. Broecker, G. H. Denton, Geochim. Cosmochim. Acta 53, 2465 (1989). 2. E. A. Boyle, L. Keigwin, Nature 330, 35 (1987). 3. W. Dansgaard et al., Nature 364, 218 (1993). 4. D. J. Piepgras, G. J. Wasserburg, Science 217, 207 (1982). 5. D. J. Piepgras, G. J. Wasserburg, Geochim. Cosmochim. Acta 51, 1257 (1987). 6. K. Tachikawa, V. Athias, C. Jeandel, J. Geophys. Res. 108, 3254 (2003). 7. S. L. Goldstein, S. Hemming, in Treatise on Geochemistry, H. Elderfield, Ed. (Elsevier, Oxford, 2003), vol. 6, pp. 453489. 8. F. Lacan, C. Jeandel, Earth Planet. Sci. Lett. 232, 245 (2005). 9. See supporting material on Science Online. 10. A. M. Piotrowski, S. L. Goldstein, S. R. Hemming, R. G. Fairbanks, Science 307, 1933 (2005). 11. I. N. McCave, Science 298, 1186 (2002). 12. L. D. Keigwin, G. A. Jones, Deep Sea Res. 36, 845 (1989). 13. E. E. Martin, B. A. Haley, Geochim. Cosmochim. Acta 64, 835 (2000). 14. V. Klevenz, D. Vance, D. N. Schmidt, K. Mezger, Earth Planet. Sci. Lett. 265, 571 (2008). 15. M. R. Palmer, H. Elderfield, Earth Planet. Sci. Lett. 73, 299 (1985). 16. M. R. Palmer, Earth Planet. Sci. Lett. 73, 285 (1985). 17. C. Pomis, G. R. Davies, S. M.-H. Conan, Earth Planet. Sci. Lett. 203, 1031 (2002). 18. H. Elderfield, M. J. Greaves, Nature 296, 214 (1982). 19. L. D. Keigwin, G. A. Jones, S. J. Lehman, J. Geophys. Res. 96, 16811 (1991). 20. W. B. Curry, D. W. Oppo, Paleoceanography 20, PA1017 (2005). 21. J. M. Gherardi et al., Paleoceanography 24, PA2204 (2009). 22. J. F. McManus, R. Francois, J.-M. Gherardi, L. D. Keigwin, S. Brown-Leger, Nature 428, 834 (2004). 23. E.-F. Yu, R. Francois, M. P. Bacon, Nature 379, 689 (1996). 24. Z. Chase, R. F. Anderson, M. Q. Fleisher, P. W. Kubik, Earth Planet. Sci. Lett. 204, 215 (2002). 25. L. D. Keigwin, E. A. Boyle, Paleoceanography 23, PA1101 (2008).

Fig. 4. Schematic illustration of ocean circulation over the Bermuda Rise (white circle) at (A) the LGM, and (B) Heinrich event 1. eNd is represented by grayscale from more negative (light gray) to less negative (dark gray), with the relative overturning strength of NSW (black arrows) and the relative effect of vertical mixing on water mass eNd (white arrows). records may also be influenced by variable scavenging due to variations in particle flux and sediment composition, particularly opal (24). Deep water 14C can also be influenced by the mixing of potentially varying and poorly constrained end members. The eNd-UF and 231Pa/230Th0 shifts are sharp, coincident, and correlated across the deglacial climate transitions (the Blling-Allerd Younger DryasHolocene variations). eNd-UF trends also match a composite 14CB-P apparent ventilation record produced from benthic and planktonic foraminifera and bivalves from deep sites throughout the western North Atlantic, confirming that these are basinwide changes (25, 26). The eNd-UF record (Fig. 3) allows us to make inferences about how AMOC source changes relate to ventilation and overturning rate. The Last Glacial Maximum (LGM) eNd-UF data record a bottom water composition of ~ 10.6, indicating the influence of SSW in the deep glacial Atlantic, an interpretation supported by d13C data (2, 20) and leachate Nd isotope data from the South Atlantic Cape Basin (10). Both Nd isotopes and 14CB-P remained constant from 20,000 to 14,600 years ago, indicating unchanged water mass source and overturning strength in the deep between the LGM and Heinrich event 1. Conversely, 231Pa/230Th0 shifted to higher values by 17,500 years ago, which was interpreted as a strong reduction in AMOC by McManus et al. (22). The fact that eNd-UF did not reach SSW endmember values (27), combined with a constant gradient of 3.5 e units between the Bermuda Rise and the Cape Basin (10), suggests that some proportion of northern source water (NSW) was continuously advected southward and is evidence against a total shutdown of AMOC. It also suggests that 231Pa/230Th0 ratios that approach the production ratio must at least in part be due to opal fluxes increasing the ratio (28), but does not fully elucidate the effect of opal flux. However, with evidence of moderate opal fluxes at the LGM (28), any correction on 231Pa/230Th0 necessitates LGM values that indicate overturning similar to or stronger than the Holocene or Blling-Allerd. Because the deep 14CB-P and eNd-UF records see a relatively small contribution from NSW during the LGM, 231Pa/230Th0 is likely integrating the entire water column and primarily recording fast, shallow overturning of glacial intermediate NSW (22). This is concordant with increasing evidence from other proxies supporting a fast, shallow overturning cell at the LGM (20, 23, 26, 29, 30). Our results support the necessity for relatively fast overturning at the LGM in order to maintain a steep vertical chemical gradient in the northwest Atlantic (20) under conditions of strong vertical mixing (31) (Fig. 4A). If freshwater perturbation induced by Heinrich event 1 slowed overturning, it affected only the upper ocean (30), resulting in constant deep 14CB-P ventilation rates and an increase in the 231Pa/230Th0 ratio integrated through the water column. With slower overturning, the strong vertical mixing would have weakened the chemical gradient, allowing more NSW to be transported to depth. This would have offset any effect of decreased NSW production and shoaling of AABW, and would have resulted in no significant change in the Nd isotopic signature at this site (Fig. 4B). Because there were no changes in the boundary conditions at this time, Heinrich event 1 may be likened to modeling results indicating relatively gradual changes in overturning rates during glacial freshwater perturbation (32). In contrast, all three deep water proxies, as well as the integrated 231Pa/230Th0 ratio shift during the deglacial events starting at 14,600 years ago, invoke whole-ocean change in both overturning rate and water mass source, similar to jumps in state via hysteresis behavior (32). This is evidence for two different ocean circulation responses to climate forcing, which suggests that glacial millennial-scale changes in shallow overturning are inherently different from Milankovitchforced glacial-interglacial changes, because the latter are accompanied by shifts in boundary conditions and whole water column reorganization. The large deglacial shifts in benthic d13C and the constant eNd gradient between the North and South Atlantic from 20,000 to 14,600 years ago suggest no change in the NSW end member. SCIENCE VOL 327

www.sciencemag.org

1 JANUARY 2010

77

REPORTS
26. L. F. Robinson et al., Science 310, 1469 (2005). 27. L. F. Robinson, T. van de Flierdt, Geology 37, 195 (2009). 28. I. M. Gil, L. D. Keigwin, F. G. Abrantes, Paleoceanography, 24, PA4101 (2009). 29. H. K. Evans, I. R. Hall, Geochem. Geophys. Geosyst. 9, Q03023 (2008). 30. S. K. Praetorius, J. F. McManus, D. W. Oppo, W. B. Curry, Nat. Geosci. 1, 449 (2008). 31. C. Wunsch, Quat. Sci. Rev. 22, 371 (2003). 32. A. Ganopolski, S. Ramstorf, Nature 409, 153 (2001). 33. C. Hillaire-Marcel, A. de Vernal, G. Bilodeau, A. J. Weaver, Nature 410, 1073 (2001). 34. Supported by UK NERC grant NE/D002206/1, RG43765/ LBZG021, Assessing the role of ocean circulation in rapid climate change through the novel integration of high-resolution proxy records, and by NSF and the Comer Science and Education Fund (J.M.). We thank A. Galy, D. Hodell, and L. Skinner for discussion; E. Martin for advice on fish teeth cleaning; and J. Clegg and A. Scrivner for help in the lab.

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/75/DC1 SOM Text Figs. S1 and S2 Tables S1 to S6 References 22 June 2009; accepted 26 October 2009 10.1126/science.1178068

Human Genome Sequencing Using Unchained Base Reads on Self-Assembling DNA Nanoarrays
Radoje Drmanac,1* Andrew B. Sparks,1 Matthew J. Callow,1 Aaron L. Halpern,1 Norman L. Burns,1 Bahram G. Kermani,1 Paolo Carnevali,1 Igor Nazarenko,1 Geoffrey B. Nilsen,1 George Yeung,1 Fredrik Dahl,1 Andres Fernandez,1 Bryan Staker,1 Krishna P. Pant,1 Jonathan Baccash,1 Adam P. Borcherding,1 Anushka Brownley,1 Ryan Cedeno,1 Linsu Chen,1 Dan Chernikoff,1 Alex Cheung,1 Razvan Chirita, 1 Benjamin Curson,1 Jessica C. Ebert,1 Coleen R. Hacker,1 Robert Hartlage,1 Brian Hauser,1 Steve Huang,1 Yuan Jiang,1 Vitali Karpinchyk,1 Mark Koenig,1 Calvin Kong,1 Tom Landers,1 Catherine Le,1 Jia Liu,1 Celeste E. McBride,1 Matt Morenzoni,1 Robert E. Morey,1 Karl Mutch,1 Helena Perazich,1 Kimberly Perry,1 Brock A. Peters,1 Joe Peterson,1 Charit L. Pethiyagoda,1 Kaliprasad Pothuraju,1 Claudia Richter,1 Abraham M. Rosenbaum,2 Shaunak Roy,1 Jay Shafto,1 Uladzislau Sharanhovich,1 Karen W. Shannon,1 Conrad G. Sheppy,1 Michel Sun,1 Joseph V. Thakuria,2 Anne Tran,1 Dylan Vu,1 Alexander Wait Zaranek,2 Xiaodi Wu,3 Snezana Drmanac,1 Arnold R. Oliphant,1 William C. Banyai,1 Bruce Martin,1 Dennis G. Ballinger,1* George M. Church,2 Clifford A. Reid1 Genome sequencing of large numbers of individuals promises to advance the understanding, treatment, and prevention of human diseases, among other applications. We describe a genome sequencing platform that achieves efficient imaging and low reagent consumption with combinatorial probe anchor ligation chemistry to independently assay each base from patterned nanoarrays of self-assembling DNA nanoballs. We sequenced three human genomes with this platform, generating an average of 45- to 87-fold coverage per genome and identifying 3.2 to 4.5 million sequence variants per genome. Validation of one genome data set demonstrates a sequence accuracy of about 1 false variant per 100 kilobases. The high accuracy, affordable cost of $4400 for sequencing consumables, and scalability of this platform enable complete human genome sequencing for the detection of rare variants in large-scale genetic studies.

G
1

enotyping technologies have enabled the routine assessment of common genetic variants at up to a million sites across the genome in thousands of individuals (1) and have increased our understanding of human genetic diversity and its biological and medical impact. Whole-genome sequencing costs have

Complete Genomics, Inc., 2071 Stierlin Court, Mountain View, CA 94043, USA. 2Department of Genetics, Harvard Medical School, Cambridge, MA 02115, USA. 3School of Medicine, Washington University, St. Louis, St. Louis, MO 63110, USA. *To whom correspondence should be addressed. E-mail: [email protected] (R.D.); dballinger@ completegenomics.com (D.G.B.) These authors contributed equally to this work. Present address: Ion Torrent Systems, San Francisco, CA 94158, USA. Present address: San Diego State University, San Diego, CA 92115, USA. ||Present address: Life Technologies, Carlsbad, CA 92008, USA.

dropped from the >$100 million cost of the first human genomes (2, 3) to the point where individual labs have generated genome sequences in a matter of months for material costs of as low as $48,000 (412) (table S5). Sequencing technologies, which use a variety of genomic microarray construction methodologies and sequencing chemistries (1332), can determine human genetic diversity over an entire genome and identify common as well as rare single-nucleotide polymorphisms (SNPs), insertions, and deletions. Despite these advances, improvements are still needed to enable the cost-effective characterization of the many hundreds of genomes required for genetic studies of complex diseases and for personalized disease prevention, prognosis, and treatment. We generated sequencing substrates [Fig. 1A and supporting online material (SOM)] by means of genomic DNA (gDNA) fragmentation and recursive cutting with type IIS restriction VOL 327 SCIENCE

enzymes and directional adapter insertion (Fig. 1B and fig. S1). The resulting circles were then replicated with Phi29 polymerase (RCR) (33). Using a controlled, synchronized synthesis, we obtained hundreds of tandem copies of the sequencing substrate in palindrome-promoted coils of single-stranded DNA, referred to as DNA nanoballs (DNBs) (Fig. 1C). DNBs were adsorbed onto photolithographically etched, surfacemodified (SOM) 25- by 75-mm silicon substrates with grid-patterned arrays of ~300-nm spots for DNB binding (Fig. 1C). The use of patterned arrays increased DNA content per array and image information density relative to random genomic DNA arrays (6, 9, 11, 14, 28). Highaccuracy combinatorial probe anchor ligation (cPAL) sequencing chemistry was then used to independently read up to 10 bases adjacent to each of eight anchor sites (Fig. 1D), resulting in a total of 31- to 35-base mate-paired reads (62 to 70 bases per DNB). cPAL is based on unchained hybridization and ligation technology (15, 27, 28, 31), previously used to read 6 to 7 bases from each of four adapter sites (26 total bases) (28), here extended using degenerate anchors to read up to 10 bases adjacent to each of the eight inserted adapter sites (Fig. 1D, right) with similar accuracy at all read positions (fig. S3). This increased read length is essential for human genome sequencing. Cell lines derived from two individuals previously characterized by the HapMap Project (34), a Caucasian male of European descent (NA07022) and a Yoruban female (NA19240), were sequenced. NA19240 was selected to allow for a comparison of our sequence to the sequence of the same genome currently being assembled by the 1000 Genome Project. In addition, lymphoblast DNA from a Personal Genome Project Caucasian male sample, PGP1 (NA20431) was sequenced because substantial data are available for biological comparisons (3537). Automated cluster analysis of the four-dimensional intensity data produced raw base reads and associated raw base scores (SOM). We mapped these sequence reads to the human genome reference assembly with a custom alignment algorithm that accommodates our read structure (fig. S4), resulting in between 124 and 241 Gb mapped and an overall genome coverage of 45- to 87-fold per genome. To assess representational biases during circle construction, we assayed genomic DNA and intermediate steps in the library construction process by quantitative polymerase chain reaction

78

1 JANUARY 2010

www.sciencemag.org

REPORTS
(QPCR) (fig. S2). This and mapped coverage showed a substantial deviation from Poisson expectation with excesses of both high and low coverage regions (fig. S5), but only a few percent of bases have coverage insufficient for assembly (Table 1). Much of this coverage bias is accounted for by local GC content in NA07022, a bias that was significantly reduced by improved adapter ligation and PCR conditions in NA19240 (fig. S5); the fraction of the genome with less than 15-fold coverage was accordingly reduced from 11% in NA07022 to 6.4% in NA19240, despite the latter having 25% less total coverage (Table 1). Discordance with respect to the reference genome in uniquely mapping reads from NA07022 was 2.1% (range 1.4% to 3.3% per slide). However, considering only the highest scoring 85% of base calls reduced the raw read discordance to 0.47%, including about 0.1% of true variant positions. Mapped reads were assembled into a best-fit, diploid sequence with a custom software suite employing both Bayesian and de Bruijn graph techniques (SOM). This process yielded diploid reference, variant, or no-calls at each genomic location with associated variant quality scores. Confident diploid calls were made for 86% to 95% of the reference genome (Table 1), approaching the 98% that can be reconstructed in simulations. The 2% that is not reconstructed

Fig. 1. Amplified DNA nanoarray platform. (A) Schematic flow diagram of the process used. (B) Library construction schematic (fig. S1). r1 to r8 are gDNA regions adjacent to distinct adapter ends; Ad1 to AD4 indicate adapters 1 to 4. (C) DNB production using Phi29 DNA polymerase (fig. S11) and nanoarray formation (SOM) schematics. (D) Schematic of cPAL products (SOM).

Table 1. Summary information from mapping and assembly of three genomes. All variations are with respect to the National Center for Biotechnology Information (NCBI) version 36 human genome reference assembly. Novel variations were ascertained by comparison to dbSNP [JDW, release 126; NA18507 (6), release 128; all other genomes, reMapped sequence (Gb) Average coverage depth (fold)

lease 129]. NA18507 and NA19240 are Yoruban HapMap samples, which may explain the number of SNPs and novelty rates. In partially called regions of the genome, one allele could be called confidently but not the other. The high call rate in NA19240 reflects reduced library bias (fig. S5).
SNPs Total Novel 10% 19% 10% 26% 19% 15% 18% Total 337,635 496,194 269,794 404,416 226,529 851,575 222,718 Indels Novel 37% 42% 37% 50% 33% 51% Insertion: deletion ratio

Sample

Percent of genome called Fully Partially

NA07022 (35) NA19240 (36) NA20431 (37) NA18507 (6) NA18507 (9) JCV (3) JDW (4)

241 178 124 131 87 21 21

87 63 45 46 31* 7 7

Genomes sequenced by Complete Genomics 91% 2% 3,076,869 95% 1% 4,042,801 86% 3% 2,905,517 Genomes previously published 4,139,196 3,866,085 3,213,401 - 3,322,093

1.0 0.96 1.0 0.77 0.72 0.4

*This is 18x when constrained to nonduplicated and properly mated reads, which were those used for variant calling.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

79

REPORTS
Table 2. Concordance with genotypes for NA07022 generated by the HapMap Project (release 24) and the highest quality Infinium assay subset of those genotypes, as well as genotyping on Illumina Infinium 1M assay. Discordances with reported HapMap Infinium genotypes were verified by Sanger sequencing (SOM).
Infinium 1M Published concordance* Number reported Called (%) Locus concordance (%) HapMap Homozygous reference genotype Heterozygous calls Homozygous alternate 1M 95.98% 99.89% 99.96% 99.78% 99.81% HapMap phase I and II SNPs 99.03% 3.9 M 94.39% 99.15% 99.34% 99.39% 98.14% HapMap Infinium subset 99.94% 143 K 96.00% 99.88% 99.96% 99.80% 99.84% HapMap Infinium SNPs tested for accuracy by Sanger sequencing These data correct 18 28 28 These data incorrect 2 46 12 Affirmed (%) 90% 38% 70%

NA07022

*Published concordance between Sanger sequencing and genotyping (34).

Average false negative rate is <0.19.

in simulations is composed of repeats that are longer than the ~400 base inserts used here and of high enough identity to prevent attribution of mappings to specific repeat copies. Longer matepair inserts minimize this limitation (6, 9). Similar limitations affect other short-read technologies. We identified a range of 2.91 to 4.04 million SNPs with respect to the reference genome, 81% to 90% of which are reported in the database dbSNP, as well as short indels and block substitutions (Table 1 and table S6). Because of the use of local de novo assembly, indels were detected in sizes ranging up to 50 bp. As expected, indels in coding regions tend to occur in multiples of length 3, indicating the possible selection of minimally impacting variants in coding regions (fig. S6). As an initial test of sequence accuracy, we compared our called SNPs with the HapMap phase I/II SNP genotypes reported for NA07022 (1). We fully called 94% of these positions with an overall concordance of 99.15% (Table 2) (the remaining 6% of positions were either half-called or not called). Furthermore, we fully called 96% of the Infinium (Illumina, San Diego, CA) subset of the HapMap SNPs with an overall concordance rate of 99.88%, reflecting the higher reported accuracy of these genotypes (34). Similar concordance rates with available SNP genotypes were observed in NA19240 (with a call rate of over 98%) and NA20431 (table S7). We further characterized 134 of the 168 calls that were discordant with Infinium loci and Sanger sequencing of PCR products in NA07022, demonstrating that 55% of these discordances are errors in the reported HapMap genotypes (Table 2). The relationship between detection rate and read depth for about 1M Infinium HD SNPs that we subsequently genotyped in NA07022 shows that coverage of 30-fold at a position is sufficient to detect 90% of SNPs at heterozygous loci and 97% of SNPs at homozygous loci (fig. S5). Because the whole-genome false-positive rate cannot be accurately estimated from known SNP loci, we tested a random subset of novel nonsynonymous variants in NA07022, a category that is enriched for errors (10). We extrapolated error rates from the targeted sequencing of 291

Table 3. False-positive rates and false discovery rates (FDRs) were calculated for the entire set of variations called in NA07022 by extrapolating the heterozygous (Het) FDRs calculated from comparative Sanger sequencing of 291 selected novel variants (table S8) to all variants. This is a conservative approach (detailed in SOM). The total number of all types of false-positive variants is estimated at 7.5 to 16.1 per Mb.
Variation type Total detected Novel Het Estimated Estimated false novel FDR false positives positives / Mbp (table S8) on genome 26% 814% 1118% 1129% 7k17k 5k8k 7k11k 3k9k 2.36.1 1.83.0 2.33.9 1.13.1 Estimated FDR 0.20.6% 3.05.0% 3.96.5% 5.213.9%

SNP 3,076,869 310,690 Deletion 168,726 61,960 Insertion 168,909 61,933 Block substitution 62,783 30,445

such loci and estimated the false-positive rate at about one variant per 100 kb, including <6.1 substitution-, <3.0 short deletion-, <3.9 short insertion- and <3.1 block variants per Mb (Table 3 and table S8). Aberrant mate-pair gaps may indicate the presence of length-altering structural variants and rearrangements with respect to the reference genome. A total of 2,126 clusters of such anomalous mate-pairs were identified in NA07022. We performed PCR-based confirmation of one such heterozygous 1500-base deletion (fig. S7). More than half of the clusters are consistent in size with the addition or deletion of a single Alu repeat element. Some applications of complete genome sequencing may benefit from maximal discovery rates, even at the cost of additional false positives, whereas for others, a lower discovery rate and lower false-positive rate may be preferable. We used the variant quality score to tune call rate and accuracy (fig. S8). Additionally, novelty rate (relative to the dbSNP) is also a function of variant quality score (fig. S9). We processed the NA07022 data with Traito-matic (Scalable Computer Experts, Somerville, Massachusetts, and Church Laboratory, Harvard Medical School, Cambridge, Massachusetts) automated annotation software [as in (12)], yielding 1159 annotated variants, 14 of which may have disease implications (table S10). Because the DNB sequencing substrates are produced by rolling-circle replication (33) in a uniform-temperature, solution-phase reaction with high template concentrations (>20 billion VOL 327 SCIENCE

per ml), this system avoids substantial selection bottlenecks and nonclonal DNBs. This circumvents the stochastic inefficiencies of approaches that require precise titration of template concentrations for in situ clonal amplification in emulsion (9, 14, 29) or bridge PCR (6, 19). Our patterned arrays include high-occupancy and high-density nanoarrays self-assembled on photolithography-patterned, solid-phase substrates through electrostatic adsorption of solution-phase DNBs and yield a high proportion of informative pixels (site occupancies >95%) (fig. S12A) compared with random-position DNA arrays. This results in several hundred reaction sites in the compact (~300-nm diameter) DNB that produce bright signals useful for rapid imaging of the sequences (SOM). Such small DNBs also allow for high-density arrays. The data set reported herein was generated with arrays with ~350 million spots at a pitch of 1.29 mm. Such a spot density and higher ones achieved in proofof-concept experiments (fig. S12B) result in high image efficiency and reduced reagent consumption that enable high sequencing throughput per instrument critical for high-scale human genome sequencing for research and clinical applications. Both sequencing by synthesis (SBS) and sequencing by ligation (SBL) use chained reads, wherein the substrate for cycle N + 1 is dependent on the product of cycle N; consequently, errors may accumulate over multiple cycles and data quality may be affected by errors (especially incomplete extensions) occurring in previous cycles. Thus, reactions need to be driven to

80

1 JANUARY 2010

www.sciencemag.org

REPORTS
near completion with high concentrations of expensive high-purity labeled substrate molecules and enzymes. The independent, unchained nature of cPAL avoids error accumulation and tolerates low-quality bases in otherwise highquality reads, thereby decreasing reagent costs. The average sequencing consumables cost for these three genomes was under $4400 (table S5). The raw base and variant call accuracy achieved compares favorably with other reported human genome sequences (212). Because the sequencing substrates are produced by a DNA engineering process based on modified nick-translation for directional adapter insertion (SOM), we obtained over 90% yield in adapter ligation and low chimeric rates of about 1% (SOM). DNA molecules with an inserted adapter are further enriched with PCR (SOM). This recursive process can be implemented in batches of 96 samples and extended by inserting additional adapters to read 120 bases or more per DNB (fig. S10). The current read length is comparable to other massively parallel sequencing technologies (612). The sequence data reported here achieve sufficient quality and accuracy for complete genome association studies, the identification of potentially rare variants associated with disease or therapeutic treatments, and the identification of somatic mutations. The low cost of consumables and efficient imaging may enable studies of hundreds of individuals. The higher accuracy and completeness required for clinical diagnostic applications provides an incentive for continued improvement of this and other technologies.
References and Notes
1. T. A. Manolio, L. D. Brooks, F. S. Collins, J. Clin. Invest. 118, 1590 (2008). 2. International Human Genome Sequencing Consortium, Nature 409, 860 (2001). 3. J. C. Venter et al., Science 291, 1304 (2001). 4. S. Levy et al., PLoS Biol. 5, e254 (2007). 5. D. A. Wheeler et al., Nature 452, 872 (2008). 6. D. R. Bentley et al., Nature 456, 53 (2008). 7. J. Wang et al., Nature 456, 60 (2008). 8. S. M. Ahn et al., Genome Res. 19, 1622 (2009). 9. K. J. McKernan et al., Genome Res. 19, 1527 (2009). 10. T. J. Ley et al., Nature 456, 66 (2008). 11. D. Pushkarev, N. F. Neff, S. R. Quake, Nat. Biotechnol. 27, 847 (2009). 12. J. I. Kim et al., Nature 460, 1011 (2009). 13. R. Drmanac et al., Genomics 4, 114 (1989). 14. R. Drmanac, R. Crkvenjakov, Sci. Yugosl. 16, 97 (1990); pdf accessible at www.rdrmanac.com. 15. R. Drmanac et al., Science 260, 1649 (1993). 16. P. C. Cheesman, U.S. patent 5,302,509 (1994). 17. R. Drmanac, World Intellectual Property Organization WO/1995/009248 (1995). 18. M. Ronaghi, S. Karamohamed, B. Pettersson, M. Uhlen, P. Nyren, Anal. Biochem. 242, 84 (1996). 19. C. P. Adams, S. J. Kron, U.S. Patent 5,641,658 (1997). 20. P. M. Lizardi et al., Nat. Genet. 19, 225 (1998). 21. S. C. Macevicz, U.S. Patent 5,750,341 (1998). 22. S. Drmanac et al., Nat. Biotechnol. 16, 54 (1998). 23. R. D. Mitra, G. M. Church, Nucleic Acids Res. 27, e34 (1999). 24. S. Brenner et al., Nat. Biotechnol. 18, 630 (2000). 25. I. Braslavsky, B. Hebert, E. Kartalov, S. R. Quake, Proc. Natl. Acad. Sci. U.S.A. 100, 3960 (2003). 26. R. D. Mitra, J. Shendure, J. Olejnik, O. E. Krzymanska, G. M. Church, Anal. Biochem. 320, 55 (2003). 27. R. Drmanac et al., World Intellectual Property Organization WO/2004/076683 (2004). 28. J. Shendure et al., Science 309, 1728 (2005). 29. M. Margulies et al., Nature 437, 376 (2005). 30. T. D. Harris et al., Science 320, 106 (2008). 31. A. Pihlak et al., Nat. Biotechnol. 26, 676 (2008). 32. J. Shendure, H. Ji, Nat. Biotechnol. 26, 1135 (2008). 33. L. Blanco et al., J. Biol. Chem. 264, 8935 (1989). 34. The International HapMap Consortium, Nature 449, 851 (2007). 35. K. Zhang et al., Nat. Methods 6, 613 (2009). 36. M. P. Ball et al., Nat. Biotechnol. 27, 361 (2009). 37. J. B. Li et al., Genome Res. 19, 1606 (2009). 38. We acknowledge the ongoing contributions and support of all Complete Genomics employees and R. Mercado for manuscript preparation. Some of this work was supported by PersonalGenomes.org and the National Heart, Lung, and Blood Institute. Data has been deposited at the National Center for Biotechnology Information: reads in the Short Read Archive (SRA), accession SRA008092, and SNPs in dbSNP, accessions ss161884913 to ss175323894. Employees of Complete Genomics have stock options in the company, and D.G.B. has stock in Perlegen Sciences. Complete Genomics has filed several patents on this work.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1181498/DC1 Materials and Methods Figs. S1 to S12 Tables S1 to S9 References 3 September 2009; accepted 23 October 2009 Published online 5 November 2009; 10.1126/science.1181498 Include this information when citing this paper.

Structure and Mechanisms of a Protein-Based Organelle in Escherichia coli

Shiho Tanaka,1 Michael R. Sawaya,2,3 Todd O. Yeates1,3,4* Many bacterial cells contain proteinaceous microcompartments that act as simple organelles by sequestering specific metabolic processes involving volatile or toxic metabolites. Here we report the three-dimensional (3D) crystal structures, with resolutions between 1.65 and 2.5 angstroms, of the four homologous proteins (EutS, EutL, EutK, and EutM) that are thought to be the major shell constituents of a functionally complex ethanolamine utilization (Eut) microcompartment. The Eut microcompartment is used to sequester the metabolism of ethanolamine in bacteria such as Escherichia coli and Salmonella enterica. The four Eut shell proteins share an overall similar 3D fold, but they have distinguishing structural features that help explain the specific roles they play in the microcompartment. For example, EutL undergoes a conformational change that is probably involved in gating molecular transport through shell protein pores, whereas structural evidence suggests that EutK might bind a nucleic acid component. Together these structures give mechanistic insight into bacterial microcompartments.

Other bacterial microcompartments with more complex metabolic functions have also been discovered (5, 15). One of these, which is dedicated to ethanolamine utilization (Eut), is present in several bacteria, including Salmonella enterica and Escherichia coli (4, 16). The Eut microcompartment shares a number of homologous enzymes with a propanediol utilization (Pdu) microcompartment; both metabolic pathways proceed via aldehyde intermediates, propionaldehyde in the case of Pdu and acetaldehyde in the case of Eut (3). Experiments in Salmonella have shown that the cellular function of the Eut microcompartment is to metabolize ethanolamine without allowing the release of acetaldehyde into the cytosol (Fig. 1B), thus mitigating the potentially toxic effects of excess aldehyde in the bacterial cytosol (1719) and also preventing the volatile acetaldehyde from diffusing across the cell membrane and leading to a loss of carbon (20).
1 Department of Chemistry and Biochemistry, University of California Los Angeles, 611 Charles Young Drive East, Los Angeles, CA 90095, USA. 2Howard Hughes Medical Institute, University of California Los Angeles, 611 Charles Young Drive East, Los Angeles, CA 90095, USA. 3Univeristy of California Los Angeles, Department of Energy Institute for Genomics and Proteomics, 611 Charles Young Drive East, Los Angeles, CA 90095, USA. 4Molecular Biology Institute, University of California Los Angeles 611 Charles Young Drive East, Los Angeles, CA 90095, USA.

acterial microcompartments are present in diverse bacteria, where they function as protein-based organelles (16). They range in size from just under 1000 to around 1500 , and they typically have a polyhedral shape (Fig. 1A). Each type of microcompartment contains a few different enzymes that catalyze sequential metabolic reactions. The enzymes are encapsulated by a shell formed from a few thousand shell protein

subunits. The simplest microcompartment is the carboxysome, which encapsulates the two enzymes carbonic anhydrase and ribulose-1,5bisphosphate carboxylase-oxygenase (RuBisCO) in order to enhance cellular CO2 fixation (1, 7, 8). Recent structural studies on the carboxysome and its shell proteins have provided a basic understanding of how that type of microcompartment is assembled and how it operates (914). SCIENCE VOL 327

*To whom correspondence should be addressed. E-mail: [email protected]

www.sciencemag.org

1 JANUARY 2010

81

REPORTS
Fig. 1. A model for the Eut microcompartment and its metabolic pathway. (A) A hypothetical model of the Eut microcompartment emphasizing the construction of a semiregular polyhedron primarily from hexameric shell proteins. (B) A model for the metabolism of ethanolamine in the Eut microcompartment. Ethanolamine is converted into ethanol, acetyl-phosphate, and acetylcoenzyme A (CoA), to be used in the tricarboxylic acid cycle for energy production. The volatile intermediate, acetaldehyde (boxed in orange), is consumed before it can escape the protein shell. The four homologous shell proteins belonging to the conserved BMC family (EutK, EutL, EutM, and EutS) are colored in light blue. The unrelated EutN protein (dark pink) may be a minor structural component of the shell, based on analogy to the carboxysome (11). Enzymes thought to be associated with the

microcompartment are indicated. Protein names: EutBC, ethanolamine ammonialyase; EutD, phosphotransacetylase; EutE, aldehyde dehydrogenase; EutG, alcohol dehydrogenase. The eut operon (fig. S1) encodes other enzymes and proteins involved in ethanolamine utilization, which are not shown. from microcompartments of the carboxysome type (21), and can now be identified in some 40 genera across the bacterial kingdom (2, 3). In Salmonella and E. coli, the 17-gene eut operon codes for four homologous BMC proteins: EutK, EutL, EutM, and EutS [(4) and supporting online material (SOM) text]. Genetic studies in Salmonella have shown that EutL, EutM, and EutK are required for growth on ethanolamine when it is the sole carbon source at high pH, and their deletion leads to rapid loss of acetaldehyde (20). Likewise, deleting EutS led to a measurable loss of acetaldehyde but did not cause a growth phenotype under the conditions examined. A fifth protein, EutN, which is probably a minor shell component but not homologous to the BMC shell proteins, was also shown to be essential for growth on ethanolamine; its structure was reported previously (11). Here we report the crystal structures of the BMC shell proteins from E. coli: EutM, EutS (wild-type and mutant), EutL (in two crystal forms), and a domain of EutK. The Eut shell proteins from E. coli share between 79 and 96% sequence identity with their orthologs from Salmonella (fig. S1 and SOM text), where functional studies have been performed (17, 18, 20). The crystal structure of EutM was determined to a resolution of 2.1 . The 97amino acid EutM protein adopts a 3D fold with an a/b structure that matches closely those reported earlier for proteins forming the shell of the carboxysome (9, 10). Six EutM subunits self-assemble to make a flat cyclic hexamer with a bowl-shaped depression on one side, punctuated by a narrow central pore (Fig. 2A). Similar hexameric structures have been established as the basic building blocks from which microcompartment shells are assembled by the tight packing of many hexamers side by side into a molecular layer or sheet (Fig. 1A) (9, 10, 12). Electron density, interpreted to be a sulfate ion from the crystallization mixture, was visualized in the center of the EutM pore, in accordance with previous suggestions that small molecules and ions can

Fig. 2. Structure of EutM and EutS shell protein hexamers. (A) Cartoon diagrams of the EutM (left), EutS (middle), and mutant EutS G39V (right) hexamers, shown in two views. The wild-type EutS hexamer is bent away from a flat configuration by approximately 40 (SOM text). This asymmetric configuration of wildtype EutS was converted into a flat symmetric configuration by the G39V mutation. (B) A hypothetical model showing how EutS (orange) might introduce curvature in an otherwise flat hexameric sheet of shell protein hexamers. Hexamer interfaces in the model are based on packing arrangements that are conserved in other crystal structures. From a row of EutS hexamers at an edge, flat hexamers could extend to form facets of the shell without major collisions. The major shell proteins from known bacterial microcompartments belong to a family of homologous proteins that are typically just over 100 amino acids long, referred to here as bacterial microcompartment (BMC) proteins. The amino-acid sequences of BMC proteins were determined first VOL 327 SCIENCE

82

1 JANUARY 2010

www.sciencemag.org

REPORTS
pass through the centers of the hexameric shell proteins (fig. S2) (9, 10). The relative abundances of the various shell proteins in the Eut microcompartment have not been established, but the EutM protein hexamer, with its relatively typical features, could serve as the basic assembly component of the shell, whereas the other shell proteins, whose unusual features we describe below, could perform more-specialized roles. The crystal structure of EutS was determined at 1.65 resolution. The 111amino acid protein adopts the conserved a/b core structure expected for BMC proteins (Fig. 2A), but the secondary structure elements occur in nonsequential order relative to the typical BMC fold. Similar circular permutations of the BMC protein fold have been observed previously (2224). Unlike other BMC proteins that have formed flat hexameric structures, the EutS hexamer has a bend of approximately 40 (Fig. 2 and SOM text). As a result, the six chemically equivalent protein subunits exist in three distinct structural environments. The closest parallels to this asymmetric assembly are to the proteins of certain viral capsids in which equivalent protein subunits are present in quasiequivalent, but distinct, environments (25, 26). The observation could be helpful in explaining microcompartment architecture, because previously observed (flat) BMC hexamers have not clarified how the edges of polyhedral microcompartment shells might be formed. The cause of the bent EutS structure was investigated by mutagenesis. A glycine residue (Gly39) involved in a crystal contact was mutated to a valine side chain (G39V). Valine is the corresponding amino acid in a homologous shell protein from the Pdu microcompartment, namely PduU, which forms a flat symmetric hexamer (22). The G39V mutant of EutS migrated more slowly in a native gel (despite an unaltered net charge), implicating a substantial conformational change (fig. S3). That observation was consistent with a crystal structure that demonstrated that the mutated EutS was flattened into a symmetric structure (Fig. 2A). Thus Gly39 appears to play an important role in bending EutS. Gly39 is conserved among EutS proteins from the eut operon of other bacteria but is absent from other BMC proteins. EutL is the longest of the Eut shell proteins. A recent structure (24) showed that it contains within a single polypeptide chain two tandem domains (Fig. 3A), with each domain being a circular permutation of the typical BMC fold (22), similar to EutS. As such, three copies of EutL assemble to form a pseudohexameric structure. We determined the crystal structure of EutL from two distinct crystal forms, both at a resolution of 2.3 . A comparison of the two structures revealed that protein backbone movements of nearly 15 (in loops covering residues 65 to 83 and 174 to 185) interconvert the EutL trimer between forms in which the central pore is either open or closed (Fig. 3B). The conformation for EutL identified here as being the closed form matches the one described earlier (24). The central pore is very nearly occluded. Very small openings in this structure were noted earlier as potential routes for transport, but it is likely the open form observed here that is competent for transport. The pore in the open form is triangular in shape, with an edge of approximately 11 and a diameter at its narrowest point of about 8 . This large pore is salient, because it provides a potential route for transporting bulky molecules such as cobalamin and other nucleotide cofactors that are required by reactions inside the microcompartment (3) (Fig. 1B). Model building suggests that the open pore could accommodate transport of these cofactors with only modest sidechain movements (fig. S6). The existence of a closed form is also potentially important, because the possibility of a triggered or gated opening goes partway toward addressing the paradox of how relatively large molecules might penetrate the shell without allowing rapid loss of the small acetaldehyde intermediate (3). A potential mechanism for

Fig. 3. The structure of the EutL shell protein and its gated pore. (A) A cartoon ribbon diagram of a EutL monomer in its closed form. The first and second BMC domains are colored in blue and purple, respectively. (B) The open (left) and closed (right) configurations of EutL trimers are shown in both ribbon diagram and surface representations. In both configurations, EutL trimers (or pseudohexamers) were observed to pack into tight molecular layers within their respective crystal forms (bottom). The conformational change between forms involves 15 movements of the protein backbone (fig. S4). Fig. 4. The structure of the EutK Ctail. (A) A schematic diagram of the BMC domain and the crystal structure of the C-terminal helix-turn-helix domain of EutK, interpreted as a monomer (fig. S7). (B) Superimpositions of the structure of the EutK-Ctail on four representative DNA/RNA binding domains (SOM text, colored according to fig. S9). (C) An electrostatically colored surface diagram of the EutK-Ctail domain emphasizing the positively charged (blue) patch that is characteristic of DNA binding domains. www.sciencemag.org SCIENCE VOL 327

1 JANUARY 2010

83

REPORTS
selective transport would involve a coupling between cofactor binding and pore opening. A similar conformational change was reported recently in a carboxysome shell protein, CsoS1D (23) (figs. S4 and S5), suggesting that gated transport could be a common mechanism in microcompartments. Like some other BMC proteins, EutL was found to associate into tightly packed molecular layers within both of the crystal forms characterized here (Fig. 3B). This observation lends further support to models for microcompartment architecture in which the shell is formed by the tight side-byside packing of hexameric (or pseudohexameric) protein building blocks into a molecular layer or sheet (911). The appearance of the EutL protein layer, composed of subunits in either the open or closed configuration, illustrates how strongly the porosity of the microcompartment would be affected by the conformational change in EutL (Fig. 3B). The fourth shell protein, EutK, is distinct among the shell proteins in the Eut microcompartment. First, although all other BMC proteins studied to date form hexamers (or pseudohexamers, like EutL), EutK is a monomer in solution (fig. S8). The apparent inability of EutK to assemble into a hexamer by itself suggests that different BMC paralogs might form mixed hexamers during assembly of the shell. Second, EutK has an extra protein domain of about 60 amino acids following the conserved BMC-type domain (Fig. 4A). Numerous instances of BMC-type proteins fused to other uncharacterized protein domains are evident in the protein sequence databases, but structural and functional data are limited (27). Based on sequence similarity of marginal statistical certainty (for example, a 40% probability of similarity due to random chance), the extra domain of EutK could be only tentatively assigned to a broad family of proteins bearing a well-known helixturn-helix motif, which is common among nucleic acidbinding proteins. Whereas crystals could not be grown using the full-length EutK protein, the crystal structure of the C-terminal domain was elucidated at a resolution of 2.1 . The crystal structure of this 59 amino acid fragment, referred to hereafter as EutK-Ctail, demonstrates that it is indeed a helixturn-helix domain (Fig. 4B). A computational search for similar structures in the protein structure database identified many known helix-turn-helix domains as close matches, with coordinate differences as low as 1.0 . About 90% of the helixturn-helix domain structures retrieved from the search bind nucleic acids, whereas a minority perform other varied cellular functions (SOM text). A comparison of the surface features of EutK-Ctail with previously characterized helix-turn-helix domains, including those that bind nucleic acids and those that do not, suggests strongly that EutK is a nucleic acidbinding protein. A prominent, positively charged surface is conserved in EutK-Ctail and in those domains that bind nucleic acids (Fig. 4C). Interestingly, on the basis of structure-guided alignments, EutK-Ctail shows the highest sequence similarity to the helix-turn-helix domains of the Ic1R family of transcription factors, which regulate, among other things, carbon metabolism in enterobacteria (28, 29) (fig. S9). The specific function of EutK and the identity of its cognate nucleic acid are unknown. Nonetheless, the key prediction that the Eut microcompartment shell binds a nucleic acid points up the possibility of unexpected parallels to viral capsids, which bind to and encapsulate their viral genomes. Taken together, the structures of the shell proteins from the Eut microcompartment paint a picture of a complex, highly evolved organelle. The structural features and conformational changes observed illustrate how these evolutionarily related proteins have diverged to fulfill specialized architectural and biochemical roles in a shell that participates actively in the function of the microcompartment.
References and Notes
1. G. C. Cannon et al., Appl. Environ. Microbiol. 67, 5351 (2001). 2. T. O. Yeates, C. A. Kerfeld, S. Heinhorst, G. C. Cannon, J. M. Shively, Nat. Rev. Microbiol. 6, 681 (2008). 3. S. Cheng, Y. Liu, C. S. Crowley, T. O. Yeates, T. A. Bobik, Bioessays 30, 1084 (2008). 4. E. Kofoid, C. Rappleye, I. Stojiljkovic, J. Roth, J. Bacteriol. 181, 5317 (1999). 5. P. Chen, D. I. Andersson, J. R. Roth, J. Bacteriol. 176, 5474 (1994). 6. J. M. Shively et al., Int. Rev. Cytol. 113, 35 (1988). 7. J. M. Shively, F. Ball, D. H. Brown, R. E. Saunders, Science 182, 584 (1973). 8. M. R. Badger, G. D. Price, J. Exp. Bot. 54, 609 (2003). 9. C. A. Kerfeld et al., Science 309, 936 (2005). 10. Y. Tsai et al., PLoS Biol. 5, e144 (2007). 11. S. Tanaka et al., Science 319, 1083 (2008). 12. S. Tanaka, M. R. Sawaya, M. Phillips, T. O. Yeates, Protein Sci. 18, 108 (2009). 13. C. V. Iancu et al., J. Mol. Biol. 372, 764 (2007). 14. M. F. Schmid et al., J. Mol. Biol. 364, 526 (2006). 15. G. D. Havemann, E. M. Sampson, T. A. Bobik, J. Bacteriol. 184, 1253 (2002). 16. J. M. Shively et al., Can. J. Bot. 76, 906 (1998). 17. M. R. Rondon, A. R. Horswill, J. C. Escalante-Semerena, J. Bacteriol. 177, 7119 (1995). 18. E. M. Sampson, T. A. Bobik, J. Bacteriol. 190, 2966 (2008). 19. M. R. Rondon, R. Kazmierczak, J. C. Escalante-Semerena, J. Bacteriol. 177, 5434 (1995). 20. J. T. Penrod, J. R. Roth, J. Bacteriol. 188, 2865 (2006). 21. R. S. English, S. C. Lorbach, X. Qin, J. M. Shively, Mol. Microbiol. 12, 647 (1994). 22. C. S. Crowley, M. R. Sawaya, T. A. Bobik, T. O. Yeates, Structure 16, 1324 (2008). 23. M. G. Klein et al., J. Mol. Biol. 392, 319 (2009). 24. M. Sagermann, A. Ohtaki, K. Nikolakakis, Proc. Natl. Acad. Sci. U.S.A. 106, 8883 (2009). 25. D. L. Caspar, A. Klug, Cold Spring Harb. Symp. Quant. Biol. 27, 1 (1962). 26. M. Carrillo-Tripp et al., Nucleic Acids Res. 37, D436 (2009). 27. J. B. Parsons et al., J. Biol. Chem. 283, 14366 (2008). 28. A. J. Molina-Henares, T. Krell, M. Eugenia Guazzaroni, A. Segura, J. L. Ramos, FEMS Microbiol. Rev. 30, 157 (2006). 29. R. G. Zhang et al., J. Biol. Chem. 277, 19183 (2002). 30. We thank T. Bobik, members of the Yeates lab, J. Escalante-Semerena, F. Guo, M. Faller, and Y. Chen for helpful discussions and M. Phillips for equilibrium sedimentation experiments. This work was supported by the Biological and Environmental Research program of the Department of Energy Office of Science and by NSF grant MCB-0843065. Coordinates and structure factors of the EutK-Ctail, EutL open form, EutL closed form, EutM, EutS, and EutS-G39V have been deposited in the Protein Data Bank under accession numbers 3I71, 3I87, 3I82, 3I6P, 3I96, and 3IA0, respectively.

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/81/DC1 Materials and Methods SOM Text Figs. S1 to S9 Tables S1 to S3 References 22 July 2009; accepted 22 October 2009 10.1126/science.1179513

The Tasmanian Devil Transcriptome Reveals Schwann Cell Origins of a Clonally Transmissible Cancer
Elizabeth P. Murchison,1,2* Cesar Tovar,3 Arthur Hsu,4 Hannah S. Bender,1,2 Pouya Kheradpour,5 Clare A. Rebbeck,1 David Obendorf,3 Carly Conlan,1 Melanie Bahlo,4 Catherine A. Blizzard,3 Stephen Pyecroft,6 Alexandre Kreiss,3 Manolis Kellis,5,7 Alexander Stark,5,7 Timothy T. Harkins,8 Jennifer A. Marshall Graves,2 Gregory M. Woods,3 Gregory J. Hannon,1 Anthony T. Papenfuss4 The Tasmanian devil, a marsupial carnivore, is endangered because of the emergence of a transmissible cancer known as devil facial tumor disease (DFTD). This fatal cancer is clonally derived and is an allograft transmitted between devils by biting. We performed a large-scale genetic analysis of DFTD with microsatellite genotyping, a mitochondrial genome analysis, and deep sequencing of the DFTD transcriptome and microRNAs. These studies confirm that DFTD is a monophyletic clonally transmissible tumor and suggest that the disease is of Schwann cell origin. On the basis of these results, we have generated a diagnostic marker for DFTD and identify a suite of genes relevant to DFTD pathology and transmission. We provide a genomic data set for the Tasmanian devil that is applicable to cancer diagnosis, disease evolution, and conservation biology.

evil facial tumor disease (DFTD) is a transmissible cancer affecting the Tasmanian devil (Sarcophilus harrisii), an endemic Tasmanian marsupial carnivore. First VOL 327 SCIENCE

observed in 1996 in northeastern Tasmania, DFTD has been implicated in devil population collapse (1, 2). DFTD is a rapidly fatal disease that culminates in large tumors, primarily on the face

84

1 JANUARY 2010

www.sciencemag.org

REPORTS
and mouth, which frequently metastasize to internal organs (3). There are no diagnostic tests, treatments, or vaccines available for DFTD, and models predict that the disease could lead to extinction of wild Tasmanian devils within 25 to 35 years (4). DFTD appears to be a clonal cell line, transmitted (by biting) as an allograft between devils (5, 6) and may be similar in transmission to canine transmissible venereal tumor (CTVT) and a transmissible sarcoma affecting Syrian hamsters (79). The prevalence and biology of such somatic cell parasites is generally unknown (10). Studies of captive Tasmanian devils have suggested that the species is prone to developing tumors, particularly carcinomas (11, 12). However, DFTD does not resemble previously described devil cancers (3, 13), and determining its etiology is critical for developing management strategies for the disease. Cytologically, DFTD appears as a soft tissue neoplasm consisting of undifferentiated round to spindle-shaped cells with few defining ultrastructural features (3, 13). Immunohistochemistry suggests that the tumor is derived from neuroectoderm (13). Clonal transmission of DFTD was proposed on the basis of karyotypic evidence (5) and was supported by genetic analysis of DFTD tumors at microsatellite and major histocompatibility complex loci (6). We genotyped at 14 microsatellite loci 25 paired tumor and host samples, as well as 10 samples from DFTD-unaffected devils from 16 locations in Tasmania (14) (figs. S1 and S2 and table S1). All DFTD tumors shared a similar genotype across all loci, regardless of location, sex, or age of the animal (P = 0.18, permutation test) (figs. S1 and S2). Furthermore, the tumor genotype was distinct from that of both the hosts and unaffected devils (P < 0.001, permutation test) (figs. S1 and S2). These data were consistent with previous studies (5, 6) and support the supposition that DFTD is a single clonal cell line propagated as a tumor allograft. To further assess the clonal origin of DFTD, we sequenced a 1180base pair fragment of the mitochondrial locus control region (LCR) from 14 tumors, 14 hosts, and 9 DFTD-unaffected devils (table S2). We found that all devils and tumors shared a single LCR haplotype in this region, except for one single-nucleotide polymorphism at position 15,711, which supported the idea that the tumors are clonal. Furthermore, this nucleotide variant was observed only in DFTD-free devils from western Tasmania (figs. S1 and S2 and table S2), a genetically distinct population (15). The karyotypic and genetic consistency between DFTD tumors (figs. S1 and S2) (5, 6) reinforces epidemiological evidence for a recent origin of DFTD (1, 3). We cloned and deeply sequenced microRNA (miRNA) from 10 devil tissues and five DFTD tumors, including one DFTD mammary metastasis, to gain insight into the histogenesis of DFTD. The 114 Tasmanian devil miRNAs, identified with strict conservation parameters, showed characteristic tissue-specific expression profiles (Fig. 1 and table S3). DFTD had a relatively consistent miRNA profile that was distinct from the other 10 tissues (Fig. 1 and fig. S3). Differential expression of four miRNAs in DFTD relative to a non-DFTD tissue (testis) was confirmed by quantitative polymerase chain reaction (PCR) (fig. S4). Hierarchical clustering based on Pearsons correlation statistic showed that the DFTD tumors were clustered (Fig. 1 and table S4) and that the non-DFTD miRNA profile most highly correlated with DFTD was that of brain (Fig. 1 and table S4). In cancer, miRNAs can both promote and suppress tumors, as well as regulate processes including cell proliferation, angiogenesis, and metastasis (16). It is noteworthy that the DFTD profile included a number of miRNAs commonly up-regulated in tumors, including miR-21, miR-24, and miR-19b (16), plus a miRNA that has been linked to tumor immune evasion (miR-222) (17) (Fig. 1 and fig. S3). In contrast, DFTD expresses very low levels of miR-29b and miR-126, two miRNAs suggested to suppress tumors (Fig. 1 and fig. S3) (16). To create a catalog of genes expressed in DFTD, we sequenced the transcriptome of DFTD and testis (chosen because of its expression of a broad range of genes) from an individual Tasmanian devil resulting in 13,665 and 16,438

Watson School of Biological Sciences, Howard Hughes Medical Institute, Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY 11724, USA. 2Research School of Biology and the ARC Centre of Excellence in Kangaroo Genomics, The Australian National University, Canberra, ACT 0200, Australia. 3Menzies Research Institute, The University of Tasmania, Hobart, Tasmania 7000, Australia. 4Bioinformatics Division, The Walter and Eliza Hall Institute of Medical Research, Parkville, Melbourne, Victoria 3052, Australia. 5Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02139, USA. 6Mt. Pleasant Animal Health Laboratories, Department of Primary Industries and Water, Kings Meadows, Tasmania 7249, Australia. 7 Broad Institute of MIT and Harvard, Cambridge, MA 02141, USA. 8Roche Applied Sciences, Indianapolis, IN 46250, USA. *Present address: Cancer Genome Project, Wellcome Trust Sanger Institute, Hinxton, Cambridge CB10 1SA, UK. To whom correspondence should be addressed. E-mail: [email protected] Present address: Research Institute of Molecular Pathology (IMP), Dr. Bohr-Gasse 7, A-1030, Vienna, Austria.

Fig. 1. miRNA profiling of DFTD. Heat map of normalized miRNA reads for 114 miRNAs cloned and sequenced from 10 Tasmanian devil tissues; four DFTD facial tumors (DFTD1, 2, 3, and 20); and one DFTD mammary metastasis (DFTD2,met). miRNAs were clustered on the basis of Pearsons correlation statistic, and bootstrap values (percentage) are indicated. The DFTD miRNA profile is shown in greater detail. miRNAs were annotated on the basis of conservation with comparison species hsa, human; mdo, opossum. SCIENCE VOL 327 1 JANUARY 2010

www.sciencemag.org

85

REPORTS
unique transcripts identified on the basis of similarity with human proteins and opossum transcripts, respectively (see SOM text). Of the annotated transcripts, 0.4% were differentially represented between tumor and testis libraries (fig. S5). The 31 transcripts with the most significant enrichment in the tumor library (tumorto-testis ratio of at least 2.5, chi-squared test P 0.05) were validated by semiquantitative PCR (Fig. 2A). Of these, 20 transcripts were at least twice as highly expressed in tumor as in testis (Fig. 2A). The gene with the highest expression in DFTD relative to a housekeeping gene, GAPDH (encoding glyceraldehyde-3-phosphate dehydrogenase), was MBP, the gene that encodes myelin basic protein (Fig. 2A). Indeed, 9 of the 20 validated tumor genes (45%) were found to be involved in the myelination pathway (table S6). Myelin, an insulating membranous layer that ensheathes nerve axons, is produced by oligodendrocytes in the central nervous system and by Schwann cells in the peripheral nervous system. Components of a molecular network that controls the differentiation of Schwann cells, including transcription factors (SOX10, SOX2, POU5F1, and JUN) and structural myelin genes (MPZ, PRX, MBP, and PMP22), were apparent in the differentially expressed DFTD transcriptome (Fig. 2A) (18). We also measured the expression of 31 tumorenriched genes from a panel of 10 Tasmanian devil tissues by semiquantitative PCR. Hierarchical clustering of this data set grouped DFTD together with peripheral nerves, a tissue enriched for Schwann cells (Fig. 2B). To confirm the expression of myelin proteins in DFTD, we stained DFTD tumor tissues with the antibody against Schwann cellspecific myelin protein, periaxin (PRX) (Fig. 3). All tumor cells in DFTD lesions, as well as the myelinated sheaths of peripheral nerve bundles, were strongly and specifically positive for PRX (Fig. 3). We also detected protein expression in DFTD of myelin proteins MBP, MPZ, and PMP22, as well as NES, NGFR, and S100 (figs. S6 and S7, and table S7). These experiments confirm that DFTD expresses genes found in myelinating cells, including Schwann cellspecific markers (PRX, PMP22, and MPZ). At present, diagnosis of DFTD is based on clinical features and histology. This can pose particular difficulties for the diagnosis of atypical DFTD such as nonfacial DFTD or DFTD metastases. We tested PRX as a potential diagnostic marker for DFTD by staining a panel of both DFTD and non-DFTD Tasmanian devil tumors with PRX. All (n = 20) of the DFTD tumors were positive for PRX, whereas none of the nine non-DFTD tumors tested were positive (table S7). In addition, all (n = 10) of the DFTD metastases collected from a variety of organs were positive for PRX (table S7). Thus, PRX is a strong and specific marker for DFTD and is suitable for diagnostic evaluation.

Fig. 2. DFTD transcriptome. (A) Semiquantitative RT-PCR expression profiling of 31 genes with enriched expression in tumor relative to testis [454read count fold change 2.5, P 0.05, chisquared test; (green points in fig. S5)]. Log values of the mean expression difference of DFTD genes relative to testis (blue bars) and relative to GAPDH (red bars) are shown. Error bars represent standard deviation. Significant differences between tumor and testis expression levels (P 0.05) are indicated by an asterisk (two-sample t test, Holms correction for multiple testing). (B) Heat map of semiquantitative PCR expression profiles of 31 genes across a panel of tissues including peripheral nerve (PN), a Schwann cellenriched tissue. Panel color represents the mean gene expression level, standardized across tissues ( z score). Hierarchical clustering based on Pearsons correlation statistic is indicated by dendrograms. For each tissue three biological replicates were performed (n = 3). ND, not determined. It is striking that DFTD, a cytologically undifferentiated tumor, expresses markers of highly differentiated Schwann cells because human Schwann cell tumors rarely express a complete set of terminally differentiated myelin genes (19). Although it is possible that Schwann cell genes have been activated in DFTD cells after carcinogenic transformation, VOL 327 SCIENCE it is more likely that the myelin program observed here reflects the DFTD cell of origin. We therefore propose that DFTD is a peripheral nerve sheath tumor that arose from a Schwann cell or Schwann cell precursor; this is supported by the miRNA profile of DFTD (Figs. 1 and 2). Schwann cells participate in nerve repair after injury and also modulate

86

1 JANUARY 2010

www.sciencemag.org

REPORTS
H&E PRX

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21. 22.

Fig. 3. Identification of a diagnostic marker for DFTD. (A to D) Hematoxylin and eosin (H&E) and (E to H) PRX antibody stains used for DFTD tumor histology. Arrowheads, DFTD tissue; arrows, peripheral nerve bundles (containing Schwann cells). Magnification, 4 (A and E; scale bar, 200 mm); 20 (B and F; scale bar, 50 mm); and 100 (C and G) peripheral nerve bundle (scale bar, 20 mm) and (D and H) DFTD tumor (scale bar, 20 mm). Boxed areas indicate approximate locations of areas magnified in lower panels. local immune reactions in the peripheral nervous system (20, 21). The plasticity and immunocompetence of Schwann cells may be significant in the evolution of DFTD as a transmissible cancer. Interestingly, proopiomelanocortin (POMC), a gene encoding a peptide hormone precursor, was expressed in DFTD (Fig. 2). This transcript varies by more than 100-fold between tumors (fig. S8). Technical limitations, due to lack of antibody cross-reactivity with devil, prevented the detection of physiological levels of adrenocorticotropic hormone (ACTH), a labile POMC cleavage product. However, compared with healthy individuals, diseased devils had higher, but not significantly different, levels of serum cortisol (fig. S9, P = 0.06, Students t test). Cortisol is a steroid hormone, released by ACTH activity, that is involved in stress responses and has immunosuppressive activity (22). These studies raise the intriguing possibility that DFTD may secrete factors that could alter host physiology and/or behavior, an interesting area for further study. Our study highlights the activities of two miRNA gene networks that may function in DFTD. PMP22, a myelin gene expressed in DFTD, is a target of miR-29, a miRNA that is down-regulated in DFTD (Figs. 1 and 2 and fig. S3) (23). The expression of transcription factor ZEB2 is negatively correlated with low expression of the miR-200 family, which controls tumor invasiveness in a negative-feedback loop (24)(Figs. 1 and 2 and fig. S3). Furthermore, this work suggests that commonly mutated Schwann cell cancer genes, such as NF1, are interesting candidates for further analysis in DFTD (25). DFTD is believed to be of neuroendocrine origin on the basis of the expression of the vimentin, S100, neuron-specific enolase, chromogranin A, and synaptophysin markers (13). As Schwann cells and neuroendocrine cells are both derived from the neural crest and overlap in gene expression, it is possible that DFTD has elements of both tissue types. A Schwann cell origin for DFTD contrasts with that of the canine clonally transmissible cancer, which has been proposed to be of histiocytic origin (26, 27). It will be of interest to define common and unique features of these two cancers and to determine how the histogenesis of transmissible cancers may influence their occurrence, evolution, and biology. Our catalog of genes provides a framework for this work, as well as for use in the effort to develop a DFTD preclinical test and vaccine.
References and Notes
1. C. E. Hawkins et al., Biol. Conserv. 131, 307 (2006). 2. S. Lachish, M. Jones, H. McCallum, J. Anim. Ecol. 76, 926 (2007).

23. 24. 25. 26.

27.

28.

R. Loh et al., Vet. Pathol. 43, 890 (2006). H. McCallum et al., EcoHealth 4, 318 (2007). A. M. Pearse, K. Swift, Nature 439, 549 (2006). H. V. Siddle et al., Proc. Natl. Acad. Sci. U.S.A. 104, 16221 (2007). H. L. Cooper, C. M. Mackay, W. G. Banfield, J. Natl. Cancer Inst. 33, 691 (1964). C. Murgia, J. K. Pritchard, S. Y. Kim, A. Fassati, R. A. Weiss, Cell 126, 477 (2006). C. A. Rebbeck, R. Thomas, M. Breen, A. M. Leroi, A. Burt, Evolution 63, 2340 (2009). H. McCallum, Trends Ecol. Evol. 23, 631 (2008). P. J. Canfield, A. A. Cunningham, J. Zoo Wildl. Med. 24, 158 (1993). L. A. Griner, J. Natl. Cancer Inst. 62, 589 (1979). R. Loh et al., Vet. Pathol. 43, 896 (2006). Materials and methods are available as supporting material on Science Online. M. E. Jones, D. Paetkau, E. Geffen, C. Moritz, Mol. Ecol. 13, 2197 (2004). R. Schickel, B. Boyerinas, S. M. Park, M. E. Peter, Oncogene 27, 5959 (2008). R. Ueda et al., Proc. Natl. Acad. Sci. U.S.A. 106, 10746 (2009). J. Svaren, D. Meijer, Glia 56, 1541 (2008). H. B. Clark, J. J. Minesky, D. Agrawal, H. C. Agrawal, Am. J. Pathol. 121, 96 (1985). L. B. Dahlin, Scand. J. Surg. 97, 310 (2008). R. Gold, J. J. Archelos, H. P. Hartung, Brain Pathol. 9, 343 (1999). D. Greco, G. H. Stabenfeldt, in Textbook of Veterinary Physiology, J. G. Cunningham, Ed. (Saunders, Philadelphia, 1997). J. D. Verrier et al., Glia 57, 1265 (2009). P. A. Gregory, C. P. Bracken, A. G. Bert, G. J. Goodall, Cell Cycle 7, 3112 (2008). S. L. Carroll, N. Ratner, Glia 56, 1590 (2008). E. Mozos, A. Mndez, J. C. Gmez-Villamandos, J. Martn De Las Mulas, J. Prez, Vet. Pathol. 33, 257 (1996). T. Marchal, L. Chabanne, C. Kaplanski, D. Rigal, J. P. Magnol, Vet. Immunol. Immunopathol. 57, 1 (1997). We thank C. Harmsen and E. Noonan (DPIPWE) and J. Smith (Fort Wayne Childrens Zoo) for assistance with sample collection. We thank K. Claudio Campos, T. Dickson, S. Peck, V. Parameswaran, J. Harris, L. Stimmler, E. Hatchwell, R. Sachidanandam, Z. Xuan, M. Rasmussen, and the students and instructors of the 2006 and 2007 Cold Spring Harbor Laboratory deep-sequencing courses. The 454 sequencing was made possible by a grant from Roche Applied Sciences. E.P.M. was supported by a Sir Keith Murdoch Fellowship from the American Australian Association and an Overseas Postdoctoral Biomedical Fellowship from the Australian National Health and Medical Council (NHMRC). A.T.P., M.B., and A.H. were supported by grants from the NHMRC. A.S. was funded by a fellowship from the Human Frontier Science Program Organization. G.J.H. was supported by grants from the National Institutes of Health. H.S.B. was funded by an Australian Research Council Linkage Grant (LP0562190). E.P.M., C.T., A.K., and G.M.W. were supported by Dr. Eric Guiler Tasmanian Devil Research Grants. Sequences generated by this study have been deposited in GEO, www.ncbi.nlm.nih.gov/geo/ accession GSE18352 (miRNA) and the Short Read Archive http://www.ncbi.nlm.nih.gov/sites/entrez?db=sra accession SRA009772 (cDNA).

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/84/DC1 Materials and Methods Figs. S1 to S9 Tables S1 to S11 References 14 August 2009; accepted 29 October 2009 10.1126/science.1180616

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

87

REPORTS

O-Mannosyl Phosphorylation of Alpha-Dystroglycan Is Required for Laminin Binding

Takako Yoshida-Moriguchi,1,2,3,4 Liping Yu,5 Stephanie H. Stalnaker,6 Sarah Davis,1,2,3,4 Stefan Kunz,7 Michael Madson,8 Michael B. A. Oldstone,9 Harry Schachter,10 Lance Wells,6 Kevin P. Campbell1,2,3,4* Alpha-dystroglycan (a-DG) is a cell-surface glycoprotein that acts as a receptor for both extracellular matrix proteins containing laminin-G domains and certain arenaviruses. Receptor binding is thought to be mediated by a posttranslational modification, and defective binding with laminin underlies a subclass of congenital muscular dystrophy. Using mass spectrometry and nuclear magnetic resonance (NMR)based structural analyses, we identified a phosphorylated O-mannosyl glycan on the mucin-like domain of recombinant a-DG, which was required for laminin binding. We demonstrated that patients with muscle-eye-brain disease and Fukuyama congenital muscular dystrophy, as well as mice with myodystrophy, commonly have defects in a postphosphoryl modification of this phosphorylated O-linked mannose, and that this modification is mediated by the like-acetylglucosaminyltransferase (LARGE) protein. These findings expand our understanding of the mechanisms that underlie congenital muscular dystrophy. iverse posttranslational modifications influence the structure and function of many proteins. Dystroglycan (DG) is a membrane protein that requires extensive posttranslational processing in order to function as an extracellular matrix receptor. It is composed of an extracellular a-DG subunit and a transmembrane b-DG subunit (1). a-DG serves as a receptor for extracellular matrix laminin G domaincontaining ligands such as laminin (1) and agrin (2) in both muscle and brain, and these interactions depend on an unidentified posttranslational a-DG modification. a-DG is also the cellular receptor for lymphocytic choriomeningitis virus (LCMV), Lassa fever virus (LFV), and clade C New World arenaviruses (3, 4). Although the binding sites for LCMV

Howard Hughes Medical Institute, University of Iowa Roy J. and Lucille A. Carver College of Medicine, 4283 Carver Biomedical Research Building, 285 Newton Road, Iowa City, IA 52242-1101, USA. 2Department of Molecular Physiology and Biophysics, University of Iowa Roy J. and Lucille A. Carver College of Medicine, 4283 Carver Biomedical Research Building, 285 Newton Road, Iowa City, IA 52242-1101, USA. 3 Department of Neurology, University of Iowa Roy J. and Lucille A. Carver College of Medicine, 4283 Carver Biomedical Research Building, 285 Newton Road, Iowa City, IA 522421101, USA. 4Department of Internal Medicine, University of Iowa Roy J. and Lucille A. Carver College of Medicine, 4283 Carver Biomedical Research Building, 285 Newton Road, Iowa City, IA 52242-1101, USA. 5Medical Nuclear Magnetic Resonance Facility, University of Iowa Roy J. and Lucille A. Carver College of Medicine, B291 Carver Biomedical Research Building, 285 Newton Road, Iowa City, IA 52242-1101, USA. 6 Complex Carbohydrate Research Center, University of Georgia, 315 Riverbend Road, Athens, GA 30602, USA. 7 Institute of Microbiology, University Hospital Center and University of Lausanne, Lausanne, Switzerland. 8Bio Logistics, 2416 North Shore Drive, Clear Lake, IA 50428, USA. 9The Scripps Research Institute, Department of Immunology and Microbial Science, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA. 10The Hospital for Sick Children, Toronto, Ontario M5G 1X8, Canada. *To whom correspondence should be addressed. E-mail: [email protected]

and LFVon a-DG have not yet been identified, they are thought to overlap with the modification recognized by laminin (5, 6). Glycosyltransferase-mediated glycosylation is one type of posttranslational modification with two main forms in mammals: N- and O-glycosylation, which are distinguished by how the oligosaccharide moiety links to the amino acid. Mutations in six known or putative glycosyltransferase genesprotein O-mannosyl transferase 1 (POMT1) (7), POMT2 (8), protein O-mannose beta-1,2-Nacetylglucosaminyltransferase 1 (POMGnT1) (9), fukutin (10), fukutin-related protein (FKRP) (11), and LARGE (12)have been identified in patients with congenital muscular dystrophy (CMD). These disorders affect the brain, eye, and skeletal muscle to different extents, the most severe being Walker-Warburg syndrome [WWS; Online Mendelian Inheritance in Man (OMIM) identification number (ID) 236670], with less severe phenotypes seen in muscle-eye-brain disease (MEB; OMIM ID 253280) and Fukuyama CMD (FCMD; OMIM ID 253800). In these diseases, the ability of a-DG to bind laminin is markedly reduced (13), suggesting that these (putative) glycosyltransferases participate in the posttranslational modification that enables a-DG to bind laminin. Whereas the molecular functions of LARGE, fukutin, and FKRP remain unclear, POMT1 and -2 (14) and POMGnT1 (9) are known to catalyze two steps in the biosynthesis of an O-mannosyl tetrasaccharide (NeuNAc-a-2,3-Gal-b-1,4-GlcNAc-b1,2-Man) that is found in high abundance on both brain and muscle a-DG (15, 16). However, this glycan itself is probably not the lamininbinding moiety, because glycosidase-mediated removal of the glycan does not reduce a-DG binding to laminin (17). To determine which posttranslational modification is necessary for the a-DG/laminin interaction, we processed wheat germ agglutininVOL 327 SCIENCE

enriched proteins (glycoproteins) from C57BL/6J (wild-type, WT) muscle using various enzymatic and chemical treatments. Treatment with cold aqueous hydrofluoric acid (HFaq), which specifically cleaves phosphoester linkages (18), resulted in the reduction of the a-DG relative molecular mass (Mr) from 150 to 70 kD, the loss of IIH6 immunoreactivity and laminin binding (Fig. 1A), and the loss of binding to LFVand LCMV (Fig. 1B). Because the Mr of N-glycosylated b-DG did not change (Fig. 1A), these effects were not caused by the degradation of either peptide or glycosyl linkages. A quantitative solid-phase assay revealed a 97% reduction in total highaffinity binding to laminin (Fig. 1C). HFaq treatment also abolished the laminin-receptor activity of a-DG in the heart, brain, and kidney (fig. S1A). We next tested whether N-glycan and/or the two O-glycans known to modify the laminin-binding form of a-DGCore1 O-glycan and the O-mannosyl tetrasaccharide (in either the sialylated or fucosylated form) (15, 16)are sensitive to HFaq treatment. Immunoblotting of WT muscle glycoproteins treated with several cocktails of glycosidases that degrade these three glycans showed that the glycosidase-mediated reduction in a-DG glycosylation was impervious to HFaq treatment (Fig. 1D). A similar experiment using muscle glycoproteins from the CMDmodel mouse LARGE myd, in which a mutation in LARGE prevents the a-DG modification necessary for laminin-binding (19), revealed that HFaq treatment did not significantly reduce the Mr of a-DG (Fig. 1D). Thus, HFaq specifically degrades the laminin-binding moiety on a-DG. Further, functional modification of a-DG appeared to involve an internal phosphoryl linkage rather than a monoester-linked phosphate, because digesting WT muscle glycoproteins with alkaline phosphatase did not reduce the lamininbinding ability (fig. S1B). To verify that a-DG is phosphorylated, we labeled human embryonic kidney (HEK293) cells expressing Fc-tagged a-DG recombinants (Fig. 2A) that are secreted into the medium with [32P]-orthophosphate. Phosphorimaging showed that secreted DGFc4, which contains only the mucin-like region of a-DG (20), was phosphorylated (Fig. 2B). Hydrolysis of [32P]-DGFc4 under conditions that are conducive to the dissolution of polypeptide and phosphoester linkages to carbohydrates, but not to linkages to amino acids (21), generated inorganic phosphate but not phosphoamino acids (fig. S2), suggesting that phosphorylation does not occur directly on the peptide. To test whether the phosphorylation depends on glycosylation, we expressed DGFc5 in [32P]-orthophosphatelabeled human cells derived from POMT1-mutated WWS, POMGnT1mutated MEB, fukutin-mutated FCMD, and control cells, as well as in fibroblasts from Largemyd and WT mice (Fig. 2C). All except the POMT1mutated WWS cells secreted [32P]-phosphorylated DGFc5 into the medium, strongly suggesting that phosphorylation occurs on the O-linked mannose

88

1 JANUARY 2010

www.sciencemag.org

REPORTS
Fig. 1. Chemical dephosphorylation by HFaq treatment abolishes laminin and virus binding to a-DG in tissues from WT mice. (A and B) Treated glycoproteins prepared from WT muscle were subjected to the following: (A) immunoblotting with antibodies against the a-DG core protein (CORE) or the laminin-binding form a-DG epitope (IIH6) and b-DG, or to laminin overlay assay; (B) virus overlay assays with g-inactivated LFV or LCMV cl-13. (C) WT muscle glycoproteins with and without HFaq treatment were subjected to a solid-phase laminin-binding assay (n = 3). Open circles, treated; solid circles, untreated. Error bars indicate SD. (D) Muscle glycoproteins from WT and Largemyd (Myd) mice were digested with cocktails of glycosidases that degrade sialylated and/or fucosylated N-glycan, Core 1 O-glycan, and O-mannosyl glycan, before (first four lanes) and after (last four lanes) HFaq treatment. The products were subjected to either immunoblotting with CORE antibody or laminin overlay assay.

Fig. 2. The mucin-like domain of a-DG is phosphorylated in an O-mannosylation dependent manner. (A) Structures of recombinant a-DG constructs used in the study. (B and C) [32P]-orthophosphate labeling of (B) Fc-Ctrl or DGFc4expressing HEK293 cells and (C) DGFc5-expressing cultured cells from CMD patients (WWS, MEB, or FCMD) and control humans, and from WT and Largemyd (Myd) mice. Fc-tagged recombinant a-DG was isolated from the www.sciencemag.org SCIENCE

culture medium with protein-A agarose, separated by SDSpolyacrylamide gel electrophoresis, stained with Coomassie brilliant blue (CBB), and analyzed by phosphorimaging ([32P]). Phosphorylation of a-DG required prior Omannosylation. Asterisks indicate contaminating proteins derived from fetal bovine serum. (D) IMAC-binding assay testing glycoproteins from WT and Largemyd mice, and from FCMD, MEB, and control human muscle (SkM). VOL 327 1 JANUARY 2010

89

REPORTS
of a-DG. By measuring inorganic phosphate after acid hydrolysis, we confirmed that native a-DG purified from rabbit skeletal muscle is phosphorylated at 4.7 mol of phosphate per mole of protein (SD = 0.22, n = 3 trials). To assess whether CMD cells that synthesize phosphorylated a-DG can further modify the phosphate residue, we immunoprecipitated glycoproteins from mouse Largemyd and WT muscle, as well as from human POMGnT1-mutated MEB, fukutin-mutated FCMD, and control muscle, by using immobilized metal affinity chromatography (IMAC) beads that bind to monoester-linked, but not diester-linked, phosphorylated compounds. Only fukutin-mutated FCMD and Largemyd muscle a-DG were captured by the beads, revealing that the phosphate residue does not undergo further modification in these CMD cells (Fig. 2D). This finding suggests that fukutin and LARGE participate in a common pathway to assemble the laminin-binding moiety onto the phosphorylated O-linked mannose. This speculation is compatible with the fact that a-DG prepared from Largemyd Fig. 3. NMR analysis of phosphorylated O-glycan on HEK293-produced DGFc4. (A) HMQC spectrum where the assigned cross peaks are labeled with a letter for the subunit designated in (E) and a number for the position on that subunit. The folded cross peaks are indicated in blue, and the cross peaks derived from sample impurities are marked by asterisks. (B) TOCSY-HSQC spectrum obtained using a selective excitation pulse at the subunit CH1 proton and a selective TOCSY mixing time of 113 ms. (C) HMBC (red) and HMQC (black and blue) spectra for the assignment of interglycoside linkages. (D) 31P/1H COSY spectrum. (E) Structure of the O-glycan, with the sugar subunits labeled A to C. ppm, parts per million. muscle that was rescued by adenovirus-mediated expression of LARGE regains laminin-receptor activity and concomitantly loses its affinity for IMAC beads (fig. S3). In the case of POMGnT1mutated MEB patient muscle, several forms of aDG were observed; the majority of these were captured by the beads, although a certain amount of a-DG with laminin-binding activity was detected in the void fraction (Fig. 2D). This finding suggests that a defect in POMGnT1 partially inhibits modification on the phosphoryl branch chain of the O-mannosyl glycan on a-DG. DGFc4 that was produced by HEK293 cells bound to IMAC beads (fig. S4A) and gained laminin-binding activity when it was coexpressed with LARGE (fig. S4B); such a gain in activity has also been observed in FCMD, MEB, and Largemyd cells, both in this study and elsewhere (22). To determine the structure of the phosphorylated O-mannosyl glycan that was necessary to assemble the laminin-binding moiety, we prepared O-glycans from HEK293-expressed DGFc4 by reductive b elimination, and we isolated the phosphorylated O-glycan using IMAC beads. Linear trap quadrupole (LTQ) mass spectrometrybased analyses detected prominent ions at mass-to-charge ratios m/z = 667 ([M-H]) and m/z = 333 ([M-2H]2) that are assigned as a phosphorylated trisaccharide composed of HexNAc2Hexitol1 (fig. S5), and analysis by highperformance anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD) revealed the compositional sugars to be GlcNAc, GalNAc, and mannitol (fig. S6). Homo- and heteronuclear NMR techniques were used to assign the 13C/1H heteronuclear multiple quantum coherence (HMQC) spectrum of the reduced O-glycan (Fig. 3A). The GlcNAc (subunit B) was assigned using double quantum-filtered correlation spectroscopy (DQF-COSY) and total correlation spectroscopy (TOCSY) spectra with a series of mixing times (fig. S7). The GalNAc (subunit C) was partially assigned based on a selective TOCSYHSQC spectrum (Fig. 3B). The GalNAc (subunit C) is linked via a b1-3 linkage to the C3 position on GlcNAc (subunit B), which is in turn con-

90

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

REPORTS

Fig. 4. Mapping of phosphorylated trisaccharide on HEK293-produced DGFc4 and characterization of mannosyl phosphorylation. (A) Collision-induced dissociation (CID)MS/MS spectra from 780 to 1320 m/z (upper panel) and 375 to 2000 m/z (lower panel), revealing a neutral loss pattern (upper panel) and peptide-derived b and y ions (lower panel) of the selected precursor ions at m/z = 1318.63. The full Fourier transform mass spectrum is shown in fig. S8. Squares, HexNAc; circles, hexose. (B) Peptide and monosaccharide unit identification based on fragmentation of the phosphorylated glycopeptide. (C) Glycoproteins prepared from cell lysates of fibroblasts derived from mucolipidosis II patients were subjected to immunoblotting with CORE antibody and laminin overlay assay. nected via a b1-4 linkage to the C4 position on mannitol (subunit A), as evidenced by the observed heteronuclear multiple-bond correlation (HMBC) cross peaks CH1/BC3 and BH1/AC4 (Fig. 3C). The phosphate group is attached to the C6 position of the mannitol (subunit A) as determined from the cross peaks 31P/AH6 and 31P/A6H detected in the 31P/1H COSY spectrum (Fig. 3D). The complete NMR resonance assignments of the reduced O-glycan and its inter-residue correlations detected in the nuclear Overhauser effect spectroscopy (NOESY) and HMBC spectra are summarized in table S1, and the determined structure is shown in Fig. 3E. To verify that this phosphorylated trisaccharide modifies the mucin-like domain of DGFc4, we enriched the trypsinized peptides using Wisteria floribunda agglutinin-lectin and analyzed the GalNAc-terminated peptides by liquid chromatographymass spectrometry (LC-MS)/MS. MS/MS fragmentation patterns at m/z = 1318.63 (Fig. 4A), 1420.17 (fig. S9), and 1501.19 (fig. S10) identified a peptide (amino acids 374 to 389 of a-DG; GenBank ID CAA45732) bearing these modifications: the phosphorylated trisaccharide in conjunction with Hex-HexNAc-Hex, HexNAcHex, or Hex. The presence of nonphosphorylated mannose-initiated structures on y3, y4, and y10 ions revealed that Thr379 is modified by the phosphorylated trisaccharide in all cases (Fig. 4, A and B, and figs. S9 and S10). Additional studies showed that a-DG is phosphorylated within the Golgi complex (fig. S11) and that this phosphorylation occurs independently from the mannose-6-phosphate synthetic pathway that is required for lysosomal protein modification (23); fibroblasts derived from patients with mucolipidosis II (OMIM ID 252500), which have a defect in GlcNAc-1-phosphotransferase, can synthesize the laminin-binding form of a-DG (Fig. 4C). We demonstrated that MEB, FCMD, and Largemyd cells, which have genetically distinct abnormalities, show a similar defect in postphosphoryl modification on the O-mannosyl glycan (fig. S12). These convergent mechanisms to pathology offer an explanation for previous reports that forced expression of LARGE can circumvent defects in a-DG modification in these CMD cells (22). We speculate that LARGE, a putative glycosyltransferase with catalytic domains sharing homology with b1,3-N-acetylglucosaminyltransferase and bacterial glycosyltransferase (19) participates in postphosphoryl glycosylation, because the forced expression increases the affinity of the cell surface for both the IIH6 antibody (fig. S13A) and the Vicia villosa lectin (fig. S13B). To our knowledge, we provide the first evidence that a vertebrate non glycosylphosphatidylinositol-anchored glycoprotein is modified by a phosphodiester linkage. Glycoproteins in the cell walls of yeasts and fungi bear phosphodiester-linked glycans that are generated by a process involving phosphorylation on the C6 hydroxyl of mannose (24). a-DG, which is well conserved as an epithelial cell-surface protein in species ranging from lower vertebrates to mammals, is likewise modified by this ancient type of SCIENCE VOL 327 glycosylation. A recent study has shown that the most severe form of CMDWWSis a genetically heterogeneous disease. Moreover, only 40% of WWS cases are explained by mutations in known CMD-causative genes (25). Thus, a defect in the phosphorylation of an O-linked mannose may be responsible for severe CMD, indicating that the discovery of mutations in new genes responsible for WWS may not be far off.
References and Notes
1. O. Ibraghimov-Beskrovnaya et al., Nature 355, 696 (1992). 2. S. H. Gee, F. Montanaro, M. H. Lindenbaum, S. Carbonetto, Cell 77, 675 (1994). 3. W. Cao et al., Science 282, 2079 (1998). 4. C. F. Spiropoulou, S. Kunz, P. E. Rollin, K. P. Campbell, M. B. Oldstone, J. Virol. 76, 5140 (2002). 5. S. Kunz, N. Sevilla, D. B. McGavern, K. P. Campbell, M. B. Oldstone, J. Cell Biol. 155, 301 (2001). 6. S. Kunz, J. M. Rojek, M. Perez, C. F. Spiropoulou, M. B. Oldstone, J. Virol. 79, 5979 (2005). 7. D. Beltran-Valero de Bernabe et al., Am. J. Hum. Genet. 71, 1033 (2002). 8. J. van Reeuwijk et al., J. Med. Genet. 42, 907 (2005). 9. A. Yoshida et al., Dev. Cell 1, 717 (2001). 10. K. Kobayashi et al., Nature 394, 388 (1998). 11. M. Brockington et al., Am. J. Hum. Genet. 69, 1198 (2001). 12. C. Longman et al., Hum. Mol. Genet. 12, 2853 (2003). 13. D. E. Michele et al., Nature 418, 417 (2002). 14. H. Manya et al., Proc. Natl. Acad. Sci. U.S.A. 101, 500 (2004). 15. T. Sasaki et al., Biochim. Biophys. Acta 1425, 599 (1998). 16. A. Chiba et al., J. Biol. Chem. 272, 2156 (1997). 17. A. C. Combs, J. M. Ervasti, Biochem. J. 390, 303 (2005). 18. T. Ilg et al., J. Biol. Chem. 271, 21583 (1996). 19. P. K. Grewal, P. J. Holzfeind, R. E. Bittner, J. E. Hewitt, Nat. Genet. 28, 151 (2001). 20. M. Kanagawa et al., Cell 117, 953 (2004).

www.sciencemag.org

1 JANUARY 2010

91

REPORTS
21. T. Ilg et al., J. Biol. Chem. 269, 24073 (1994). 22. R. Barresi et al., Nat. Med. 10, 696 (2004). 23. N. M. Dahms, P. Lobel, S. Kornfeld, J. Biol. Chem. 264, 12115 (1989). 24. J. E. Gander, Annu. Rev. Microbiol. 28, 103 (1974). 25. M. C. Manzini et al., Hum. Mutat. 29, E231 (2008). 26. This study was supported in part by a Paul D. Wellstone Muscular Dystrophy Cooperative Research Center grant (1U54NS053672 to K.P.C.). K.P.C. is an Investigator of the Howard Hughes Medical Institute. Additional acknowledgments and funding sources are provided in the supporting online material (SOM) text. SOM text Figs. S1 to S13 Table S1 References

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/88/DC1 Materials and Methods 12 August 2009; accepted 20 October 2009 10.1126/science.1180512

The Rate and Molecular Spectrum of Spontaneous Mutations in Arabidopsis thaliana

Stephan Ossowski,1* Korbinian Schneeberger,1* Jos Ignacio Lucas-Lled,2* Norman Warthmann,1 Richard M. Clark,3 Ruth G. Shaw,4 Detlef Weigel,1 Michael Lynch2 To take complete advantage of information on within-species polymorphism and divergence from close relatives, one needs to know the rate and the molecular spectrum of spontaneous mutations. To this end, we have searched for de novo spontaneous mutations in the complete nuclear genomes of five Arabidopsis thaliana mutation accumulation lines that had been maintained by single-seed descent for 30 generations. We identified and validated 99 base substitutions and 17 small and large insertions and deletions. Our results imply a spontaneous mutation rate of 7 109 base substitutions per site per generation, the majority of which are G:CA:T transitions. We explain this very biased spectrum of base substitution mutations as a result of two main processes: deamination of methylated cytosines and ultraviolet lightinduced mutagenesis. ost of what we know about molecular evolution comes from the comparison of biological sequences that have survived many cycles of natural selection. In order to infer the properties of the original source of variation and to detect the signature of natural selection from such data sets, we need to assume that variants affecting certain types of sites, such as the last base of fourfold redundant codons or pseudogenes, are not subject to natural selection. This pervasive assumption is very rarely tested and difficult to avoid, because of the slow pace of spontaneous mutagenesis. However, with the advent of high-throughput sequencing technologies, some estimates of the rate of spontaneous mutations have begun to appear (13). Here, we report a direct estimate of the spontaneous base substitution rate in Arabidopsis thaliana, a plant species with extensive DNA methylation. As a result, we reduce the uncertainty associated with key aspects of the evolutionary history of this species, including the time since divergence from A. lyrata and the effect of methylation on the probability of mutation. We sequenced the genomes of five individuals derived by 30 generations of single-

Department of Molecular Biology, Max Planck Institute for Developmental Biology, 72076 Tbingen, Germany. 2Department of Biology, Indiana University, Bloomington, IN 47405, USA. 3 Department of Biology, University of Utah, Salt Lake City, UT 84112, USA. 4Department of Ecology, Evolution, and Behavior, University of Minnesota, St. Paul, MN 55108, USA. *These authors contributed equally to this work. To whom correspondence should be addressed. E-mail: [email protected] (J.I.L.-L.); [email protected] (D.W.)

seed descent from the reference strain Col-0 (4), for which a high-quality genome was published in 2000 (5). We used the Illumina (Illumina, San Diego, CA) Genome Analyzer platform to obtain a depth of sequence coverage of between 23 and 31 in each mutation accumulation (MA) line. Sequencing reads between 36 and 43 base pairs (bp) in length were aligned to the reference genome, from which over 1000 errors had been removed in a previous study (6). We identified single-base substitutions using two complementary methods: a consensus approach and the single-nucleotide polymorphism (SNP) caller function of SHORE (http://1001genomes.org/ downloads/shore.html) (7). In the consensus approach, base substitutions were called if one of the MA lines differed from all others. We estimated a frequency of sequencing errors of ~0.3% per site per read (8). We assumed a binomial distribution of errors to derive the probabilities of false positives and false negatives, and we corrected our estimates accordingly (9). Because sequencing and mapping errors are not randomly distributed among sites (3), we used strict quality filters to exclude from analysis sites suspected to have higher error rates (8). Between 93 and 95 million sites out of the 120 millionbp reference genome matched the quality requirements in each line. Across all five lines, 85 single-base substitutions were called by this method, 83 of which were confirmed by Sanger sequencing. In the other two sites, two or three lines had a nonreference base, whereas the rest matched the reference, and we interpret this to be a result of differential fixation of the two alleles present in ancestrally VOL 327 SCIENCE

heterozygous sites rather than as parallel mutations, although the latter cannot be ruled out. In addition to this conservative approach, we used SHORE to detect single-base substitutions, short insertions and deletions (indels) of up to 3 bp, and long deletions. The algorithms implemented in SHORE are more sensitive (8), and between 98.8 and 100.9 million sites in each line had sufficient read information for calling either a mutation or the reference base. We detected 99 single-base substitutions (98 of which were confirmed by Sanger sequencing, and 1 was rejected because the reference base was revealed), 9 short deletions (8 confirmed, 1 rejected), 5 short insertions of 1 bp (all confirmed), and 8 long deletions covering 11 to over 5000 bp (4 confirmed, 2 ambiguous, and 2 rejected). A 2-bp deletion was shown to be present in two lines, suggesting that it was heterozygous in the ancestral line. The chromosomal positions of all validated mutations are shown in Fig. 1. Both false positives and false negatives are expected to be absent from the final set of simple base-substitution mutations (8). Fifteen sites where all MA lines had a common composition, but different from the reference, including 13 single-base substitutions and two deletions of 1 and 2 bp, were interpreted as fixed mutations in the ancestral line. We estimated the overall mutation rate to be 5.9 109 T 0.6 109 base substitutions per site per generation according to the consensus approach and 6.5 109 T 0.7 109 according to SHORE. In addition, joint maximum-likelihood estimates of the overall mutation rate and the sequencing error frequency were obtained following a recently developed method (9). With this approach, a slightly higher mutation rate of 7.1 109 T 0.7 109 at a slightly lower error frequency of 0.2% was estimated. Mutations were evenly distributed among MA lines (Table 1). Within chromosomes, a significantly higher base-substitution mutation rate for intergenic regions was observed closer to the centromere (within 3.0 106 bp, for example) than farther away (Fishers exact test, P value = 0.01, for nonmethylated sites). The estimated rates of 1- to 3-bp deletions and insertions are 0.6 109 T 0.2 109 and 0.3 109 T 0.1 109 per site per generation, respectively. Out of the 13 short indels that we observed, 6 were found in complex sequences, corresponding to a mutation rate of 4.0 1010 T 1.6 1010 indels per site per generation, or about 0.05 T 0.02 indels per haploid genome per generation, excluding homopolymers and microsatellites. This estimate should be considered a

92

1 JANUARY 2010

www.sciencemag.org

REPORTS
lower bound, given the unknown level of false negatives among indels. The short-read sequencing approach is less well-suited for the analysis of dinucleotide repeats, because the most frequent class of dipolymers (AT/TA) has low read coverage, most likely resulting from the sequencing library construction protocol (10) (fig. S1). Deletions larger than 3 bp occurred at a frequency of 0.5 109 T 0.2 109 per site per generation and removed on average 800 T 1900 bp per event. The preceding estimates, together with the recently published rate of mutation at dinucleotide microsatellites (11), give an almost complete view of the spectrum of spontaneous mutations in A. thaliana (Table 2). Our estimate of the mutation rate is close to the lower bound of an indirect estimate based on the divergence between monocots and dicots (12). If we assume that only nonsynonymous mutations and indels affecting coding regions are likely to affect fitness, the diploid genomic rate of mutations affecting fitness would be 0.2 T 0.1 per generation, which is not significantly different from previous estimates based on the evaluation of fitness components in MA lines of A. thaliana (13, 14). Alternatively, the proportion of deleterious mutations among mutations in coding regions can be estimated from sequence comparisons between A. thaliana and A. lyrata (8, 15). Using that information, the estimated genomic deleterious mutation rate is 0.14 T 0.04 per generation. We did not detect any difference between the unpolarized (8) spectrum of base-substitution mutations and the genome-wide spectrum of polymorphisms at synonymous sites surveyed in natural populations by two independent studies (6, 16) (fig. S2; Fishers exact test, P = 0.4 and P = 0.7, respectively). Transitions were 2.4 times more frequent than transversions, and G:CA:T transitions, most of which are silent at the third codon position, were by far the most frequent type of mutations (Fig. 2). Under the observed mutational spectrum, the base-composition equilibrium achieved only by mutation would be 85% A+T, which is far from the current 68% observed in intergenic and intronic regions and from the 65% in fourfold redundant coding sites across the A. thaliana genome. Whether selection is preventing a further increase in A+T content, or whether the genome is still evolving toward a higher A+T content, is not known. Spontaneous deamination of methylated cytosine, which leads to thymine substitution (1719), is thought to be a major source of mutations. Thus, we exploited a single baseresolution methylation map of the A. thaliana genome (20) to test whether cytosine methylation can account for the overabundance of G:CA:T transitions. G:C sites where the cytosine has been Table 1. Number of mutations inferred by SHORE and validated by Sanger sequencing, their distributions among functional classes in each MA line, and totals.
MA line 29 49 59 69 119 Total Intergenic 8 11 12 Mobile elements 3 3 4 UTR 2 0 0 Intron 3 0 2 Synonymous 0 0 3 Nonsynonymous 1 4 2 All mutations 17 18 23 Mutation rate 5.7 6.0 7.6 (109) Standard error 1.4 1.4 1.6 (109) 10 4 0 1 1 2 18 5.9 1.4 14 3 2 1 0 2 22 7.4 1.6 54 17 4 7 4 11 98 6.5 0.7

Table 2. Haploid mutation rates per genome per generation and standard errors (SEM). Estimates for indels in dinucleotide repeats comes from Marriage and colleagues (11) and are the product of their per-locus per-generation mutation rate and the number of perfect repeats in the genome.
Mutation type A:TG:C C:GT:A A:TT:A C:GA:T A:TC:G C:GG:C Complex sequence AT repeats AG repeats AC repeats Large deletions (>3 bp) Rate 0.09 0.41 0.04 0.06 0.04 0.05 0.05 19.12 2.40 0.13 0.03 SEM 0.03 0.06 0.02 0.02 0.02 0.02 0.02 1.77 0.55 0.09 0.02

Fig. 1. Distribution of spontaneous mutations across chromosomes. Labels indicate the type of mutation and colors their functional context or predicted effect. Short insertions and deletions are shown by the letters representing the affected bases preceded by a plus or a minus sign, respectively. Long deletions are depicted by a minus sign and the number of deleted base pairs in parentheses. An asterisk next to a C or a G means that the cytosine of the mutant base pair is known to be methylated (20). The definitions for colors are as follows: red, intergenic region; yellow, intron; dark blue, nonsynonymous substitution, shift of reading frame for short indels, or gene deletion for large deletions; green, synonymous substitution; purple, UTR; and light blue, transposable element. www.sciencemag.org SCIENCE VOL 327

1 JANUARY 2010

93

REPORTS
Fig. 2. Conditional mutation rates per A:T or G:C site per generation. Complementary mutations, such as AC and TG, are pooled. Error bars indicate standard errors of the mean. The overall mutation rate, which is the average of the total mutation rates at A:T and G:C sites, and its standard error in gray are shown in the background. Only estimates from the consensus method are shown.
3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. P. D. Keightley et al., Genome Res. 19, 1195 (2009). R. G. Shaw, D. L. Byers, E. Darmo, Genetics 155, 369 (2000). Arabidopsis Genome Initiative, Nature 408, 796 (2000). S. Ossowski et al., Genome Res. 18, 2024 (2008). K. Schneeberger et al., Genome Biol. 10, R98 (2009). Materials and methods are available as supporting material on Science Online. M. Lynch, Mol. Biol. Evol. 25, 2409 (2008). M. A. Quail et al., Nat. Methods 5, 1005 (2008). T. N. Marriage et al., Heredity 103, 310 (2009). K. H. Wolfe, W.-H. Li, P. M. Sharp, Proc. Natl. Acad. Sci. U.S.A. 84, 9054 (1987). S. T. Schultz, M. Lynch, J. H. Willis, Proc. Natl. Acad. Sci. U.S.A. 96, 11393 (1999). F. H. Shaw, C. J. Geyer, R. G. Shaw, Evolution 56, 453 (2002). S. I. Wright, B. Lauga, D. Charlesworth, Mol. Biol. Evol. 19, 1407 (2002). R. M. Clark et al., Science 317, 338 (2007). T. Lindahl, B. Nyberg, Biochemistry 13, 3405 (1974). C. Coulondre, J. H. Miller, P. J. Farabaugh, W. Gilbert, Nature 274, 775 (1978). B. K. Duncan, J. H. Miller, Nature 287, 560 (1980). S. J. Cokus et al., Nature 452, 215 (2008). R. Lister et al., Cell 133, 523 (2008). E. C. Friedberg et al., DNA Repair and Mutagenesis [ASM (American Society for Microbiology) Press, Washington, DC, 2006]. A. Kawabe, A. Forrest, S. I. Wright, D. Charlesworth, Genetics 179, 985 (2008). We thank C. Lanz for generating the Illumina data, S. E. Jacobsen for providing the methylation data, and P. Tiffin for valuable comments. Funded by the Deutsche Forschungsgemeinschaft (DFG) (ERA-PG ARelatives), a Gottfried Wilhelm Leibniz Award (DFG), the Max Planck Society (D.W.), NIH grant GM36827 to M. L. and W. Kelly Thomas, Pioneer Hi-Bred International to E. Darmo, and NSF grants DEB 9629457 and 9981891 to R.G.S.

reported to be at least partially methylated had a higher probability of mutation to A:T than nonmethylated sites (Fishers exact test, P = 3.2 107). However, 91% of G:C sites in A. thaliana were not reported to be methylated, and they too had a higher rate of transition (but not transversion) than A:T sites (Fishers exact test, P = 1.2 108). G:C sites in CpG contexts are known to be more frequently methylated (20, 21). However, transitions at G:C sites not known to be methylated do not happen in CpG contexts more often than expected by chance (Fishers exact test, P = 0.6). This suggests that factors in addition to methylation contribute to the high rate of transitions at G:C sites. In both prokaryotes and eukaryotes, most of the mutations caused by ultraviolet (UV) light are G: CA:T transitions at sites where the C is adjacent to another C or to a T (dipyrimidine sites) (22). Among the 33 observed transition mutations at nonmethylated G:C sites, 31 were in dipyrimidine contexts, which is more than expected by chance at the P = 0.02 level (Fishers exact test). Thus, we conclude that the increased rate of transitions at G:C sites, relative to A:T sites, can be largely explained by the combined effect of UV-induced mutagenesis and deamination of methylated cytosines. This implies that the mutation rate in nature could be higher than that reported here, because UVradiation during the MA experiment was probably lower than in natural conditions. We used the Arabidopsis Information Resource (TAIR) 8 annotation (www.arabidopsis.org) to group all analyzed sites into the functional classes: intergenic, intronic, untranslated region (UTR), coding, pseudogene, mobile element, and noncoding. There is no deficit of nonsynonymous mutations (G test, P = 0.4), supporting the notion that the mutation rate we observed is not affected by selection. We did, however, observe an excess of intergenic mutations, relative to mutations in coding regions, introns, and UTRs (fig. S3). These differences were still significant after taking into account the effects of base composition and methylation (G test, P = 0.00025). To test whether the lack of mutations in genic regions was due to undetected levels of selection, we compared the intergenic

mutation rate with the rate at synonymous sites and introns, which are less likely to be under strong selection, and we still detected a significant deficit of mutations in the latter (Fishers exact test, P = 0.001, for nonmethylated sites). We attribute the deficit of genic mutations to our observation of a higher mutation rate in pericentromeric regions (see above), where gene density is lower (5), although transcription-coupled DNA repair could also contribute to the pattern. Lastly, the finding of a higher mutation rate in pericentromeric regions provides an explanation of the Arabidopsis-specific pattern of higher polymorphism levels near the centromeres (16, 23), although the underlying mechanism of such a mutational bias remains to be explained.
References and Notes
1. M. Lynch et al., Proc. Natl. Acad. Sci. U.S.A. 105, 9272 (2008). 2. D. R. Denver et al., Proc. Natl. Acad. Sci. U.S.A. 106, 16310 (2009).

23. 24.

Supporting Online Material

www.sciencemag.org/cgi/content/full/327/5961/92/DC1 Materials and Methods SOM Text Figs. S1 to S8 Tables S1 to S3 References 17 August 2009; accepted 27 October 2009 10.1126/science.1180677

Targeted 3 Processing of Antisense Transcripts Triggers Arabidopsis FLC Chromatin Silencing

Fuquan Liu,* Sebastian Marquardt,* Clare Lister, Szymon Swiezewski, Caroline Dean Noncoding RNA is emerging as an important regulator of gene expression in many organisms. We are characterizing RNA-mediated chromatin silencing of the Arabidopsis major floral repressor gene, FLC. Through suppressor mutagenesis, we identify a requirement for CstF64 and CstF77, two conserved RNA 3-endprocessing factors, in FLC silencing. However, FLC sense transcript 3 processing is not affected in the mutants. Instead, CstF64 and CstF77 are required for 3 processing of FLC antisense transcripts. A specific RNA-binding protein directs their activity to a proximal antisense polyadenylation site. This targeted processing triggers localized histone demethylase activity and results in reduced FLC sense transcription. Targeted 3 processing of antisense transcripts may be a common mechanism triggering transcriptional silencing of the corresponding sense gene. xtensive noncoding RNA has been found in many organisms (1, 2). The degree and mechanism of processing of these transcripts is unclear, as studies of RNA processing

have so far focused on coding (messenger) RNA transcripts in mammalian and yeast cells. The large protein complex mediating 3-end processing and polyadenylation has recently been de-

94

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

REPORTS
scribed (3, 4); however, its role in processing noncoding transcripts is unknown. Analysis of 3 processing of unstable noncoding RNAs has identified independent termination pathways with different components (5). We have been studying the function of RNA processing in chromatin regulation through analysis of the gene encoding the major Arabidopsis floral repressor FLOWERING LOCUS C (FLC) (6). The autonomous floral promotion pathway results in FLC transcriptional silencing through the activities of two RNA-binding proteins, FCA and FPA; a member of a cleavage and polyadenylation specificity factor (CPSF) RNA 3processing complex, FY, which interacts directly with FCA (7, 8); and FLD, a dimethylated histone H3 at lysine 4 (H3K4me2) demethylase and homolog of human LSD1 (9). To better understand how RNA 3-processing links to histone modification to result in FLC silencing, we sought to identify all the components necessary for FCA-mediated FLC repression through suppressor mutagenesis. We generated an Arabidopsis line sensitized to FCA action expressing 35S::FCAga transgene overexpressing FCA, FRIGIDAa strong activator of FLC expression and FLC::LUC, so FLC expression could be monitored with a bioluminescence detection assay (10, 11). Mutations that suppressed the ability of FCA to repress FLC, together with an epistasis analysis of their interaction with fca mutants, enabled us to identify components required for FCA action (fig. S1). These mutations, named suppressors of overexpressed FCA (sof ) (9), identified additional alleles of known FCA pathway components FY and FLD (fig. S2) and new complementation groups sof2 and sof19 (Fig. 1A and fig. S2). Genetic mapping and sequencing revealed sof2 and sof19 carried mutations in CstF77 (At1g17760) and CstF64 (At1g71800), respectively (fig. S3, A to C). CstF77 and CstF64 are the Arabidopsis homologs of two of the three components of the CstF RNA 3-processing complex, conserved from yeast to plants and humans (12). CstF77 is a single-copy gene in the Arabidopsis genome. In sof2 (carrying the cstf77-1 allele), the splice acceptor site in intron 12 is mutated, which leads to an in-frame deletion of Tyr339 to Lys385 in exon 13 (fig. S3A). CstF77 is an essential gene in other organisms (13, 14), yet cstf77-1 plants are viable, but their flowering is delayed. cstf77-1 is likely to be a hypomorphic allele, because homozygous mutant progeny could not be identified from a mutation caused by a transferred DNA insertion into the 5 end of the gene (cstf 77-2) (fig. S3A). Further analysis suggested that cstf 77-2 causes female gametophytic lethality (fig. S4, A and B),
Department of Cell and Developmental Biology, John Innes Centre, Norwich NR4 7UH, UK. *These authors contributed equally to this work. Present address: Institute of Biochemistry and Biophysics, Polish Academy of Sciences, 02-106 Warsaw, Poland. To whom correspondence should be addressed. E-mail: [email protected]

consistent with an essential function for CstF77. The sof19 mutation introduces a premature stop codon in the only full-length homolog of CstF64 in the Arabidopsis genome (fig. S3A). This allele (cstf64-1) also reduces fertility of the plants. A second, strong allele in Columbia (Col) background (cstf64-2) (fig. S3A) leads to reduced organ size, pale leaves, and sterility (fig. S4, C to E). Therefore, CstF64 function is generally required in plant growth and development. In mammals, CstF77 interacts physically with CstF64 through a proline-rich region (15). This interaction was found to be conserved in plants and to be unaffected by the in-frame deletion in cstf 77-1 (fig. S4F). We further analyzed the involvement of CstF64 and CstF77 in FLC repression in genotypes not sensitized to FCA action. cstf64-1, cstf64-2, and cstf77-1 had elevated FLC mRNA levels and flowered later than their respective controls (Fig. 1, B and C, and fig. S5). We also undertook an epistasis analysis to determine whether the CstF components functioned in the same pathway with known FCA pathway components, namely FCA, FY, and FLD. cstf64-2 was not additive with fca-9, fy-2, or fld-4 mutations but was additive with fve-3 on the basis of analysis of both flowering time and FLC expression levels (Fig. 1C and fig. S6). FVE functions to repress FLC expression by histone deacetylation in an FCA-independent manner (16). In summary, our analysis demonstrates that CstF components do function in the same pathway as FCA in the repression of FLC. Because the genetic analysis established that CstF components function with the histone demethylase FLD to down-regulate FLC, we tested whether the increased FLC in cstf mutants was an effect of defective transcriptional repression. cstf64 and cstf77 mutants showed elevated FLC nascent transcript levels (Fig. 2B and fig. S7);

higher RNA polymerase II (Pol II) association at FLC (Fig. 2C); and higher levels of H3K4me3, a histone modification associated with active transcription (17) (Fig. 2D). The CstF complex is generally required for canonical mRNA 3-end formation (12), but its loss did not affect FLC sense RNA 3 processing; indeed, FLC sense levels accumulate and are functional as evidenced by the late flowering of cstf64 double mutants (Fig. 1C). Thus, CstF components mediate transcriptional repression of FLC levels but are unlikely to be required for FLC mRNA 3-end formation. FCA directly associates with FLC chromatin (9), so one explanation for our observations is that the substrates of CstF64, CstF77, FCA, and FY are other nascent transcripts from the FLC locus. We have previously described alternatively polyadenylated FLC antisense transcripts (9, 18). One polyadenylation site of the FLC antisense RNA coincides with the location of FCA on FLC chromatin, whereas a distal polyadenylation site overlaps with the FLC sense promoter (9). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis revealed increased levels of FLC antisense transcription in fca, fy, and cstf mutants (fig. S8A). Further analysis has also revealed additional complexity in the antisense transcripts, so we developed an assay to detect RNA 3 processing and polyadenylation at the specific polyadenylation sites shown in Fig. 3. qRT-PCR revealed that FLC antisense RNA 3 processing was reduced at both proximal and distal polyadenylation sites in cstf64 and cstf77 (Fig. 3B), but FLC sense RNA polyadenylation was not reduced (fig. S8B). Thus, 3 processing and polyadenylation of FLC antisense transcripts, but not FLC sense transcripts, appear sensitive to CstF complex activity. Rapid amplification of cDNA 3 ends (RACE) analysis revealed that the same polyadenylation sites are

Fig. 1. CstF64 and CstF77 regulate FLC levels. (A) Northern and flowering time analysis (total leaf number) of sof2 (cstf77-1 in FCA-sensitized background) and sof19 (cstf64-1 in FCA-sensitized background). Asterisk indicates FLC::LUC; APT is the loading control. (B) Northern analysis of FLC levels in cstf64 carrying no transgenes or cstf77 carrying the linked FRI transgene. b-TUB or APT is the loading control. (C) cstf64-2 is not additive with fca-9, fy-2, or fld-4 (all Col genotype). (A) and (C) graph values are means T SD (n = 20). SCIENCE VOL 327 1 JANUARY 2010

www.sciencemag.org

95

REPORTS
used in the fca mutant and wild type; however, qRT-PCR, supported by strand-specific Northern analysis, showed that the relative usage of each site differed. In fca and fy, proximal site usage was reduced, and in fca, distal site usage increased (Fig. 3C and fig. S8C). This is not a feature of general FLC de-repression, because proximal site usage was not affected in fve (Fig. 3C) (16). To investigate the generality of this finding, we analyzed the effects of FPA, a second RNA recognition motif (RRM) protein that functions in the autonomous floral promotion pathway. FPA also requires FLD to repress FLC (19), but it functions independently of FCA (19). As in fca, proximal site usage in the FLC antisense transcript was reduced, and distal site usage increased in fpa-7 (Fig. 3C). Because the two RRM proteins trigger FLC silencing independently, we predicted that disruption of FPA function would cause increased FLC expression even in the sensitized FCA background. This was confirmed through the identification of sof34 as a novel fpa mutation (fig. S9). Taken together, these results reveal that both RNA-binding proteins independently function to promote 3 processing at the proximal site in the FLC antisense transcript and that this activity triggers FLC sense strand transcriptional silencing. The position of the proximal 3-processing site on the antisense transcript coincides with the site where FCA associates with FLC chromatin (9). This suggests a model whereby FCA, interacting with a component of the CPSF complex, targets CstF-dependent 3 processing to the proximal site on FLC antisense transcripts (fig. S10). An indirect effect, via CstF regulation of an intermediary regulator, is possible, but unlikely, because of the direct association of FCA with FLC. Proximal FLC antisense transcript 3 processing is promoted by FCA and FY (Fig. 3), which is reminiscent of the FCA-FY interactiondependent 3-processing site choice in FCA negative-feedback regulation (7). FPA also enhances usage of this proximal 3-processing site. These activities then appear to converge to trigger FLD-dependent demethylation of H3K4me2 in the body of the gene, downstream of the proximal polyadenylation site. We speculate that this may involve termination-associated processes, perhaps cotranscriptional decay of the antisense RNA downstream of the cleavage site (20). Whatever the mechanism, down-regulation of both sense and antisense transcription is the net result. We suspect the activity of DICER-LIKE3, previously shown to function in this process (9), is required to couple the 3 processing to FLD histone demethylase function (21); however, RNA polymerases involved in silencing mediated by small interfering RNA (siRNA) (18) only weakly suppress FCA activity (fig. S11). A role for small RNAs in the targeting of the histone demethylase activity to the central part of the locus may lead to trans effects on homologous sequences in the genome, perhaps explaining the small effect of an fca mutation on an FLC transgene lacking the 3 region (22). The proteins involved in this silencing mechanismtwo RRM proteins, a factor associated with 3 processing components, and a conserved histone demethylasealso silence transposons and transgenes in Arabidopsis (20, 23). This mechanism may therefore play a more general role, rather than only regulating the floral repressor gene, FLC. This is consistent with the role for CPSF complex components in suppressing viral amplicon-induced gene silencing in Arabidopsis (24). This mechanism may also be very important in nonplant genomes. Genome-wide antisense transcripts, identified in

Fig. 2. cstf-64 and cstf-77 increase FLC transcription. (A) Schematic of FLC, vertical bars denote exons. Horizontal bars (A, B, C, and G) denote regions analyzed by qPCR in chromatin immunoprecipitation (ChIP). Horizontal bars (a and b) denote the intron-containing FLC nascent transcripts analyzed by qRT-PCR. (B) Nascent FLC transcript analysis of fragment a (containing intron 1) and fragment b (containing introns 2 and 3). (C) Pol II ChIP assay in the different FLC regions in sof2 (gray) and sof19 (white) normalized to parental line C2 (black). (D) H3K4me3 ChIP assay in FLC regions in sof2 (gray) and sof19 (white) normalized to parental line C2 (black). Error bars represent standard errors derived from at least two biological and two technical repeats.

Fig. 3. The 3 end processing of FLC antisense RNA. (A) In this FLC schematic, vertical bars denote exons; transcription start site of FLC sense RNA is indicated by an arrow. FLC sense (top) and antisense (bottom) transcripts with dashed lines indicating introns, arrows with symbol (A) are polyadenylation sites, and shading shows the region of multiple antisense RNA start sites. (B) The 3 processing of FLC antisense transcripts is reduced at both distal and proximal sites in sof2 and sof19. FLC antisense transcript levels are normalized to total FLC antisense transcript and given as fold change compared with the parental line C2. (C) The 3 processing of FLC antisense transcripts at the proximal (top) and the distal sites (bottom) in different mutants (all in the Col genotype without transgene). FLC antisense transcript levels are normalized to total FLC antisense transcript and given as fold change compared with the Col wild type. The raw qPCR data are given in table S1. In (B) and (C), averages and standard errors of three biological repeats with two technical repeats are shown.

96

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org

REPORTS
many organisms, function in chromatin silencing (2, 25, 26) and RNA 3-processing components mediate gene silencing in Caenorhabditis elegans (27). Therefore, the 3 processing of antisense transcripts may be a general mechanism triggering chromatin silencing in eukaryotes.
References and Notes
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. B. T. Wilhelm et al., Nature 453, 1239 (2008). C. P. Ponting, P. L. Oliver, W. Reik, Cell 136, 629 (2009). M. J. Moore, N. J. Proudfoot, Cell 136, 688 (2009). Y. Shi et al., Mol. Cell 33, 365 (2009). L. Vasiljeva, M. Kim, N. Terzi, L. M. Soares, S. Buratowski, Mol. Cell 29, 313 (2008). G. G. Simpson, Curr. Opin. Plant Biol. 7, 570 (2004). G. G. Simpson, P. P. Dijkwel, V. Quesada, I. Henderson, C. Dean, Cell 113, 777 (2003). D. Manzano et al., Proc. Natl. Acad. Sci. U.S.A. 106, 8772 (2009). F. Liu et al., Mol. Cell 28, 398 (2007). J. Mylne, T. Greb, C. Lister, C. Dean, Cold Spring Harb. Symp. Quant. Biol. 69, 457 (2004). 11. Materials and methods are available as supporting material on Science Online. 12. J. Zhao, L. Hyman, C. Moore, Microbiol. Mol. Biol. Rev. 63, 405 (1999). 13. M. Simonelig, K. Elliott, A. Mitchelson, K. O'Hare, Genetics 142, 1225 (1996). 14. L. Minvielle-Sebastia, B. Winsor, N. Bonneaud, F. Lacroute, Mol. Cell. Biol. 11, 3075 (1991). 15. Y. Bai et al., Mol. Cell 25, 863 (2007). 16. I. Ausin, C. Alonso-Blanco, J. A. Jarillo, L. Ruiz-Garcia, J. M. Martinez-Zapater, Nat. Genet. 36, 162 (2004). 17. B. Li, M. Carey, J. L. Workman, Cell 128, 707 (2007). 18. S. Swiezewski et al., Proc. Natl. Acad. Sci. U.S.A. 104, 3633 (2007). 19. I. Burle, C. Dean, PLoS One 3, e2733 (2008). 20. M. Kim et al., Nature 432, 517 (2004). 21. T. Iida, J.-i. Nakayama, D. Moazed, Mol. Cell 31, 178 (2008). 22. C. C. Sheldon, A. B. Conn, E. S. Dennis, W. J. Peacock, Plant Cell 14, 2527 (2002). 23. I. Baurle, L. Smith, D. C. Baulcombe, C. Dean, Science 318, 109 (2007). 24. A. J. Herr, A. Molnar, A. Jones, D. C. Baulcombe, Proc. Natl. Acad. Sci. U.S.A. 103, 14994 (2006). 25. J. Camblong, N. Iglesias, C. Fickentscher, G. Dieppois, F. Stutz, Cell 131, 706 (2007). 26. T. A. Volpe et al., Science 297, 1833 (2002). 27. J. K. Kim et al., Science 308, 1164 (2005). 28. We thank our colleagues for comments and advice and G. Szittya for comments on the manuscript. Supported by a UK Biotechnology and Biological Sciences Research Council (BBSRC) Core Strategic grant to John Innes Centre; BBSRC grant BB/D010799/1 (C.D.); European Union (EU) Marie Curie studentship MEST-CT-2005-019727 to S.M.; EU Framework VI program Integrated Project LSHG-CT-2006-037900.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1180278/DC1 Materials and Methods SOM Text Figs. S1 to S11 Table S1 References 7 August 2009; accepted 27 October 2009 Published online 3 December 2009; 10.1126/science.1180278 Include this information when citing this paper.

Reproducibility Distinguishes Conscious from Nonconscious Neural Representations

Aaron Schurger,1,2* Francisco Pereira,1,2 Anne Treisman,1 Jonathan D. Cohen1,2 What qualifies a neural representation for a role in subjective experience? Previous evidence suggests that the duration and intensity of the neural response to a sensory stimulus are factors. We introduce another attributethe reproducibility of a pattern of neural activity across different episodesthat predicts specific and measurable differences between conscious and nonconscious neural representations indepedently of duration and intensity. We found that conscious neural activation patterns are relatively reproducible when compared with nonconscious neural activation patterns corresponding to the same perceptual content. This is not adequately explained by a difference in signal-to-noise ratio. hough once controversial, it is now widely accepted that sensory-perceptual information can be processed by the brain, even at the semantic level, without that information reaching or entering awareness (13). But what does it mean for neural information to reach awareness? Once the information has been encoded in neural activity, what else has to happen for it to become part of ones subjective reality? A growing body of evidence suggests that the intensity of activation in areas that encode the contents of perception (such as the ventral-temporal cortex) is one determinant of whether or not that information contributes directly to subjective experience (47). However, local enhancement of a cortical sensory signal is also associated with attention (8), which can be independent of awareness (911). Therefore, there may be additional features other than the intensity of neural activity that distinguish conscious from nonconscious neural information.

Department of Psychology, Princeton University, Princeton, NJ 08540, USA. 2Center for the Study of Brain, Mind, and Behavior, Princeton University, Princeton, NJ 08540, USA. *To whom correspondence should be addressed. E-mail: [email protected]

Kinsbourne (12) proposes three interacting properties that collectively determine whether or not a neural representation will contribute directly to subjective experience: (i) the duration and (ii) the intensity of a pattern of activity and (iii) the coherence of that pattern of activity with the dominant configuration of neural activity at the global level. Here, we propose that another attribute of neural activity patterns, reproducibility, characterizes conscious representations. We define reproducibility as the similarity of patterns of neural activity across different instances of the same percept. We focused specifically on reproducibility because it is measurable and therefore empirically testable. A corollary of our proposal that conscious representations are more reproducible is that unconscious representations are more variable, even as they may carry information within a given episode. We used functional magnetic resonance imaging (fMRI) to measure brain activity while subjects performed a simple visual categorydiscrimination task (n = 12 subjects) (13). Stimuli were simple line drawings of faces and houses (12 of each), rendered in two opposing but isoluminant colors (Fig. 1) (13). Visibility of SCIENCE VOL 327

the stimuli was manipulated by using dichoptic color masking (DCM) (Fig. 1) (7). Subjects were asked to identify the category of the stimulus (face or house) on each trial, guessing if necessary, and to wager (high or low for monetary rewards) on the accuracy of each of their perceptual decisions (1416). Wagering was used as a collateral index of subjects awareness of the object. For visible stimuli, performance was at or near 100% correct for all 12 subjects, and all wagers were high. For invisible stimuli, task performance was only marginally different from chance (54 T 2.5[SEM]% correct; P < 0.06, one-tailed t test), and sensitivity of high wagers to correct responses [wagering d-prime, or d (13)] was not different from zero (mean d = 0.015 T 0.11[SEM]; P = 0.45, one-tailed t test). For invisible stimuli, wagering d and overall willingness to place high wagers were not significantly correlated across subjects [correlation coefficient (r) = 0.33, P > 0.30, n = 12 subjects]. This reassures against the possibility that wagering d was artificially low because of an interaction with a wagering bias (16). The proportion of high wagers (for invisible stimuli) was similar for faces and houses (0.20 and 0.19, respectively). Subjects were always aware of a visual event a yellowish flickering squareand this provoked substantial activation in and of itself. What varied was subjects awareness of an object embedded in the square. We used multivariate pattern analysis to ascertain how the encoding of perceptual information differs depending on whether or not that information is present in subjective experience (17). Thus, in our analyses we focused specifically on the patterns of activation corresponding to the perceptual information of which the subject was or was not aware: the category of the object. To verify the neural representation of categoryspecific information for both visible and invisible stimuli, we attempted to discriminate the category of the stimulus (faces versus houses) on the basis of the spatial pattern of neural activity in the temporal lobes [derived statistically from each run of

www.sciencemag.org

1 JANUARY 2010

97

REPORTS
functional data (13)]. We did this independently for the visible and invisible stimuli using a Gaussian nave Bayes (GNB) classifier (18). We focused our analyses on the temporal lobes because these are widely viewed as being critical for high-level perceptual representation of visual information (19). Mean accuracy of the classifier (percent correct averaged across 12 subjects) was significantly different from chance (50%) for both visible [63% correct; Students t test, t = 3.82, P < 0.002] and invisible (58% correct; t = 2.53, P < 0.02) stimuli (Table 1). The difference in accuracy for visible versus invisible stimuli was not significant (P < 0.2, one-tailed paired-samples t test). It might be expected that as long as the classifier performed above chance on both types of stimuli, then it would also perform well when trained on one type and tested on the other (20). However, this was not the case for these stimuli (Table 1). Each round of training and testing of the classifier involved a dimensionality-reduction step, in which we determined which voxels (features) varied most consistently as a function of stimulus category (feature selection) separately for visible and invisible stimuli (13). Training and testing of the classifier was then performed on these smaller feature spaces (selections). Our approach involved the examination of the patterns of activity within these selections of voxels on the assumption that these would reveal properties of information encoding under conditions of conscious and nonconscious perception. Treating patterns of activation as vectors allows us to test hypotheses about the properties of neural information independently of specific loci and their level of activity. The angle between two activation vectors reflects differences in the contents of perception, whereas the norm of each vector corresponds to the intensity of the information being encoded. We can then define reproducibility as the similarity in the pattern of activity across different instances of the same stimulus category among voxels that carry relevant information. This can be measured by computing the trial-to-trial variability of the vector angle in the space of the voxels selected as informative for classification. We predicted that activation vectors associated with conscious perception (visible stimuli) would exhibit less trial-to-trial variability in their angle than those associated with nonconscious perception (reflecting greater reproducibility) without necessarily any difference in the norm (that is, in intensity). To assess the reproducibility of representations, we measured the variability in the angle between pairs of vectors (both from the same run and same stimulus category), as well as the norm of each vector, separately for visible and invisible stimuli (13, 21). We repeated this in both the visible and the invisible selections (22). This resulted in four sets of data: responses to visible and invisible stimuli in the visible selection and responses to visible and invisible stimuli in the invisible selection. To avoid confounds that were likely to arise from comparing properties of vectors in different subsets of voxels (and hence different regions of cortex), we restricted our comparisons to vectors within the same selection (23). We used the mean within-category within-run angular deviation as an index of reproducibility. Within the invisible selection, the variability of the vector angle (dVA) is significantly less for visible than for invisible stimuli (P < 0.01, pairedsamples two-sided signed rank test) (Fig. 2B). There was no difference in dVA between visible and invisible stimuli in the visible selection (Fig. 2A), suggesting that the variability is found primarily in the subset of voxels that carry nonconscious information and that this subset is distinct from that within which conscious information is found (for this particular combination of stimuli and task). This is consistent with the failure of the classifier to generalize across the two levels of visibility. When dVA for the invisible selection was compared with the baseline level 4 s earlier (at the time of stimulus onset), there was a significant interaction (P = 0.021, two-sided signed rank test on the deviation from baseline): dVA was below baseline in response to visible stimuli and higher than baseline in response to invisible stimuli (Fig. 2B). There was no difference in the mean or variance of the vector norm for visible versus invisible stimuli, either in the visible or invisible selection (means, P > 0.35, paired-samples two-sided signed rank test; variances, P > 0.7, Levenes test) (Fig. 2, C and D). Thus, a difference in signal-to-noise ratio is not sufficient to explain the effect. Because measurable category-specific information had been identified separately for both visible and invisible stimuli, we examined where in the brain the information tended to coalesce in each case (Fig. 3). For any given subject, reliably informative voxels could be found throughout the temporal lobes (Fig. 3A). Averaging across subjects (24) revealed two clusters in the right ventral temporal cortex, one for visible and the other for invisible stimuli, with minimal spatial overlap, which is consistent with the failure of the classifier trained on one type of stimulus to generalize to the other (Fig. 3, B and C). The anterior-posterior relationship of the two clusters (visible and invisible selections, respectively) coincides with previous observations (25). Conscious and nonconscious neural activation patterns coexist within the cerebral cortex, side by side at the same time, but presumably they differ in several ways. Proposed differences include duration, intensity, and coherence. Here, we show that they also differ in their relative reproducibility across presentations of similar stimuli. Why might reproducibility distinguish conscious from nonconscious representations? One possibility is that conscious information is represented in a more discrete form (26), making it more durable and robust, but also more stereotypical (and therefore more reproducible). Another possibility is that conscious information manifests itself in relatively stable neural firing patterns, corresponding to the settled states of recurrent network interactions (27). There are a number of plausible theories regarding the neural correlates of consciousness but relatively little data concerning the nature of conscious versus nonconscious encoding. Further work is required to understand the difference (or differences) in the way perceptual information is encoded in the brain depending on whether or not that information is present in subjective experience. Such work is likely to have profound importance in a variety of arenas, including the assessment of consciousness under presumed Table 1. Performance of a GNB classifier. The objective of the classifier was to discriminate the category of the stimulus based on the pattern of beta weights [a general linear model (GLM) was applied separately to each run of functional data (13)]. A voxel-wise analysis of variance and nested cross-validation (18) were used for dimensionality reduction on each round of training and testing. For within-condition classification (visible-visible and invisible-invisible) a leave-one-run-out crossvalidation was performed. For between-condition classification, we trained on all the data from one condition and tested on the other and vice-versa. All t tests are one-tailed with df = 11.
Train Visible Visible 63 T 3.5 t = 3.8, P < 0.002* 52 T 3.0 t = 0.69, P = 0.25 Test Invisible 48 T 2.3 t = 0.78, P = 0.77 58 T 3.1 t = 2.5, P < 0.02*

Fig. 1. Dichoptic-color masking. This method of manipulating awareness, originally devised by (7), relies on the phenomenon of dichoptic color fusion. The same color mode corresponds to the visible condition, and the opposite color mode corresponds to the invisible condition. In order to achieve disappearance of the image in the opposite color mode, the two colors must be approximately isoluminant and the object boundaries slightly blurred. Before the experiment, subjects were trained to maintain steady fixation and were cued to do so during each trial with the appearance of the fixation point (500 ms before stimulus onset). Stimuli were presented stereoscopically in the fMRI scanner by using a cardboard divider and prism lenses (28). VOL 327 SCIENCE

Invisible

*Statistically significant; P < 0.05.

98

1 JANUARY 2010

www.sciencemag.org

REPORTS
Fig. 2. Variability in the angle of activation vectors in the visible and invisible selections (A and B) and mean vector norm (C and D). In both (A) and (B), t0 corresponds to the TR (2 s) on which the stimulus was presented before the hemodynamic response had begun to rise. t2 corresponds to 2 TRs (4 s) after the stimulus was presented at the (approximate) peak of the hemodynamic response (n = 12 subjects). This analysis was performed by using a leave-onerun-out procedure: Voxel selection was performed on data from n 1 runs, and the norm and angular deviation were computed on data from the run that had been left out (13). Comparisons between the two selections [(A) versus (B) or (C) versus (D)] are not valid (23). Fig. 3. Spatial distribution of informative voxels. (A and B) Voxels that were selected as informative for classification (face versus house) on 6 or more (out of 12) runs for a subject with comparable classification accuracy (72% correct) for visible and invisible stimuli. (C and D) The mean across subjects (24) projected onto the AFNI TT_N27 template brain (right hemisphere) at a statistical threshold of P < 0.05 (corrected). The oblique white line serves as a visual landmark. The cluster in (C) encompasses a portion of the fusiform and parahippocampal gyri in the area of the fusiform face area (FFA) and parahippocampal place area (PPA). The cluster in (D) lies along the posterior fusiform gyrus. anesthesia or coma and the investigation of brain function in conditions such as schizophrenia, autism, and dissociation disorders.
References and Notes
1. S. Dehaene et al., Nature 395, 597 (1998). 2. P. M. Merikle, D. Smilek, J. D. Eastwood, Cognition 79, 115 (2001). 3. S. Kouider, S. Dehaene, Philos. Trans. R. Soc. B Bio. Sci. 362, 857 (2007). 4. Y. Jiang, S. He, Curr. Biol. 16, 2023 (2006). 5. G. Rees et al., Brain 123, 1624 (2000). 6. P. Vuilleumier et al., Proc. Natl. Acad. Sci. U.S.A. 98, 3495 (2001). 7. K. Moutoussis, S. Zeki, Proc. Natl. Acad. Sci. U.S.A. 99, 9527 (2002). 8. L. Pessoa, S. Kastner, L. G. Ungerleider, J. Neurosci. 23, 3990 (2003). 9. B. Bahrami, N. Lavie, G. Rees, Curr. Biol. 17, 509 (2007). 10. V. Wyart, C. Tallon-Baudry, J. Neurosci. 28, 2667 (2008). 11. A. Schurger, A. Cowey, J. D. Cohen, A. Treisman, C. Tallon-Baudry, Neuropsychologia 46, 2189 (2008). 12. M. Kinsbourne, in Scientific Approaches to Consciousness, J. D. Cohen, J. W. Schooler, Eds. (Lawrence Erlbaum Associates, Inc., Hillsdale, NJ, 1997), pp. 335355. 13. Materials and methods are available as supporting material on Science Online. 14. Post-decision wagering has been proposed as an independent measure of awareness under the assumption that if the subject is aware of the relevant sensory information, then there will be a correspondence between high wagers and correct responses. If no such correspondence is found, then under this assumption we deduce that the subject was not aware of the relevant sensory information (in this case, information sufficient to discern the category of the stimulus) (15). 15. N. Persaud, P. McLeod, A. Cowey, Nat. Neurosci. 10, 257 (2007). 16. A. Schurger, S. Sher, Trends Cogn. Sci. 12, 209, author reply 210 (2008). 17. J. D. Haynes, Trends Cogn. Sci. 13, 194 (2009). 18. F. Pereira, T. Mitchell, M. Botvinick, Neuroimage 45 (suppl.), S199 (2009). 19. D. L. Sheinberg, N. K. Logothetis, Proc. Natl. Acad. Sci. U.S.A. 94, 3408 (1997). 20. P. Sterzer, J. D. Haynes, G. Rees, J. Vis. 8, 10 (2008). 21. While voxel selection was based on coefficients derived statistically from each functional run (13), the activation patterns among these voxels were taken trial by trial from the minimally processed fMRI signal data [at t0 + 2TR, where t0 is time of stimulus onset and 1 TR (time of repetition) = 2 s]. This was done in a leave-one-run-out fashion: The selection was chosen on the basis of data from n 1 runs, and then the activity vectors from the left-out run (2 visible/invisible face/house per run) were projected into that space (13). 22. The visible selection comprises the voxels that were maximally informative as to the category of visible stimuli. Likewise, the invisible selection comprises the voxels that were maximally informative as to the category of invisible stimuli. 23. Because the visible selection and the invisible selection occupy separate and largely non-overlapping regions of cortex, comparisons between their functional properties are confounded with differences between the hemodynamic and magnetic-field properties of the regions they inhabit. 24. To produce spatial maps of reliably informative voxels, each voxel was coded with either a 1, if selected on a majority of runs, or a 0 otherwise (Fig. 3, A and B). In order to uncover regional tendencies in the average across subjects, maps for each subject were blurred by ~10 mm and then discretized again (ceiling). The probability distribution of the average map under the null hypothesis was estimated by using a permutation test (number of voxels held constant for each subject per selection, but locations randomized) and used to set a statistical threshold. 25. M. Bar et al., Neuron 29, 529 (2001). 26. J. Sackur, S. Dehaene, Cognition 111, 187 (2009). 27. D. Balduzzi, G. Tononi, O. Sporns, PLOS Comput. Biol. 4, e1000091 (2008). 28. A. Schurger, J. Neurosci. Methods 177, 199 (2009). 29. A.S. was supported by a grant from the Mind Science Foundation and by a Ruth L. Kirschstein National Research Service Award from the National Institute of Mental Health (MH075342). Special thanks to S. Sher for helpful discussions, S. Dehaene and two anonymous reviewers for comments, M. Kim for help with behavioral testing, and L. Nystrom for advice and assistance with data analysis.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1180029/DC1 Materials and Methods Fig. S1 References 3 August 2009; accepted 19 October 2009 Published online 12 November 2009; 10.1126/science.1180029 Include this information when citing this paper.

www.sciencemag.org

SCIENCE

VOL 327

1 JANUARY 2010

99

LIFE SCIENCE TECHNOLOGIES

AAAS/Science Business Office

NEW PRODUCTS

TOXICITY ASSAYS
The MaxDiscovery Toxicity Assay Kits monitor in vivo toxicity by measuring liver and heart function in serum samples.These simple and convenient tests allow the rapid, colorimetirc determination of well-characterized functional markers of liver and heart toxicity. They can be used to monitor toxic effects caused by administration of small-interfering RNA, DNA plasmids, transfection agents, or other therapeutic treatments during preclinical studies. The assays can be used in a range of mammals, including mice and rats, providing a powerful tool to assess and reduce the toxicity of drug formulations and to improve drug delivery methods. Bioo Scientic For info: 512-707-8993
www.biooscientic.com

SUBCELLULAR PROTEIN FRACTIONATION

The Protein Fractionation Kit enables the stepwise separation and preparation of cytoplasmic, membrane, nuclear soluble, chromatin-bound, and cytoskeletal protein extracts from mammalian cultured cells or tissue for protein localization studies or enrichment of specic cellular compartments. The simple protocol is completed in less than three hours and requires no ultracentrifugation over gradients during any of the steps. The subcellular extracts obtained are compatible with a variety of downstream applications, including protein assays, protein immunoblotting, electrophoretic mobility assays, and enzyme activity assays. The kit includes four extraction buffers, a stabilized nuclease, and the Halt Protease Inhibitor Cocktail. Thermo Fisher Scientic
For info: 815-968-0747 www.thermo.com/pierce

open up new chemical spaces, and this service is designed to help in that effort. The service allows researchers to screen the selectivity of their allosteric compounds for more than 150 GPCRs. Millipore can also provide custom services to help researchers identify new positive allosteric modulators of a GPCR target of interest. Millipore
For info: 800-548-7853 www.millipore.com/gpcr

FLAME STERILIZATION CAROUSEL

The AutoloopPRO is a fully automatic carousel for ame sterilizing inoculation loops. The stable housing of the AutoloopPRO enables comfortable and easy access to inoculation loops. Removal positions on both sides make the carousel equally suitable for right- and left-handers. It can be used with up to four inoculation loops. Flaming and cooling time can be adjusted to the second. The carousel rotates and controls aming automatically. It is fabricated of stainless steel and anodized aluminum and features a graphic display protected by heat-resistant glass. WLD-TEC
For info: +49-(0)-551-793789 www.WLD-TEC.com

PROTEIN QUANTIFICATION
Gyrolab xP is a second generation of a platform for miniaturized protein quantication at all stages of drug development. The system offers enhanced ease-of-use, 21 CFR part 11 compliance, and extended validation support for smooth implementation in good laboratory practiceregulated environments. Gyros
For info: +46-18-56-64-00 www.gyros.com

VIBRATION CANCELLATION PLATFORM

The Stacis FP active vibration cancellation oor platform system is designed for use with scanning electron microscopes (SEMs). It features subhertz vibration cancellation in an active hard-mount oor platform that ts most commercial SEMs. SEMs generally incorporate an internal vibration isolation system, and Stacis FP is compatible with all of them. The system is only seven inches tall, measures three inches by 40 inches, weighs less than 400 pounds, and can support 2,500 pounds or more with no soft air suspension. It is available with three or six degrees of freedom. Technical Manufacturing Corporation
For info: 978-532-6330 www.techmfg.com

GPCR SERVICE
The AllostericProler service is aimed at researchers developing new drugs using G-protein coupled receptors (GPCRs). The service provides drug developers with cell-based assays to evaluate the functional selectivity of drugs against GPCRs. Although 40 percent of all marketed drugs are targeted to GPCRs, the similarity of GPCR binding sites coupled with saturated intellectual property around this chemical space have hampered GPCR drug discovery efforts. Increasingly, drug makers are instead focusing on novel allosteric GPCR sites to

Electronically submit your new product description! Go to www.sciencemag.org/products/newproducts.dtl for more information. Newly offered instrumentation, apparatus, and laboratory materials of interest to researchers in all disciplines in academic, industrial, and governmental organizations are featured in this space. Emphasis is given to purpose, chief characteristics, and availabilty of products and materials. Endorsement by Science or AAAS of any products or materials mentioned is not implied. When you seek additional information from the manufacturer or supplier, tell them you saw it here.

100

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencemag.org/products

Release The Power of Science

Science Careers Classied Advertising

For full advertising details, go to ScienceCareers.org and click For Employers, or call one of our representatives. Tracy Holmes Worldwide Associate Director Science Careers Phone: +44 (0) 1223 326525

UNITED STATES & CANADA

E-mail: [email protected] Fax: 202-289-6742 Daryl Anderson US Sales Manager Phone: 202-326-6543 Tina Burks Midwest/Canada Phone: 202-326-6577 Alexis Fleming East Coast Phone: 202-326-6578 Nicholas Hintibidze West Coast/South Central Phone: 202-326-6533 Online Job Posting Questions Phone: 202-326-6577

EUROPE & REST OF WORLD

E-mail: [email protected] Fax: +44 (0) 1223 326532 Alex Palmer Phone: +44 (0) 1223 326527 Dan Pennington Phone: +44 (0) 1223 326517 Susanne Kharraz Tavakol Phone: +44 (0) 1223 326529 Lisa Patterson Phone: +44 (0) 1223 326528

FULL TENURE-TRACK FACULTY POSITION, ONCOLOGIST-INVESTIGATOR The Meharry Medical College/Vanderbilt-Ingram Cancer Center Partnership (MMC/VICC Partnership) is currently soliciting competitive applications for one oncologist as a physician-investigator for a full tenuretrack faculty position at the level of either ASSISTANT or ASSOCIATE POSITION at Meharry Medical College with a joint appointment at Vanderbilt-Ingram Cancer Comprehensive Center (VICC). The qualified candidate will also have either FULL or an ASSOCIATE MEMBERSHIP status at the VICC. The candidate will have primary appointment in a clinical department with a secondary appointment in the Department of Biochemistry and Cancer Biology at Meharry. The candidate will have 70 percent research time and 30 percent clinical time and will be expected to develop and maintain a funded competitive cancer research. The candidate will enjoy collaboration and/or mentoring by senior cancer research scientists within the partnership. Initially, the position will be fully funded by the National Cancer Institute_s U54 Cancer Center Partnership Grant to the MMC/ VICC Partnership, whose principal investigators are Dr. Harold Moses, Emeritus Director of VICC, and Dr. Samuel Evans Adunyah, Chairman of Department of Biochemistry and Cancer Biology at Meharry Medical College. The position also comes with full laboratory support which includes a startup package for a new laboratory, supplies, and three laboratory support personnel for the investigator. Surgical or medical oncologists (M.D. or M.D./Ph.D.) with cancer research interests are encouraged to apply by December 30, 2009. Please, send an electronic file on CD-rom containing the following information to the address below: (1) A letter which describes your interest in our position and your current position. (2) Research description and future research goals. (3) Current full curriculum vitae. (4) References (at least four). Dr. Samuel Evans Adunyah Chairman, Department of Biochemistry and Cancer Biology Co-P.I., MMC/VICC Partnership Department of Cancer Biology 1005 D.B. Todd Boulevard Nashville, TN 37208 Dr. Adunyah can be reached at telephone: 615327-6345; fax: 615-327-6440; e-mail: sadunyah@ mmc.edu. Or send your materials electronically to Ketia Barnes, MMC/VICC Cancer Partnership Program Coordinator, e-mail: [email protected].

ASSISTANT to FULL PROFESSOR Neuropharmacology The Medical College of Georgia is seeking an investigator with outstanding research accomplishments and potential in neuropharmacology and neuroscience to complement and enhance the diversity of basic and clinical neuroscience research strengths in our Department and Institution. MCG has an eminent, well-funded neuroscience research community across the Departments of Pharmacology and Toxicology, Physiology, Cell Biology, Molecular Medicine, Psychiatry and Health Behavior, and Neurology. Faculty members in pharmacology and toxicology conduct research that is relevant to a range of neurological and psychiatric disorders including Alzheimer_s disease, Parkinson_s disease, drug abuse, schizophrenia, and mental retardation. We seek applicants with a Ph.D. or M.D. degree with a record of productivity and extramural funding. Applicants with expertise in cell signaling, drug discovery, systems neurobiology, or neurotoxicology are particularly encouraged to apply. Strong consideration will be given to applicants with a clear collaborative potential with current faculty members in the Department of Pharmacology and Toxicology. We offer a competitive salary and startup package commensurate with the level of appointment, excellent laboratory space and outstanding core facilities for microarray technology, genetically modified animals, cell imaging, electron microscopy, small animal and nonhuman primate behavior, and clinical collaborations. The successful applicant will participate in teaching programs for professional and graduate students. Please send curriculum vitae, summary of professional and research goals, and the names and addresses of three references to: Search Committee, c/o Dr. Alvin V. Terry, Department of Pharmacology and Toxicology, Medical College of Georgia, Augusta, GA 30912-2300; e-mail: aterry@mail. mcg.edu. Visit the department homepage at website: http://www.mcg.edu/SOM/phmtox/index.html. Application review will begin January 15, 2010. MCG is an Equal Employment Opportunity/Affirmative Action/Equal Access Employer. ACH56836

JAPAN
ASCA Corporation Jie Chin Phone: +81-3-6802-4616 Fax: +81-3-6802-4615 E-mail: [email protected]
To subscribe to Science: In US call 866 434-2227 In the rest of the world call +1 202 326-6417 All ads submitted for publication must comply with applicable US and non-US laws. Science reserves the right to refuse any advertisement at its sole discretion for any reason, including without limitation for offensive language or inappropriate content, and all advertising is subject to publisher approval. Science encourages our readers to alert us to any ads that they feel may be discriminatory or offensive.

TENURED or TENURE TRACK FACULTY POSITION Drug Delivery to the Central Nervous System Department of Pharmaceutics University of Minnesota The Department of Pharmaceutics, College of Pharmacy, at the University of Minnesota, invites applications and nominations for the position of ASSISTANT PROFESSOR (tenure-track) or ASSOCIATE PROFESSOR (tenured or tenuretrack) specializing in drug delivery to the central nervous system (CNS). Requirements include an earned doctorate (Ph.D. or equivalent) in pharmaceutics, biomedical engineering, neuroscience, neurobiology or similar field and preferred completion of a postdoctoral research fellowship with research expertise in drug delivery and the barriers of the central nervous system such as the blood brain barrier, blood-CSF barrier, and intraparenchymal transport barriers. Additional information and applicant information may be found at website: http://www.pharmacy. umn.edu./employment. For specific questions about this position, please contact the search chair, Dr. William Elmquist at e-mail: [email protected]. The University of Minnesota is an Equal Opportunity Educator and Employer.

TENURE-TRACK FACULTY POSITION Assistant Professor in Structural Bioinformatics/Computational Structural Biology The College of Letters, Arts and Science of the University of Southern California invites applications for a tenure-track faculty position at the level of Assistant Professor, beginning fall 2010. Candidates in all areas of structural bioinformatics and computational structural biology are welcome to apply. The individual will assist in further strengthening and expanding the existing strong computational biology and bioinformatics program. The position is in the interdisciplinary Program in Molecular and Computational Biology (MCB) in the Department of Biological Sciences. The successful candidate will be expected to participate in undergraduate and graduate teaching and to establish a vigorous, externally funded research program. Review of applications will begin immediately. Interested candidates should electronically send curriculum vitae and research plans and should arrange for three letters of reference to be sent to e-mail: [email protected] or, if necessary, Eleni Yokas, Computational Biology Search Committee, Department of Biological Sciences, RRI201, University of Southern California, Los Angeles, CA 90089-2910. For additional information about our program, visit our website: http://www.cmb. usc.edu/. USC strongly values diversity and is committed to equal opportunity in employment. Women and men, and members of all racial and ethnic groups, are encouraged to apply.

www.sciencecareers.org

SCIENCE

VOL 327

1 JANUARY 2010

101

online @sciencecareers.org

POSITIONS OPEN

POSITIONS OPEN

Department of Health and Human Services National Institut es of Health National Cancer Institute, Center for Cancer Research Postdoctoral Fellowship positions at the National Cancer Institute Postdoctoral fellowship positions are available in the Lung Carcinoma Section, Medical Oncology Branch at the National Cancer Institute (NCI), Bethesda, MD. Projects involve studying molecular pathways required to support oncogenesis, identifying new therapeutic targets for Ras-driven cancers and developing new RNAi tools for mammalian genetics. Applicants must have a PhD or MD degree and less than ve years postdoctoral experience. Candidates should have a strong background in molecular biology, genetics or biochemistry, and a strong interest in cancer biology and translational research. For further information about the MOB, NIH or NCI programs, faculty and training, please visit our respective Web sites: http://ccr.cancer. gov/labs/lab.asp?labid=753, http://ccr.nci.nih.gov, http://www.nih.gov Anticipated starting date for this position is summer 2010. Candidates should email a cover letter, CV, including publications in peer reviewed journals, and the names/phone numbers of three references to: Dr. Ji Luo c/o Allyson Cole: [email protected] Clinical Research ARC, OM, NCI 10 Center Drive, Bldg. 10, Room 12N226 Bethesda, MD 20892-1904 DHHS, NIH and NCI are Equal Opportunity Employers

Division of Construction/ARRA Construction Team Health Scientist Administrator The National Center for Research Resources, a component of the National Institutes of Health, has been charged by the President of the United States to administer a $1 billion program under the American Recovery and Reinvestment Act (ARRA) to provide funding for construction, renovation, and/or repair of facilities used for biomedical and behavioral research. In order to accomplish this goal, NCRR has created a Division of Construction to oversee and administer grant awards along with other activities required to ensure the transparency and accountability in the program. The ARRA construction team will be reviewing construction designs, overseeing the progress of construction, and reporting the status of the program to a variety of audiences, including all levels of Federal Government and the American taxpayers. At full strength, the Division will have 7 members and additional advertisements for a construction project manager, professional engineers, and a program analyst will be announced soon. Please check the job opportunities listed at the NCRR website, www.ncrr.nih.gov, for these specic announcements. As part of the team, NCRR is currently seeking applications for Health Scientist Administrators. The salary range is $73,100 to $133,543 per annum, commensurate with qualications and professional experience. A full benets package is available, which includes retirement, Thrift Savings Plan participation, health, life, and long-term care insurance. For qualication requirements, evaluation criteria, and application instructions, view the vacancy announcements at http://www.jobs.nih.gov/hsa/. Please contact Michelle Lipinski at 301-594-2286 if you have any questions.

Clinical Tenure-Track Position Division of Intramural Research (DIR)

The National Institute of Allergy & Infectious Diseases (NIAID), Division of Intramural Research (DIR), is seeking an outstanding tenure-track investigator to develop a clinical research program to better understand, treat, and ultimately prevent infectious, immunologic, and/ or allergic diseases. The successful candidate will implement and direct an independent clinical research program with an emphasis on clinical studies but which may include translational and basic research. The incumbent can choose the laboratory with which he or she would prefer to be afliated. Any clinical protocols developed should complement the research goals of the selected laboratory. In addition, the incumbent will be paired with a senior investigator, who will serve as a clinical mentor. An outstanding postdoctoral record of research accomplishment and an M.D., M.D. /Ph.D., or equivalent degree is required for this position; board eligibility/board certication is also required. The incumbent will be expected to meet the requirements for authorization of patient care privileges by the Credentialing Services of the NIH Clinical Center. Candidates will be assigned independent resources to include clinical and/or laboratory support personnel, equipment, space, and an allocated annual budget for services, supplies, and salaries sufcient to foster success. This is a tenure-track appointment under Title 42. Salary is dependent on experience and qualications. Interested candidates may contact Dr. Karyl Barron, DIR deputy director at 301-496-3006 or [email protected] for additional information about the position. To apply for the position, e-mail your curriculum vitae, bibliography, and an outline of your proposed research program (no more than two pages), by January 14, 2010 to Ms. Yushekia Hill at [email protected]. In addition, send three letters of recommendation to Chair, NIAID DIR Clinical Tenure Track Search Committee, c/o Ms. Yushekia Hill at [email protected] or 10 Center Drive MSC 1356, Building 10, Room 4A-22, Bethesda, MD 20892-1356. E-mail is preferred. Please note search #027 when sending materials.

National Institute of Allergy and Infectious Diseases

Further information about DIR laboratories is available at www.niaid.nih.gov/about/organization/dir and information about working at NIAID is available at www.niaid.nih.gov/careers/sctt.

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES National Institutes of Health

National Institute of Allergy and Infectious Diseases

Proud to be Equal Opportunity Employers

NIAID

online @sciencecareers.org

STANFORD UNIVERSITY SCHOOL OF MEDICINE

LUCILE PACKARD CHILDRENS HOSPITAL AT STANFORD

Iowa Center for Muscular Dystrophy Research and Department of Molecular Physiology and Biophysics University of Iowa Carver College of Medicine

CENTER FOR EXCELLENCE IN PULMONARY BIOLOGY

Divisions of Pulmonary, Asthma, and Critical Care Medicine

Faculty Positions Skeletal Muscle Biology and Disease

The Iowa Center for Muscular Dystrophy Research seeks outstanding candidates for tenure-track positions at any rank. Successful candidates are expected to establish independent laboratories focusing on skeletal muscle biology and disease. We are particularly interested in individuals who would complement existing strengths in the Center. These positions feature outstanding research space with state-of-the-art shared instrumentation, as well as substantial startup funds for equipment, personnel support and supplies. The University of Iowa is located in Iowa City, an affordable college community with many cultural amenities. All applicants must have a relevant doctoral degree, productive research experience, and a strong record of research accomplishment. Candidates are judged on their potential to initiate and maintain a creative, independent research program, and their desire to train students and postdoctoral fellows. To apply for this position, please visit the University of Iowa website at http://jobs.uiowa.edu/faculty/view/57388. Applicants should include a CV, letter of interest, and the names of three references. Consideration of completed applications will begin on February 15, 2010. Questions may be directed to Dr. Kevin P. Campbell, Director of the ICMDR at [email protected]. The University of Iowa is an Equal Opportunity/Afrmative Action Employer. Women and minorities are strongly encouraged to apply.

Stanford University Pediatric Pulmonary Center of Excellence in Pulmonary Biology

The newly established Center for Excellence in the Department of Pediatrics at Stanford University seeks an outstanding faculty (UTL) member in Pediatric Pulmonary Medicine. Essential Requirement for UTL candidates: Be an MD with the Board Certication in Pediatrics, and be Board Eligible or Board Certied in Pediatric Pulmonary Medicine, have medical licensure in California by starting date, and a record of accomplishment in investigator-initiated, hypothesis-driven research. It is anticipated that the successful faculty candidate will be an excellent researcher, committed teacher, and an outstanding clinician and have an emerging regional/national reputation in Pediatric Pulmonary related research, and have the skills to establish an investigator-initiated research program focused primarily upon discoveries that can be used to inform patient care. The specic focus of the research program will be on translating the clinical observation and laboratory based discoveries into clinical programs. The overriding requirement for faculty appointment, reappointment and promotion within the UTL must be distinguished performance, or (in the case of junior faculty) the promise of distinguished performance. There should be a major commitment to research and teaching. There must be outstanding accomplishments in research and excellent overall performance in teaching, as well as in clinical care and institutional service appropriate to the programmatic need the individual is expected to fulll. Such programmatic need, including nancial viability, should be evaluated and must be established for each appointment, reappointment and promotion. Contingent on professional accomplishment, the candidate will be appointed as an Assistant or Associate Professor (University Tenure Line) in the Stanford University School of Medicine. Rank and salary will be commensurate with qualications and experience. Applications will be reviewed beginning July 1st, 2009 and accepted until position is lled. Submit applications and CV to: David Corneld, M.D. Chair, Search Committee The Anne T. and Robert M. Bass Professor of Pulmonary Medicine Director Center for Excellence in Pulmonary Biology Service Chief of Pulmonary, Allergy and Critical Care Medicine Suite 350, 770 Welch Road Stanford, California, 94304 dcorneld.edu Stanford University is an Equal Opportunity Employer and is committed to increasing the diversity of its faculty. It welcomes nominations of and applications from women and members of minority groups as well as others who would bring additional dimensions to the universitys research, teaching, and clinical missions.

SOOCHOW UNIVERSITY

Faculty Positions
The Soochow University, founded in 1900 in Suzhou, Jiangsu, is a premier university in China. The University is committed to excellence in education, science and innovation. A major expansion of its research programs is underway across a broad spectrum of areas in mathematics, physics, chemistry, mechanical and electronic engineering, optical and energy engineering, material and computer sciences, urban planning, environmental science, textile engineering and design, life sciences, and medicine. A complete listing of open positions is available at http://rsc.suda.edu.cn/zpxx.asp?type=15. We seek distinguished professors, academic pacesetters and laboratory heads to join us at the Institute of Neuroscience. Successful candidates should have a Ph.D. degree with more than 3 years of research experience abroad and an outstanding record of academic achievements with high quality publications in top international journals. We provide competitive relocation, salary packages, generous start-up funds and state-of-the-art facilities. Interested individuals should send curriculum vitae, a summary of accomplishments, future research plans, and a list of three references to LI Jun, Institute of Neuroscience, Soochow University, email: [email protected], campus of Soochow University, 199 Ren-Ai Road, Suzhou, Jiangsu, 215123, China, http: //www.suda.edu.cn/.

SCIENCE Editorial
Science and AAAS seek a talented scientist to serve as an Associate Editor for our new interdisciplinary journal, Science Translational Medicine. This position is designed for an individual with broad interests, a lively curiosity, and experience with cutting-edge research in at least one, but preferably more than one, biomedical or clinical research eld. The tasks include, but are not limited to: Manage the review, selection, and editing of submitted manuscripts; Judge the scientic value of research; Foster relationships and communication with the scientic community through literature reviews, meetings and professional contacts; Select reviewers for submitted manuscripts; Discuss and make recommendations regarding manuscripts and reviews with other staff, advisers, authors; Write summaries of research results for publication; Guide authors on manuscript revisions; Edit the manuscripts for scientific content and style before and after revisions; Follow the manuscript through production process to ensure material is published in a timely manner; Travel to scientic meetings. The minimum qualications to be competitive and considered for the position are: Mastery of a professional eld typically acquired through completion of a doctoral degree in at least one biomedical or clinical research eld; 3-5 years experience, including post doctoral research experience and multiple publications; Ability to work constructively as a member of a team; Experience with cutting-edge research in one of the elds mentioned above; Comprehensive knowledge of scientic research methods in order to discuss technical issues with authors; Exceptional writing, communication, and listening skills in order to communicate with authors and reviewers in evaluating, editing and modifying manuscripts. Previous editorial experience is not required. If you would like to be a member of the AAAS team, please visit our Job Information website at http://www.aaas.org/ careercenter/employmentataaas/ to get more information and to apply online today.

Faculty Positions in Bioinformatics and Computational Biology at the Jan and Dan Duncan
Neurological Research Institute

Texas Childrens Hospital and Baylor College of Medicine are seeking several outstanding computational biologists to join the faculty of the Jan and Dan Duncan Neurological Research Institute (NRI).The NRI will bring together world experts in neurology, neuroscience, pediatrics, genetics, physiology, developmental biology, chemistry, behavior, imaging and applied mathematics to pursue collaborative, interdisciplinary basic and translational research on a variety of neurological and neurodevelopmental disorders. The new 13-story NRI, is nestled between Baylor College of Medicine,Texas Childrens Hospital and M.D.Anderson Cancer Center and will include state-of-the-art computational resources as well as core facilities overseen by experts in imaging, physiology, neurobehavior, neuropathology and animal models of human disease among others. The successful faculty will have appointments in the Departments of Pediatrics and other basic science departments (according to interest and expertise) at Baylor College of Medicine; joint appointments at Rice University will be considered when appropriate.The primary responsibility will be to establish a cutting-edge Bioinformatics and Computational Biology Program within the NRI that will promote the research interests of the appointees as well as foster collaborations with other NRI faculty members.We seek exceptional research scientists that can capitalize on our highly collaborative environment.We envision the formation of interdisciplinary teams of investigators focused on unraveling the mechanisms responsible for brain disorders using systems approaches.The candidates should complement existing strengths at the NRI with their expertise in quantitative analysis such as statistical pattern recognition, machine learning, data mining, neuroinformatics, and algorithm optimization, and will be expected to contribute to the goals of the Institute by developing methods for high-throughput image analysis and for integrating -omics data.A generous start-up package and state-of-the-art computational resources will be provided. Candidates must possess a Ph.D., M.D. or equivalents in Neuroscience,Applied Mathematics, Computational Biology, or Biophysics etc. Academic rank will be commensurate with experience. Interested individuals should send a curriculum vitae and personal statement of their professional goals to: Huda Y. Zoghbi, M.D., Director, Neurological Research Institute; Baylor College of Medicine One Baylor Plaza, MS BCM225, Room T807 Houston,TX 77030. Baylor College of Medicine is an Equal Opportunity, Affirmative Action, Equal Access Employer.

2009 Texas Childrens Hospital. All rights reserved.

online @sciencecareers.org

online @sciencecareers.org

Director of Proteomics and Mass Spectrometry

University of Massachusetts Medical School

The University of Massachusetts Medical School seeks an outstanding individual to serve as Director of its Proteomics Facility. Over the past decade, the medical school has expanded its research program by establishing new departments and programs, by bringing over 120 new faculty to the campus and by constructing state-of-the-art facilities. The next growth phase will focus on translational research including RNA therapeutics, stem cell biology and gene therapy, and will be housed in an adjacent building now under construction. Exciting opportunities in proteomics for the existing and developing programs require a highly qualied individual to lead the Proteomics Facility. Competitive candidates will have a PhD and at least ve years of professional experience in the application of a variety of sophisticated mass spectrometric methods to biology and/or medicine. Candidates will also have demonstrated leadership abilities critical for managing a multi-component facility and its staff. The individual will be responsible for training students and postdoctoral fellows, establishing collaborations with our faculty, and advancing the capabilities of the facility via competitive funding from external sources. Information on our current proteomics facilities, which will be combined into an integrated facility under the leadership of the new director, can be found under Core Facilities at http: //www.umassmed.edu/research. Applicants should submit a cover letter explaining their qualications and their interest in this position, a curriculum vitae and the names and contact information for three individuals who could provide letters of recommendation to http://www.academicjobsonline.org. Minorities and women are especially encouraged to apply. Inquiries but not application materials may be addressed to [email protected]. As an Equal Opportunity and Afrmative Action Employer, UMMS recognizes the power of a diverse community and encourages applications from individuals with varied experiences, perspectives and backgrounds.

Assistant Professor (tenure-track) Faculty Positions Department of Biochemistry, School of Medicine

West Virginia University Morgantown, WV
The Department of Biochemistry is expanding its faculty with positions at the Assistant Professor rank (tenure-track) in the area of Cancer Biology. The department has existing strengths in the areas of regulation of gene or protein expression, analysis of oncogene signaling networks, metabolism, and mechanisms of cell death by apoptosis or anoikis. Investigators who complement these strengths, or with research programs in emerging areas of interest are especially encouraged to apply. Innovative programs addressing fundamental questions related to cancer will be given priority. Minimal qualications are a PhD plus three years of highly productive postdoctoral research. The successful candidate will be expected to develop an extramurally funded research program, demonstrate excellence in the departmental teaching mission, and be an active participant in department and school-wide research forums and seminars. West Virginia University is a comprehensive, land grant university with 185 undergraduate, graduate and professional programs and an enrollment of 28,840 students, including 6,910 professional and graduate students. The Robert C. Byrd Health Sciences Center includes the Schools of Medicine, Pharmacy, Dentistry and Nursing. Recent expansion at the Health Sciences Center includes two new research buildings, which collectively provide approximately 200,000 sq. ft. of additional research space, to accommodate our growth in research programs and core facilities. With a metro area population of over 115,000, Morgantown is consistently rated as one of the best small cities in the U.S., with an affordable cost of living, excellent schools, ample job opportunities, a picturesque countryside, and many outdoor recreational opportunities. Review of applications will begin immediately and continue until the position is lled. Please submit as a single electronic le your C.V., summary of past research accomplishments and future research plans, and the names of three references to: Ms. Lana Yoho, [email protected]. WVU is an Equal Opportunity/Afrmative Action Employer.

National Institute for Basic Biology, Okazaki, Japan

Three Professorships
The National Institute for Basic Biology (NIBB), Okazaki, Japan, invites applications for three professorships. Since its establishment following a recommendation by the Science Council of Japan in 1977, NIBB has conducted research activities as a center of excellence in biological sciences. The research activities at NIBB are among the highest ranked in the ield of biological sciences in Japan, as judged by factors such as citation index per paper. To continue our efforts to serve as a national center for biological sciences, NIBB recruits leading scientists not only from Japan but also from abroad. We seek candidates with excellent research records who can be international leaders in the ield of basic biology. In particular, those using biological approaches to answer fundamental questions in such ields as (1) Neuroscience, (2) Plant science, and (3) System sciences for environmental responses of biological organisms. In addition, we welcome applications from candidates of any ield that have excellent potential to advance basic biology. To apply, please see the website below for a list of required documents; which must be received no later than January 18, 2010. Online application is possible for overseas applicants. For more information, please visit the website at http://www.nibb.ac.jp/ profapplication/ Inquires may be directed to Professor Tetsuo Yamamori, National Institute for Basic Biology, 38 Myodaiji, Okazaki, 444-8585, Japan, Tel/ Fax: 81-564-55-7615, E-mail: [email protected]

VCU

SENIOR POSITION IN HUMAN AND MOLECULAR GENETICS (HMG) VCU INSTITUTE OF MOLECULAR MEDICINE (VIMM) AND VCU MASSEY CANCER CENTER (MCC)

VIRGINIA COMMONWEALTH UNIVERSITY (VCU) SCHOOL OF MEDICINE, HMG, VIMM and MCC

Under the leadership of Dr. Paul B. Fisher the Department of Human and Molecular Genetics (HMG) and the VCU Institute of Molecular Medicine (VIMM) in collaboration with the VCU Massey Cancer Center (MCC) in Richmond, Virginia seeks to recruit a seasoned investigator who focuses on cancer development and progression and whose research bridges the gap between laboratory discovery and clinical trials. Research that employs current genomic discoveries in medicine with an emphasis on translating these ndings into improved approaches for diagnosis and treatment of neoplastic diseases are areas of high priority of this institutional initiative. The ideal candidate will be an experienced investigator who has demonstrated consistent research excellence, with a sustained track record of research funding, high-level publications and administrative experience. We are particularly interested in candidates that have managed multifaceted, interactive research programs using hypothesis-based, innovative approaches to address important health-related areas. Candidates with a sustained record of NIH funding will be given the highest priority. The appropriate candidate will be recruited at the level of Associate/Full Professor with qualications commensurate with tenure. This individual will play a major role in VCU SOM and will be considered for appointment to a senior administrative position in HMG. An appropriate candidate will also be considered for appointment as the Associate Director of the VIMM and/or Co-Program Leader in the MCC. HMG, VIMM and MCC provide an interactive and collaborative research and educational environment that will facilitate the training of the next generation of research scientists, clinicians and academicians, and will provide a direct conduit for the translation of genetic information from bench-to-bedside. Outstanding state-of-the-art core research facilities with a generous start-up and support package available for the qualied applicant. Richmond, VA provides an ideal rural living and cultural environment with affordable housing, outstanding school systems and ready access to other metropolitan areas (including Washington, DC, Baltimore, Philadelphia and New York). Moreover, the City of Richmond and surrounding areas offer a diverse and rich cultural heritage that engenders a high quality of living for its residents. Interested candidates should provide by e-mail (preferably as a single PDF le): a letter of interest, a curriculum vitae, a description of administrative philosophy, and an outline of research interests and future research directions, with contact information for three to four references to: Dr. Paul B. Fisher ([email protected]) Department of Human and Molecular Genetics, Virginia Commonwealth University, School of Medicine 1101 East Marshall Street, Sanger Hall Building, Room 11-015, Richmond, VA 23298-0033 Review of Applications will begin February 1, 2010 and will continue till the position is lled. VCU is an EEO/AA Employer. Female, Minorities and persons with disabilities are encouraged to apply.

www.uidaho.edu

DEAN, COLLEGE OF NATURAL RESOURCES

MOSCOW, IDAHO

The University of Idaho, a leading research institution and landgrant university, seeks as its next dean of the College of Natural Resources, an experienced executive and research professional who will provide strategic and operational leadership. The university recently has implemented a strategic action plan to guide its ongoing renewal and expansion for the future and is poised to develop the next phase of strategic innovation (http://www.uidaho.edu/provost/strategicactionplan.aspx). The dean is the chief administrative ofcer of the college and reports to the provost and executive vice president. The dean has authority and responsibility for all aspects of curriculum planning and development, faculty and staff evaluation and development, budget and facilities. Minimum Qualications: The candidate must possess an earned doctorate/terminal degree from an accredited institution and credentials to hold a tenured professorship. Completed applications should include a letter of interest and qualications relative to the position description, statements on the visions for education and research appropriate for the college of a land grant university, vitae, and contact information for ve references (including names, title, address, phone number and email address). Nominations are welcome. All applications will be given full consideration. Review of applications will begin February 1, 2010 and the search process will continue until an appointment is made. Search Chairperson is Dean Scott Wood, University of Idaho, College of Science, P.O. Box 443025, Moscow, Idaho 83844-3025. Electronic online application: http://www.hr.uidaho.edu. An Equal Opportunity/Afrmative Action Employer.

UNIVERSITY OF VIRGINA HEALTH SYSTEM Center for Membrane Biology in the Department of Molecular Physiology and Biological Physics, is seeking to ll two tenure-track faculty positions. Rank will depend on qualications. Candidates must have a PhD or MD degree, at least two years of postdoctoral experience, and are expected to be competitive at a national level by recognition through peer-reviewed publications and demonstrated ability to secure competitive national grant support. The Center housed in newly constructed space is based in the Department of Molecular Physiology and Biological Physics in the School of Medicine and also draws faculty from several other Departments. UVa has a history of outstanding research in the structure and function of membranes, and current strengths include (1) the structural biology of membrane channels, transporters and receptors, (2) signal transduction in membranes, (3) viral and intracellular membrane fusion, (4) trafcking of membrane proteins, (5) cell adhesion, and (6) lipid-protein interactions. We especially encourage applicants with expertise that complements and broadens these areas, including translational research more closely related to disease. The positions offer access to state-of-the-art facilities for membrane protein production, structure determination by crystallography, NMR spectroscopy, and electron microscopy, as well as other biophysical techniques. In addition to membrane biology, the Department of Molecular Physiology and Biological Physics also has strengths in cardiovascular physiology, cancer biology, and structural biology in general. To apply, visit https://jobs.virginia.edu and search on Posting Number 0604829, complete a Candidate Prole online and attach Curriculum Vitae with publication list, cover letter, a statement of signicant research accomplishments and future research plans, and a list of three references with name, email address and phone number. Review of applications will begin January 25, 2010 and the positions will remain open to applications until lled. For additional information about the positions, please contact the chair of the search committee, Dr. Lukas K. Tamm, Director of the Center for Membrane Biology, Department of Molecular Physiology and Biological Physics, University of Virginia Health System, P.O. Box 800886, Charlottesville, VA 22908-0886. For further information about the application process, please contact Howard Phipps at [email protected]. We also recommend you send a pdf copy of your application to this email address. The University of Virginia is an Equal Opportunity/Afrmative Action Employer.

online @sciencecareers.org

online @sciencecareers.org

Arizona Biltmore | Phoenix, Arizona

march 8-9, 2010

personalized medicine in the clinic:

policy, legal, and ethical implications

This national conference with top experts will examine the impact of personalized medicine on the delivery of healthcare in the future. Conference highlights:
patient rights medical privacy and condentiality ethics individualized medical care economics liability issues for physicians

For CLE and CME information and to register, visit www.law.asu.edu/personalizedmedicine2010. To become a conference supporter, call 480.965.2465.

Conference co-sponsors:
CENTER FOR THE STUDY OF

L AW, SCIENCE , & TECHNOLOGY

AWARDS
ROBERT J. AND CLAIRE PASAROW FOUNDATION 23rd Annual MEDICAL RESEARCH AWARDS CALL FOR NOMINATIONS
Cancer, Cardiovascular Disease, Neuropsychiatry
The Foundation has established three yearly medical prizes for distinguished accomplishment in research in order to increase public awareness of vital areas of investigation. This is the twenty-third year of the awards program. Each award is for $50,000, presented directly to the awardee. The three prizes one in each of the three elds are given for extraordinary basic and/or clinical or translational research. Cancer: including basic cellular processes and the various forms of cancer. Past awardees: Peter K. Vogt, PhD; Irving L. Weissman, MD; George F. Vande Woude, PhD; Erkki Ruoslahti, MD; Harold N. Weintraub, MD, PhD; Ronald M. Evans, PhD; Stanley J. Korsmeyer, MD; Carlo M. Croce, MD; Alfred G. Knudson, Jr., MD, PhD; Robert A. Weinberg, MD; Eric S. Lander, D.Phil.; Paul L. Modrich, PhD; Anthony S. Fauci, MD; Alexander J. Varshavsky, PhD; Tom Maniatis, PhD; Roger D. Kornberg, PhD; Elizabeth H. Blackburn, PhD; Fred W. Alt, PhD; Bert OMalley, MD; Tony Hunter, PhD; and Bert Vogelstein, MD Cardiovascular Disease: including disorders of the heart and vascular system. Past awardees: Burton E. Sobel, MD; Harvey Feigenbaum, MD; Bernardo Nadal-Ginard, MD, PhD; Mordecai P. Blaustein, MD; Jonathan Seidman, PhD and Christine Seidman, MD; Glenn A. Langer, MD; Philip Majerus, MD; Jan L. Breslow, MD; Kenneth R. Chien, MD, PhD; Michael A. Gimbrone, Jr., MD; Masashi Yanagisawa, MD, PhD; Mark T. Keating, MD; Eric N. Olson, PhD; Richard P. Lifton, MD, PhD; Robert J. Lefkowitz, MD; Shaun Coughlin, MD, PhD; Judah Folkman, MD; Barry S. Coller MD; Douglas C. Wallace, PhD; Daniel Steinberg, MD, PhD; and Richard O. Hynes, PhD Neuropsychiatry: including neurologic and mental disorders. Past awardees: Nancy Wexler, PhD; Eric R. Kandel, MD; Floyd E. Bloom, MD; Solomon H. Snyder, MD; Michael E. Phelps, PhD; Patricia S. Goldman-Rakic, PhD; Huda Akil, PhD and Stanley Watson, MD, PhD; Arvid Carlsson, MD, PhD and Philip Seeman, MD, PhD; Stanley B. Prusiner, MD; Joseph T. Coyle, MD; Eric J. Nestler, MD, PhD; Fred H. Gage, PhD; Michael I. Posner, PhD and Marcus E. Raichle, MD; Pasko Rakic, MD, PhD; Seymour Benzer, PhD; Tomas Hkfelt, MD, PhD; Thomas M. Jessell, PhD; Judith L. Rapoport, MD; Bruce McEwen, PhD; Huda Y. Zoghbi, MD; and Aaron T. Beck, MD
This booklet is brou the AAAS/Science ght to you by Business Ofce

Download your free copy today.

ScienceCareers.org/ booklets

CAREER Careers Away TRENDS from the Bench

Advice and Options for Scientists

The criterion for the Pasarow Medical Research Awards is evidence of extraordinary accomplishment in medical science. Nominators for the 23rd Award should provide a letter of no more than one page stating the rationale for the nomination and a copy of the nominees curriculum vitae and bibliography in two-page NIH format. Applications will be reviewed by the Board of Directors in consultation with various medical scholars. Members of the Board of Directors are Michael Pasarow, Chairman; Anthony H. Pasarow, Co-Treasurer; Susan A. Pasarow, MSW, Co-Treasurer; Jack D. Barchas, MD, President; Shaun Coughlin, MD, PhD University of California, San Francisco; Ronald M. Evans, PhD The Salk Institute; Brian E. Henderson, MD University of Southern California; Joseph P. Van Der Meulen, MD University of Southern California, and Alexander J. Varshavsky, PhD CalTech. Nominations should be sent to: Robert J. and Claire Pasarow Foundation, c/o Jack D. Barchas, MD; Weill Cornell Medical College, 1300 York Avenue, Box 171, Room F1231, New York, NY 10065. For more information, please contact Jack D. Barchas, MD at (212) 7463770 or [email protected]. Nominations should be received by Monday, February 10, 2010. The Pasarow Foundation congratulates two past Pasarow award winners for their stunning achievements in 2009: Elizabeth H. Blackburn, PhD (recipient of the 17th Pasarow Award in Cancer Research) - Along with Drs. Carol W. Greider and Jack W. Szostak, recipient of The Nobel Prize in Physiology or Medicine 2009 for their fundamental research which led to the discovery of how chromosomes are protected by telomeres and the enzyme telomerase. Michael I. Posner, PhD (recipient of the 13th Pasarow Award in Neuropsychiaty with Marcus Raichle, MD) received the 2009 Presidents National Medal of Science for incisive and accurate modeling of functional tasks, and his development of methodical and conceptual tools to help understand the mind and the development of brain networks of attention.

From technology specialists to patent attorneys to policy advisers, learn more about the types of careers that scientists can pursue and the skills needed in order to succeed in nonresearch careers.

meetings and announcements

online @sciencecareers.org

online @sciencecareers.org

POSITIONS OPEN

POSITIONS OPEN

POSITIONS OPEN

Marshall University is seeking a tenure-track ASSISTANT PROFESSOR of vertebrate evolutionary biology in the Department of Biological Sciences, College of Science. Required qualifications include a Ph.D. in biology, paleontology or related discipline; relevant postdoctoral experience; experience teaching human anatomy; and a record of research productivity. Demonstrated excellence in teaching and outstanding communication skills are desired. The successful candidate will participate in undergraduate and graduate (M.S/M.A) teaching and research mentorship, and will be expected to establish an active, externally funded research program. The primary teaching responsibility will be a service course in human anatomy without cadaver dissection. We also seek a colleague who will support our interdisciplinary programs and contribute to the university_s general education curriculum with its emphasis on First-Year Seminar and core courses, which are intended to enhance students_ critical thinking. The desired area of research is broadly defined as vertebrate evolutionary biology, but should complement existing departmental strengths in vertebrate paleobiology, paleontology, mammalogy, and herpetology. Outstanding applications from other related fields will also be considered. Interested applicants should send cover letter, current curriculum vitae, statements of research interests and teaching philosophy, selected electronic reprints, university transcripts (official transcripts will be requested of finalists), and request three letters of reference sent to: e-mail: [email protected]. Please assemble all documents, except letters of reference, into a single PDF file. Electronic submission is preferred, but paper application materials may be sent to: Dr. F. Robin O_Keefe, Department of Biological Sciences, Marshall University, One John Marshall Drive, Huntington, WV 25755. Application deadline: February 15, 2010. Marshall University is an Affirmative Action Equal Opportunity Employer dedicated to increasing the diversity of its faculty and students. ASSISTANT PROFESSOR BIOLOGY Saint Francis University invites applications for an Assistant Professor to begin August 2010. The candidate should hold a Ph.D. in physiology or related field. Applicants must have a strong commitment to undergraduate education and research. Preference will be given to candidates with teaching experience. The selected applicant is expected to teach courses in human anatomy and physiology, vertebrate physiology, introductory biology, and other courses relating to candidate_s background. Candidates should send their curriculum vitae, undergraduate and graduate transcripts, three letters of recommendation, statements of teaching philosophy and research interests to: Search Committee for Assistant Professor of Biology, c/o Office of Human Resources, Saint Francis University, P.O. Box 600, Loretto, PA 15940. E-mail: [email protected]. Review of applicants will begin January 8, 2010. Affirmative Action/Equal Opportunity Employer. POSTDOCTORAL RESEARCHER POSITIONS in Cardiovascular Disease and Tuberculosis Candidates with molecular, cell biological, or biochemical skills needed to work on problems relevant to tuberculosis and cardiovascular disease at the Burnett School of Biomedical Sciences, College of Medicine, University of Central Florida, Orlando, Florida. Send curriculum vitae and e-mail addresses of at least three references to P.E. Kolattukudy, e-mail: pk@mail. ucf.edu. The University of Central Florida is an Equal Opportunity, Equal Access, and Affirmative Action Employer. As a member of the Florida State University System, all application materials and selection procedures are available for public review.

MEDICINAL CHEMIST The University of Tennessee College of Pharmacy has a 12-month, tenure-track (ASSISTANT, ASSOCIATE, or FULL PROFESSOR) position available. Candidates must have a Ph.D. degree in medicinal chemistry or a related area and an externally funded research program. The candidate must have an interest in teaching medicinal chemistry at the Pharm.D. and graduate levels and possess good communication skills. All areas of medicinal chemistry and related disciplines will be considered, but preference will be given to candidates with research programs in chemical biology and anti-infective agents. The position offers startup funds, an attractive research incentive plan, and a competitive salary and benefits package commensurate with experience. The College will be moving into a new building (183,857 gross square feet) in 2010. Applications are being accepted. Applicants should send a letter, including a summary of future research plans, curriculum vitae, and three letters of reference to: Isaac Donkor, Ph.D., Professor and Vice Chair, Department of Pharmaceutical Sciences, 847 Monroe Avenue, Suite 327, Memphis, TN 38163. Applications will be accepted until the position is filled. The University of Tennessee is an Equal Employment Opportunity/Affirmative Action/Title VI/Title IX/Sect. 504/ADA/ADEA Employer.

VACCINE DEVELOPMENT AND BACTERIAL PATHOGENESIS POSTDOCTORAL FELLOW position is available immediately to join the research group of Dr. James E. Samuel studying Coxiella burnetii. Individual will be primarily responsible for conducting collaborative research on vaccine development and publication of results. Research will emphasize the immune response to protective antigens in Q fever using mouse and guinea pig virulence models. Additional opportunities for investigation include novel genetic approaches for molecular pathogenesis with the agent. Ph.D. required and a record of productive experience in immunology, animal models, and/or with bacterial pathogens preferred. Send curriculum vitae and names and addresses of three references to: Dr. James E. Samuel, Department Microbial and Molecular Pathogenesis, Texas A&M Health Science Center, MS 1114, 407 Reynolds Medical Building, College Station, TX 77843-1114. Fax: 979-845-3479; e-mail: jsamuel@ medicine.tamhsc.edu. Contact Dr. Samuel, telephone: 979-862-1684 for additional information. The Texas A&M Health Science Center is an Affirmative Action/ Equal Opportunity Employer.

Washington State University Tri-Cities invites applications for a full-time, tenure-track ASSISTANT PROFESSOR position in applied microbiology with an appointment expected in the School of Molecular Biosciences at WSU. The successful candidate will be expected to teach at the undergraduate and graduate levels and should have a background that will allow broad participation in the teaching program. Expected research contributions in applied microbiology are summarized below. We encourage applications from individuals with demonstrated expertise, as evidenced by peer-reviewed publications in the fields of cellular and molecular microbiology (e.g., applied, environmental, pathogenic). Preference will be given to applicants whose research complements current research strengths within the Center for Bioproducts and Bioenergy and to the university_s Viticulture and Enology program. The campus also has a strong partnership with the Pacific Northwest National Laboratory. A competitive candidate will have a Ph.D. in microbiology or a closely related field, ability to teach undergraduate and graduate courses, expertise in applied microbiology, a record of research productivity with the potential to obtain extramural support, and ability to work with multidisciplinary teams. Additional information is available for position #5272 at website: http://www. wsujobs.com. Interested individuals should submit electronically: (1) a cover letter discussing training and experience as related to the required and desired qualifications; (2) curriculum vitae; and (3) the names and contact information for three references. Review of completed applications will begin on January 29, 2010. Applications should be submitted to: Microbiology Search, c/o Bonnie Bates (e-mail: bbates@ tricity.wsu.edu), Washington State University TriCities, 2710 University Drive, Richland, WA 99354. Washington State University is an Equal Opportunity/ Affirmative Action Educator and Employer. Members of groups underrepresented in science are encouraged to apply.

Find your future here.

POSTDOCTORAL FELLOWSHIP The Lillehei Heart Institute is a premier institute at the University of Minnesota, Minneapolis with state-of-the-art technologies and core facilities focused on the molecular regulation of myogenesis, stem cell and regenerative biology. We are accepting applications for highly motivated Postdoctoral Fellows to work on NIH-funded research pertaining to the role of transcription factor mediated networks to direct the fate of stem and iPS cells to a mesodermal fate (i.e., cardiac, endothelial, skeletal muscle). Ph.D. and expertise with molecular biologicalbiochemistry techniques, the use of transgenesis and knockout technologies will be required for this position. Interested applicants should apply online to: Daniel J. Garry, M.D., Ph.D., Director of Lillehie Heart Institute, University of Minnesota at website: http://employment. umn.edu (reference job search #161786) and include curriculum vitae, statement of interest, and names of three references. The University of Minnesota is an Affirmative Action/Equal Opportunity Employer.

Help employers nd you. Post your resume/cv.

www.ScienceCareers.org

www.ScienceCareers.org

110

1 JANUARY 2010

VOL 327

SCIENCE

www.sciencecareers.org

Evolution

In America today, 1 in 3 individuals does not accept evolution.1 Thats why AAAS continues to play an important role in the effort to protect the integrity of science education. AAAS is hard at work ensuring that evolution continues to be taught in science classrooms, but we need your help. Join us. Together we can make a difference. aaas.org/plusyou/evolution
1.

Pew Research Center for the People & the Press. May 2009, General Public Science Survey.