CONTENTS Volume 333 Issue 6046

page 1095
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1061
COVER
False-colored backscattered electron image of a small rocky
particle, 150 micrometers in size, recovered from the surface of
near-Earth asteroid 25143 Itokawa by the Japanese spacecraft
Hayabusa. Mineralogy and chemistry of the particle indicate that
the asteroid consists of very primitive solar system material and
has had a complex formation history. See the series of Reports
beginning on page 1113.
Image: Japanese Aerospace Exploration Agency
DEPARTMENTS
1067 This Week in Science
1072 Editors’ Choice
1074 Science Staff
1108 AAAS News & Notes
1167 New Products
1168 Science Careers
page 1084
EDITORIAL
1071 Much More Than a Telescope
Michael S. Turner
NEWS OF THE WEEK
1076 A roundup of the week’s top stories
NEWS & ANALYSIS
1079 U.S. to Leave Consortium
and Go It Alone After 2013
1080 Cambridge Seeks Global Reach,
Says Univeristy Head
1081 Hayabusa Gets to the Bottom of
Deceptive Asteroid Cloaking
>> Report p. 1121
1083 8.7 Million: A New Estimate for
All the Complex Species on Earth
NEWS FOCUS
1084 Who Were the Denisovans?
A Denisovan Legacy in the Immune System?
>> Science Express Report by
L. Abi-Rached et al.; Science Podcast
1088 Mending the Youngest Hearts
1090 Europe’s Innovation Engine,
Eager to Grow, Faces Criticism
LETTERS
1092 Endangered Wolves Fall Prey to Politics
B. J. Bergstrom
Predators’ Effects on Ecosystem Entropy
W. T. Flueck
Looking to NSF as an NIH Model
W. A. Yost
1093 TECHNICAL COMMENT ABSTRACTS
1093 CORRECTIONS AND CLARIFICATIONS
BOOKS ET AL.
1094 The Changing Body
R. Floud et al., reviewed by L. Haddad
1095 Cricket Radio
J. Himmelman, reviewed by M. Berenbaum
EDUCATION FORUM
1096 Drawing to Learn in Science
S. Ainsworth et al.
>> Science Podcast
PERSPECTIVES
1098 Bringing Part of an Asteroid
Back Home
A. N. Krot
>> Reports pp. 1113, 1116, 1119, 1121,
1125, and 1128
1099 Arranging a Cellular Checkerboard
W. Choi and M. Peifer
>> Report p. 1144
1100 How Great Wings Can Look Alike
S. B. Carroll
>> Report p. 1137
1102 Another NOTCH for Cancer
R. H. Brakenhoff
>> Reports pp. 1154 and 1157
1103 A Window on the Sophistication
of Plants
R. A. Jorgensen
>> Report p. 1141
1104 Architecturally Complex Polymers
with Controlled Heterogeneity
K. Matyjaszewski
SCIENCE PRIZE ESSAY
1106 Astronomical Perspectives for
Young Children
C. J. Ödman-Govender and D. Kelleghan
CONTENTS continued >>
Published by AAAS
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1063
CONTENTS
pages 1099 & 1144
pages 1102, 1154, & 1157
page 1131
REVIEW
1109 Mitochondria and the Autophagy–
Inflammation–Cell Death Axis
in Organismal Aging
D. R. Green et al.
REPORTS
1113 Itokawa Dust Particles: A Direct Link
Between S-Type Asteroids and Ordinary
Chondrites
T. Nakamura et al.
>> Science Podcast
1116 Oxygen Isotopic Compositions of
Asteroidal Materials Returned from
Itokawa by the Hayabusa Mission
H. Yurimoto et al.
1119 Neutron Activation Analysis of a Particle
Returned from Asteroid Itokawa
M. Ebihara et al.
1121 Incipient Space Weathering Observed
on the Surface of Itokawa Dust Particles
T. Noguchi et al.
>> News story p. 1081
1125 Three-Dimensional Structure of
Hayabusa Samples: Origin and
Evolution of Itokawa Regolith
A. Tsuchiyama et al.
1128 Irradiation History of Itokawa Regolith
Material Deduced from Noble Gases
in the Hayabusa Samples
K. Nagao et al.
Laboratory analysis of samples returned
from an asteroid establishes a direct link
between asteroids and meteorites and
provides clues to the complex history
of the asteroid and its surface.
>> Perspective p. 1098
1131 Synthesis and Structure Determination
of the Hierarchical Meso-Microporous
Zeolite ITQ-43
J. Jiang et al.
A zeolite with microporous channels (6 to 7
angstrom diameter) and mesoporous channels
(~2-nanometer diameter) was made.
1134 Unraveling the Perplexing Structure
of the Zeolite SSZ-57
C. Baerlocher et al.
X-ray analysis reveals a zeolite structure
in which 10-sided channels are periodically
disrupted by 12-sided channels.
1137 optix Drives the Repeated Convergent
Evolution of Butterfly Wing Pattern
Mimicry
R. D. Reed et al.
Heliconius butterfly wing pattern mimicry
is driven by cis-regulatory variation
of the optix gene.
>> Perspective p. 1100
1141 Chaperonins Facilitate KNOTTED1
Cell-to-Cell Trafficking and
Stem Cell Function
X. M. Xu et al.
Active transfer of a transcription factor
between plant cells is required for
stem cell maintenance.
>> Perspective p. 1103
1144 Nectins Establish a Checkerboard-Like
Cellular Pattern in the Auditory Epithelium
H. Togashi et al.
Interactions between adhesion molecules
support the checkerboard-like patterning
of cells in the cochlea.
>> Perspective p. 1099
1147 Impacts of Fishing Low–Trophic Level
Species on Marine Ecosystems
A. D. M. Smith et al.
High harvest levels of low–trophic level fishes
may have cascading marine ecosystem effects.
1151 Expanding the Genetic Code of
Escherichia coli with Phosphoserine
H.-S. Park et al.
Engineered bacterial translation can be
used to direct site-specific insertion
of an amino acid into proteins.
1154 Exome Sequencing of Head and Neck
Squamous Cell Carcinoma Reveals
Inactivating Mutations in NOTCH1
N. Agrawal et al.
1157 The Mutational Landscape of Head
and Neck Squamous Cell Carcinoma
N. Stransky et al.
The mutational profile of head and neck
cancer is complex and may pose challenges
to the development of targeted therapies.
>> Perspective p. 1102
1161 MED23 Mutation Links Intellectual
Disability to Dysregulation of
Immediate Early Gene Expression
S. Hashimoto et al.
A single missense mutation is linked
to intellectual impairment.
CONTENTS continued >>
Published by AAAS
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1065
CONTENTS
SCIENCEXPRESS
www.sciencexpress.org
Promoting the Peace Process by Changing Beliefs
About Group Malleability
E. Halperin et al.
A belief that the beliefs of other groups are
changeable rather than fixed is conducive
to negotiation.
10.1126/science.1202925
The Shaping of Modern Human Immune Systems
by Multiregional Admixture with Archaic Humans
L. Abi-Rached et al.
Viral defense and embryo implantation mechanisms
have been shaped by contributions from Neandertal
and Denisovan genes.
10.1126/science.1209202
>> News story p. 1084
MED12, the Mediator Complex Subunit 12
Gene, Is Mutated at High Frequency
in Uterine Leiomyomas
N. Mäkinen et al.
Uterine fibroids frequently harbor mutations
in a specific gene that has been implicated
in transcriptional regulation.
10.1126/science.1208930
Transformation of a Star into a Planet
in a Millisecond Pulsar Binary
M. Bailes et al.
Timing observations of a millisecond pulsar reveal
a planet that is far denser than any known planet.
10.1126/science.1208890
A Reservoir of Ionized Gas in the Galactic Halo
to Sustain Star Formation in the Milky Way
N. Lehner and J. C. Howk
Clouds of ionized gas located inside our galaxy
provide a major supply of matter for fueling
ongoing star formation.
10.1126/science.1209069
TECHNICALCOMMENTS
Comment on “Drought-Induced Reduction
in Global Terrestrial Net Primary Production
from 2000 Through 2009”
A. Samanta et al.
Full text at www.sciencemag.org/cgi/content/
full/333/6046/1093-c
Comment on “Drought-Induced Reduction
in Global Terrestrial Net Primary Production
from 2000 Through 2009”
B. E. Medlyn
Full text at www.sciencemag.org/cgi/content/
full/333/6046/1093-d
Response to Comments on “Drought-Induced
Reduction in Global Terrestrial Net Primary
Production from 2000 Through 2009”
M. Zhao and S. W. Running
Full text at www.sciencemag.org/cgi/content/
full/333/6046/1093-e
SCIENCEONLINE
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 © 2011 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): $149 ($74 allocated to subscription).
Domestic institutional subscription (51 issues): $990; 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 20090–6178. 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 $25.00 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
Reader’s Guide to Periodical Literature and in several specialized indexes.
SCIENCENOW
www.sciencenow.org
Highlights From Our Daily News Coverage
‘Eco-Friendly’ Chilean Sea Bass
May Not Be So Green
A genetic study finds the first evidence of
mislabeling in fish certified as sustainable.
http://scim.ag/ecoseabass
Immune System Protects Female Bedbugs
From Traumatic Sex
Insects anticipate infections from violent
male insemination.
http://scim.ag/bedbugbacteria
Lager Beer’s Mystery Yeast
Scientists identify a microbe that is key
to the brewing process.
http://scim.ag/lageryeast
SCIENCESIGNALING
www.sciencesignaling.org
The Signal Transduction Knowledge Environment
23 August issue: http://scim.ag/ss082311
RESEARCH ARTICLE: Tumor Progression Locus 2
Mediates Signal-Induced Increases in Cytoplasmic
Calcium and Cell Migration
M. Hatziapostolou et al.
The kinase Tpl2 triggers calcium signaling and cell
migration downstream of various receptors implicated
in cancer and inflammation.
PERSPECTIVE: TBK1 Mediates Crosstalk Between
the Innate Immune Response and Autophagy
H. Weidberg and Z. Elazar
Phosphorylation of an autophagy adaptor by a
kinase involved in innate immune responses
limits pathogenic bacterial growth.
PODCAST
H. D. Mansvelder and A. M. VanHook
β2 subunit–containing nicotinic acetylcholine
receptors in the prefrontal cortex are sufficient
for acetylcholine-mediated enhancement of
attentional focus.
SCIENCECAREERS
www.sciencecareers.org/career_magazine
Free Career Resources for Scientists
Experimental Error: This Is Only a Test
A. Rubin
As we are training to become fully fledged
scientists, we ourselves are the test subjects.
http://scim.ag/EE_Testing
Creating Your Own Job: From Engineer
to Entrepreneur
S. A. Holgate
A layoff set engineer Diana Hodgins on a new
career path, leading eventually to managing
her own company.
http://scim.ag/Diana_Hodgins
In Person: A Sabbatical in Namibia
L. A. Levin and D. M. Checkley
On the heels of our feature on doing science in
Namibia, two scientists describe their sabbatical
experiences there.
http://scim.ag/Namibia_Sabbatical
SCIENCETRANSLATIONAL MEDICINE
www.sciencetranslationalmedicine.org
Integrating Medicine and Science
24 August issue: http://scim.ag/stm082411
COMMENTARY: Herding CATS
G. A. Petsko
The National Center for Advancing Translational
Sciences seems primed to polarize scientists while
possibly failing to drive biomedical research and
development forward.
RESEARCH ARTICLE: Long-Term Persistence of a
Polyclonal T Cell Repertoire After Gene Therapy
for X-Linked Severe Combined Immunodeficiency
H. B. Gaspar et al.
RESEARCH ARTICLE: Hematopoietic Stem Cell
Gene Therapy for Adenosine Deaminase–
Deficient Severe Combined Immunodeficiency
Leads to Long-Term Immunological Recovery
and Metabolic Correction
H. B. Gaspar et al.
PERSPECTIVE: A Tale of Two SCIDS
K. L. Shaw and D. B. Kohn
Gene therapy can restore immune and metabolic
function in immunodeficient patients.
RESEARCH ARTICLE: Potent Kinetic Stabilizers
That Prevent Transthyretin-Mediated
Cardiomyocyte Proteotoxicity
M. M. Alhamadsheh et al.
A screen for transthyretin ligands that inhibit protein
aggregation identifies agents that prevent amyloid
toxicity in cardiomyocytes.
SCIENCEPODCAST
www.sciencemag.org/multimedia/podcast
Free Weekly Show
On the 26 August Science Podcast: dust particles
from asteroid Itokawa, new insights into the
Denisovans, drawing to learn in science, and more.
SCIENCEINSIDER
news.sciencemag.org/scienceinsider
Science Policy News and Analysis
Published by AAAS
1067
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

A
R
N
A
U
D

M
A
R
T
I
N
;

B
A
E
R
L
O
C
H
E
R

E
T

A
L
.
Continued on page 1069
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011
Mature Mitochondria?
Mitochondria are the essential source of meta-
bolic energy for the cell. These organelles direct
crossroads from which cells can move down a
path that leads to cell death or to inflammatory
damage to an organism. They are thus important
factors of certain diseases and, more universally,
of the aging process. Green et al. (p. 1109)
review recent insights into mitochondrial func-
tion, their roles in the control of cell death and in
chronic inflammation, and the role of autophagy
or, more specifically, mitophagy in protecting
cells from mitochondrial damage.
Extraterrestrial
Dust Collection
The Japanese Hayabusa spacecraft returned to
Earth on 13 June 2010 carrying within it a pre-
cious but invisible cargo—1534 particles of dust
from the surface of the asteroid Itokawa, which
the spacecraft visited in 2005. These particles, up
to 180 micrometers in size, are the first material
ever fetched from an asteroid and returned to
Earth for laboratory analysis (see the Perspec-
tive by Krot; see the cover). Nakamura et al. (p.
1113) describe the mineralogy of the particles;
Yurimoto et al. (p. 1116) report measurements
of the oxygen isotopic composition of minerals
in 28 dust grains; and Ebihara et al. (p. 1119)
discuss trace-element data for one of the largest
grains in the collection. These studies confirm that
the material is indeed extraterrestrial in origin
and is similar to material in stony meteorites of
the LL type, thus establishing an unequivocal link
between asteroids and meteorites. Previously,
meteorites were thought to originate from
asteroids, based on spectroscopic studies of aster-
oid surfaces. However, these surfaces suffer from
the effects of space weathering, which modifies
their spectral properties compared with those of
meteorites. Noguchi et al. (p. 1121) show that
5 of the 10 particles analyzed show evidence of
surface alteration related to their exposure to the
space environment on the surface of Itokawa and
conclude that the process is different from the one
operating on the Moon’s surface. The particles col-
lected by Hayabusa sample the asteroid’s regolith,
the layer of unconsolidated material that covers
rocky surfaces. Based on the texture and shapes of
the particles and on noble gas analyses of three
grains, Tsuchiyama et al. (p. 1125) and Nagao
et al. (p. 1128) trace the history of Itokawa’s
regolith, the second extraterrestrial regolith to
have been sampled after the Moon’s.
Zeolites Up Close
In the preparation of microporous zeolitic
materials for catalysis, the introduction of hier-
archical porosity has been
of great interest. Because
pores have multiple-size
scales, it should be possible
to enhance catalytic activ-
ity by controlling dif fusion
characteristics. Jiang et
al. (p. 1131) developed a
zeolitic topology in which
this hierarchical porosity is a
natural feature of the material through the use
of organic and inorganic structure-directing
agents. Zeolites vary widely in the size and
arrangement of their pore structures, but their
lattice geometries can generally be described by
a relatively simple replicated core arrangement.
Baerlocher et al. (p. 1134) used sophisticated
analysis of x-ray diffraction data to uncover the
structure of an unexpectedly complex zeolite
variant, in which the unit cell comprises 99
silicon atoms (compared with ~20 or fewer
observed in conventional zeolites).
Getting There Is
Only Half the Fun
Although most transcription factors are syn-
thesized in the same cell in which they act, in
plants some transcription factors get shipped
into other cells, where they take action on the
genes of the destination cell. The transcrip-
tion factors KNOTTED1 (KN1) of maize and the
closely similar SHOOTMERISTEMLESS (STM)
of Arabidopsis, as well as SHORTROOT (SHR)
of Arabidopsis, are a few of these peripatetic
regulators. These factors need to travel through
the plasmodesmata that connect one cell to
another. Xu et al. (p. 1141; see the Perspective
by Jorgensen) used a reporter system
in Arabidopsis that depends on the
transport of the transcription factor to
direct trichome development. A spe-
cific chaperonin complex was required
in the recipient cell for full function of
KN1 or STM, which highlights the need
for refolding of these transcription
factors after transport.
Playing Checkers
Developmental patterning is very important in
the arrangements of cells within organs and tis-
sues. The organ of Corti in the inner ear contains
<< Seeing Red Across Species
The phenomenon of mimicry in butterflies has generated wide-spread
interest among naturalists and evolutionary biologists, but a devel-
opmental and genetic understanding of the process requires the
identification of the individual genes involved Reed et al. (p. 1137,
published online 21 July; see the Perspective by Carroll) identified
the optix gene as the major determinant of red color patterns in mul-
tiple Heliconius butterfly species by tying gene expression data to
color-pattern variation. Furthermore, cis-regulatory variation is likely
responsible for the diversity of red patterns across species. The optix
gene encodes a transcription factor that is also associated with the
butterfly visual system and may thus have been co-opted from a dif-
ferent ancestral function.
EDITED BY STELLA HURTLEY
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

This Week in Science
C
R
E
D
I
T
:

P
A
R
K

E
T

A
L
.
mechanosensory hair cells that are interdigitated with nonsensory supporting cells. During devel-
opment, these two cell types are arranged in a checkerboard-like pattern. Togashi et al. worked
in mice (p. 1144, published online 28 July; see the Perspective by Choi and Peifer) to show that
the immunoglobulin-like cell-cell adhesion molecules nectin-1 and nectin-3 play a key role in this
checkerboard-like patterning through their heterophilic interactions. Mice lacking either nectin-1 or
-3 fail to organize their auditory epithelia properly. Thus, differential cell adhesion can control the
spatial pattern of cells in a tissue.
Little Fish, Big Fish
Low–trophic level species in marine ecosystems are responsible for transmitting energy from primary
producing phytoplankton to higher trophic levels. These species are also heavily harvested and
make up 30% of global fisheries’ production. Smith et al. (p. 1147, published online 21 July)
used ecosystem models to explore the impacts that an increased harvest of low–tropic level species
could have on marine ecosystems across the globe. Fishing these species at levels considered to
be the maximum for sustaining the species can have large impacts on other ecosystem members
(including fish, seabirds, and marine mammals). Halving exploitation rates has the potential to
provide an adequate allocation for human use and to greatly reduce the ecosystem impacts of
harvesting these essential species.
Protein Phosphorylation by Design
O-phosphoserine (Sep), the most abundant phosphoamino
acid, is not genetically encoded but synthesized posttransla-
tionally by a complex network of kinases and phosphatases.
Park et al. (p. 1151) engineered an Escherichia coli strain
that was able to incorporate Sep directly into proteins. The
strain harbors a Sep-accepting transfer RNA (tRNA
Sep
), its
cognate Sep–tRNA synthetase, and an elongation factor–Tu
engineered to permit Sep-tRNA
Sep
binding. The system was
used to synthesize a human kinase with one or two Sep resi-
dues inserted at specific sites. The ability to biosynthesize specific phosphoproteins directly will be
very helpful for detailed studies of their enzyme and substrate properties.
The Downside of Diversity
Head and neck squamous cell carcinoma (HNSCC) affects about 600,000 individuals each year
and has a mortality rate of about 50%. Environmental factors such as tobacco and alcohol use and
human papillomavirus (HPV) infection are key participants in pathogenesis (see the Perspective by
Brakenhoff). In independent studies aimed at exploring the molecular genetics of these tumors,
Agrawal et al. (p. 1154, published online 28 July) and Stransky et al. (p. 1157, published online
28 July) sequenced the protein-coding genes of multiple tumors. Tumors from smokers had more
mutations than those from nonsmokers, and tumors that were HPV-positive had fewer mutations
than HPV-negative tumors. HNSCCs harbored mutations in a diverse array of genes, including genes
implicated in squamous differentiation such as NOTCH1. Notably, the pattern of NOTCH1 mutations
suggests that this gene acts as a tumor suppressor in HNSCC, in direct contrast to its role as an on-
cogene in other tumor types. The diverse mutational etiology of HNSCC and the dearth of activating
mutations in established oncogenes suggest that targeted therapies for the disease will be especially
challenging, which emphasizes the importance of prevention and early detection.
Mediator Missense
The Mediator complex is a multiprotein complex that functions as a transcriptional coactivator. Mis-
sense mutations in the Mediator complex can alter its activity as a regulator of gene transcription
without eliminating such interactions. Hashimoto et al. (p. 1161) identified an alteration in the
MED23 gene that encodes one of the Mediator subunits, which correlates with inherited intellectual
disability in the affected family. The mutated MED23 protein causes altered regulation of the JUN
and FOS genes, which have been linked to neuronal function, thus building a chain of events from
transcriptional dysregulation to intellectual disability.
Continued from page 1067
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Much More Than a Telescope
IMAGINE THAT THE HUBBLE SPACE TELESCOPE NEVER EXISTED—THAT IT FELL VICTIM TO BUDGET
overruns, schedule slips, and the tragedy of the Challenger disaster, and was ultimately
canceled. Breakthroughs in our understanding of the universe would not have occurred,
countless young people around the world would not have been inspired, and U.S. leadership
in science and space would be far less. The United States is at a similar crossroads today.
Last month, an appropriations committee in the House of Representatives voted to termi-
nate Hubble’s successor, the James Webb Space Telescope, as part of the 2012 budget for
the National Aeronautics and Space Administration (NASA). The Senate has yet to be heard
from, but the numbers in the Budget Control Act of 2011 that passed earlier this month sug-
gest that the Senate could make the funds available to restore funding for the Webb, to the
benefit of astronomy and the nation.
The James Webb Space Telescope is a worthy successor to Hubble.
It will have 100 times Hubble’s sensitivity and operate a million miles
from Earth, allowing us to see what we can only imagine right now: the
first stars and galaxies that formed a few hundred million years after
the Big Bang, the assembly of galaxies from gas and dark matter, and
evidence of life on distant exoplanets. Like Hubble, the Webb Tele-
scope will rewrite the textbooks, inspire the next generation of scien-
tists and the public worldwide, and be a source of national pride. For
these reasons, the U.S. National Academy of Sciences’ Astronomy and
Astrophysics Decadal Survey made this project its top priority in 2000,
amd the 2010 survey made it a cornerstone for the current decade.
Last year, because of growing concerns about the Webb Telescope
project, Senator Barbara Mikulski (D-MD) called for a review. In
October, the Independent Comprehensive Review Panel reported that
the project would cost at least $1.6 billion more (assuming launch in
2015) than the NASA estimate and that there were serious management problems. On the
bright side, the panel found that money had not been wasted and that technical progress
had been excellent. So far, about $3.5 billion dollars have been spent (about half the total
cost), 75% of the parts are in fabrication or have been completed, and NASA has responded
positively to criticism in the panel’s report by formulating a new plan for launch in 2018.
To put things in perspective, the cost of Hubble in comparable dollars was about $13 billion
(not including five servicing missions).
From most of NASA’s science missions, we only hear exciting results, as it should be.
However, the real game-changers such as Hubble and the Webb Telescope have been differ-
ent. In both cases, their unplanned budget growth has not only cost the taxpayer but has also
squeezed out other important space science (including astrophysics). NASA must continue
to think big, but to avoid a repetition of this budgetary calamity in the future, it needs to rec-
ognize that these large, complex missions require special management; early investment in
technology; sufficient project reserves to cover the unexpected; and most of all, honesty with
themselves and Congress about the cost and schedule.
Following in the footsteps of the Hubble, the Webb Telescope will awe us with its discov-
eries about our place in the universe. Although the United States has key international part-
ners (Canada and the European Space Agency are contributing close to a billion dollars in
total to the Webb Telescope), and the capabilities of other nations in space are on the rise, at
present only the United States can carry out an undertaking this ambitious. The Webb Tele-
scope is more than an instrument of scientific discovery—it is a powerful symbol of U.S.
leadership in science and space. Terminating it now would save a few billion dollars, but
would be both wasteful and short-sighted. By restoring funding for the Webb in this difficult
budgetary time, Congress can let the world know that the United States still has the vision
and audacity to do things that inspire both the nation and the world.
10.1126/science.1212119
– Michael S. Turner
1071
EDITORIAL
C
R
E
D
I
T
S
:

(
T
O
P

L
E
F
T
)

J
O
E

T
U
R
N
E
R
;

(
R
I
G
H
T
)

N
A
S
A
/
E
S
A
/
M
.

L
I
V
I
O

&

H
U
B
B
L
E

2
0
T
H

A
N
N
I
V
E
R
S
A
R
Y

T
E
A
M

(
S
T
S
C
I
)
Michael S. Turner is the
director and Rauner
Distinguished Service
Professor at the Kavli
Institute for Cosmologi-
cal Physics, University
of Chicago, Chicago,
IL, and is a member of
Science’s Senior Edi-
torial Board. E-mail:
mturner@uchicago.edu.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

1072
EDITED BY KRISTEN MUELLER AND JAKE YESTON
EDITORS’CHOICE
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

S
O
E

Z
E
Y
A

T
U
N
/
R
T
R
/
N
E
W
S
C
O
M
;

F
O
T
O
S
E
A
R
C
H
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org
ing data from five national surveys conducted
between 1975 and 2007, Siikamäki attempted
to assess the value of one aspect of state parks:
recreation. About a third of the time people cur-
rently spend on nature recreation in the United
States is accounted for in state parks. Although
use of state parks has declined over the time pe-
riod covered in the surveys, the cost of running
the parks remains easily outdistanced by the rec-
reation value derived from the parks. Following
the conventional approach of valuing recreation
time at one-third of paid wage time, the value
of nature recreation in state parks reaches about
$14 billion annually, whereas annual opera-
tion costs total about $2.3 billion. Thus, these
sometimes unassuming parks build a network
that delivers ecosystem and economic value, as
well as just giving us a nice day. — PJH
Proc. Natl. Acad. Sci. U.S.A. 108, 10.1073/
pnas.1108688108 (2011).
C H E MI S T R Y
Stuffing Your Struts
The potential for using metal-organic frame-
works (MOFs) as catalysts is often frustrated by
a fundamental tradeoff. The synthetic routes for
S O C I O L O G Y
Not Contagious After All?
Studies of social influences on behavior have
led to the idea that a range of characteristics
from loneliness to obesity might be contagious.
A significant problem for the field has been to
distinguish effects due to similarities between
people (homophily) from social influence. One
strategy for doing this has been to look at
changes that occur over time. However, such
studies have been the subject of consider-
able debate, and Noel and Nyhan now add a
cautionary note. Their analyses of a model used
in past social contagion studies suggest that
previous investigations have not fully controlled
for the possibility that friendship formation and
termination are dynamic processes, and friend-
ships between people who are more similar may
tend to be more stable over time. Or to put it in
Facebook terms, friendships that are between
people who are less similar may be less stable,
and therefore may result in “unfriending.”
Homophily might thus be having a larger effect
than appreciated, and under certain conditions
could account for most of the contagion effects
observed. They conclude that this unfriending
problem renders a determination of causality
much more complicated in longitudinal social
network data. — BJ
Soc. Networks 33, 211 (2011).
P O L I C Y
The Value of State Parks
State parks in the United States can be a refuge
for wildlife, a reservoir of plant diversity, or a
source of untamed water. They are also a popular
destination for recreation. But recent economic
belt-tightening means that budgets for state
parks are going under the knife. By analyz-
G E O P H Y S I C S
To Catch a Quake
Large earthquakes leave affected populations disoriented and emergency
responders scrambling. Having fast, reliable estimates of the location and
magnitude of the mainshock and its aftershocks is vital to an effective
rapid response plan. In developing countries with the potential for large
earthquakes, installing and maintaining seismic networks to collect such
data can be expensive. The Quake-Catcher Network is a volunteer-based
seismic network that employs personal computers as low-cost seismic sta-
tions by sending seismic data collected with a small USB accelerometer
through the user’s Internet connection. Although this network had yet to
be installed in Chile before the 2010 M 8.8 Maule earthquake, volunteers
rapidly installed nearly 100 accelerometers within weeks in and around
the mainshock area, including stations set up in local government build-
ings, health stations, and homes. Chung et al. describe the network’s
ability to accurately collect and discriminate aftershock data and shaking
intensity estimates—most within 30 s of an individual event. Improve-
ments in accelerometer technology and data processing will probably
improve data quality to the point that future networks of hundreds or
thousands of stations installed in high-seismicity regions will aid first
responders in determining where most help is needed after a large earth-
quake strikes. — NW
Seismol. Res. Lett. 82, 526 (2011).
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

EDITORS’CHOICE
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011
making MOFs tend to saturate all of the coor-
dination sites of metals, even ones contained
within the linkers or struts, but catalysis by met-
als is usually enabled by the presence of unsatu-
rated sites. Shultz et al. developed a strategy
that addresses this dilemma for metallosalen
struts to use as catalysts within MOFs. Although
Mn(salen) can be incorporated into MOFs for
asymmetric epoxidations, other metals, such
as Co(salen) for asymmetric ring opening, are
unsaturated and do not form MOFs directly.
The authors describe how Mn(salen) MOFs (in
which zinc ions link the struts in a paddlewheel
arrangement) can be demetalated with H
2
O
2
and
replaced isostructurally with a wide variety of
dicationic transition-metal ions, including Co,
Cr, Cu, Zn, and reduced Mn (which recovered its
catalytic activity after reoxidation). — PDS
J. Am. Chem. Soc. 133, 10.1021/
ja204820d (2011).
C E L L B I O L O G Y
Small Changes Have
Big Consequences
Protein folding and the maintenance of a
healthy proteome often involve molecular chap-
erones of the heat shock protein 70 (Hsp70)
family. The basis of functional distinctions of
highly homologous and functionally redundant
Hsp70s is unknown. Nearly identical yeast
cytosolic Hsp70s, Ssa1p and Ssa2p, function dif-
ferently in the propagation of yeast prions and
in a vacuole-import and protein-degradation
pathway involved in regulating levels of gluco-
neogenesis enzymes. By swapping amino acids,
Sharma and Masison show that a single amino
acid difference in the nucleotide-binding regula-
tory domain of the Ssa proteins is the basis for
the functional distinction in both processes.
It seems that this small structural difference
affects regulation of Hsp70 by cofactors rather
than by altering intrinsic Hsp70 activity. Thus,
subtle changes in Hsp70 structure may have
evolved to confer functional specificity without
affecting overall Hsp70 function. — SMH
Proc. Natl. Acad. Sci. U.S.A. 10.1073/
pnas.1107421108 (2011).
P O L I C Y
Basic Implications
In 1980, the United States adopted the Bayh-
Dole Act, which allowed researchers to retain
intellectual property rights to federally funded
research, as a way to encourage the development
of technologies that might otherwise languish
in labs. Since then, the number of university
technology transfer offices has boomed, along
with patent applications and licensing income.
Despite this apparent success, some argue that
the act has co-opted universities and undermined
the overall scientific and economic enterprise,
with long-term academic pursuit of knowledge
through basic research abandoned in favor of
nearer-term profits. Thursby and Thursby exam-
ined whether basic research has indeed suffered
relative to applied research since Bayh-Dole.
They analyzed data on thousands of science
and engineering faculty from eight major U.S.
research universities spanning the years 1983
to 1999. The “basicness” of faculty research
was explored via publication analysis, on the
reasoning that “basic” journals are cited more
heavily by “applied” journals than vice versa.
They assessed faculty interest in commercial-
izing research by tracking the history of formally
disclosing potential inventions. Their model also
accounted for research funding, age, tenure, and
other influences. They found no evidence that
interest in commercialization detracted from basic
research, and in fact saw that basic research effort
increased in light of the incentive of potential
commercialization profits. — BW
Res. Policy 40, 1077 (2011).
P S Y C H O L O G Y
Human Justice
Are decisions made by judges, which are sup-
posed to be made on the basis of facts and legal
reasoning, also influenced by outside variables
such as ethnicity or even how recently the judge
had a meal? Shayo and Zussman examined
1750 small claims court rulings handed down
by more than 100 judges in Israel from 2000
to 2004; these court cases were fender-bender
traffic accidents in which the plaintiff and defen-
dant were of different ethnicities. They found a
pattern of ingroup bias: Jewish judges favored
Jewish plaintiffs, and Arab judges favored Arab
plaintiffs. By considering the times and loca-
tions of civilian fatalities during this period, the
authors decompose these decisions into half
that were arguably ethnicity-neutral and a more
biased half associated with spatial-temporal
proximity to terrorism, highlighting that a
person’s ability to favor their own social group
may affect seemingly rational judgments.
Danziger et al. analyzed 1100 parole board
decisions made by eight judges in Israel in a
10-month period. They document a bias toward
retaining the status quo (that is, denying parole)
in successive cases adjudicated in a single day.
This tendency was eliminated after each of two
food breaks, suggesting that cognitive balance
can be restored after resource depletion. — GJC
Q. J. Econ. 126, 10.1093/qje/qjr022 (2011);
Proc. Natl. Acad. Sci. U.S.A. 108, 6889 (2011).
17050 Montebello Road
Cupertino, California 95014
Email: AAASInfo@betchartexpeditions.com
www.betchartexpeditions.com
For a detailed brochure,
please call (800) 252-4910
All prices are per person twin share + air
Costa Rica
&Nicaragua
On board Star Flyer
Nov. 27–Dec. 4, 2011
Discover SanJuandel Sur
and Lago Nicaragua (the Atlantic-
Pacific crossing before the Panama
Canal) and national parks along the
coast of Costa Rica. Discover birds,
turtles, sloths and snorkel in pristine
santuaries. From $2,995 + air
Antarctica on Sale!
On board M/V Plancius
Dec. 30, 2011–Jan. 11, 2012
Explore the greatest frontier on
earth, a magnificent wilderness with
a profusion of penguins, whales,
and seals. See spectacular
mountains &glaciers which
lunge to the sea. Explore
by zodiac, foot & kayak.
Simply breathtaking!
From $5,995 quad + air
BETCHART EXPEDITIONS Inc.
AAAS is here –
bringing educational infrastructure
to the developing world.
AAAS is helping the Rwandan government
rebuild its educational infrastructure. By
providing materials such as the Project
2061 Atlas of Science Literacy, lesson
plans from Science NetLinks, and access
to Science digital libraries, AAAS is helping
the people of Rwanda work toward a future
built around science and technology. If
you’re not yet a AAAS member, join us.
Together we can make a difference.
To learn more, visit
aaas.org/plusyou/rwanda
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

1074 26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org
www.sciencemag.org
SENIOR EDITORIAL BOARD
Cori Bargmann, The Rockefeller Univ.
John I. Brauman, Chair, Stanford Univ.
Richard Losick, Harvard Univ.
Michael S. Turner, University of Chicago
BOARD OF REVIEWING EDITORS
Adriano Aguzzi, Univ. Hospital Zürich
Takuzo Aida, Univ. of Tokyo
Sonia Altizer, Univ. of Georgia
Sebastian Amigorena, Institut Curie
Angelika Amon, MIT
Kathryn Anderson, Memorial Sloan-Kettering Cancer Center
Siv G. E. Andersson, Uppsala Univ.
Peter Andolfatto, Princeton Univ.
Meinrat O. Andreae, Max Planck Inst., Mainz
John A. Bargh, Yale Univ.
Ben Barres, Stanford Medical School
Marisa Bartolomei, Univ. of Penn. School of Med.
Jordi Bascompte, Estación Biológica de Doñana, CSIC
Facundo Batista, London Research Inst.
Ray H. Baughman, Univ. of Texas, Dallas
David Baum, Univ. of Wisconsin
Yasmine Belkaid, NIAID, NIH
Philip Benfey, Duke Univ.
Stephen J. Benkovic, Penn State Univ.
Gregory C. Beroza, Stanford Univ.
Peer Bork, EMBL
Bernard Bourdon, Ecole Normale Superieure de Lyon
Ian Boyd, Univ. of St. Andrews
Paul M. Brakefield, Univ. of Cambridge
Christian Büchel, Universitätsklinikum Hamburg-Eppendorf
Joseph A. Burns, Cornell Univ.
William P. Butz, Population Reference Bureau
Gyorgy Buzsaki, Rutgers Univ.
Mats Carlsson, Univ. of Oslo
Mildred Cho, Stanford Univ.
David Clapham, Children’s Hospital, Boston
David Clary, Univ. of Oxford
J. M. Claverie, CNRS, Marseille
Jonathan D. Cohen, Princeton Univ.
Alan Cowman, Walter & Eliza Hall Inst.
Robert H. Crabtree, Yale Univ.
Wolfgang Cramer, Potsdam Inst. for Climate Impact Research
F. Fleming Crim, Univ. of Wisconsin
Jeff L. Dangl, Univ. of North Carolina
Tom Daniel, Univ. of Washington
Stanislas Dehaene, Collège de France
Emmanouil T. Dermitzakis, Univ. of Geneva Medical School
Robert Desimone, MIT
Claude Desplan, New York Univ.
Ap Dijksterhuis, Radboud Univ. of Nijmegen
Dennis Discher, Univ. of Pennsylvania
Scott C. Doney, Woods Hole Oceanographic Inst.
Jennifer A. Doudna, Univ. of California, Berkeley
Julian Downward, Cancer Research UK
Bruce Dunn, Univ. of California, Los Angeles
Christopher Dye, WHO
David Ehrhardt, Carnegie Inst. of Washington
Michael B. Elowitz, Calif. Inst. of Technology
Tim Elston, Univ. of North Carolina at Chapel Hill
Gerhard Ertl, Fritz-Haber-Institut, Berlin
Barry Everitt, Univ. of Cambridge
Paul G. Falkowski, Rutgers Univ.
Ernst Fehr, Univ. of Zurich
Tom Fenchel, Univ. of Copenhagen
Alain Fischer, INSERM
Wulfram Gerstner, EPFL Lausanne
Karl-Heinz Glassmeier, For Geophysics & Extraterrestrial
Physics, TU Braunchweig
Diane Griffin, Johns Hopkins Bloomberg School of
Public Health
Elizabeth Grove, Univ. of Chicago
Taekjip Ha, Univ. of Illinois at Urbana-Champaign
Christian Haass, Ludwig Maximilians Univ.
Steven Hahn, Fred Hutchinson Cancer Research Center
Gregory J. Hannon, Cold Spring Harbor Lab.
Martin Heimann, Max Planck Inst., Jena
Isaac Held, NOAA
James A. Hendler, Rensselaer Polytechnic Inst.
Janet G. Hering, Swiss Fed. Inst. of Aquatic
Science & Technology
Ray Hilborn, Univ. of Washington
Michael E. Himmel, National Renewable Energy Lab.
Kai-Uwe Hinrichs, Univ. of Bremen
Kei Hirose, Tokyo Inst. of Technology
David Hodell, Univ. of Cambridge
Ove Hoegh-Guldberg, Univ. of Queensland
David Holden, Imperial College
Lora Hooper, UT Southwestern Medical Ctr at Dallas
Jeffrey A. Hubbell, EPFL Lausanne
Steven Jacobsen, Univ. of California, Los Angeles
Kai Johnsson, EPFL Lausanne
Peter Jonas, Universität Freiburg
William Kaelin, Dana-Farber Cancer Inst.
Barbara B. Kahn, Harvard Medical School
Daniel Kahne, Harvard Univ.
Bernhard Keimer, Max Planck Inst., Stuttgart
Joel Kingsolver, Univ. of North Carolina at Chapel Hill
Robert Kingston, Harvard Medical School
Alberto R. Kornblihtt, Univ. of Buenos Aires
Leonid Kruglyak, Princeton Univ.
Mitchell A. Lazar, Univ. of Pennsylvania
David Lazer, Harvard Univ.
Virginia Lee, Univ. of Pennsylvania
Ottoline Leyser, Cambridge Univ.
Olle Lindvall, Univ. Hospital, Lund
Marcia C. Linn, Univ. of California, Berkeley
John Lis, Cornell Univ.
Jianguo Liu, Michigan State Univ.
Richard Losick, Harvard Univ.
Jonathan Losos, Harvard Univ.
Ke Lu, Chinese Acad. of Sciences
Laura Machesky, CRUK Beatson Inst. for Cancer Research
Andrew P. MacKenzie, Univ. of St Andrews
Anne Magurran, Univ. of St Andrews
Oscar Marin, CSIC & Univ. Miguel Hernández
Charles Marshall, Univ. of California, Berkeley
Martin M. Matzuk, Baylor College of Medicine
Grahma Medley, Univ. of Warwick
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.
Pär Nordlund, Karolinska Inst.
Helga Nowotny, European Research Advisory Board
Luke O'Neill, Trinity College, Dublin
Stuart H. Orkin, Dana-Farber Cancer Inst.
Christine Ortiz, MIT
Elinor Ostrom, Indiana Univ.
Andrew Oswald, Univ. of Warwick
Jane Parker, Max-Planck Inst. of Plant Breeding Research
Donald R. Paul, Univ. of Texas at Austin
P. David Pearson, Univ. of California, Berkeley
Reginald M. Penner, Univ. of California, Irvine
John H. J. Petrini, Memorial Sloan-Kettering Cancer Center
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
Mike Ryan, Univ. of Texas, Austin
Shimon Sakaguchi, Kyoto Univ.
Miquel Salmeron, Lawrence Berkeley National Lab
Jürgen Sandkühler, Medical Univ. of Vienna
Randy Seeley, Univ. of Cincinnati
Christine Seidman, Harvard Medical School
Vladimir Shalaev, Purdue Univ.
Joseph Silk, Univ. of Oxford
Denis Simon, Univ. of Oregon
Alison Smith, John Innes Centre
Davor Solter, Inst. of Medical Biology, Singapore
John Speakman, Univ. of Aberdeen
Allan C. Spradling, Carnegie Institution of Washington
Jonathan Sprent, Garvan Inst. of Medical Research
Elsbeth Stern, ETH Zürich
Ira Tabas, Columbia Univ.
Yoshiko Takahashi, Nara Inst. of Science and Technology
John Thomas, Duke Univ.
Herbert Virgin, Washington Univ.
Bert Vogelstein, Johns Hopkins Univ.
Cynthia Volkert, Univ. of Gottingen
Bruce D. Walker, Harvard Medical School
Douglas Wallace, Leibniz Inst. of Marine Sciences
Ian Walmsley, Univ. of Oxford
David A. Wardle, Swedish Univ. of Agric Sciences
Detlef Weigel, Max Planck Inst., Tübingen
Jonathan Weissman, Univ. of California, San Francisco
Sue Wessler, Univ. of California, Riverside
Ian A. Wilson, The Scripps Res. Inst.
Timothy D. Wilson, Univ. of Virginia
Jan Zaanen, Leiden Univ.
Mayana Zatz, University of Sao Paolo
Jonathan Zehr, Ocean Sciences
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
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
INSTITUTIONAL SITE LICENSES 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.
officedepot.com/portalLogin.do; Seabury & Smith Life Insurance 800-
424-9883; Subaru VIP Program 202-326-6417; VIP Moving Services
www.vipmayflower.com/domestic/index.html; Other Benefits: AAAS
Member Services 202-326-6417 or www.aaasmember.org.
science_editors@aaas.org (for general editorial queries)
science_letters@aaas.org (for queries about letters)
science_reviews@aaas.org (for returning manuscript reviews)
science_bookrevs@aaas.org (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 Science—including
editorials, news and comment, and book reviews—are signed and reflect
the individual views of the authors and not official points of view adopted
by AAAS or the institutions with which the authors are affiliated.
AAAS was founded in 1848 and incorporated in 1874. Its mission is to
advance science, engineering, and innovation throughout the world for
the benefit 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.
INFORMATION FOR AUTHORS
See pages 784 and 785 of the 11 February 2011 issue or
access www.sciencemag.org/about/authors
EDITOR-IN-CHIEF Bruce Alberts
EXECUTIVE EDITOR NEWS EDITOR
Monica M. Bradford Colin Norman
MANAGING EDITOR, RESEARCH JOURNALS Katrina L. Kelner
DEPUTY EDITORS R. Brooks Hanson, Barbara R. Jasny, Andrew
M. Sugden
EDITORIAL SENIOR EDITORS/COMMENTARY Lisa D. Chong, Brad Wible; SENIOR
EDITORS 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, Laura M. Zahn (San Diego); ASSOCIATE EDITORS Kristen L.
Mueller, Jelena Stajic, Sacha Vignieri, Nicholas S. Wigginton; 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, Lauren Kmec, Barbara P. Ordway, Trista Wagoner; COPY EDITOR
Chris Filiatreau; SENIOR EDITORIAL COORDINATORS Carolyn Kyle, Beverly
Shields; EDITORIAL COORDINATORS Joi S. Granger, Anita Wynn; PUBLICATIONS
ASSISTANTS Ramatoulaye Diop, Emily Guise, Jeffrey Hearn, Michael Hicks,
Lisa Johnson, Scott Miller, Jerry Richardson, Brian White; EDITORIAL
ASSISTANT Patricia M. Moore; EXECUTIVE EDITORIAL ASSISTANT Yolanda O'Bannon
(San Francisco); EXECUTIVE ASSISTANT Alison Crawford; ADMINISTRATIVE SUPPORT
Maryrose Madrid; EDITORIAL FELLOW Melissa R. McCartney
EDITORIAL DIRECTOR, WEB AND NEW MEDIA Stewart Wills; SENIOR WEB EDITOR
Tara S. Marathe; RESEARCH ASSOCIATE Corinna Cohn; WEB DEVELOPMENT
MANAGER Martyn Green; WEB DEVELOPER Andrew Whitesell; INTERNS
Andrew Green, Kerry Klein
NEWS DEPUTY NEWS EDITORS Robert Coontz, David Grimm (Online), Eliot
Marshall, Jeffrey Mervis, Leslie Roberts, John Travis; CONTRIBUTING EDITORS
Elizabeth Culotta, Polly Shulman; NEWS WRITERS Yudhijit Bhattacharjee,
Adrian Cho, Jennifer Couzin-Frankel, Carolyn Gramling, Jocelyn Kaiser,
Richard A. Kerr, Eli Kintisch, Greg Miller, Elizabeth Pennisi,
Robert F. Service (Pacific NW), Erik Stokstad; WEB DEVELOPER
Daniel Berger; INTERNS Daniel Strain, Yasmin Ogale, Natalie Villacorta;
CONTRIBUTING CORRESPONDENTS Jon Cohen (San Diego, CA), Daniel
Ferber, Ann Gibbons, Sam Kean, Andrew Lawler, Mitch Leslie,
Charles C. Mann, Virginia Morell, Gary Taubes; COPY EDITORS Linda
B. Felaco, Melvin Gatling, Melissa Raimondi; ADMINISTRATIVE SUPPORT
Scherraine Mack; BUREAUS San Diego, CA: 760-942-3252, FAX 760-942-
4979; Pacific Northwest: 503-963-1940
PRODUCTION DIRECTOR Wendy K. Shank; ASSISTANT MANAGER Rebecca
Doshi; SENIOR SPECIALISTS Steve Forrester, Chris Redwood, Anthony
Rosen; PREFLIGHT DIRECTOR David M. Tompkins; MANAGER Marcus Spiegler;
SPECIALISTS Jason Hillman, Tara Kelly
ART DIRECTOR Yael Fitzpatrick; ASSOCIATE ART DIRECTOR Laura Creveling;
SENIOR ILLUSTRATORS Chris Bickel, Katharine Sutliff; ILLUSTRATOR Yana
Hammond; SENIOR ART ASSOCIATES Holly Bishop, Preston Huey,
Nayomi Kevitiyagala, Matthew Twombly; ART ASSOCIATE Kay Engman;
PHOTO EDITOR Leslie Blizard
SCIENCE INTERNATIONAL
EUROPE (science@science-int.co.uk) EDITORIAL: INTERNATIONAL MANAGING
EDITOR Andrew M. Sugden; SENIOR EDITOR/COMMENTARY Julia Fahrenkamp-
Uppenbrink; SENIOR EDITORS Caroline Ash, Stella M. Hurtley, Ian S. Osborne,
Peter Stern; ASSOCIATE EDITOR Maria Cruz; LOCUM EDITOR Helen Pickersgill;
EDITORIAL SUPPORT Samantha Hogg, Alice Whaley; ADMINISTRATIVE
SUPPORT John Cannell, Janet Clements, Louise Hartwell; NEWS: DEPUTY
NEWS EDITOR, U.K. Daniel Clery; NEWSWRITER Sara Reardon; CONTRIBUTING
EDITOR, EUROPE Martin Enserink; CONTRIBUTING CORRESPONDENTS Michael Balter
(Paris), John Bohannon (Vienna), Gretchen Vogel (Berlin)
LATIN AMERICA CONTRIBUTING CORRESPONDENT Antonio Regalado
ASIA Japan Office: 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, inquiry@sciencemag.jp;
ASIA NEWS EDITOR Richard Stone (Beijing: rstone@aaas.org); CONTRIBUTING
CORRESPONDENTS Dennis Normile [Japan: +81 (0) 3 3391 0630, FAX
+81 (0) 3 5936 3531; dnormile@gol.com]; Hao Xin [China:
cindyhao@gmail.com]; Mara Hvistendahl [China: mara@
marahvistendahl.com]; Pallava Bagla [South Asia: +91 (0) 11 2271
2896; pbagla@vsnl.com]
EXECUTIVE PUBLISHER Alan I. Leshner
PUBLISHER Beth Rosner
FULFILLMENT SYSTEMS AND OPERATIONS (membership@aaas.org); CUSTOMER
SERVICE SUPERVISOR Pat Butler; SPECIALISTS Latoya Casteel, LaVonda
Crawford, Vicki Linton, April Marshall; DATA ENTRY SUPERVISOR Cynthia
Johnson; SPECIALISTS Shirlene Hall, Tarrika Hill, William Jones
BUSINESS OPERATIONS AND ADMINISTRATION DIRECTOR Deborah Rivera-
Wienhold; BUSINESS SYSTEMS AND FINANCIAL ANALYSIS DIRECTOR Randy Yi;
MANAGER, FULFILLMENT SYSTEMS Frits Buningh; MANAGER, BUSINESS ANALYSIS Eric
Knott; MANAGER, BUSINESS OPERATIONS Jessica Tierney; BUSINESS ANALYSTS
Priti Pamnani, Celeste Troxler; Christine Wehrli; RIGHTS AND
PERMISSIONS: ADMINISTRATOR Emilie David; ASSOCIATE Elizabeth Sandler;
MARKETING DIRECTOR Ian King; MARKETING MANAGERS Allison Pritchard, Alison
Chandler, Julianne Wielga, Samantha Smith; MARKETING ASSOCIATES
Aimee Aponte, Mary Ellen Crowley; SENIOR MARKETING EXECUTIVE Jennifer
Reeves; DIRECTOR, SITE LICENSING Tom Ryan; DIRECTOR, CORPORATE RELATIONS
Eileen Bernadette Moran; SENIOR PUBLISHER RELATIONS SPECIALIST
Kiki Forsythe; PUBLISHER RELATIONS MANAGER Catherine Holland; PUBLISHER
RELATIONS, EASTERN REGION Phillip Smith; PUBLISHER RELATIONS, WESTERN REGION
Ryan Rexroth; CUSTOMER RELATIONS MANAGER Iquo Edim; CUSTOMER RELATIONS
COORDINATOR David Lee; MARKETING MANAGER Christina Schlecht; MARKETING
ASSOCIATES Laura Tutino, Chad Johnson; ELECTRONIC MEDIA: DIRECTOR Lizabeth
Harman; ASSISTANT MANAGER Lisa Stanford; SENIOR PRODUCTION SPECIALIST Ryan
Atkins; PRODUCTION SPECIALISTS Antoinette Hodal, Nichele Johnston, Kimberly
Oster; DIRECTOR, WEB AND NEW MEDIA Will Collins; PROJECT MANAGER Trista Snyder;
SENIOR PRODUCTION SPECIALIST Christopher Coleman; COMPUTER SPECIALISTS
Walter Jones, Kai Zhang
ADVERTISING DIRECTOR, WORLDWIDE AD SALES Bill Moran
COMMERCIAL EDITOR Sean Sanders: 202-326-6430
ASSISTANT COMMERCIAL EDITOR Tianna Hicklin 202-326-6463
PRODUCT ( sci ence_advert i si ng@aaas. org) ; MI DWEST Ri ck
Bongiovanni: 330-405-7080, FAX 330-405-7081; EAST COAST/
E. CANADA Laurie Faraday: 508-747-9395, FAX 617-507-8189;
WEST COAST/W. CANADA Lynne Stickrod: 415-931-9782, FAX 415-520-
6940; UK/EUROPE/ASIA Roger Goncalves: TEL/FAX +41 43 243 1358;
JAPAN ASCA Corporation, Makiko Hara: +81 (0) 3 6802 4616, FAX
+81 (0) 3 6802 4615; ads@sciencemag.jp; CHINA/TAIWAN Ruolei Wu:
+86 1367 1015 294 rwu@aaas.org
WORLDWIDE ASSOCIATE DIRECTOR OF SCIENCE CAREERS Tracy Holmes: +44 (0)
1223 326525, FAX +44 (0) 1223 326532
CLASSIFIED (advertise@sciencecareers.org); U.S.: MIDWEST/WEST COAST/
SOUTH CENTRAL/CANADA Tina Burks: 202-326-6577; EAST COAST/INDUSTRY
Elizabeth Early: 202-326-6578; SALES ADMINISTRATOR: Marci Gallun;
EUROPE/ROW SALES: Susanne Kharraz, Dan Pennington, Alex Palmer; SALES
ASSISTANT Lisa Patterson; JAPAN ASCA Corporation, Makiko Hara +81 (0)
3 6802 4616, FAX +81 (0) 3 6802 4615; careerads@sciencemag.jp;
CHINA/TAIWAN Ruolei Wu: +86 1367 1015 294 rwu@aaas.org;
ADVERTISING SUPPORT MANAGER Karen Foote: 202-326-6740; ADVERTISING
PRODUCTION OPERATIONS MANAGER Deborah Tompkins; SENIOR PRODUCTION
SPECIALIST/GRAPHIC DESIGNER Amy Hardcastle; PRODUCTION SPECIALIST
Yuse Lajiminmuhip; SENIOR TRAFFIC ASSOCIATE Christine Hall; SALES
COORDINATOR Shirley Young
AAAS BOARD OF DIRECTORS RETIRING PRESIDENT, CHAIR Alice Huang;
PRESIDENT Nina Fedoroff; PRESIDENT-ELECT William Press; TREASURER
David E. Shaw; CHIEF EXECUTIVE OFFICER Alan I. Leshner; BOARD Nancy
Knowlton, Stephen Mayo, Raymond Orbach, Julia M. Phillips, Sue V.
Rosser, David D. Sabatini, Inder Verma, Thomas A. Woolsey
Published by AAAS
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1076
NEWS OF THE WEEK
C
R
E
D
I
T
:

N
O
A
A
Lomonosov Ridge called the Alpha Ridge
that runs north of the Yukon coast, which has
remained virtually unexplored.
The Canadian research team is joined
by American researchers on a second ice-
breaker. Russia is performing similar map-
ping missions and is expected to submit its
claim this year. Denmark plans to make one
in 2014. http://scim.ag/CanadaArctic
Kandakadu, Sri Lanka 3
Bananas a Flashpoint for
Human-Elephant Conflict
Endangered Asian elephants may face a new
threat, conservationists say: bananas. Dole
Food Company and local partner Letsgrow
recently established a banana plantation on
about 890 hectares of scrubland near Kanda-
kadu, at the edge of Somawathiya National
Park in Sri Lanka’s war-ravaged North Cen-
tral Province. Conservationists claim that up
to half of the plantation’s land is protected
forest—a claim Dole denies.
Dole contends that the Sri Lanka Army,
which partners with Letsgrow on agricul-
tural ventures, granted them the rights to
the Kandakadu land. But conservationists
insist that the army does not own the land,
and that part of the plantation belongs to the
government’s National Livestock Develop-
ment Board.
How the plantation in Somawathiya will
affect elephants is unclear. Marty Ordman, a
spokesperson for Dole, asserts that elephants
are “not very common” in the area around
Dole’s plantation. But lands surrounding
national parks are important to elephants in
those parks, explains Shermin de Silva, direc-
tor of the Uda Walawe Elephant Research
Project in southern Sri Lanka. “Large ani-
mals need a large area to roam in,” she says.
“They don’t just stay where people say it’s
OK for them to be.” http://scim.ag/_Dole
Mumbai, India 4
Hints of Higgs Boson
Appear Weaker
Last month physicists working with the
world’s highest-energy atom-smasher—the
Large Hadron Collider (LHC) at the Euro-
pean particle physics laboratory, CERN,
near Geneva, Switzerland—reported pos-
sible evidence of the Higgs boson, the key to
physicists’ explanation of how all particles
get their mass. At the biannual Lepton Pho-
ton conference this week, however, the same
two teams reported that, with more data,
those signs appear weaker. That suggests the
signals could be statistical fluctuations in the
“background” produced by decays of famil-
iar particles.
“If you add 50% more data, you expect
the signal to grow, and it does not,” says
Vivek Sharma of the University of Califor-
nia, San Diego, who reported the new results
of the team working with the CMS particle
detector. Still, it’s not certain that the signals
are just background fluctuations, says Bill
Murray of Rutherford Appleton Labora-
tory near Didcot, U.K., who works with the
ATLAS detector, also fed by the LHC. “The
old data and the new data are telling different
stories,” he says. “Which one is telling us the
right story, we can’t yet say.”
Researchers should be able to spot the
particle or rule it out by year’s end, Sharma
says: “If it’s not there, it will be known to be
science fiction by December.”
http://scim.ag/Higgshints
Washington, D.C. 1
FDA Approves Gene-Targeted
Melanoma Drug
A new kind of melanoma drug that targets a
molecular weak spot in cancer cells received
approval last week from the U.S. Food and
Drug Administration.
The drug, called vemurafenib or Zelbo-
raf, blocks a protein made by a mutated gene
called BRAF that spurs tumor growth. In a
clinical trial published last spring, advanced
melanoma patients with the BRAF mutation
who took vemurafenib lived significantly
longer than patients on a standard drug.
Only the 50% or so of advanced melanoma
patients whose tumors carry the BRAF muta-
tion will receive the new drug.
Vemurafenib, which is made by Roche
and Plexxikon (now owned by Daiichi
Sankyo), is the latest example of several
remarkably effective new gene-targeted can-
cer therapies. However, often these drugs
eventually stop working because tumors
develop resistance. Researchers hope that
combining vemurafenib with a second drug
will prolong patient survival.
Kugluktuk, Nunavut, Canada 2
Geologists Hope to Claim
Piece of Arctic for Canada
A geological survey team of Canadian sci-
entists set out 18 August to map the last
uncharted regions of the Arctic sea floor.
This is the last leg of Canada’s project to
collect geological data to support their
claim to the Arctic Ocean’s sunken moun-
tain range, the Lomonosov Ridge, and the
potential natural resources it holds. Canada
is facing a 2013 deadline to make their
case to the United Nations.
Under the U.N. Convention on the Law of
2
4
3
1, 5
AROUND THE WORLD
Break up. Icebreakers
in the Arctic
the Sea, a country can claim land up to
200 nautical miles off its shore. Canada,
Russia, and Denmark (through Greenland)
are each trying to determine whether the ridge
is an extension of their continental shelf.
Canada has spent 10 years and $109 mil-
lion on the project so far. The 6-week cruise
will try to place the last piece of the puzzle,
a 2000-kilometer-long extension of the
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1077
NEWS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

B
E
C
K
Y

K
A
G
A
N
/
A
D
V
A
N
C
E
D

I
M
A
G
I
N
G

A
N
D

V
I
S
U
A
L
I
Z
A
T
I
O
N

L
A
B
O
R
A
T
O
R
Y
,

W
O
O
D
S

H
O
L
E

O
C
E
A
N
O
G
R
A
P
H
I
C

I
N
S
T
I
T
U
T
I
O
N
;

P
O
L
Y
T
E
C
H
N
I
C

U
N
I
V
E
R
S
I
T
Y

O
F

T
U
R
I
N
;

D
A
V
I
D

W
A
C
E
Y
/
U
N
I
V
E
R
S
I
T
Y

O
F

W
E
S
T
E
R
N

A
U
S
T
R
A
L
I
A
Witnessing a Watery Junkyard
During World War II, German U-boats sank dozens of U.S. ships off the coast of North
Carolina, creating a watery junkyard known as the Graveyard of the Atlantic. Those ships
are now the target of scientists from Woods Hole Oceanographic Institution (WHOI), who
are using a remotely operated vehicle outfitted with underwater 3D cameras to capture
images of sunken ships previously located with side-scan sonar.
In mid-August, Evan Kovacs, director of 3D photography for WHOI’s Advanced Imag-
ing and Visualization Lab, and his team took new 3D and 2D images of the ships (shown
here), as part of an ongoing project with the National Oceanic and Atmospheric Admin-
istration and the National Park Service to study their deterioration and impact on the
environment. “Video and imagery have always been key for marine archaeology,”
Kovacs says, “but before 3D, it’s [been] difficult to pull hard, quantitative data out of
it.” The 3D data, he says, allow scientists to analyze rates of collapse and corrosion. And,
he adds, 3D images are a great outreach tool: “It’s an insanely immersive environment.”
THEY SAID IT
“ To be clear. I believe in evo-
lution and trust scientists on
global warming. Call me crazy.”
—Tweet by 2012 Republican
presidential candidate Jon Huntsman on
18 August after Texas governor and
Republican candidate Rick Perry said that
evolution is “a theory that’s out there”
and questioned whether climate scientists
manipulated data for money.
FINDINGS
NEWSMAKER
Washington, D.C. 5
New U.S. Conflict of Interest Rules
The U.S. government this week issued new
rules aimed at cracking down on financial
conflicts of interest in biomedical research.
The update of a 16-year-old regulation will
require researchers funded by the Pub-
lic Health Service to report more of their
income from drug companies and other out-
side sources to their institutions. The regula-
tions also require that institutions, rather than
investigators, determine if any payments con-
stitute a conflict. http://scim.ag/_CoI
Reformer Takes Reins in Rome
The Italian National Research Council
(CNR)—a €1 billion basic research agency
with 100 institutes around the country—may
be headed for some major changes with the
appointment on 13 August of Francesco
Profumo, the successful, reform-minded rec-
tor of the Polytechnic University of Turin,
Fossil Cells Are World’s Oldest
Scientists say they have discovered 3.4-bil-
lion-year-old cells, possibly the oldest fossils
ever found. Geologists Martin Brasier of the
University of Oxford in the United Kingdom
and David Wacey of the University of West-
ern Australia in Crawley discovered the fossil
cells between cemented sand grains from an
ancient beach in Western Australia.
The fossil cells were hollow, and some
were clustered together in groups sur-
rounded by what looked like a membrane.
“The morphology is very cell-like,” Brasier
says. The cells were also patchily distrib-
uted in the sediment, just as modern bacteria
tend to congregate near sources of food, the
researchers reported online 21 August in
Nature Geoscience.
Old cells. Ancient life (inset) found in Australia.
as CNR’s new president. Italian scientists
say he is likely to modernize and revitalize
the agency, which has been
plagued by budget cuts.
While at the Polytech-
nic University, Profumo,
58, worked hard to build
international ties, turn scien-
tific research into economic
development and innova-
tion, and forge tight collaborations with
industry. Profumo says he’ll try some of the
same recipes at CNR and will spur Italy to
get more research money from the European
Union. Italy currently contributes 14% of the
E.U.’s research funds but receives only 9% of
the grants the union distributes.
“Profumo has proved his exceptional
skills in coordinating research, technology,
and industry, and he will give CNR a central
role in applied research,” says Luigi Donato,
founder and former head of the CNR Insti-
tute of Clinical Physiology in Pisa. “The
latter aspect is crucial in this moment of
financial crisis.” http://scim.ag/_Profumo
The two scientists say their chemical anal-
yses of the minerals near the cells suggest
the microorganisms depended on sulfur for
fuel. Such a beach might have been life’s first
breeding ground, Brasier says.
The work represents “some of the best
evidence for the nature of Earth’s earliest
life,” says Bruce Runnegar, a geologist at the
University of California, Los Angeles, who
was not involved in the study.
http://scim.ag/fossilcells
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1076
NEWS OF THE WEEK
C
R
E
D
I
T
:

N
O
A
A
Lomonosov Ridge called the Alpha Ridge
that runs north of the Yukon coast, which has
remained virtually unexplored.
The Canadian research team is joined
by American researchers on a second ice-
breaker. Russia is performing similar map-
ping missions and is expected to submit its
claim this year. Denmark plans to make one
in 2014. http://scim.ag/CanadaArctic
Kandakadu, Sri Lanka 3
Bananas a Flashpoint for
Human-Elephant Conflict
Endangered Asian elephants may face a new
threat, conservationists say: bananas. Dole
Food Company and local partner Letsgrow
recently established a banana plantation on
about 890 hectares of scrubland near Kanda-
kadu, at the edge of Somawathiya National
Park in Sri Lanka’s war-ravaged North Cen-
tral Province. Conservationists claim that up
to half of the plantation’s land is protected
forest—a claim Dole denies.
Dole contends that the Sri Lanka Army,
which partners with Letsgrow on agricul-
tural ventures, granted them the rights to
the Kandakadu land. But conservationists
insist that the army does not own the land,
and that part of the plantation belongs to the
government’s National Livestock Develop-
ment Board.
How the plantation in Somawathiya will
affect elephants is unclear. Marty Ordman, a
spokesperson for Dole, asserts that elephants
are “not very common” in the area around
Dole’s plantation. But lands surrounding
national parks are important to elephants in
those parks, explains Shermin de Silva, direc-
tor of the Uda Walawe Elephant Research
Project in southern Sri Lanka. “Large ani-
mals need a large area to roam in,” she says.
“They don’t just stay where people say it’s
OK for them to be.” http://scim.ag/_Dole
Mumbai, India 4
Hints of Higgs Boson
Appear Weaker
Last month physicists working with the
world’s highest-energy atom-smasher—the
Large Hadron Collider (LHC) at the Euro-
pean particle physics laboratory, CERN,
near Geneva, Switzerland—reported pos-
sible evidence of the Higgs boson, the key to
physicists’ explanation of how all particles
get their mass. At the biannual Lepton Pho-
ton conference this week, however, the same
two teams reported that, with more data,
those signs appear weaker. That suggests the
signals could be statistical fluctuations in the
“background” produced by decays of famil-
iar particles.
“If you add 50% more data, you expect
the signal to grow, and it does not,” says
Vivek Sharma of the University of Califor-
nia, San Diego, who reported the new results
of the team working with the CMS particle
detector. Still, it’s not certain that the signals
are just background fluctuations, says Bill
Murray of Rutherford Appleton Labora-
tory near Didcot, U.K., who works with the
ATLAS detector, also fed by the LHC. “The
old data and the new data are telling different
stories,” he says. “Which one is telling us the
right story, we can’t yet say.”
Researchers should be able to spot the
particle or rule it out by year’s end, Sharma
says: “If it’s not there, it will be known to be
science fiction by December.”
http://scim.ag/Higgshints
Washington, D.C. 1
FDA Approves Gene-Targeted
Melanoma Drug
A new kind of melanoma drug that targets a
molecular weak spot in cancer cells received
approval last week from the U.S. Food and
Drug Administration.
The drug, called vemurafenib or Zelbo-
raf, blocks a protein made by a mutated gene
called BRAF that spurs tumor growth. In a
clinical trial published last spring, advanced
melanoma patients with the BRAF mutation
who took vemurafenib lived significantly
longer than patients on a standard drug.
Only the 50% or so of advanced melanoma
patients whose tumors carry the BRAF muta-
tion will receive the new drug.
Vemurafenib, which is made by Roche
and Plexxikon (now owned by Daiichi
Sankyo), is the latest example of several
remarkably effective new gene-targeted can-
cer therapies. However, often these drugs
eventually stop working because tumors
develop resistance. Researchers hope that
combining vemurafenib with a second drug
will prolong patient survival.
Kugluktuk, Nunavut, Canada 2
Geologists Hope to Claim
Piece of Arctic for Canada
A geological survey team of Canadian sci-
entists set out 18 August to map the last
uncharted regions of the Arctic sea floor.
This is the last leg of Canada’s project to
collect geological data to support their
claim to the Arctic Ocean’s sunken moun-
tain range, the Lomonosov Ridge, and the
potential natural resources it holds. Canada
is facing a 2013 deadline to make their
case to the United Nations.
Under the U.N. Convention on the Law of
2
4
3
1, 5
AROUND THE WORLD
Break up. Icebreakers
in the Arctic
the Sea, a country can claim land up to
200 nautical miles off its shore. Canada,
Russia, and Denmark (through Greenland)
are each trying to determine whether the ridge
is an extension of their continental shelf.
Canada has spent 10 years and $109 mil-
lion on the project so far. The 6-week cruise
will try to place the last piece of the puzzle,
a 2000-kilometer-long extension of the
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1077
NEWS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

B
E
C
K
Y

K
A
G
A
N
/
A
D
V
A
N
C
E
D

I
M
A
G
I
N
G

A
N
D

V
I
S
U
A
L
I
Z
A
T
I
O
N

L
A
B
O
R
A
T
O
R
Y
,

W
O
O
D
S

H
O
L
E

O
C
E
A
N
O
G
R
A
P
H
I
C

I
N
S
T
I
T
U
T
I
O
N
;

P
O
L
Y
T
E
C
H
N
I
C

U
N
I
V
E
R
S
I
T
Y

O
F

T
U
R
I
N
;

D
A
V
I
D

W
A
C
E
Y
/
U
N
I
V
E
R
S
I
T
Y

O
F

W
E
S
T
E
R
N

A
U
S
T
R
A
L
I
A
Witnessing a Watery Junkyard
During World War II, German U-boats sank dozens of U.S. ships off the coast of North
Carolina, creating a watery junkyard known as the Graveyard of the Atlantic. Those ships
are now the target of scientists from Woods Hole Oceanographic Institution (WHOI), who
are using a remotely operated vehicle outfitted with underwater 3D cameras to capture
images of sunken ships previously located with side-scan sonar.
In mid-August, Evan Kovacs, director of 3D photography for WHOI’s Advanced Imag-
ing and Visualization Lab, and his team took new 3D and 2D images of the ships (shown
here), as part of an ongoing project with the National Oceanic and Atmospheric Admin-
istration and the National Park Service to study their deterioration and impact on the
environment. “Video and imagery have always been key for marine archaeology,”
Kovacs says, “but before 3D, it’s [been] difficult to pull hard, quantitative data out of
it.” The 3D data, he says, allow scientists to analyze rates of collapse and corrosion. And,
he adds, 3D images are a great outreach tool: “It’s an insanely immersive environment.”
THEY SAID IT
“ To be clear. I believe in evo-
lution and trust scientists on
global warming. Call me crazy.”
—Tweet by 2012 Republican
presidential candidate Jon Huntsman on
18 August after Texas governor and
Republican candidate Rick Perry said that
evolution is “a theory that’s out there”
and questioned whether climate scientists
manipulated data for money.
FINDINGS
NEWSMAKER
Washington, D.C. 5
New U.S. Conflict of Interest Rules
The U.S. government this week issued new
rules aimed at cracking down on financial
conflicts of interest in biomedical research.
The update of a 16-year-old regulation will
require researchers funded by the Pub-
lic Health Service to report more of their
income from drug companies and other out-
side sources to their institutions. The regula-
tions also require that institutions, rather than
investigators, determine if any payments con-
stitute a conflict. http://scim.ag/_CoI
Reformer Takes Reins in Rome
The Italian National Research Council
(CNR)—a €1 billion basic research agency
with 100 institutes around the country—may
be headed for some major changes with the
appointment on 13 August of Francesco
Profumo, the successful, reform-minded rec-
tor of the Polytechnic University of Turin,
Fossil Cells Are World’s Oldest
Scientists say they have discovered 3.4-bil-
lion-year-old cells, possibly the oldest fossils
ever found. Geologists Martin Brasier of the
University of Oxford in the United Kingdom
and David Wacey of the University of West-
ern Australia in Crawley discovered the fossil
cells between cemented sand grains from an
ancient beach in Western Australia.
The fossil cells were hollow, and some
were clustered together in groups sur-
rounded by what looked like a membrane.
“The morphology is very cell-like,” Brasier
says. The cells were also patchily distrib-
uted in the sediment, just as modern bacteria
tend to congregate near sources of food, the
researchers reported online 21 August in
Nature Geoscience.
Old cells. Ancient life (inset) found in Australia.
as CNR’s new president. Italian scientists
say he is likely to modernize and revitalize
the agency, which has been
plagued by budget cuts.
While at the Polytech-
nic University, Profumo,
58, worked hard to build
international ties, turn scien-
tific research into economic
development and innova-
tion, and forge tight collaborations with
industry. Profumo says he’ll try some of the
same recipes at CNR and will spur Italy to
get more research money from the European
Union. Italy currently contributes 14% of the
E.U.’s research funds but receives only 9% of
the grants the union distributes.
“Profumo has proved his exceptional
skills in coordinating research, technology,
and industry, and he will give CNR a central
role in applied research,” says Luigi Donato,
founder and former head of the CNR Insti-
tute of Clinical Physiology in Pisa. “The
latter aspect is crucial in this moment of
financial crisis.” http://scim.ag/_Profumo
The two scientists say their chemical anal-
yses of the minerals near the cells suggest
the microorganisms depended on sulfur for
fuel. Such a beach might have been life’s first
breeding ground, Brasier says.
The work represents “some of the best
evidence for the nature of Earth’s earliest
life,” says Bruce Runnegar, a geologist at the
University of California, Los Angeles, who
was not involved in the study.
http://scim.ag/fossilcells
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

1078 26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

(
P
H
O
T
O
I
L
L
U
S
T
R
A
T
I
O
N

I
M
A
G
E
S
)

I
S
T
O
C
K
P
H
O
T
O
.
C
O
M
;

A
R
S
/
U
S
D
A
;

F
O
T
O
S
E
A
R
C
H
8.7 million +/-1.3 million
Number of eukaryote species on
Earth, including 6.5 million on
land and 2.2 million in the oceans,
according to a report by Census of
Marine Life scientists in PLoS Biology.
€239 million Cost of unused
H1N1 vaccines in Germany that are
expired and being destroyed. Three-
quarters of 34 million doses ordered
went unused due to distribution
problems and confusion over how
many doses conferred protection.
Random Sample
A(nother) Modest Proposal
Lions roaming the streets of Athens, gorillas swinging
from traffic lights in Spain—sure, it might be dangerous,
but think of the ecotourism dollars! That revenue could
provide a badly needed boost to troubled nations’ econo-
mies, says the (so-called) Coalition of Financially
Challenged Countries With Lots of Trees (CoFCCLoT).
In their “Modest Proposal,” published in Biotropica
on 16 August, CoFCCLoT also suggests that wealthy Western nations reforest their land, which
would help slow climate change and provide new habitat for displaced and endangered species
such as bears and wolves.
Such a proposal may sound unreasonable, but it is the kind of environmentally conscious
action that the West expects of countries that harbor tropical forest, says conservation scien-
tists Erik Meijaard and Douglas Sheil, who created the fictitious coalition. Conservationists in
wealthy countries shouldn’t complain about tropical deforestation, but then drink a cup of
Brazilian coffee every morning and buy cheap palm oil, the authors say.
Meijaard and Sheil hope to use satire to highlight such double standards. “What we’re trying
to show is the viewpoints that people are having are very often black and white. … If you want
to make political progress it’s a natural give and take,” Meijaard says. They modeled their paper
on Jonathan Swift’s “A Modest Proposal,” which suggested an ironic solution to poverty and
population growth in 18th century Ireland: Eat the children. “Humor can reveal us as holding
dual standards—as most of us do,” Sheil says. “Once you can laugh at your own viewpoint it’s
easier to step outside [it] and see how narrow-minded you’ve been.”
>>FINDINGS
Helping Fat Mice Live Longer
They work in mice—but will they work in
people? That’s one question hanging over
a class of compounds intended to boost
SIRT1, a protein thought to mimic the
effects of calorie restriction. In a host of
animals, slashing calories extends life and
preserves health—and a study in Scientific
Reports published 18 August shows that a
particular SIRT1 booster helps obese mice
live longer. The drug cut the amount of fat
in the animals’ livers and made them more
responsive to insulin. It also prolonged their
lives by 44% compared with control mice—
but the obese animals still died earlier than
normal mice. Closely related compounds are
currently in clinical trials in people.
The new study is at odds with an earlier
study by a group at Pfizer, which reported
that this compound and others like it don’t
activate the SIRT1 protein, and that they have
many unintended targets in the body, suggest-
ing they’re unlikely to make good drugs.
Disease-Fighting Mosquitoes
Proliferate in Field Test
A mosquito strain specially
created to fight dengue, a viral
disease that causes head-
aches, rash, and excruciating
muscle and joint pains, has
shown promise in its first open field study
in two small towns in northern Australia.
The experimental mosquitoes—a spe-
cies called Aedes aegypti, which is dengue’s
main vector—are infected with a strain of
Wolbachia pipientis, an intracellular bacte-
rium that makes the insects virtually unable
to transmit dengue but also has a way of
spreading rapidly through their populations
by playing strange tricks with their sex lives
(Science, 10 December 2010, p. 1460). Led
by Scott O’Neill from Monash University
in Melbourne, the team released between
10,000 and 20,000 infected mosquitoes
NEWS OF THE WEEK
weekly in each of the towns for
10 weeks. Six weeks after the final
release, the entire population in both
towns had become infected, the team
reported 25 August in Nature.
The study wasn’t designed to have an
effect on dengue, which is rare in Aus-
tralia. Field tests in Vietnam, Indonesia,
Thailand, and Brazil, all endemic coun-
tries, are awaiting approval; O’Neill says
the first released could happen within a
year. http://scim.ag/_Dengue
Iberian Lynx Not Jinxed by Genetics
A study of ancient DNA has given scientists more hope
that the world’s most endangered cat species can be sal-
vaged. Habitat destruction and the decline of its main
prey, the European rabbit, have caused the population of
the Iberian lynx (Lynx pardinus) to plummet below 300
individuals in two isolated areas in Spain. Scientists are
trying to help with a breeding program, but some believe
a lack of genetic diversity—which leads to inbreeding
problems and an inability to adapt to change—may doom
the species. But a new study of DNA found in fossil bones
shows that the Iberian lynx has had very low genetic diver-
sity, and presumably small populations, for at least 50,000
years. For reasons that are unclear, it always got by, the research-
ers conclude in their paper in Molecular Ecology. If the lynx is lost,
they say, don’t blame its genes; blame the lack of political will to save it.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1077
NEWS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

B
E
C
K
Y

K
A
G
A
N
/
A
D
V
A
N
C
E
D

I
M
A
G
I
N
G

A
N
D

V
I
S
U
A
L
I
Z
A
T
I
O
N

L
A
B
O
R
A
T
O
R
Y
,

W
O
O
D
S

H
O
L
E

O
C
E
A
N
O
G
R
A
P
H
I
C

I
N
S
T
I
T
U
T
I
O
N
;

P
O
L
Y
T
E
C
H
N
I
C

U
N
I
V
E
R
S
I
T
Y

O
F

T
U
R
I
N
;

D
A
V
I
D

W
A
C
E
Y
/
U
N
I
V
E
R
S
I
T
Y

O
F

W
E
S
T
E
R
N

A
U
S
T
R
A
L
I
A
Witnessing a Watery Junkyard
During World War II, German U-boats sank dozens of U.S. ships off the coast of North
Carolina, creating a watery junkyard known as the Graveyard of the Atlantic. Those ships
are now the target of scientists from Woods Hole Oceanographic Institution (WHOI), who
are using a remotely operated vehicle outfitted with underwater 3D cameras to capture
images of sunken ships previously located with side-scan sonar.
In mid-August, Evan Kovacs, director of 3D photography for WHOI’s Advanced Imag-
ing and Visualization Lab, and his team took new 3D and 2D images of the ships (shown
here), as part of an ongoing project with the National Oceanic and Atmospheric Admin-
istration and the National Park Service to study their deterioration and impact on the
environment. “Video and imagery have always been key for marine archaeology,”
Kovacs says, “but before 3D, it’s [been] difficult to pull hard, quantitative data out of
it.” The 3D data, he says, allow scientists to analyze rates of collapse and corrosion. And,
he adds, 3D images are a great outreach tool: “It’s an insanely immersive environment.”
THEY SAID IT
“ To be clear. I believe in evo-
lution and trust scientists on
global warming. Call me crazy.”
—Tweet by 2012 Republican
presidential candidate Jon Huntsman on
18 August after Texas governor and
Republican candidate Rick Perry said that
evolution is “a theory that’s out there”
and questioned whether climate scientists
manipulated data for money.
FINDINGS
NEWSMAKER
Washington, D.C. 5
New U.S. Conflict of Interest Rules
The U.S. government this week issued new
rules aimed at cracking down on financial
conflicts of interest in biomedical research.
The update of a 16-year-old regulation will
require researchers funded by the Pub-
lic Health Service to report more of their
income from drug companies and other out-
side sources to their institutions. The regula-
tions also require that institutions, rather than
investigators, determine if any payments con-
stitute a conflict. http://scim.ag/_CoI
Reformer Takes Reins in Rome
The Italian National Research Council
(CNR)—a €1 billion basic research agency
with 100 institutes around the country—may
be headed for some major changes with the
appointment on 13 August of Francesco
Profumo, the successful, reform-minded rec-
tor of the Polytechnic University of Turin,
Fossil Cells Are World’s Oldest
Scientists say they have discovered 3.4-bil-
lion-year-old cells, possibly the oldest fossils
ever found. Geologists Martin Brasier of the
University of Oxford in the United Kingdom
and David Wacey of the University of West-
ern Australia in Crawley discovered the fossil
cells between cemented sand grains from an
ancient beach in Western Australia.
The fossil cells were hollow, and some
were clustered together in groups sur-
rounded by what looked like a membrane.
“The morphology is very cell-like,” Brasier
says. The cells were also patchily distrib-
uted in the sediment, just as modern bacteria
tend to congregate near sources of food, the
researchers reported online 21 August in
Nature Geoscience.
Old cells. Ancient life (inset) found in Australia.
as CNR’s new president. Italian scientists
say he is likely to modernize and revitalize
the agency, which has been
plagued by budget cuts.
While at the Polytech-
nic University, Profumo,
58, worked hard to build
international ties, turn scien-
tific research into economic
development and innova-
tion, and forge tight collaborations with
industry. Profumo says he’ll try some of the
same recipes at CNR and will spur Italy to
get more research money from the European
Union. Italy currently contributes 14% of the
E.U.’s research funds but receives only 9% of
the grants the union distributes.
“Profumo has proved his exceptional
skills in coordinating research, technology,
and industry, and he will give CNR a central
role in applied research,” says Luigi Donato,
founder and former head of the CNR Insti-
tute of Clinical Physiology in Pisa. “The
latter aspect is crucial in this moment of
financial crisis.” http://scim.ag/_Profumo
The two scientists say their chemical anal-
yses of the minerals near the cells suggest
the microorganisms depended on sulfur for
fuel. Such a beach might have been life’s first
breeding ground, Brasier says.
The work represents “some of the best
evidence for the nature of Earth’s earliest
life,” says Bruce Runnegar, a geologist at the
University of California, Los Angeles, who
was not involved in the study.
http://scim.ag/fossilcells
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

1078 26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

(
P
H
O
T
O
I
L
L
U
S
T
R
A
T
I
O
N

I
M
A
G
E
S
)

I
S
T
O
C
K
P
H
O
T
O
.
C
O
M
;

A
R
S
/
U
S
D
A
;

F
O
T
O
S
E
A
R
C
H
8.7 million +/-1.3 million
Number of eukaryote species on
Earth, including 6.5 million on
land and 2.2 million in the oceans,
according to a report by Census of
Marine Life scientists in PLoS Biology.
€239 million Cost of unused
H1N1 vaccines in Germany that are
expired and being destroyed. Three-
quarters of 34 million doses ordered
went unused due to distribution
problems and confusion over how
many doses conferred protection.
Random Sample
A(nother) Modest Proposal
Lions roaming the streets of Athens, gorillas swinging
from traffic lights in Spain—sure, it might be dangerous,
but think of the ecotourism dollars! That revenue could
provide a badly needed boost to troubled nations’ econo-
mies, says the (so-called) Coalition of Financially
Challenged Countries With Lots of Trees (CoFCCLoT).
In their “Modest Proposal,” published in Biotropica
on 16 August, CoFCCLoT also suggests that wealthy Western nations reforest their land, which
would help slow climate change and provide new habitat for displaced and endangered species
such as bears and wolves.
Such a proposal may sound unreasonable, but it is the kind of environmentally conscious
action that the West expects of countries that harbor tropical forest, says conservation scien-
tists Erik Meijaard and Douglas Sheil, who created the fictitious coalition. Conservationists in
wealthy countries shouldn’t complain about tropical deforestation, but then drink a cup of
Brazilian coffee every morning and buy cheap palm oil, the authors say.
Meijaard and Sheil hope to use satire to highlight such double standards. “What we’re trying
to show is the viewpoints that people are having are very often black and white. … If you want
to make political progress it’s a natural give and take,” Meijaard says. They modeled their paper
on Jonathan Swift’s “A Modest Proposal,” which suggested an ironic solution to poverty and
population growth in 18th century Ireland: Eat the children. “Humor can reveal us as holding
dual standards—as most of us do,” Sheil says. “Once you can laugh at your own viewpoint it’s
easier to step outside [it] and see how narrow-minded you’ve been.”
>>FINDINGS
Helping Fat Mice Live Longer
They work in mice—but will they work in
people? That’s one question hanging over
a class of compounds intended to boost
SIRT1, a protein thought to mimic the
effects of calorie restriction. In a host of
animals, slashing calories extends life and
preserves health—and a study in Scientific
Reports published 18 August shows that a
particular SIRT1 booster helps obese mice
live longer. The drug cut the amount of fat
in the animals’ livers and made them more
responsive to insulin. It also prolonged their
lives by 44% compared with control mice—
but the obese animals still died earlier than
normal mice. Closely related compounds are
currently in clinical trials in people.
The new study is at odds with an earlier
study by a group at Pfizer, which reported
that this compound and others like it don’t
activate the SIRT1 protein, and that they have
many unintended targets in the body, suggest-
ing they’re unlikely to make good drugs.
Disease-Fighting Mosquitoes
Proliferate in Field Test
A mosquito strain specially
created to fight dengue, a viral
disease that causes head-
aches, rash, and excruciating
muscle and joint pains, has
shown promise in its first open field study
in two small towns in northern Australia.
The experimental mosquitoes—a spe-
cies called Aedes aegypti, which is dengue’s
main vector—are infected with a strain of
Wolbachia pipientis, an intracellular bacte-
rium that makes the insects virtually unable
to transmit dengue but also has a way of
spreading rapidly through their populations
by playing strange tricks with their sex lives
(Science, 10 December 2010, p. 1460). Led
by Scott O’Neill from Monash University
in Melbourne, the team released between
10,000 and 20,000 infected mosquitoes
NEWS OF THE WEEK
weekly in each of the towns for
10 weeks. Six weeks after the final
release, the entire population in both
towns had become infected, the team
reported 25 August in Nature.
The study wasn’t designed to have an
effect on dengue, which is rare in Aus-
tralia. Field tests in Vietnam, Indonesia,
Thailand, and Brazil, all endemic coun-
tries, are awaiting approval; O’Neill says
the first released could happen within a
year. http://scim.ag/_Dengue
Iberian Lynx Not Jinxed by Genetics
A study of ancient DNA has given scientists more hope
that the world’s most endangered cat species can be sal-
vaged. Habitat destruction and the decline of its main
prey, the European rabbit, have caused the population of
the Iberian lynx (Lynx pardinus) to plummet below 300
individuals in two isolated areas in Spain. Scientists are
trying to help with a breeding program, but some believe
a lack of genetic diversity—which leads to inbreeding
problems and an inability to adapt to change—may doom
the species. But a new study of DNA found in fossil bones
shows that the Iberian lynx has had very low genetic diver-
sity, and presumably small populations, for at least 50,000
years. For reasons that are unclear, it always got by, the research-
ers conclude in their paper in Molecular Ecology. If the lynx is lost,
they say, don’t blame its genes; blame the lack of political will to save it.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

NEWS & ANALYSIS
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1079
NEWS & ANALYSIS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

N
E
W
S
C
O
M
;

©

I
O
D
P
/
J
A
M
S
T
E
C
Pleading poverty, the U.S. National Science
Foundation (NSF) is ending a decade-long
era of international collaboration in ocean
drilling, leaving Japan and Europe to fend
for themselves.
Last week, NSF officials quietly posted
a notice to the community that the United
States would be pulling out of the Integrated
Ocean Drilling Program (IODP) when the
program expires in September 2013. Instead
of renewing the IODP relationship, which
involves three drill ships and 26 countries,
the United States hopes to reduce costs by
going it alone and attracting enough foreign
contributions to return its own drill ship to
full-time operation. “Given the likely budget
outlook,” says Timothy Killeen, NSF assis-
tant director for geosciences, “this is the best
way to support U.S. investigators.”
The proposed realignment harkens back
to an earlier era of scientific ocean drilling.
Between 1985 and 2003, the United States
led the Ocean Drilling Program that oper-
ated the JOIDES Resolution, also known as
the JR (Science, 24 October 2008, p. 512).
Other countries contributed 40% of the
funding in return for the opportunity to par-
ticipate. Since 2003, IODP, co-led by Japan
and the United States, has used a refurbished
JOIDES Resolution along with the newer,
larger, and far more capable Chikyu built
and operated by Japan (Science, 22 Febru-
ary 2008, p. 1037). On occasion a European
consortium has provided special platforms,
such as one leased for drilling in ice-choked
Arctic waters.
The Chikyu allows scientists to drill
deeper into the sea floor’s most challeng-
ing targets, such as the fault that gener-
ates great earthquakes off of Japan. But it
is much more expensive to operate than the
JR. Indeed, all manner of higher costs—for
fuel, drilling pipe, and unexpected repairs—
have been a perennial problem for the drill-
ing platforms. As a consequence, Chikyu
has spent only 3 to 4 months a year doing
scientific drilling, with some of the balance
of its time hired out to industry for oil and
gas exploration.
A budget squeeze has also forced NSF to
trim 2 months off the JR’s scientific sched-
ule in 2012, to only 6 months of operation.
A 28 June update from NSF’s Division of
Ocean Sciences Director David Conover
notes that operating costs for the JR have
also skyrocketed, and that NSF is still pay-
ing for cost overruns from the JR refur-
bishment, completed in early 2009. Some
$25 million in one-time funding for IODP
from the 2009 stimulus package is almost
exhausted, and NSF’s overall budget this
year is 1% lower than in 2010, when it spent
$63 million on IODP.
In their letter, NSF’s Killeen and Conover
announce “a new operating model that
streamlines operating costs and generates
new external sources of revenue to support
the JR.” While short on details, the proposed
realignment would have NSF operate the JR
independently of any other lead agency, but
the agency would solicit direct contributions
from other countries. All told, “we believe
this will get the JR close to if not at full-duty
operation,” Killeen says.
U.S. scientists who have built their
careers on rock and mud retrieved by ocean
drilling say NSF’s decision to go it alone
makes sense given its gloomy financial out-
look. “I view the proposal as an encourag-
ing sign,” says geochemist Steven D’Hondt
of the University of Rhode Island, Narra-
gansett. “What we’re hearing is that NSF is
going to protect the capability we provide to
the international community.” To Theodore
Moore Jr., a professor emeritus at the Uni-
versity of Michigan, Ann Arbor, the options
were “either dissolve the partnership and go
our separate ways or kill everything. There’s
lots of good science that remains to be done”
that is within the JR’s capabilities.
Japanese officials say that they aren’t sur-
prised by NSF’s decision to pull out but that
the timing is unfortunate. “I expected some-
thing like this was going to happen,” says
Asahiko Taira, an executive director of the
Japan Agency for Marine-Earth Science and
Technology. Both sides have long been frus-
trated with a rigid structure set up 15 years
ago when the research environment was quite
different, he says. Since then, budgets have
shrunk. There is also growing public scrutiny
of scientific efforts and increasing expecta-
tions that research be more relevant to social
needs. It didn’t help, Taira says, that NSF
ocean drilling program officials had a “dif-
ficult” time communicating with the career
civil servants at Japan’s Ministry of Educa-
tion (MEXT).
Still, the decision comes just weeks after
the consortium unveiled a new 10-year
plan for scientific exploration that would
have relied upon drilling platforms from
the United States, Japan, and Europe. “The
NSF announcement was completely at odds
with the discussions among the participat-
ing countries,” says Shingo Shibata, director
of deep-sea research for MEXT. Taira says
Japan will continue ocean drilling and will
look to the international community not so
much for funding but for “scientific proposals
and scientific participation.” He would still
like to see “an overarching science advisory
structure” to maintain an international vision
for ocean drilling and to avoid duplication.
Although Shibata says that “we understand
NSF’s decision given its severe financial situa-
tion,” Taira worries that “the U.S. is becoming
more inward-looking” because of budget con-
straints. Such an attitude, he says, would be a
blow to global scientific cooperation.
–RICHARD A. KERR AND DENNIS NORMILE
U.S. to Leave Consortium
And Go It Alone After 2013
SCI ENTI FI C OCEAN DRI LLI NG
Sailing away. The U.S. JOIDES Resolution (top) and
Japan’s Chikyu are parting company.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1080
NEWS&ANALYSIS
C
R
E
D
I
T
:

©

U
N
I
V
E
R
S
I
T
Y

O
F

C
A
M
B
R
I
D
G
E
After 3 years in charge of the Medical
Research Council, the primary govern-
ment funder of biomedical research in the
United Kingdom, Leszek Borysiewicz, who
is universally known as “Borys,” switched
back to the academy last October. He
joined the University of Cambridge as vice-
chancellor, the ancient university’s top admin-
istrative position. Just after he got there, the
Conservative–Liberal Democrat coalition
government announced unexpectedly that a
cap on annual student tuition fees would rise
from £3000 to £9000 (about $15,000) per
annum. At the same time, most public support
for university teaching will be withdrawn in
England (other parts of the United Kingdom
have autonomy in this sphere and will not
follow suit). The change has rocked the U.K.
university system and even prompted student
riots, but is now being swiftly implemented—
alongside a freeze in research funding.
An immunologist who made his scien-
tific reputation developing a vaccine for cer-
vical cancer, Borysiewicz early in his career
had a brief spell lecturing at Cambridge—
often rated as Europe’s leading research
university—but was educated at the Uni-
versity of Wales in Cardiff and earned his
administrative stripes as deputy rector of
Imperial College London. With his impos-
ing presence, amiable manner, and Polish-
tinged Welsh accent—both his parents were
refugees from Poland, and he spoke no Eng-
lish until the age of 5—Borysiewicz stands
out sharply from the Oxbridge-educated
men who still dominate the British estab-
lishment. As his first academic year in the
position drew to a close, the vice-chancellor
sat down with Science in his office in the Old
Schools building for a frank discussion of
the future of Cambridge and other U.K. uni-
versities. This interview has been edited for
brevity and clarity; an extended version can
be found online at http://scim.ag/Borys.
–COLIN MACILWAIN
Colin Macilwain is a writer in Edinburgh, U.K.
Q: Radical changes in funding for univer-
sity teaching are being implemented in the
United Kingdom, and it has been argued
that the universities’ response has been
supine. Why have they been unable to rally
support behind public funding of univer-
sity teaching?
L.B.: The whole sector is 137 independent
institutions, and they each have to look at
their own positions. Cambridge came to a
unanimous view; it didn’t hit the headlines,
but it was a firm statement of the universi-
ty’s position. In essence, it said that we do
not agree with reduction of state support for
teaching, we will continue to take in under-
graduate students with the same approach to
selection as before, and that we believe in
the autonomy of institutions and not setting
external quotas [on student numbers]. Those
were very firm statements that represent the
position of the university.
Q: If Britain’s universities are as good as
they say they are, and fundamental to U.K.
business and society, why is the country, as
some would argue, going to hell?
L.B.: I could suggest three reasons: Universi-
ties work in the medium- and long-term, they
don’t do quick fixes. Secondly, top universi-
ties focus on major problems that are difficult
to solve; again, these are long-term problems
with no quick fixes. And thirdly, while uni-
versities are repositories of knowledge, and it
is important that these repositories are turned
to the public good, we know that this transla-
tion takes time. The Medical Research Coun-
cil’s report on biomedical science, Medical
Research: What’s it Worth?, estimated the
translation time at 17 years.
Q: Isn’t the German university system,
with its emphasis on breadth—rather than
excellence at the very top—and vocational
education, proving to be more strategically
effective?
L.B.: In certain sectors, Germany is in the
lead. In others, particularly life sciences and
biomedicine, it is not. That may depend on
particular skills in the university system.
Q: How are you cutting your cloth in
response to the research funding outlook?
L.B.: If we just look at the revenue funding
from the research councils, they are looking
to make some reductions. We’ll be trying to
attract as much of the pot as we can from them,
and we will continue to diversify our research
base. We believe that strategy will work appro-
priately for Cambridge. But the capital budget
is a very real problem. We’re going to see a
reduction in capital budget coming through
SIF [the Higher Education Funding Coun-
cil for England’s Strategic Investment Fund]
from £33 million to £11 million per annum.
There are also cutbacks in equipment budgets.
For Cambridge, maintaining the best pos-
sible infrastructure is absolutely vital, so
that our academics can compete and interact
with industry, which is at the cutting edge.
If we fall below that, it is going to be a core
weakness. I suspect what the government
may be thinking is that a couple of years isn’t
going to make much difference—but any
sustained reduction in capital investment in
universities can do damage in the long term,
particularly to the most internationally com-
petitive parts of our university system.
Q: What are your research priorities?
L.B.: Cambridge is going to maintain a very
broad research direction, for two reasons.
We attract academics by providing them
with time and space and opportunity to fol-
low their own instincts; and I also believe
that major paradigm shifts in thinking, in
the humanities or in sciences, can actu-
ally occur. But this is going to be difficult,
because funders have more and more dif-
ficulty sustaining the investigator-led pro-
grams, which support this kind of science.
They are instead moving towards “grand
challenges,” and Cambridge is going to have
to respond to that. This year, we’ve identified
strategic priorities to help us compete for
large-scale funding in areas such as energy,
cancer, infectious disease, neuroscience,
and linguistics. In these areas, we’re going
Cambridge Seeks Global Reach, Says University Head
NEWSMAKER I NTERVI EW: LESZEK BORYSI EWI CZ
Back to school. Borysiewicz has returned to uni-
versity administration at a challenging time for the
United Kingdom’s higher education sector.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1081
NEWS&ANALYSIS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

T
.

N
O
G
U
C
H
I

E
T

A
L
.
,

S
C
I
E
N
C
E

3
3
3
,

6
0
4
6

(
2
6

A
U
G
U
S
T

2
0
1
1
)
;

C
O
U
R
T
E
S
Y

O
F

J
A
X
A

Astronomers had a problem. They knew that
meteorites start out as parts of bigger rocks in
space, mostly asteroids. But when they com-
pared the spectral colors of asteroids with
those of the most common kind of meteorite
that falls to Earth, they couldn’t find a match.
Now Japanese researchers have found the
answer in the first bits of asteroid returned to
Earth by a spacecraft. Dissecting the samples,
they have nailed the source of Earth’s most
common meteorites and gotten to the root of
astronomers’ color problem. The blast of the
solar wind and perhaps other ”space weath-
ering” processes, it turns out,
altered the spectral color of
the most common type of
asteroids in the inner aster-
oid belt—the “S types”—and
masked their true nature.
Most planetary scien-
tists, after close-up but still
not hands-on inspection of
asteroids, had concluded
that exposure to the rigors of
space must alter the surface
of S-type asteroids—but the
hard proof still wasn’t there.
Then, in June 2010, Japan’s Hayabusa space-
craft returned a sample of the S-type asteroid
Itokawa. In March at the Lunar and Plane-
tary Science Conference (Science, 15 April,
p. 302), Japanese researchers presented their
elemental, isotopic, and mineralogical anal-
yses of 50 Itokawa particles smaller than
100 micrometers. They report those and
additional analyses in detail in this issue of
Science (pp. 1113-1130).
The match between Itokawa and the
most common type of meteorite on Earth,
the ordinary chondrites, is right on, just as
astronomers predicted. That result brings
an “irrefutable closure to the long-standing
‘S asteroid conundrum,’ ” says asteroid spe-
cialist Clark Chapman of the Southwest
Research Institute in Boulder, Colorado.
Remote sensing, however, couldn’t
reveal what created the deceptive altera-
tion. For that, Hayabusa analysts used scan-
ning transmission electron microscopy
(STEM) to examine vanishingly thin slices
of 10 Itokawa rock particles that averaged
52 micrometers in diameter. They were
looking for “nanoblobs” that are opaque
and about the same size as the wavelength
of light so they could scatter light and make
Itokawa look redder, at least to the supersen-
sitive eye of a spectrometer. That sort of red-
dening is what misled the astronomers.
On the rims of five of the particles,
Takaaki Noguchi of Ibaraki University in
Japan and 17 colleagues found two thin lay-
ers. The outer, surface layer was only 5 to
15 nanometers thick and contained 1- to
2-nanometer-wide blobs of iron sulfide. To
judge by the STEM images, the outer layer
and its iron-sulfide nanoblobs formed when
micrometeorite impacts, solar wind, or both
vaporized nearby minerals and the vapor
condensed onto the
rock particle.
I mme d i a t e l y
beneath the outer
layer of iron-rich
Itokawa particles
was another layer,
20 to 50 nano meters
thick, containing
only metallic iron nanoblobs. These blobs
must have been formed by solar wind when
charged particles such as protons penetrated
Itokawa rock particles. Unless the outer lay-
er’s iron-sulfide nanoblobs have an unex-
pectedly large effect on light, these more
abundant metallic iron nanoblobs created by
solar wind “may be the major cause of the
spectral modification of Itokawa,” writes
Noguchi in an e-mail. So S-type asteroids
seem to have gotten a sunburn from sitting
too long in the wind of deep space. Their skin
was so reddened that for decades the parents
of terrestrial meteorites went unrecognized.
–RICHARD A. KERR
Hayabusa Gets to the Bottom of
Deceptive Asteroid Cloaking
PLANETARY SCI ENCE
to make sure we have the interdisciplinarity
to make us competitive.
Q: The European Research Council has a
good reputation, but the quality of other
European Commission research programs
has been criticized. Is it improving?
L.B.: There are issues that I would have
with some of the directions Europe is tak-
ing: for example, how on Earth the priori-
ties for the European Institute of Technology
are being set without seemingly any consul-
tation with the academic community (see
p. 1090). But engagement with Europe, both
with the commission and through bipartite
arrangements with universities in Europe, is
a very important way forward. We’re among
the lead universities in Europe, and I think
we should take on that leadership role and
extend it into Europe to work hand-in-glove
with partners in Germany and France and
other European countries.
Q: What’s your strategy for Asia? Will you
open campuses there?
L.B.: We’ve tied our research strategy and
our international strategy together. We do
not see that that strategy is going to involve
a wholesale expansion of undergraduate
education, nor do we see extra undergrad-
uate campuses as being necessary. But our
efforts in research are going to be far more
inter national; they are going to be focused
on direct interactions with the Far East, just
as much as with North America and Europe.
Q: Will you look at remote research
campuses?
L.B.: Yes, we will—when we have a critical
mass of investigator-led research that has
reached a point where institutional-level
support is appropriate. We are now identi-
fying areas where we should move towards
institution-level interaction. I’ve been to
China twice and Singapore twice this year,
once to India in spring and I’m going back
again in September, to look at where such
opportunities exist. I believe top interna-
tional universities will gradually be working
together. They can make a dent into these
global problems, but it will be done on the
basis of international cooperation. We’re not
going to find solutions to global warming
instantaneously.
Q: Haven’t previous attempts to build inter-
national partnerships between universities
tended to fizzle?
L.B.: Historically what you say may be cor-
rect; what I’m looking to do is to buck that
trend.
Thin-skinned. A rock particle (top) from asteroid
Itokawa (bottom) has layers (I and II) whose “nano-
blobs” (light smudges) redden the asteroid.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1081
NEWS&ANALYSIS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

T
.

N
O
G
U
C
H
I

E
T

A
L
.
,

S
C
I
E
N
C
E

3
3
3
,

6
0
4
6

(
2
6

A
U
G
U
S
T

2
0
1
1
)
;

C
O
U
R
T
E
S
Y

O
F

J
A
X
A

Astronomers had a problem. They knew that
meteorites start out as parts of bigger rocks in
space, mostly asteroids. But when they com-
pared the spectral colors of asteroids with
those of the most common kind of meteorite
that falls to Earth, they couldn’t find a match.
Now Japanese researchers have found the
answer in the first bits of asteroid returned to
Earth by a spacecraft. Dissecting the samples,
they have nailed the source of Earth’s most
common meteorites and gotten to the root of
astronomers’ color problem. The blast of the
solar wind and perhaps other ”space weath-
ering” processes, it turns out,
altered the spectral color of
the most common type of
asteroids in the inner aster-
oid belt—the “S types”—and
masked their true nature.
Most planetary scien-
tists, after close-up but still
not hands-on inspection of
asteroids, had concluded
that exposure to the rigors of
space must alter the surface
of S-type asteroids—but the
hard proof still wasn’t there.
Then, in June 2010, Japan’s Hayabusa space-
craft returned a sample of the S-type asteroid
Itokawa. In March at the Lunar and Plane-
tary Science Conference (Science, 15 April,
p. 302), Japanese researchers presented their
elemental, isotopic, and mineralogical anal-
yses of 50 Itokawa particles smaller than
100 micrometers. They report those and
additional analyses in detail in this issue of
Science (pp. 1113-1130).
The match between Itokawa and the
most common type of meteorite on Earth,
the ordinary chondrites, is right on, just as
astronomers predicted. That result brings
an “irrefutable closure to the long-standing
‘S asteroid conundrum,’ ” says asteroid spe-
cialist Clark Chapman of the Southwest
Research Institute in Boulder, Colorado.
Remote sensing, however, couldn’t
reveal what created the deceptive altera-
tion. For that, Hayabusa analysts used scan-
ning transmission electron microscopy
(STEM) to examine vanishingly thin slices
of 10 Itokawa rock particles that averaged
52 micrometers in diameter. They were
looking for “nanoblobs” that are opaque
and about the same size as the wavelength
of light so they could scatter light and make
Itokawa look redder, at least to the supersen-
sitive eye of a spectrometer. That sort of red-
dening is what misled the astronomers.
On the rims of five of the particles,
Takaaki Noguchi of Ibaraki University in
Japan and 17 colleagues found two thin lay-
ers. The outer, surface layer was only 5 to
15 nanometers thick and contained 1- to
2-nanometer-wide blobs of iron sulfide. To
judge by the STEM images, the outer layer
and its iron-sulfide nanoblobs formed when
micrometeorite impacts, solar wind, or both
vaporized nearby minerals and the vapor
condensed onto the
rock particle.
I mme d i a t e l y
beneath the outer
layer of iron-rich
Itokawa particles
was another layer,
20 to 50 nano meters
thick, containing
only metallic iron nanoblobs. These blobs
must have been formed by solar wind when
charged particles such as protons penetrated
Itokawa rock particles. Unless the outer lay-
er’s iron-sulfide nanoblobs have an unex-
pectedly large effect on light, these more
abundant metallic iron nanoblobs created by
solar wind “may be the major cause of the
spectral modification of Itokawa,” writes
Noguchi in an e-mail. So S-type asteroids
seem to have gotten a sunburn from sitting
too long in the wind of deep space. Their skin
was so reddened that for decades the parents
of terrestrial meteorites went unrecognized.
–RICHARD A. KERR
Hayabusa Gets to the Bottom of
Deceptive Asteroid Cloaking
PLANETARY SCI ENCE
to make sure we have the interdisciplinarity
to make us competitive.
Q: The European Research Council has a
good reputation, but the quality of other
European Commission research programs
has been criticized. Is it improving?
L.B.: There are issues that I would have
with some of the directions Europe is tak-
ing: for example, how on Earth the priori-
ties for the European Institute of Technology
are being set without seemingly any consul-
tation with the academic community (see
p. 1090). But engagement with Europe, both
with the commission and through bipartite
arrangements with universities in Europe, is
a very important way forward. We’re among
the lead universities in Europe, and I think
we should take on that leadership role and
extend it into Europe to work hand-in-glove
with partners in Germany and France and
other European countries.
Q: What’s your strategy for Asia? Will you
open campuses there?
L.B.: We’ve tied our research strategy and
our international strategy together. We do
not see that that strategy is going to involve
a wholesale expansion of undergraduate
education, nor do we see extra undergrad-
uate campuses as being necessary. But our
efforts in research are going to be far more
inter national; they are going to be focused
on direct interactions with the Far East, just
as much as with North America and Europe.
Q: Will you look at remote research
campuses?
L.B.: Yes, we will—when we have a critical
mass of investigator-led research that has
reached a point where institutional-level
support is appropriate. We are now identi-
fying areas where we should move towards
institution-level interaction. I’ve been to
China twice and Singapore twice this year,
once to India in spring and I’m going back
again in September, to look at where such
opportunities exist. I believe top interna-
tional universities will gradually be working
together. They can make a dent into these
global problems, but it will be done on the
basis of international cooperation. We’re not
going to find solutions to global warming
instantaneously.
Q: Haven’t previous attempts to build inter-
national partnerships between universities
tended to fizzle?
L.B.: Historically what you say may be cor-
rect; what I’m looking to do is to buck that
trend.
Thin-skinned. A rock particle (top) from asteroid
Itokawa (bottom) has layers (I and II) whose “nano-
blobs” (light smudges) redden the asteroid.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1083
NEWS&ANALYSIS
C
R
E
D
I
T
:

A
D
A
P
T
E
D

F
R
O
M

C
A
M
I
L
O

M
O
R
A

With apologies to Jonathan Swift’s famous
flea-ridden 1733 poem, biologists typically
do not believe life on Earth proceeds “ad
infinitum.” The planet must hold a limited,
not infinite, number of plants, animals,
fungi, and other complex species. But as sci-
entists have tried to pin down that number
for close to 2 centuries, their estimates have
ranged wildly from fewer than a million to
about 100 million. Now, a novel type of anal-
ysis, which takes advantage of an apparently
natural mathematical pattern in the biodiver-
sity produced by evolution, has settled on a
tally of about 8.7 million eukaryotic species,
give or take a million.
Many researchers find the case for this
latest estimate, outlined online 23 August in
PLoS Biology, convincing, though not the
final word. “It’s definitely an imaginative
new method,” says Robert May of the Uni-
versity of Oxford in the United Kingdom,
who wrote a commentary accompanying the
study and has previously put forth his own
predictions regarding the extent of the plan-
et’s biodiversity. “But all methods are neces-
sarily approximate.”
Biologist Edward O. Wilson of Harvard
University suggests that the study might
undershoot the mark, however, although
he admires the attempt to provide an inno-
vative way of calculating the planet’s total
biodiversity. “We should keep doing it, but
we should be always suspicious,” he says.
Undiscovered species tend to be undiscov-
ered for a reason, Wilson notes—they’re
rare, isolated, and, consequently, probably a
lot more diverse than many suspect. Explor-
ing this unknown diversity will be “one of
the greatest areas of biology awaiting us this
century,” Wilson adds, especially because
scientists are in a race to discover species
before they go extinct.
One of the first to pursue scientific esti-
mates of species numbers was Oxford ento-
mologist John Obadiah Westwood in 1833.
He surveyed the extrapolations of many of
his early colleagues, based on current taxo-
nomic collections, reporting an estimate that
about half a million insect species possibly
lived on Earth.
In 1982, Terry Erwin, a taxonomist at the
Smithsonian National Museum of Natural
History, took a new, experimental tack and
got a much bigger number: He and his col-
leagues gassed 19 trees in a jungle in Pan-
ama, collecting about 1000 unique beetle
species. Using a set of assumptions still
hotly debated today, Erwin calculated that
tropical beetles and their arthropod relatives
alone constituted 30 million species.
Erwin launched a boom of research into
biodiversity, but his analysis and similar
ones that followed suffered from the same
flaw, says marine biologist Boris Worm of
Dalhousie University in Halifax, Canada:
They couldn’t be tested. There’s “no way
unless you go out and count all the beetles,”
Worm says.
While trying to tally the number of
organisms in the sea as part of the Census of
Marine Life, Worm and his colleagues stum-
bled on a taxonomic pattern that seemed to
resolve this dilemma. With each step down
in Linnaeus’s famous classification sys-
tem, the number of classes, orders, and so
on tended to creep up in a predictable man-
ner, his team realized. The kingdom Anima-
lia, for instance, hosts 32 recorded phyla,
90 classes, 493 orders, 5404 families, and
94,240 genera. Extending that growth curve
out, Worm and his colleagues predicted
there should exist some 7,770,000 animal
species—even though taxonomists have
identified only about 950,000 so far. Plants,
fungi, and other eukaryotes total about
1 million other species, they also predict.
The researchers validated their taxonomic
approach by comparing estimates made this
way with real species counts from well-
known evolutionary branches, such as mam-
mals and birds. “Our predictions were very,
very close to the true numbers,” says study
co-author Camilo Mora, a marine ecologist
at the University of Hawaii, Manoa. If their
8.7 million species estimate is valid, Mora
and his colleagues note, taxonomists have
yet to catalog nearly 90% of land- and ocean-
dwelling eukaryotes. (The team strayed
away from crunching a total for bacteria and
archaea because scientists still puzzle over
how to categorize species for the microbes.)
Mora and Worm’s study highlights that
scientists are getting closer to pinpoint-
ing Earth’s vast biodiversity, says Nigel
Stork, a biologist at Griffith University in
Queensland, Australia. Last year, he and
colleagues used a combination of beetle
counts made in New Guinea and complex
mathematical modeling to estimate that
there are some 3.7 million arthropod spe-
cies on Earth. Despite the radically differ-
ent method, much more akin to Erwin’s early
work, Stork’s estimate falls in line with what
Mora and Worm found: “We are beginning
to home in,” Stork says.
Many still argue that a tally of 8.7 mil-
lion species sells our planet short. Taxono-
mists have only brushed the surface of the
diversity of many microscopic organisms
such as yeasts, cautions Steven Stephen-
son, an ecologist who specializes in fungi at
the University of Arkansas, Fayetteville. He
suspects that the PLoS Biology study under-
estimates the diversity of these and other
eukaryotic microbes.
And the analysis may have a more funda-
mental flaw, some suggest. The relationship
between levels in Linnaeus’s classification
system is a statistical pattern, not one based
on sound scientific principles, says Lucas
Joppa, a conservation ecologist at Microsoft
Research in the United Kingdom. “There’s
no inherent ecological rationale for why
kingdom should be to phylum as phylum is
to class,” Joppa says.
Erwin agrees and says that he’s inclined
not to set a limit on species numbers at all.
The more he digs into the Amazon, the more
it looks like the insect species there, like
Swift’s fleas, don’t have a finite end. “Bio-
diversity is infinite,” he says. “And there is
no way to estimate the infinite.”
–DANIEL STRAIN
8.7 Million: A New Estimate for All
The Complex Species on Earth
BI ODI VERSI TY
32
90
493
5404
94,240
953,434
10,000,000
1,000,000
100,000
10,000
1,000
100
10
ADDING UP ANIMALS
Number of described groups
1. Phylum
2. Class
3. Order
4. Family
5. Genus
6. Species
Phylum
Class
Order
Predicted
Described/Cataloged
Family
Genus
Species
K
I
N
G
D
O
M

A
N
I
M
A
L
I
A
1 2 3 4 5 6
Predictive pyramid. Given the number of animal
phyla, orders, families, and genera, researchers
predict that the kingdom has 7,770,000 species.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1084
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

A
N
N

G
I
B
B
O
N
S
;

B
E
N
C
E

V
I
O
L
A
/
M
P
I

E
V
A
;

A
N
N

G
I
B
B
O
N
S
;

B
E
N
C
E

V
I
O
L
A
/
M
P
I

E
V
A DENISOVA CAVE, SIBERIA—Bence Viola
first saw the ancient molar last summer, just
after a piece of it was dug out of layers full
of brown dirt, gray rock, animal bones, stone
tools, and goat feces. He considered the tooth
fragments too big and weirdly shaped to be
human. “I thought it must belong to a cave
bear,” he says.
Several fossils were found that summer
in this remote cave in the Altai Mountains.
Some, including a toe bone, looked human
and were to be sent for DNA analysis to
paleogeneticist Svante Pääbo at the Max
Planck Institute for Evolutionary Anthro-
pology in Leipzig, Germany. Viola, a post-
doc at Max Planck, almost didn’t include the
molar. But he and Pääbo decided to play it
safe and test all the new fossils. The layer
that held the molar in Denisova Cave was
also the resting place of a girl’s finger bone,
which was so well preserved that Pääbo’s
lab was able to sequence its nuclear genome
and identify it as belonging to a previously
unknown type of archaic human. The team
called them the Denisovans. For the first
time, researchers had a genome in search of
a fossil record, so every possible new bone
was significant.
Back in Leipzig, graduate student
Susanna Sawyer was charged with extract-
ing DNA from the animal bones. In June, she
stopped Pääbo in the hall. “I think I found
another Denisovan,” she said. Preliminary
analysis suggested that the molar’s DNA was
similar to that of the cave girl’s. Pääbo shook
Sawyer’s hand—this was only the third fossil
ever found of a Denisovan, the others being
the bit of finger bone and another molar, also
from Denisova cave.
Who Were the Denisovans?
At an unusual meeting at a Siberian cave, researchers find that these mysterious archaic
humans lived in the same place as both modern humans and Neandertals—though not
necessarily at the same time—and their range probably stretched into east Asia
Cave treasure.
Researchers have
found the tooth of
a Denisovan, plus a
sophisticated stone
bracelet and tools,
in Denisova Cave.
NEWSFOCUS
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1085
NEWSFOCUS
B
i
y
a
C
h
u
l
y
s
h
m
a
n
K
a
t
u
n
C
h
a
r
y
s
h
K
o
k
s
a
B
a
s
h
k
a
u
s
C
h
u
ya
Anui
P
e
s
c
h
a
n
a
y
a
Ob
Denisova
Ust-Karakol
Kara-Bom
Karama
Okladnikov
Chagyrskaya
Strashnaya
0 60 km
Cave site
Open air site
Biysk
Gorno-Altaysk
Novosibirsk
K A Z A K H S TA N
MO NG O L I A
C H I NA
A
l
t
a
i

M
o
u
n
t
a
i
n
s
What’s more, preliminary analysis of the
mitochondrial DNA from the toe bone sug-
gests that it belonged not to a Denisovan
but to a Neandertal. That means both types
of archaic humans lived in the same cave.
And the large, three-room cave also holds
sophisticated stone tools and bone artifacts
that appear to have been crafted by our own
species, Homo sapiens. “The one place
where we are sure all three human forms
have lived at one time or another is here in
Denisova Cave,” Pääbo said.
Today the cave is off the beaten path, in
southern Siberia, 350 kilometers north of
the Russian border with both Kazakhstan
and Mongolia, and closer to Beijing than
Moscow. Now the Denisovan discoveries
have shifted the spot-
light from ancient
humans in Africa,
the Middle East, and
Europe to those in this
remote corner of Asia.
As Russian Academy of Sci-
ences (RAS) archaeologist Anatoly
Derevianko puts it: “The world is
looking eastward.”
To that end, Derevianko and his
Russian colleagues invited Pääbo
and a select group of human origins
researchers from different disciplines
and countries to a remarkable sym-
posium at an archaeological camp
near Denisova Cave in July. Their
goal was to try to solve the mys-
tery of the cave girl’s identity, to find
more of her people, and to explore
how the discovery is challenging
models of modern human origins.
In lively discussions sometimes cat-
alyzed by vodka toasts, they com-
pared what archaeology, genetics,
and fossils reveal about the world the
Denisovans inhabited 30,000 to 50,000 years
ago. Genomic data have already shown that
our ancestors mingled with archaic humans,
who may have given us valuable immune cell
types (see sidebar, p. 1086). But it’s not clear
when and where this happened.
Invisible human
The gathering gave Derevianko, director of
the Institute of Archaeology and Ethnogra-
phy at the RAS in Novosibirsk, a chance to
showcase some of the region’s impressive
archaeological sites.
Driving off dirt roads in
troop movers and along
rutted roads in inde-
structible UAZ vans,
the Russians took their
visitors to a dozen digs.
Some were caves at the
edge of alpine forests of
silver birch and Siberian
larch; others were open-
air sites in grassy mead-
ows of bee balm, wild
mint, and edelweiss.
The trail of ancient
humans starts with H.
erectus, which left prim-
itive “pebble” tools in
the Altai almost 800,000
years ago. After a hiatus
when the climate was
frigid, the descendants
of H. erectus returned
by 300,000 years ago, leaving more tools
behind. Some kind of human has lived here
ever since.
Starting 80,000 to 70,000 years ago,
archaic humans began to use more modern
methods to make tools at sites called Kara
Bom and Ust-Karakol, where 10% of the
tools were blades or burins (a tool used to
chisel wood); the Russians see this as the first
stirrings of modern human behavior here.
From 50,000 to 30,000 years ago,
the archaic people hunted bear, lynx, and
wild boar in the Altai
Mountains, where they
set up seasonal camps
in summer, said RAS
archaeologist Mikhail
Shunkov as he led the
tours. They retreated to
limestone caves such as Denisova in winter.
“With a natural opening for a chimney, the cave
was quite a cozy place,” Shunkov said, point-
ing to an opening in the ceiling at Denisova.
With a clear view of the Anui River—and any
humans or animals passing below—Denisova
must have been choice housing, said Pääbo,
noting how sunlight streaming through the
opening overhead lit the cave like a chapel.
“It is kind of cool to imagine that the person
whose genome was sequenced had seen these
walls,” he said.
At about this time, at least two different
types or local cultures of artifacts appear, one
at Kara Bom and one at Ust-Karakol. The
Russians consider both to be sophisticated C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

M
A
P

A
D
A
P
T
E
D

F
R
O
M

A
.

D
E
R
E
V
I
A
N
K
O

A
N
D

M
.

S
H
U
N
K
O
V
,

P
A
L
E
O
N
T
O
L
O
G
I
C
A
L

J
.

4
3
,

8

(
2
0
0
9
)
;

B
E
N
C
E

V
I
O
L
A
/
M
P
I

E
V
A

(
2
)
Online
sciencemag.org
Podcast interview
with author
Ann Gibbons.
Meeting of the minds. Archaeologist Anatoly Derevianko (top) and paleo-
geneticist Svante Pääbo worked together to discover the Denisovans.
Room with a view. Denisova Cave was such prime
real estate, it attracted three kinds of humans.
a.
f Sci-
gist AAnatoly
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1086
NEWSFOCUS
C
R
E
D
I
T
:

A
B
I
-
R
A
C
H
E
D

E
T

A
L
.
,

S
C
I
E
N
C
E

(
A
D
V
A
N
C
E
D

O
N
L
I
N
E

E
D
I
T
I
O
N
)

cultures traditionally associated with only
H. sapiens. Similarly advanced artifacts
appear at the same time in Denisova, with
stone bladelets used on spears; pendants
made of teeth of fox, bison, and deer; and
even a bracelet made of a mineral found hun-
dreds of kilometers away. Until recently, the
archaeologists had “no doubts that people
associated with this industry were anatomi-
cally modern,” Derevianko says. But now,
thanks to the genomic results, it’s possible
that some were Denisovans, Shunkov says.
To identify the toolmakers, researchers
need fossils, but they are few and far between.
As a result, “it remains unknown what the
Denisovan looked like or how he behaved,”
says biological anthropologist Maria
Mednikova of the RAS in Moscow. So
Viola’s talk at the meeting, describing the
single new tooth, drew intense interest. Like
the first molar found, it is very large and
lacks specialized features found in Neander-
tals. Nor does the tooth resemble a modern
human molar, as it has many unusual cusps,
Viola says. The finger bone fragment that
first yielded Denisovan DNA was so small
that it yielded little information other than
it was a child’s because the growth plate was
not fused.
In addition to the few Denisovan fossils,
Neandertals also left fossils and characteris-
tic Mousterian stone points and scrapers in
Denisova and other caves. At the meeting,
Russian researchers described new finds of
Neandertal tools and fossils in caves just
100 and 150 kilometers away from Denisova
Cave, dated to 45,000 years ago. Mednikova
adds that the toe bone from Denisova looks
most like a Neandertal toe from Iraq, fitting
well with the preliminary DNA finding. And
yet Derevianko thinks Neandertals didn’t
stay long here, because their bones and arti-
facts disappear by 40,000 years ago. He
views them as brief visitors, probably com-
ing from the west in Kazakhstan.
Neighbors, or successors?
It is now clear that Neandertals, Deniso-
vans, and modern humans once occupied
the Altai—but were they all there at the
same time? This is hard to answer because
there are questions about the dating of cru-
cial layer 11 in Denisova Cave. This meter-
thick layer held the Denisovan finger and
molars, the Neandertal toe, and the mod-
ern human artifacts, although some were
found in different galleries of the cave. The
bones and teeth are too fragmentary to be
dated directly. But radiocarbon dating of
seven animal bones with cut marks from
layer 11 provides dates of 50,000 years or
older in both galleries. Yet the layer’s young-
est sediments date to as late as 16,000 to
30,000 years ago, as reported in December
in Nature. Thus layer 11 has artifacts from at
least two different periods. And, in the south
gallery near the spot where the finger bone
was found, an obvious wedge of disturbed
sediment suggests some mixing.
For now, Derevianko and colleagues pro-
pose sequential occupations: The Deniso-
vans were in the cave about 50,000 years
ago, Neandertals came in briefly about
A Denisovan Legacy in the Immune System?
Everybody knows about the dangers of inbreeding (see Hapsburg dynasty, collapse of). In fact the
reproductive strategies of many animals are based on avoiding it, as when female chimpanzees
move out of their birth groups to mate. Last year, researchers showed that human ancestors took
that strategy to its limits by breeding with the now-extinct Neandertals and Denisovans (Science,
28 January, p. 392). Now a study published online in Science this week (http://scim.ag/Abi-Rached)
suggests that such mating was beneficial, boosting the immune systems of early Europeans and
Asians and leaving a valuable legacy in the genes of many people alive today. “This is the first sug-
gestion that something that came from archaic hominins into modern humans conferred an advan-
tage,” says paleogeneticist Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology
in Leipzig, Germany.
Genomic data from fossils thus far suggest that living people carry only small amounts of
archaic DNA. Only 2% to 7% of the DNA of today’s Europeans and Asians apparently came from
the ancient Denisovans and Neandertals
(see main text). The new paper exam-
ines Europeans and Asians and finds that
archaic people contributed more than half
of the alleles that code for proteins made
by the human leukocyte antigen system
(HLA), which helps the immune system
recognize pathogens. “Archaic alleles
have significantly shaped modern human
immune systems,” wrote Peter Parham
and Laurent Abi-Rached of Stanford Uni-
versity in Palo Alto, California.
Immunogeneticist Parham has spent
16 years puzzling over the evolution of
one rare HLA allele, called HLA-B*73. This
variant is quite different from others but
is similar to alleles in the same position in
the genomes of chimps and gorillas. So it seems to be ancient, perhaps arising long before our
ancestors split from gorillas about 16 million years ago. Yet today, B*73 is concentrated in western
Asia, where modern humans have lived for less than 90,000 years, and it is absent from African
tribes who usually carry the most ancient gene lineages.
While studying this allele, Parham’s team got a big break last year when Pääbo’s team pub-
lished the complete genome of the Denisovan cave girl. She didn’t carry B*73—and it hasn’t been
found in Siberia—but she carried two other linked HLA-C variants, which occur on the same stretch
of chromosome 6. If living people have any of these variants, they almost always carry at least two
of the three variants—as did the cave girl. So even though she lacked B*73, the researchers pro-
pose that all three variants were inherited, often in pairs, from archaic humans in Asia. The Deniso-
vans are the prime suspects, given their presumed distribution in Asia.
The team also examined other HLA alleles in three Neandertals and one Denisovan and found
several other ancient variants that today show up in living Asians or Europeans. Parham thinks
these variants were beneficial and so, once acquired from archaic people, spread rapidly in small
but expanding modern populations. “The fact [that these genes] may have been parachuted into
modern humans is an attractive interpretation,” says immunologist John Trowsdale of the Univer-
sity of Cambridge in the United Kingdom.
However, others are not quite convinced that the alleles came from archaic humans. Parham’s
team hasn’t completely ruled out other explanations for the gene distributions, such as certain
types of selection, says geneticist David Reich of Harvard University. Regardless, he says, “I am
happy to see people using archaic genomes for different kinds of analyses.” –A.G.
Ancient roots. The allele HLA-B*73, today mostly
seen in west Asia, may come from Denisovans.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1087
NEWSFOCUS
C
R
E
D
I
T
S

(
T
O
P

T
O

B
O
T
T
O
M
)
:

A
N
N

G
I
B
B
O
N
S
;

M
A
R
I
A

D
O
B
R
O
V
O
L
S
K
A
Y
A
;

A
N
N

G
I
B
B
O
N
S
45,000 years ago, and modern humans fol-
lowed. But the researchers agree that the
microstratigraphy of the cave needs more
analysis. They are redating layer 11 with radio-
carbon on more cut-marked animal bones.
Overall, Derevianko and his colleagues
see a gradual, local evolution of H. erectus
into H. sapiens in the Altai, with a brief intru-
sion of Neandertals and Denisovans. This
fits a minority view of human origins, called
multiregionalism, which posits that the
descendants of H. erectus evolved into
Neandertals and modern humans—and,
apparently, Denisovans—in different
regions. Then humans coming out of Africa
mingled with the other groups and H. sapi-
ens emerged worldwide.
As Russian and Chinese archaeologists
raised their glasses to toast regional conti-
nuity, however, several geneticists shifted
uncomfortably or even quietly demurred:
That theory is in contrast to the long-
prevailing view that H. sapiens was born
in Africa and swept the globe, wiping out
local archaic peoples. And in light of the
genomic data, most geneticists now hold
a middle-of-the-road view that modern
humans arose in and spread out of Africa,
then interbred with local archaic peoples to a
limited degree (Science, 28 January, p. 392).
“If you write that I drank a toast to [regional]
continuity, I’ll kill you,” one geneticist told
a reporter.
But the geneticists do agree with the
Russians that modern humans mingled with
both Neandertals and Denisovans. Pääbo’s
team found in 2010 that living Europe-
ans and Asians have inherited about 2.5%
of their DNA from Neandertals (Science,
7 May 2010, pp. 680 and 710) and
that living Melanesians carry an
additional 5% of Denisovan DNA.
If modern humans interbred with
Neandertals, researchers speculated
that fossils of each group, about the
same age and found close to each
other in Israeli caves, represented the
groups who mixed sometime before
90,000 years ago. Those modern
people carrying a small amount of
Neandertal DNA then split into at
least two groups—one that headed
into Europe to replace the Neander-
tals there, and a second group that
headed into Asia to mix with the
Denisovans, says population geneti-
cist David Reich of Harvard Medical
School in Boston.
At the meeting, the DNA research-
ers offered some new insights into
this story. They found that the three
Denisovans, all from one cave, had more vari-
ation in their mtDNA than did seven
Neandertals from western Europe
to Siberia, Sawyer reported. This
and another report at the meeting—
that Australian Aborigines, like
Melanesians, have inherited 5%
of their DNA from Denisovans—
suggests that the Denisovan home
range once stretched far beyond the
Altai, into eastern Asia. “This tells
us that the Denisovans had large
population sizes,” despite their
puny fossil record, Pääbo says. It
also shows that Denisovans and the
ancestors of Melanesians must have
interbred before 40,000 to 60,000
years ago, when Aborigines first
settled Australia.
As for the timing of the Nean-
dertal-human mixing, the newest
analyses tend to push that younger.
Population geneticist Montgom-
ery Slatkin of the University of
California, Berkeley, said that
his model runs gave him a wide
range of preliminary results, from
65,000 years to 45,000 years ago,
but he’s still working the numbers.
Reich reported that his independent
analyses also suggest a younger
date. If the mixing happened more
recently than 90,000 years ago, it
rules out the Israeli fossils as repre-
sentatives of the groups who mixed.
Others, such as Derevianko and
paleoanthropologist John Hawks of
the University of Wisconsin, Mad-
ison, interpret the genetic data dif-
ferently. They think that even small amounts
of interbreeding confirm the regional conti-
nuity model, and that there was more mixing
in the past, but its traces were erased by later
waves of immigrants who swamped out the
archaic genes.
To help decide among these models, sev-
eral groups are searching for Denisovans
beyond Denisova, as far east as China, where
Pääbo is now analyzing fossil DNA. As Pääbo
climbed down a ladder into a floodlit pit at
Denisova and bent his lanky frame low to get
a good look at layer 11, a colleague shouted:
“Grab a trowel, Svante.” Pääbo didn’t. But
like the others, he is convinced that all
types of data—genetic, archaeological, and
fossil—will have to be integrated in order
to tell the story of the Denisovans and so of
our own species. “We’re beginning to clarify
history in eastern Eurasia,” Pääbo said, “and
I’m sure that in the next few years, there will
be more discoveries.”
–ANN GIBBONS
On tour. Archaeologist Mikhail Shunkov showcased the
many archaeological sites of the Altai Mountains.
Teamwork. Anthropologist Maria Mednikova (top) analyzed
fossils, and geneticists Susanna Sawyer and David Reich
studied the DNA of the ancient Denisovans.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1086
NEWSFOCUS
C
R
E
D
I
T
:

A
B
I
-
R
A
C
H
E
D

E
T

A
L
.
,

S
C
I
E
N
C
E

(
A
D
V
A
N
C
E
D

O
N
L
I
N
E

E
D
I
T
I
O
N
)

cultures traditionally associated with only
H. sapiens. Similarly advanced artifacts
appear at the same time in Denisova, with
stone bladelets used on spears; pendants
made of teeth of fox, bison, and deer; and
even a bracelet made of a mineral found hun-
dreds of kilometers away. Until recently, the
archaeologists had “no doubts that people
associated with this industry were anatomi-
cally modern,” Derevianko says. But now,
thanks to the genomic results, it’s possible
that some were Denisovans, Shunkov says.
To identify the toolmakers, researchers
need fossils, but they are few and far between.
As a result, “it remains unknown what the
Denisovan looked like or how he behaved,”
says biological anthropologist Maria
Mednikova of the RAS in Moscow. So
Viola’s talk at the meeting, describing the
single new tooth, drew intense interest. Like
the first molar found, it is very large and
lacks specialized features found in Neander-
tals. Nor does the tooth resemble a modern
human molar, as it has many unusual cusps,
Viola says. The finger bone fragment that
first yielded Denisovan DNA was so small
that it yielded little information other than
it was a child’s because the growth plate was
not fused.
In addition to the few Denisovan fossils,
Neandertals also left fossils and characteris-
tic Mousterian stone points and scrapers in
Denisova and other caves. At the meeting,
Russian researchers described new finds of
Neandertal tools and fossils in caves just
100 and 150 kilometers away from Denisova
Cave, dated to 45,000 years ago. Mednikova
adds that the toe bone from Denisova looks
most like a Neandertal toe from Iraq, fitting
well with the preliminary DNA finding. And
yet Derevianko thinks Neandertals didn’t
stay long here, because their bones and arti-
facts disappear by 40,000 years ago. He
views them as brief visitors, probably com-
ing from the west in Kazakhstan.
Neighbors, or successors?
It is now clear that Neandertals, Deniso-
vans, and modern humans once occupied
the Altai—but were they all there at the
same time? This is hard to answer because
there are questions about the dating of cru-
cial layer 11 in Denisova Cave. This meter-
thick layer held the Denisovan finger and
molars, the Neandertal toe, and the mod-
ern human artifacts, although some were
found in different galleries of the cave. The
bones and teeth are too fragmentary to be
dated directly. But radiocarbon dating of
seven animal bones with cut marks from
layer 11 provides dates of 50,000 years or
older in both galleries. Yet the layer’s young-
est sediments date to as late as 16,000 to
30,000 years ago, as reported in December
in Nature. Thus layer 11 has artifacts from at
least two different periods. And, in the south
gallery near the spot where the finger bone
was found, an obvious wedge of disturbed
sediment suggests some mixing.
For now, Derevianko and colleagues pro-
pose sequential occupations: The Deniso-
vans were in the cave about 50,000 years
ago, Neandertals came in briefly about
A Denisovan Legacy in the Immune System?
Everybody knows about the dangers of inbreeding (see Hapsburg dynasty, collapse of). In fact the
reproductive strategies of many animals are based on avoiding it, as when female chimpanzees
move out of their birth groups to mate. Last year, researchers showed that human ancestors took
that strategy to its limits by breeding with the now-extinct Neandertals and Denisovans (Science,
28 January, p. 392). Now a study published online in Science this week (http://scim.ag/Abi-Rached)
suggests that such mating was beneficial, boosting the immune systems of early Europeans and
Asians and leaving a valuable legacy in the genes of many people alive today. “This is the first sug-
gestion that something that came from archaic hominins into modern humans conferred an advan-
tage,” says paleogeneticist Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology
in Leipzig, Germany.
Genomic data from fossils thus far suggest that living people carry only small amounts of
archaic DNA. Only 2% to 7% of the DNA of today’s Europeans and Asians apparently came from
the ancient Denisovans and Neandertals
(see main text). The new paper exam-
ines Europeans and Asians and finds that
archaic people contributed more than half
of the alleles that code for proteins made
by the human leukocyte antigen system
(HLA), which helps the immune system
recognize pathogens. “Archaic alleles
have significantly shaped modern human
immune systems,” wrote Peter Parham
and Laurent Abi-Rached of Stanford Uni-
versity in Palo Alto, California.
Immunogeneticist Parham has spent
16 years puzzling over the evolution of
one rare HLA allele, called HLA-B*73. This
variant is quite different from others but
is similar to alleles in the same position in
the genomes of chimps and gorillas. So it seems to be ancient, perhaps arising long before our
ancestors split from gorillas about 16 million years ago. Yet today, B*73 is concentrated in western
Asia, where modern humans have lived for less than 90,000 years, and it is absent from African
tribes who usually carry the most ancient gene lineages.
While studying this allele, Parham’s team got a big break last year when Pääbo’s team pub-
lished the complete genome of the Denisovan cave girl. She didn’t carry B*73—and it hasn’t been
found in Siberia—but she carried two other linked HLA-C variants, which occur on the same stretch
of chromosome 6. If living people have any of these variants, they almost always carry at least two
of the three variants—as did the cave girl. So even though she lacked B*73, the researchers pro-
pose that all three variants were inherited, often in pairs, from archaic humans in Asia. The Deniso-
vans are the prime suspects, given their presumed distribution in Asia.
The team also examined other HLA alleles in three Neandertals and one Denisovan and found
several other ancient variants that today show up in living Asians or Europeans. Parham thinks
these variants were beneficial and so, once acquired from archaic people, spread rapidly in small
but expanding modern populations. “The fact [that these genes] may have been parachuted into
modern humans is an attractive interpretation,” says immunologist John Trowsdale of the Univer-
sity of Cambridge in the United Kingdom.
However, others are not quite convinced that the alleles came from archaic humans. Parham’s
team hasn’t completely ruled out other explanations for the gene distributions, such as certain
types of selection, says geneticist David Reich of Harvard University. Regardless, he says, “I am
happy to see people using archaic genomes for different kinds of analyses.” –A.G.
Ancient roots. The allele HLA-B*73, today mostly
seen in west Asia, may come from Denisovans.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1088
NEWSFOCUS
The notion that tissue engineers can provide
a stock of lab-grown body parts to replace
faulty tissues is still, for the most part, a
dream. Ready-made hearts, livers, or kid-
neys that could ease the shortage of donor
organs will not be available in the clinic any-
time soon. But recent progress suggests the
dream is not completely beyond reach. Lab-
grown bladders are functioning in dozens
of patients in the United States, and doctors
in Europe have implanted lab-grown tra-
cheas into several patients. In Japan, several
dozen children and young adults born with
severe heart defects are living with tissue-
engineered cardiac blood vessels. The first
received implants 10 years ago. They go to
school, hold full-time jobs, play sports—
in short, says Christopher Breuer, one of
the implants’ developers, they live active,
healthy lives. This month, after an arduous
approval process, surgeons are testing the
blood vessels in the first U.S. patients.
The U.S. trial marks “an important
signpost for the whole field,” says Joseph
Vacanti, a transplant surgeon and tissue
engineer at Massachusetts General Hospital
in Boston. The implants, which are used to
connect a major cardiac vein and the artery
that carries blood to the lungs, are made of
a synthetic scaffold seeded with cells from
the patient’s own bone marrow. In the body,
the graft develops into a living blood vessel
that grows with the patient.
The engineered vessels were devel-
oped by a group at Yale University led by
Breuer, a pediatric surgeon, and Toshiharu
Shinoka, a cardiosurgeon. Although getting
approval for the trial from the U.S. Food and
Drug Administration (FDA) took more than
4 years and generated more than 3000 pages
of documents, the process paid off, Breuer
says: Recent animal studies, arising in part
from questions the FDA asked, have turned
some of Breuer and Shinoka’s assumptions
upside down, leading to a better understand-
ing of how the graft works and ideas for how
to improve it.
The new experiments suggest that
inflammation, long seen as an enemy of
transplants and artificial implants alike,
seems to play a key role in the transforma-
tion of the cell-filled scaffold into a healthy
blood vessel. And stem cells, which have
been seen as the stars of tissue engineering,
play a less significant role than expected.
The results are prompting tissue engineers
to rethink the role of inflammation and stem
cells, says Anita Driessen-Mol, a tissue
engineer at Eindhoven University of Tech-
nology in the Netherlands. “It’s very inspir-
ing work,” she says.
Replumbing the heart
The lab-made blood vessels are meant for
children whose severely malformed hearts
are unable to supply their bodies with
enough oxygen. At birth the children are
known as “blue babies” for the skin tint that
results. Unlike children with a normal heart,
which has two blood-pumping chambers,
or ventricles, these children have only one
working ventricle. Without a repair, Breuer
says, 70% of children with such defects will
die before their first birthday.
In the late 1960s, the surgeon Francis
Fontan and his colleagues developed a tech-
nique to make such hearts more efficient.
They rearranged the organ’s plumbing to
concentrate pumping in the single func-
tioning ventricle. Over the years, surgeons
have improved the procedure by adding a
length of blood vessel to better connect the
heart’s inferior vena cava, which collects
blood from veins in the lower body, to the
pulmonary artery, which leads to the lungs,
bypassing the heart (see diagram). In a few
cases, surgeons can build this diversionary
vessel from the patient’s own tissue. But
often there isn’t enough tissue available, and
surgeons use tubes of synthetic materials
such as Gore-Tex.
Such artif icial blood vessels have
significant drawbacks, Breuer says. The
grafts can become calcified, trigger blood
clots, and, if cells build up on the inside,
can develop stenosis, a dangerous narrow-
ing of the vessel. And because the synthetic
graft doesn’t grow with the child, surgeons
C
R
E
D
I
T
S
:

(
T
O
P
)

A
D
A
P
T
E
D

F
R
O
M

J
.

D
.

R
O
H

E
T

A
L
.
,

P
N
A
S

1
0
7
,

1
0

(
9

M
A
R
C
H

2
0
1
0
)

©

N
A
T
I
O
N
A
L

A
C
A
D
E
M
Y

O
F

S
C
I
E
N
C
E
S
;

(
B
O
T
T
O
M
)

B
.

S
T
R
A
U
C
H
/
S
C
I
E
N
C
E
Mending the Youngest Hearts
Researchers have begun implanting tissue-engineered blood vessels into toddlers with
heart defects, and new studies of the grafts in animals show they work in unexpected ways
TI SSUE ENGI NEERI NG
Lifesaving implant. A graft bypasses the heart,
redirecting low-oxygen blood from the inferior
vena cava directly to the pulmonary artery.
Biodegradable scaffold. Bone marrow cells seeded
on a synthetic frame attract immune cells; these signal
nearby vessels to grow into and over the graft.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1089
NEWSFOCUS
must either delay surgery until the heart
has grown larger or implant a graft that is
initially too big.
For more than a decade, Breuer and
Shinoka have been working to develop lab-
grown vessels that act like a patient’s own
tissue. The team uses a scaffold made of a
biodegradable polyester tube, which they
incubate briefly with a patient’s bone mar-
row mononuclear cells (BMCs)—a mix of
cells including immune cells and blood-
and vessel-forming stem cells. Some of the
patient’s cells stick, seeding the scaffold.
Preclinical experiments in lambs showed
that the grafts soon formed a normal-
looking blood vessel, with endo-
thelial cells lining the inside and
smooth muscle cells surrounding
them. As expected, the scaffold
degraded within a few months;
new collagen fibrils, the connec-
tive tissue that helps blood vessels
hold their shape, replaced it. And
just like a normal blood vessel,
the new tissue could grow. Based
on those positive results, the clin-
ical trial in Japan was approved
and went ahead. Results have
been very promising. The only
complications were a few cases
of stenosis, Shinoka and his col-
leagues have reported.
Still, researchers weren’t
sure exactly how the new blood vessel
formed. Were the seeded cells growing and
differentiating? Or were new cells migrat-
ing into the graft? To better understand what
happens after implantation—and to prove to
the FDA “that each step of the way we were
doing what we thought we were,” Breuer
says, he and his colleagues went back to the
lab. Using new, more precise fabrication
techniques, they developed a mouse-sized
version of their blood vessel scaffold. They
seeded it with human BMCs and implanted
the vessel in mice. To their surprise, though
the vessel remained intact, the human cells
disappeared within a week.
To test whether the human cells dis-
appeared because they were attacked by
the mouse immune system, the research-
ers seeded a vessel scaffold with mouse
cells genetically matched to the recipient
but tagged with green fluorescent protein
(GFP). In a paper in The FASEB Journal this
month, they confirm their earlier results: A
week after implantation, almost all of the
GFP-tagged cells had disappeared.
The observation “was a huge eye opener
for the field,” Driessen-Mol says. “We
thought that the cells you put in there would
still be around for weeks.” Instead, it seems,
the blood vessel that forms somehow comes
from cells in the host’s body. Although the
seeded cells don’t stick around for long, they
do provide an advantage to the implant, the
researchers reported last year in the Pro-
ceedings of the National Academy of Sci-
ences. They secrete a protein that attracts
monocytes, immune cells that modulate
inflammation and can help prompt the for-
mation of new blood vessels. Breuer’s team
showed that the seeded cells “are essential
to initiate the proper kind of inflammation
response. Somehow they attract the right
kind of initial cells,” Driessen-Mol says.
To pinpoint the origin of the cells that
build the new blood vessel, the Yale research-
ers created special chimeric mice. They
gave female mice a lethal dose of radiation
to destroy the rodents’ immune and blood-
forming stem cells, then rescued the mice
with bone marrow stem cells from males
that carried the GFP gene. The donated cells
repopulated the animals’ bone marrow, and
soon the female mice had GFP-tagged male
cells in their blood.
The researchers implanted their cell-
seeded polyester tube into the chimeric mice
and tracked what happened. In the first few
weeks after implantation, the researchers
found male, GFP-expressing immune cells
in the grafts. But by 6 months, the endo-
thelial cells and smooth muscle cells that
formed the new stable blood vessel were all
female. They also found no sign of stem cell
markers in the vessel, suggesting that the
cells growing into the graft were differen-
tiated cells from the female host—not her
stem cells at all. Further experiments with
tagged cells showed that the new cells come
from the adjacent blood vessels.
Taken together, Breuer says, the evi-
dence suggests that the graft prompts the
adjacent vessels to expand into and over the
implanted scaffold in a process similar to
normal blood vessel growth. Robert Nerem,
a tissue engineer at the Georgia Institute of
Technology in Atlanta, says these experi-
ments represent “exactly the kind of study
that needs to be done so we understand
what’s happening mechanistically in these
therapeutic approaches.” Still, he would like
the results confirmed in larger animals.
Engineered regeneration
The results from these animal studies over-
turn the belief, held by many tissue engi-
neers, that rare stem cells in the seeded
BMCs would differentiate and
grow on the implanted scaf-
fold to form new tissue. The
results are consistent with
other evidence that BMCs can
prompt tissue repair without
contributing to the new tis-
sue directly; something simi-
lar seems to happen when the
cells are injected into diseased
hearts. Breuer says that a more
precise understanding may
help researchers design safer
and more effective lab-grown
tissues. He and Shinoka are
working to develop grafts that
wouldn’t need seeded cells but
instead would contain a combi-
nation of signaling molecules to attract the
needed response from monocytes.
That could make the implants easier to
produce and much less expensive, Vacanti
says, increasing the chance that they would
be widely adopted. He says the new under-
standing might also simplify efforts to
construct the tissue engineer’s ultimate
challenge: whole organs, with multiple cell
types and a full set of blood vessels. For
researchers trying to construct entire hearts,
for example, “you could imagine that you
wouldn’t need to preseed with vascular cells,
just with muscle cells,” he says.
In the meantime, the new trial will track
the six U.S. patients after they receive their
implants, following up with regular mag-
netic resonance imaging scans to watch for
signs of stenosis. The team proposed start-
ing small, Breuer says, to signal to the FDA
that “we’re willing to get started very slowly
and carefully.” That’s the right approach,
Vacanti says. “It’s terrific that they are going
ahead,” he says. “Their work is thoughtful,
rigorous, and very carefully done.” Setting a
positive precedent is crucial for the field, he
says. “They’ll do it properly.”
–GRETCHEN VOGEL C
R
E
D
I
T
:

©
R
O
B
E
R
T

A
.

L
I
S
A
K
,

2
0
1
1
Engineer surgeons.
Shinoka (left) and
Breuer launched a
U.S. trial this month.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1090
C
R
E
D
I
T
S

(
L
E
F
T

T
O

R
I
G
H
T
)
:

©

B
U
S
I
N
E
S
S
/
A
L
A
M
Y
;

©

B
E
R
N
H
A
R
D

C
L
A
S
S
E
N
/
A
L
A
M
Y
;

©

I
A
I
N

M
A
S
T
E
R
T
O
N
/
A
L
A
M
Y
When José Manuel Barroso, the president of
the European Commission, first suggested
building the European Institute of Technol-
ogy (EIT) in 2005, perhaps to rival the Mas-
sachusetts Institute of Technology on the
global stage, he got a skeptical response.
Many argued that existing universities
would be better placed than a new institute
to bolster Europe’s competitiveness.
The result was a fudge. EIT was indeed
established—and slightly rebranded as the
European Institute of Innovation and Tech-
nology—but rather than a brick-and-mortar
institute, it became a network of dozens of
universities and companies across Europe
with a small headquarters in Budapest.
Its budget was just €309 million over the
period 2007–13.
Now, the European Union is consider-
ing a major expansion of EIT, which in June
issued a strategic plan that would cost €4 bil-
lion over the next budget round, from 2014
to 2020. But the young institute is beset by
problems. A scathing external review has
criticized its track record, universities say
they feel out of the loop about its targets and
activities, and industry appears divided on its
usefulness and reluctant to pay its share of the
bill. Managing the institute has been compli-
cated by the rapid turnover of senior staff.
EIT was meant to give Europe a leg up
in the innovation competition with countries
such as the United States, India, and China.
“Europe is struggling with innovation by
any measure,” says Martin Schuurmans,
former executive vice president of Philips
Research, the research arm of the Dutch
electronics corporation, and the first chair
of the EIT board. “We have been spending
billions on R&D programs, and these have
helped, but we are still struggling, and we
need new approaches.”
EIT’s approach seeks to reinforce all
three sides of the so-called innovation tri-
angle of business, education, and research
through newly created consortia of univer-
sities and companies across Europe known
as Knowledge and Innovation Communities
(KICs). The first three KICs—dealing with
climate change, information technology, and
energy—have been established over the past
year, and up to a dozen more could follow.
The KICs receive funding to train grad-
uate students in science, technology, and
business; conduct collaborative research;
and cultivate new high-tech companies.
Each distributes its work among half a dozen
“co-location centers” hosted by a university
partner (see sidebar).
But the way EIT’s governing board has set
up the first three KICs has rankled European
universities that are meant to be its partners.
“There is huge frustration from these univer-
sities, including those who are participating
and those who are not,” over EIT’s failure
to consult with them, says Karin Markides,
president of the Conference of European
Schools for Advanced Engineering Edu-
cation and Research (CESAER). She
says EIT’s strategic plan is vague as well.
“Most of us really believe in this concept,”
Markides says, “but it has been described in
far too general a way.”
The external review, undertaken by con-
sulting company Ecorys for the European
Commission and released in June, echoed
CESAER’s criticism. It said EIT “has not
yet engaged extensively with the wider audi-
ence of organisations involved in promoting
innovation within the E.U.” and criticized
the Budapest headquarters for its subservi-
ence to EIT’s governing board.
But the board says it was essential to
forge ahead with the KICs. “The European
style is just to keep talking, but we have had
to make choices. There will always be peo-
ple who see things differently,” Schuurmans
says. “The European Union discussed this
idea for 4 years—and we implemented it in
18 months.” Now the KICs need the time to
show what they can do, he says.
Rooted in the past
The KICs’ first concrete actions have been in
education, for which the ambition is to com-
bine science or engineering with the kind of
dynamic entrepreneurship associated with
Silicon Valley or the greater Boston area.
Each of the KICs is planning distinctive mas-
ter’s and Ph.D. programs; students will spend
time at universities in at least two nations,
learn business skills, and work in industry.
So far, interest seems high. For instance,
the KIC InnoEnergy has received 900 appli-
cations for the first 200 master’s places,
which start next month, says Chief Executive
Officer Diego Pavia, an electrical engineer
with a background in the computer industry.
And even EIT critics agree that this kind
of course is needed in Europe, where few
postgraduate degrees in science or engineer-
ing include business training. “I’m pretty
convinced that what they are doing in educa-
tion will add value,” says Peter Tindemans,
chair of the policy committee at the lobby
group Euroscience and a longtime skeptic
about EIT’s organization model.
Europe’s Innovation Engine,
Eager to Grow, Faces Criticism
A European attempt to remain competitive by boosting innovation is off to a shaky start
EUROPEAN UNI VERSI TI ES Ready for the future? Europe’s leaders fret that
their high-tech industries are slow to innovate.
NEWSFOCUS
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1091
NEWSFOCUS
It’s not clear whether industry is equally
interested. Rolls-Royce and Philips, two of
the three major companies to comment on a
European Commission consultation exercise
on EIT this spring, were lukewarm in their
public submissions. A third—Cambridge-
based chipmaker ARM, one of Europe’s
most striking high-tech success stories—
said the effectiveness of EIT was “destined
to be greatly disappointing,” arguing that
its co-location model was “impossible” and
“unnecessary” in a world in which research
partners no longer need to be close together to
collaborate. “Their implementation is rooted
in the past,” ARM’s principal engineer, Ian
Phillips, told Science.
Industry is supposed to help finance
EIT’s activities. The KICs’ multiyear busi-
ness plans state that 25% of funding will
come from EIT, about 25% from private
sources, and the rest from other public
sources, including regional, national, and
E.U.-wide research programs. The hope is
that companies will be attracted by the pool
of research expertise and student talent at
the co-location centers.
Schuurmans and other EIT leaders P
H
O
T
O

C
R
E
D
I
T
S
:

I
S
T
O
C
K
P
H
O
T
O
.
C
O
M
The Knowledge and Innovation Com-
munity (KIC) called InnoEnergy, dedi-
cated to sustainable energy sources,
is distributed over six co-location
centers focusing on nuclear energy,
renewables, energy efficiency, clean
coal, “smart grids,” and energy from
chemical fuels. The network will
receive €30 million from the Euro-
pean Institute of Technology this year
and €80 million from its partners,
which include 11 universities and 34
energy companies, including major
ones such as Sweden’s ABB. “In bud-
getary terms, [industry’s] contribu-
tion doesn’t go above 26%—but in
terms of its real value, we’re getting
billions of euros,” says the KIC’s chief
executive officer, Diego Pavia.
Can These Networks KIC-Start European Innovation?
declined to estimate the total current value
of corporate pledges, saying that it was “too
early to say.” But Pavia says he’s confident that
his energy KIC will have no trouble meeting
its target of getting 26% of its income from
the 34 companies involved.
Traumatic negotiations
Meanwhile, critics say the rapid personnel
changes at EIT signal an organization in tur-
moil. Jan van der Eijk, the first chief executive
of the climate KIC and a former chief technol-
ogy officer of Shell, left last December, after
just 6 months on the job, amid tough negotia-
tions to constitute the KIC. The first director
of EIT, Gérard de Nazelle, left last September
after 12 months in the position. His succes-
sor, Spanish space technologist José Manuel
Leceta, didn’t start until July.
Supporters and critics agree on one
thing: If EIT is to have any noticeable
impact on European innovation, it will have
to markedly expand. In June, the European
Commission said it hopes to secure €80 bil-
lion for Horizon 2020, the 7-year research
and innovation program starting in 2014.
The EIT governing board has asked for a
€4 billion slice of that pie; one European
Parliament source says the number in the
European Commission’s budget proposal,
due later this year, may be closer to €2 bil-
lion. “If they just double funding [to €600
million], they should close it tomorrow,”
says Willem Jonker, who heads ICT Labs,
the information technology KIC.
The decision will depend in part on E.U.
member states and the European Parlia-
ment. Although initially skeptical, coun-
tries are now warming to EIT, says Maria
da Silva Carvalho, a Portuguese member of
the European Parliament who, as an adviser
to Barroso, played a key role in devising
EIT’s structure. “It’s not an easy model to set
up, but what has been established is a good
basis, and we have top universities and com-
panies involved.”
Pavia says Europe has no choice but to
continue the experiment if it is serious about
staying competitive. “This is the only way
European innovation can make a break-
through,” he says. “Otherwise, in 20 years’
time, we will just be living in a theme park.”
–COLIN MACILWAIN
Colin Macilwain is a writer in Edinburgh, U.K.
ICT Labs has perhaps the broadest of all of the KICs’ remits,
dealing with what European policymakers regard as the conti-
nent’s persistent failure to compete effectively with the United
States in information technology. The consortium includes 21
universities, six research centers, and 20 companies, includ-
ing household names such as Siemens, Philips, Nokia, and
Ericsson. Chief Executive Officer Willem Jonker says that it’ll
be previous advances in ICT itself—such as Skype confer-
encing—that will enable the KIC to operate as a single unit.
“Twenty years ago it would not have been possible,” he says.
ENERGY
Kraków
Stuttgart
Darmstadt
Barcelona
Katowice
Wroclaw
Grenoble
Zurich
Cadarache
Lyon
Marseille
Eindhoven
Leuven
Karlsruhe
Trento
Bologna
London
Budapest
Paris
Utrecht
Valencia
Birmingham
Stockholm
Uppsala
Madrid
Lisbon
Helsinki
Berlin
National
Regional
Nodes
Associated
Partners
Energy
Climate
ICT
Knowledge
and Innovation
Communities
CLIMATE
The Climate KIC will develop
approaches to counteract climate
change, such as smarter energy and
water distribution systems. That
means it will serve a young indus-
try with few well-established players.
University partners include elite insti-
tutions such as ETH Zurich and Impe-
rial College London; among the five
corporate participants are energy sup-
plier EDF and Amsterdam’s Schiphol
Airport. Because publicly funded
organizations play such a prominent
role in responding to climate change,
the KIC also has public partners, says
Chief Executive Mary Ritter, including
the Dutch province of Utrecht, which
is hosting a pilot project on climate-
resilient, low-carbon cities.
INFORMATION AND
COMMUNICATION
TECHNOLOGY
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1092
LETTERS
edited by Jennifer Sills
LETTERS I BOOKS I POLICY FORUM I EDUCATION FORUM I PERSPECTIVES
Learning by
drawing
1096 1095
Insects that sing
in the night
COMMENTARY
C
R
E
D
I
T
:

G
A
R
Y

K
R
A
M
E
R
/
U
.
S
.

F
I
S
H

A
N
D

W
I
L
D
L
I
F
E

S
E
R
V
I
C
E

Endangered Wolves Fall Prey to Politics
IN MAY 2011, THE NORTHERN ROCKY MOUNTAIN GRAY WOLF WAS REMOVED FROM THE FEDERAL
List of Endangered and Threatened Wildlife in the first legislative delisting of an endangered
species since the 1973 U.S. Endangered Species Act (1) was passed into law. The delisting
decision excluded Wyoming because proposed management by that state was deemed insuffi-
cient to sustain its wolf population. In early August, Interior Secretary Salazar agreed to delist
the wolf in Wyoming after the state agreed to sustain a minimum of 150 wolves in northwestern
Wyoming (or at least 100 outside Yellowstone National Park, whichever is greater) and remove
them everywhere else by any means necessary (2). This decision provides insufficient protec-
tion to Wyoming’s wolves. The terms of the agreement—similar to those opposed in the past by
federal courts as unjustified according to the
best available science and the plain language
of the Endangered Species Act (3, 4)—are a
response to pressure by western politicians,
ranchers, and sport hunters (3).
These interest groups claim that continued
protection will allow wolf populations to grow
exponentially, serially extirpating elk popu-
lations and preying on livestock and people
(5). Data do not support this argument; elk in
Yellowstone National Park increased substan-
tially before wolf reintroduction in 1995 and
have decreased substantially since, but they
have not been extirpated. In fact, state biolo-
gists say that Wyoming’s elk herd should be reduced in order to maintain the health of the herd
and the habitat (6). Meanwhile, the park’s wolf population has not increased exponentially but
rather declined 44% since 2003 and stabilized at fewer than 100 (7). Critics’ concerns about
livestock are also unfounded: Confirmed cattle deaths due to wolves in Wyoming, as well
as the number of wolf packs responsible for cattle deaths, have declined steadily from 2006
through mid-2011, even as Wyoming’s wolf population has increased (7). Overall, protecting
the wolves has positively affected the ecosystem (8), which was a key intent of the 1973 Endan-
gered Species Act (1).
Recent wolf population increases have resulted largely from wolf movement into new areas
of public land with sufficient prey. Ecologists believe that such expansions will facilitate wolf-
driven ecosystem restoration, as occurred in
Yellowstone National Park (8), and that this
would benefit western lands degraded by over-
browsing elk. Wyoming’s removal plan and
goal of reducing its 350 wolves by more than
40% (2) will reduce the degree and total geo-
graphic area of such trophic restoration. The
fact that Idaho and Montana killed 525 wolves,
or 32% of the population, by regulated hunt-
ing and control actions during the year (2009 to
2010) they were delisted (7) shows that Wyo-
ming’s goal is achievable. Such losses will
likely result in ecologically ineffective—and
possibly unsustainable—wolf populations (3).
There is increasing recognition that
removing predators from natural ecosystems
comes with serious consequences (“Trophic
downgrading of planet Earth,” J. A. Estes
et al., Review, 15 July, p. 301). Rather than
bowing to political pressure as the Interior
Department’s short-sighted plan does, we
should focus on the many benefits of wolf
restoration and expansion that established
science has shown (3).
BRADLEY J. BERGSTROM
Biology Department, Valdosta State University, Valdosta,
GA 31698, USA. E-mail: bergstrm@valdosta.edu
References
1. Endangered Species Act of 1973, 16 U.S.C. 1531–1544,
87 Stat. 884.
2. U.S. Department of the Interior, “Salazar, Ashe finalize
agreement with Wyoming on revised Gray Wolf manage-
ment plan” (3 August 2011).
3. B. J. Bergstrom, S. Vignieri, S. R. Sheffield, W. Sechrest,
A. A. Carlson, BioScience 59, 991 (2009).
4. K. Ridler, “Judge Molloy rejects settlement to lift wolf
protections in Montana, Idaho,” Missoulian, 9 April
2011.
5. J. Hollenhorst, “Hunting group turning up heat on
wolves,” KSL Newsradio, 14 March 2010.
6. Wyoming Game and Fish Department, “2010 Annual
Report” (2010); http://gf.state.wy.us/downloads/pdf/
annualreports/2010/FY10%20Annual%20Report.pdf.
7. U.S. Fish and Wildlife Service, “Gray Wolves in the North-
ern Rocky Mountains” (2011); www.fws.gov/
mountain-prairie/species/mammals/wolf/.
8. R. L. Beschta, W. J. Ripple, Biol. Conserv. 142, 2401
(2009).
Predators’ Effects
on Ecosystem Entropy
IN THEIR REVIEW “TROPHIC DOWNGRADING
of planet Earth” (15 July, p. 301), J. A.
Estes and coauthors highlight the far-
reaching changes to ecosystems that result
from the loss of apex consumers. In particu-
lar, they point out large reductions in vegeta-
tion brought about by unregulated herbivore
populations in such systems. This leads me
Letters to the Editor
Letters (~300 words) discuss material published in
Science in the past 3 months or matters of gen-
eral interest. Letters are not acknowledged upon
receipt. Whether published in full or in part, Let-
ters are subject to editing for clarity and space.
Letters submitted, published, or posted elsewhere,
in print or online, will be disqualified. To submit a
Letter, go to www.submit2science.org.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1093
1098 1106
Spore Prize
Essay
Sampling an
asteroid
to point out yet another important conse-
quence of losing large terrestrial predators
and their top-down effects (1). When pho-
tosynthetic capacity is reduced, not only
does atmospheric CO
2
increase, but heat
dissipation through evapotranspiration is
reduced. Some models estimate that evapo-
transpiration’s contribution to global warm-
ing is greater than that of CO
2
by one to
two orders of magnitude (2, 3). The trophic
downgrading discussed by Estes et al. thus
not only alters ecosystems as they describe,
but reduces the rate of entropy production,
which increases surface temperature varia-
tion. These indirect effects may have sub-
stantial consequences for biodiversity and
ecosystem functions.
WERNER T. FLUECK
Swiss Tropical and Public Health Institute, University
Basel, CH-4002 Basel, Switzerland, and Institute of Nat-
ural Resources Analysis-Patagonia (Universidad Atlantida
Argentina-CONICET), C.C. 592, 8400 Bariloche, Argentina.
E-mail: wtf@deerlab.org
References
1. W. T. Flueck, Eur. J. Wildl. Res. 46, 139 (2000).
2. E. D. Schneider, J. J. Kay, Futures 24, 626 (1994).
3. N. A. Brunsell et al., Earth Syst. Dynam. 2, 87 (2011).
Looking to NSF
as an NIH Model
M. ROSBASH’S EDITORIAL “A THREAT TO MEDI-
cal innovation” (8 July, p. 136) discusses
the financial challenges faced by the U.S.
National Institutes of Health (NIH). NIH
must figure out how to fund a higher per-
centage of meritorious proposals and how to
ensure that the funded proposals represent
the best science performed by the best sci-
entists. In addition to Rosbash’s suggestions,
NIH could look to the U.S. National Science
Foundation (NSF) for solutions.
NIH should consider restricting the
amount of investigator salaries charged to a
grant, especially if the investigator already
receives a salary. At NSF, funded salaries
are usually limited to 2 months (1), whereas
funded salaries at NIH are limited only by
the NIH Salary Cap (2). Decreasing the
share of grant budgets devoted to salaries
would lead to substantial savings, which
could be devoted to proposals that would
go unfunded otherwise. Many would fight
such salary restrictions, but the alternative
of funding only 10 to 15% of grant applica-
tions, as cited by Rosbash, is far worse.
Most NIH Study Sections often do not
have the breadth of expertise needed to
cover all of the topics of the submitted pro-
posals. As a result, ad hoc reviewers are
added, usually one per proposal. Too much
of the review decision is based on this one
expert. At NSF, 3 to 10 experts (ad hoc
reviewers) evaluate a proposal, and it is then
reviewed by an NSF Program Officer and
in many cases a review panel (3). Adopting
a similar system at NIH would ensure that
grant selections rewarded the best science.
WILLIAM A. YOST
Department of Speech and Hearing Science, Arizona State
University, Tempe, AZ 85287, USA. E-mail: william.yost@
asu.edu
References
1. NSF, “Proposal and award policies and procedures guide,
part II: Award and administration guide” (2011), ch. V:
Allowability of Costs, section B1a(ii)(a); www.nsf.gov/
pubs/policydocs/pappguide/nsf11001/aag_5.jsp#VB1.
2. NIH, “NOT-OD-11-073: Salary limitations on grants,
cooperative agreements, and contracts” (2011).
3. NSF, “Proposal and award policies and procedures guide,
part I: Grant proposal guide” (2011), ch. III: NSF Pro-
posal Processing and Review; www.nsf.gov/pubs/
policydocs/pappguide/nsf11001/gpg_3.jsp.
CORRECTIONS AND CLARIFICATIONS
Table of Contents: (24 June, p. 1475). The teaser for the
Research Article by B. Marty et al. was incorrect. The correct
teaser should read “Earth, the inner planets, and meteorites
are about 40% enriched in the heavy nitrogen-15 isotope
compared with the Sun and Jupiter.”
Visualization Challenge: (18 February, p. 847). The hon-
orable mention illustration, “Enterobacteria Phage T4,” by J.
Heras showed the virus with eight legs. The virus only has six
legs. Heras’s corrected image is shown here.
TECHNICAL COMMENT ABSTRACTS
Comment on “Drought-Induced Reduction in Global Terrestrial Net Primary
Production from 2000 Through 2009”
Arindam Samanta, Marcos H. Costa, Edson L. Nunes, Simone A. Vieira, Liang Xu, Ranga B. Myneni
Zhao and Running (Reports, 20 August 2010, p. 940) reported a reduction in global terrestrial net primary production
(NPP) from 2000 through 2009. We argue that the small trends, regional patterns, and interannual variations that they
describe are artifacts of their NPP model. Satellite observations of vegetation activity show no statistically significant
changes in more than 85% of the vegetated lands south of 70°N during the same 2000 to 2009 period.
Full text at www.sciencemag.org/cgi/content/full/333/6046/1093-c
Comment on “Drought-Induced Reduction in Global Terrestrial Net Primary
Production from 2000 Through 2009”
Belinda E. Medlyn
Zhao and Running (Reports, 20 August 2010, p. 940) reported that global net primary production has declined over
the past decade and that this reduction was caused by drought. However, their findings are not direct measurements,
but rather are based on outcomes from models in which a strong temperature dependence is assumed. I examine the
assumptions underlying their results and show that their findings can be explained as logical consequences of these
assumptions.
Full text at www.sciencemag.org/cgi/content/full/333/6046/1093-d
Response to Comments on “Drought-Induced Reduction in Global Terrestrial
Net Primary Production from 2000 Through 2009”
Maosheng Zhao and Steven W. Running
Samanta et al. and Medlyn challenge our report of reduced global terrestrial net primary production (NPP) from 2000
through 2009. Our new tests show that other vegetation indices had even stronger negative changes through the decade,
and weakening temperature controls on water stress and respiration still did not produce a positive trend in NPP. These
analyses strengthen the conclusion of drought-induced reduction in global NPP over the past decade.
Full text at www.sciencemag.org/cgi/content/full/333/6046/1093-e
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Comment on “Drought-Induced Reduction
in Global Terrestrial Net Primary
Production from 2000 Through 2009”
Arindam Samanta,
1,2
* Marcos H. Costa,
3
Edson L. Nunes,
3
Simone A. Vieira,
4
Liang Xu,
1
Ranga B. Myneni
1
Zhao and Running (Reports, 20 August 2010, p. 940) reported a reduction in global terrestrial net
primary production (NPP) from 2000 through 2009. We argue that the small trends, regional
patterns, and interannual variations that they describe are artifacts of their NPP model. Satellite
observations of vegetation activity show no statistically significant changes in more than 85%
of the vegetated lands south of 70°N during the same 2000 to 2009 period.
Z
hao and Running (1), hereafter ZR10, re-
ported a reduction of 0.55 petagrams of
carbon (Pg C) in global terrestrial net pri-
mary production (NPP) of 535.21 Pg C over a
10-year period (2000 to 2009), or 0.1%. They
attributed this decline to a drying trend in the
Southern Hemisphere that decreased NPP by
1.83 Pg C (0.34%) and that was counteracted
by increased NPP in the Northern Hemisphere
by 1.28 Pg C(0.24%). These minute changes raise
questions about the robustness of these numbers
and the reported regional patterns. The Amazonian
forests are a good case study because these forests
play a dominant role in trends and interannual
variability (66%) reported in (1).
First, ZR10’s estimates differ from compara-
bly upscaled field measurements (2–6) by 28%
at 14 different sites and time periods in Brazil,
Colombia, and Peru (Table 1). In a majority of
these cases, ZR10 underestimate field-based mea-
surements by 31% and overestimate by 18% for
the rest. Notably, they overestimate NPP by 32%
for two sites at Tambopata (Peru), which suffered
an intense drought in 2005 when the peak dry
season (July to September) precipitation stan-
dardized anomaly was –1.51 relative to the 1998
to 2006 period. This poor model performance in
tropical forests is also evident from an earlier
article by the same authors [figure 8 in (7)], and
no further evidence of model advancement is
presented in (1). None of this imbues confidence
in the small NPP trends reported not only for the
Amazonian forests but also for other tropical for-
ests; for example, in Asia, they report a similarly
small decline (–0.562 Pg C) during 2000 to 2009.
Second, ZR10 gap-fill missing and atmosphere-
corrupted satellite-based input data of their
model—namely, leaf area index (LAI) and frac-
tion of photosynthetically active radiation (FPAR)
absorbed by vegetation—by linearly interpolating
across uncorrupted observations (7). This re-
sults in artifacts, especially in the Amazon, where
corruption from clouds and biomass-burning
aerosols is pervasive [figure S13A in (1)]. Our
analysis of the same data, properly filtered for
atmosphere-corruption effects but not gap-filled,
does not showsimilarly inflated LAI or FPARval-
ues (fig. S2), consistent with other analyses (8).
We thus question the credibility of ZR10’s results
(e.g., NPP changes of 0.1% at the global scale
and 0.24 to 0.34% at the hemispheric scale),
which were obtained from a poorly performing
model (Table 1) driven with a mix of measure-
ments and gap-filled input values containing ap-
parent artifacts.
Third, the correlation between NPP and at-
mospheric CO
2
growth rate (CGR) anomalies [Fig.
1 in (1)] is spurious because both their gross pri-
mary production (GPP) down-regulation, to sim-
ulate temperature and soil moisture stresses, and
autotrophic respiration modeling are overly sensi-
tive to temperature (section S1 in ZR10). Therefore,
ZR10’s Fig. 1 (1) is a depiction of the correlation
between temperature and CGR anomalies, not be-
tween NPP and CGR anomalies (fig. S1). Thus,
their claim that “global terrestrial NPP is a ma-
jor driver of the interannual CO
2
growth rate” is
questionable.
Fourth, although ZR10 invoke a report (9)
of aboveground biomass declines in Amazonian
forests due to the 2005 drought as supporting evi-
dence, a closer examination reveals incongruities.
The biomass declines were a consequence of in-
creased mortality of some select drought-sensitive
TECHNICALCOMMENT
1
Department of Geography and Environment, Boston Univer-
sity, Boston, MA 02215, USA.
2
Atmospheric and Environmental
Research Inc., Lexington, MA 02421, USA.
3
Federal University
of Viçosa (UFV), Avenida Peter Henry Rolfs, s/n, Viçosa, MG, CEP
36570-000, Brazil.
4
State University of Campinas–UNICAMP,
Rua dos Flamboyants, 155, Campinas, SP, CEP 13083-867, Brazil.
*To whom correspondence should be addressed. E-mail:
arindam.sam@gmail.com
Fig. 1. Spatial patterns of statistically significant (p < 0.05) trends (%/year) in annual mean Collection 5
(C5) Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) at 0.05° by
0.05° spatial resolution during the decade 2000 to 2009. Cloud-, shadow-, climatology aerosol- and high
aerosol–contaminated data are screened from analysis (supporting online material, section S3). Areas
with statistically insignificant trends are shaded gray, and barren areas are white. These data are fromthe
Moderate Resolution Imaging Spectroradiometer (MODIS).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1093-c

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

trees, whereas the forest growth rate, i.e., the NPP,
did not change in 2005 relative to the pre-2005
period (9). The NPP declines in ZR10, however,
are due to the absolute intensity of the drought,
namely, vapor pressure deficit–dependent down-
grading of calculated GPP and enhanced main-
tenance respiration due to higher air temperatures
in 2005 (section S9 in ZR10). Also noteworthy
is the poor performance of the ZR10 model at
drought-affected sites (two at Tambopata and
two at Amacayacu) (Table 1). Therefore, it is dif-
ficult to reconcile how the model could have
produced the right results for the wrong reasons
in response to the 2005 drought, while at the same
time failing to match field measurements, some
of which are from the very same drought year.
Fifth, even if we were to accept the 2005 es-
timates as accurate and due to the drought, ZR10’s
NPP estimates for the following two nondrought
years are equally low and lack corroborating evi-
dence [table S4 in (1)]. Actually, water balance
studies, based on satellite altimeter data and pre-
cipitation observations, show significant positive
hydrological anomalies in 2006 (10), and the 2005
drought itself was alleviated by a return to normal
rainfall levels starting from October 2005 (8).
Thus, ZR10’s low NPP estimates for 2006 and
2007 are difficult to explain.
Sixth, if we accept the low 2006 and 2007
ZR10 estimates as lingering effects of the 2005
drought, the difference in NPP between the pe-
riod before 2005 and the period after 2007 is only
1.1%. This suggests that the forests have recov-
eredbytheir own estimates. However, ZR10present
this short-termnegative anomaly, which was brack-
eted by periods of normal functioning, as a 10-year
declining trend (–0.424 Pg Cper decade), which is
misleading, for it implies a progressively degener-
ative anomaly caused by some fundamental shift in
the underlying processes for which there is no
evidentiary basis.
Finally, satellite vegetation indices represent
direct observations of the physiologically func-
tioning greenness level by capturing the amount
of photosynthetically active radiation absorbed
by chlorophyll in green leaves (11). These obser-
vations do not show any large-scale declines in
the greenness level of the land surface (Fig. 1);
86% of all vegetated land south of 70°N shows
no trends. About 8 to 9% show declining trends
in three nonforested regions: the Eurasian steppes,
Argentina, and central Australia. These patterns,
buttressed by trends in LAI and FPAR data (fig.
S3), do not support ZR10. Their argument that
model NPP fields capture reality while observa-
tions do not (section S9 in ZR10) is unwarranted
for at least two reasons. First, their analysis of the
same LAI and FPAR data shows gap-filling and
other artifacts (compare fig. S2 with fig. S13A in
ZR10). Second, published literature is replete with
evidence of linkages between variations in satellite
greenness index levels and climatic anomalies,
several in the context of terrestrial NPP (12–15).
Based on these arguments, we conclude that
the trends, regional patterns, and interannual var-
iations reported by ZR10 (1) are model artifacts
and that observations of vegetation activity show
no significant changes in more than 85% of the
vegetated lands south of 70°Nduring the 2000 to
2009 period.
References and Notes
1. M. S. Zhao, S. W. Running, Science 329, 940 (2010).
2. E. L. Nunes, thesis, Federal University of Vicosa (2008);
available at www.biosfera.dea.ufv.br/phtm/teses/pdf/
DS_EdsonNunes.pdf.
3. M. H. Costa, E. L. Nunes, M. C. A. Senna,
H. M. A. Imbuzeiro, Brazilian J. Meteorol. 24, 179 (2009).
4. D. P. Turner et al., Remote Sens. Environ. 102, 282
(2006).
5. S. Vieira et al., Oecologia 140, 468 (2004).
6. L. E. O. C. Aragão et al., Biogeosciences Discuss. 6, 2441
(2009).
7. M. S. Zhao, F. A. Heinsch, R. R. Nemani, S. W. Running,
Remote Sens. Environ. 95, 164 (2005).
8. A. Samanta et al., Geophys. Res. Lett. 37, L05401 (2010).
9. O. L. Phillips et al., Science 323, 1344 (2009).
10. W. Llovel, M. Becker, A. Cazenave, J. F. Cretaux,
G. Ramillien, C. R. Geosci. 342, 179 (2010).
11. R. B. Myneni, F. G. Hall, P. J. Sellers, A. L. Marshak,
IEEE Trans. Geosci. Rem. Sens. 33, 481 (1995).
12. C. J. Tucker, I. Y. Fung, C. D. Keeling, R. H. Gammon,
Nature 319, 195 (1986).
13. R. B. Myneni, C. D. Keeling, C. J. Tucker, G. Asrar,
R. R. Nemani, Nature 386, 698 (1997).
14. R. R. Nemani et al., Science 300, 1560 (2003).
15. P. Ciais et al., Nature 437, 529 (2005).
Acknowledgments: This work was funded by NASA Earth
Science Division and executed through NASA Earth
Exchange (NEX).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1093-c/DC1
SOM Text
Figs. S1 to S3
References and Notes
13 October 2010; accepted 18 July 2011
10.1126/science.1199048
Table 1. Comparison of ZR10 (1) NPP estimates with comparably upscaled observations for forest sites
in Brazil, Colombia, and Peru (2). BA712 is an Atlantic forest site in the southern state of Bahia, and the
rest are in the Amazonian region. Observed NPP is field-measured NPP (supporting online material,
sections S1 and S2). AGP, CAX, and TAMrepresent two different sites each. The average absolute difference
between measured and modeled NPP reported by ZR10 (1) is 27.93%. The average underestimation is
31.47%, and the average overestimation is 18.48%.
Observed NPP Zhao and Running (1)
Site Period
(kg-C m
−2
year
−1
)
(kg-C m
−2
year
−1
)
Error (%)
KM67 2001 1.230 0.832 –32.36
KM67 2004 1.055 0.733 –30.52
ZF-2 2001 1.063 0.779 –26.72
ZF-2 2002 1.356 0.703 –48.16
UFAC 2001 1.343 0.997 –25.76
UFAC 2002 1.299 0.925 –28.79
BA712 2006 1.366 1.519 11.20
AGP (AGP-01 and AGP-02) 2004–2006 1.148 1.000 –12.28
CAX (CAX-06 and CAX-08) 2004–2006 1.396 0.737 –47.21
TAM (TAM-05 and TAM-06) 2005 1.534 2.028 32.20
ZAR-01 2004–2006 0.930 1.042 12.04
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1093-c
TECHNICAL COMMENT

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Comment on “Drought-Induced Reduction
in Global Terrestrial Net Primary
Production from 2000 Through 2009”
Belinda E. Medlyn
Zhao and Running (Reports, 20 August 2010, p. 940) reported that global net primary production
has declined over the past decade and that this reduction was caused by drought. However, their
findings are not direct measurements, but rather are based on outcomes from models in which a
strong temperature dependence is assumed. I examine the assumptions underlying their results
and show that their findings can be explained as logical consequences of these assumptions.
Z
hao and Running (1), hereafter ZR10,
reported that global terrestrial net primary
production (NPP, the amount of atmo-
spheric carbon fixed by plants and accumulated
as biomass) has declined over the past decade
and that large-scale droughts are responsible for
the decline. However, their findings are based on
outcomes frommodels in which a strong temper-
ature dependence is assumed. A simple alterna-
tive explanation for their findings is that increased
temperatures are driving modeled results.
NPP cannot be directly measured at large
scales. Instead, the Moderate Resolution Imaging
Spectroradiometer (MODIS) NPP algorithm
used by ZR10 is a simple model that calculates
NPP from remotely sensed data and climatic
variables. The remote sensing input is the fraction
of photosynthetically active radiation (FPAR)
absorbed by the vegetation. Gross primary
productivity (GPP) is estimated by multiplying
this input data by the incoming radiation, a
constant light-use efficiency, and two modifier
terms. The two modifier terms represent assump-
tions that productivity is reduced by (i) low tem-
peratures (T
min
< 8° to 12°C) and (ii) high vapor
pressure deficit (VPD > 0.65 – 0.8 kPa). Rising
temperature can increase GPP through the ef-
fect of T
min
where temperatures are low, but de-
creases GPP at higher temperatures through the
effect of rising VPD, which is correlated with
temperature.
NPP is then calculated from GPP by sub-
tracting a respiration term that increases ex-
ponentially with temperature. The Supporting
Online Material [SOM text for (1)] indicates that
previous versions of the algorithm held the Q10
of this temperature dependence constant and equal
to 2. For this paper, they state that a “temperature-
acclimated Q10 equation” is used. However, the
new temperature dependence they use is taken
from a paper about variation in the short-term
Q10 of respiration with measurement tempera-
ture (2). This function gives a stronger effect of
temperature on respiration than previously, with a
Q10 that ranges between 2 (at 25°C) and 3.22 (at
0°C). In contrast, temperature acclimation tends
to reduce the basal rate of respiration, with the
result that long-termQ10 values are considerably
less than 2 (3–5). Measured temperature depen-
dences are further reduced when high temperature
is associated with drought because plant respira-
tion rates are reduced by drought (5, 6), an effect
not accounted for by ZR10. Thus, ZR10 are as-
suming a strong temperature dependence of res-
piration and, consequently, a strong negative effect
of temperature on modeled NPP.
Overall, therefore, NPP is assumed to in-
crease with rising temperature in cold regions
(< T
min
) but to decrease with rising temperature
in warmer regions. As one might expect from a
model based on these assumptions, ZR10 report
that modeled NPP has increased with rising tem-
perature in the cooler Northern Hemisphere but
decreased with rising temperature in the warmer
Southern Hemisphere. The reported reduction in
NPP is clearly a consequence of the chosen as-
sumptions. If the respiration assumption were to be
replaced with a weaker temperature dependence,
the calculated reduction in NPP would decrease.
ZR10 argue that the reduction they observe
in modeled NPP is induced by drought, demon-
strating a correlation between NPP anomaly and
the Palmer Drought Severity Index (PDSI) anom-
aly. Drought severity, like NPP, is not directly
measurable at large scales, so model indices such
as PDSI are commonly used to estimate it. The
PDSI is calculated as a balance fromprecipitation
inputs and evaporative losses. However, the PDSI
also incorporates a strong temperature depen-
dence, because potential evapotranspiration is
calculated as a function of temperature (7). The
PDSI therefore decreases (more severe drought)
as temperatures increase (8). This temperature
dependence of the PDSI provides a simple ex-
planation for the correlation observed by ZR10
between NPP and PDSI in the Southern Hemi-
sphere. It is not that drought is causing a reduc-
tion in NPP; rather, both NPP and drought severity
are assumed to vary with temperature.
Models have a very important role to play in
our understanding of terrestrial ecosystems. How-
ever, it is important to clearly present the assump-
tions underlying a modeling study, and correctly
describe the conclusions. ZR10 have not shown,
as claimed, that terrestrial NPP has decreased
over the last decade. Rather, they have shown
that if NPP is assumed to be affected by climate
as specified in their model, then NPP would have
declined over the past decade. It is important to
make this distinction, because otherwise we run
the risk of mistaking model outcomes for reality.
References
1. M. S. Zhao, S. W. Running, Science 329, 940 (2010).
2. M. G. Tjoelker, J. Oleksyn, P. B. Reich, Glob. Change Biol.
7, 223 (2001).
3. O. K. Atkin, D. Bruhn, V. M. Hurry, M. G. Tjoelker,
Funct. Plant Biol. 32, 87 (2005).
4. O. K. Atkin et al., Glob. Change Biol. 14, 2709 (2008).
5. K. Y. Crous et al., Glob. Change Biol. 17, 1560 (2011).
6. O. K. Atkin, D. Macherel, Ann. Bot. 103, 581 (2009).
7. W. M. Alley, J. Clim. Appl. Meteorol. 23, 1100 (1984).
8. Q. Hu, G. D. Willson, Int. J. Clim 20, 1899 (2000).
26 October 2010; accepted 18 July 2011
10.1126/science.1199544
TECHNICALCOMMENT
Department of Biological Sciences, Macquarie University,
North Ryde NSW 2109, Australia. E-mail: belinda.medlyn@
mq.edu.au
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1093-d

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Response to Comments on
“Drought-Induced Reduction in Global
Terrestrial Net Primary Production
from 2000 Through 2009”
Maosheng Zhao* and Steven W. Running
Samanta et al. and Medlyn challenge our report of reduced global terrestrial net primary
production (NPP) from 2000 through 2009. Our new tests show that other vegetation indices
had even stronger negative changes through the decade, and weakening temperature controls
on water stress and respiration still did not produce a positive trend in NPP. These analyses
strengthen the conclusion of drought-induced reduction in global NPP over the past decade.
W
e (hereafter ZR10) used satellite data
to estimate global terrestrial net primary
production (NPP) over the past decade
and found that large-scale regional droughts
and a general drying trend over the Southern
Hemisphere have caused a decline in NPP (1).
Samanta et al. (2) and Medlyn (3) challenge our
results, and their critiques focus on four areas.
First, how reliable are spectral vegetation indices
for quantifying terrestrial biospheric dynamics
and anomalies? Second, can a simple globally
generalized carbon balance algorithmusing daily
incident radiation, temperature, and vapor pres-
sure deficit (VPD) adequately reflect ecosystem
physiology of photosynthesis-respiration controls?
Third, can ground measurements be used to vali-
date a global-satellite-derived data set? Fourth,
was the decadal global trend illustrated in ZR10
significant? We address these issues below.
Samanta et al. (2) argue that there were not
any large-scale declines in global annual mean
MODIS vegetation indices [VIs; including the
normalized difference vegetation index (NDVI),
enhanced vegetation index (EVI), fraction of pho-
tosynthetically active radiation (FPAR), and leaf
area index (LAI)] over the past decade and con-
cluded there should be no NPP change as well.
However, the authors used less precise data pro-
cessing in two ways. First, because we are quan-
tifying plant growth, we screen out dormant
season data [as explained in supporting online
material (SOM) text S1] using the Moderate
Resolution Imaging Spectroradiometer (MODIS)
snow cover data set and process VIs only during
the growing season. Samanta et al. used annual
average VIs, which add many snow-covered pix-
els with corrupted spectral reflectances. Although
they further screened out lowvalues of VIs, some
values above their defined thresholds may still
be in dormant seasons, especially for evergreen
needleleaf forests.
Second, Samanta et al. used the MODIS
CMG (Climate Modeling Grid aggregated to
0.05°) rather than the native 1-km data, intro-
ducing potential inconsistency to their NDVI
and EVI data sets from within cell heterogeneity
and making direct comparison with our analysis
more difficult. To explore the potential differences
in processing details, we replicated their processing
of filtering out cloud- or aerosol-contaminated
MODIS 1-km VIs and then aggregated reliable
VIs into a 0.05° CMGgrid (SOMtext S1). How-
ever, instead of their annual average, we calcu-
lated growing-season average reliable VIs [SOM
text S1 and (1)]. We found that more than 60%
(58%) of global vegetated areas had a negative
trajectory in NDVI (EVI) for the 2000 to 2009
period (Fig. 1 and Table 1). For the Southern
Hemisphere (SH), all VIs show many more areas
with significant decreased trends than those with
significant increased trends (table S1). The reduced
NPP for the tropical rainforests are further con-
firmed by 40.58% (16.69%) of areas with sig-
nificant decreased NDVI (EVI) compared with
1.35% (1.86%) with significant increased NDVI
(EVI) (table S1 and fig. S2).
Our NPP decadal trend agrees in direction
(positive or negative) with each of the three con-
sidered spectral VIs over 62.4%of global vegetated
land area (Fig. 1 and Table 1). Geographically,
the most coherent region of disagreement is cen-
tral Africa, where the spectral indices show a
negative decadal trend but our NPP shows a pos-
itive trend. This region has the sparsest density of
weather stations in the tropics, so our confidence
in the gridded meteorology is low (4). However,
if the trends in the VIs are more correct, it would
strengthen our main conclusion, that NPP has de-
clined over the past decade.
FPAR, an input to the MODIS NPP algo-
rithm, provides essential information on ground
vegetation dynamics, and hence its trajectory can
partially influence NPP change. In contrast to
the negative trends in NDVI and EVI, the FPAR
trend is virtually flat, except in the SH(fig. S2 and
table S1). This FPAR anomaly is also exhibited in
the 2005 Amazon drought, with 28.5% of rain-
forests greener versus 16.8% browner over the
dry season compared with the previous 5 years
over rainforests south of the equator (fig. S4).
This FPAR anomaly is different from changes in
EVI (5) but consistent with our previous report
on FPAR [figure S13 in (1)]. Despite the weakly
positive FPAR anomaly, our NPP captured the
reduction of NPP (6) because the model incor-
porates biophysical constraints from meteorolog-
ical data, suggesting the usefulness of the model.
If we had used MODIS NDVI or EVI, instead of
FPAR, as an input to our NPP model, global NPP
would have exhibited a stronger decadal reduction.
As to the second critique by Samanta et al.
(2), gap-filling or temporally smoothing satellite-
based vegetation indices is a common practice
and an accepted method for unifying time-series
satellite data in cloudy climates (7). To calculate
annual total MODIS NPP using our model, gap-
filling must be employed to fill cloud-contaminated
satellite data periods simply because photosynthesis
occurs in rainforests during daylight hours regard-
less of cloud cover (fig. S3).
Samanta et al. (2) and Medlyn (3) argue that
our results are an artifact of an overly sensitive
temperature dependence in the NPP algorithm
because autotrophic respiration and VPD are
sensitive to temperature change. Medlyn also
suggests that a significant positive correlation be-
tween NPP and Palmer Drought Severity Index
(PDSI) over the SHis caused by the sensitivity of
PDSI to temperature. Plant respiration is sensitive
to air temperature (8, 9). Our respiration calcula-
tion follows the common ecological modeling
approach for estimating autotrophic respiration,
a Q
10
function assuming an exponential relation-
ship with temperature in which Q
10
is the ratio
of the rate at one temperature to that at a temper-
ature 10°Clower (10). In ZR10 (1), we adopted a
temperature-corrected Q
10
for foliage across di-
verse plant taxa to replace a constant Q
10
value
of 2.0 in earlier algorithms, which had been
found to be biased (10). A recent global-scale
analysis of autotrophic respiration reveals that
Q
10
ranges from 1.9 to 2.5 for forests (11). A
study of total ecosystemrespiration based on data
from 60 FLUXNET sites found that Q
10
aver-
aged ~1.4 (12). However, these lower Q
10
values
include heterotrophic respiration from dead tis-
sue, whereas we compute only autotrophic res-
piration of live tissue.
We performed four sensitivity tests to exam-
ine the response of our NPP algorithm to tem-
perature by modifying Q
10
alone (two tests with
constant 2.0 or 1.4) and both Q
10
(constant 1.4)
and daytime maximum VPD (VPD
max
). VPD
max
is the threshold where stomatal conductance re-
duces to zero (two tests with VPD
max
= 5000 or
10,000 Pa) and photosynthesis halts (Table 2 and
TECHNICALCOMMENT
Numerical Terradynamic Simulation Group, Department of
Ecosystem and Conservation Sciences, University of Montana,
Missoula, MT 59812, USA.
*To whom correspondence should be addressed. E-mail:
zhao@ntsg.umt.edu
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1093-e

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Fig. 3). We also performed bootstrap tests for
the global total NPP to examine the percentage
of negative NPP trends (Table 2). As expected
by Medlyn (3), if the respiration and water stress
algorithms had weaker temperature dependence,
the calculated reduction in NPP would generally
decrease. However, none of these parameter com-
binations produce a positive NPP trend through
our time period (Table 2). In addition, relaxed
VPDcontrol doesn’t necessarily induce a smaller
reduction (Table 2). Relaxed control of VPD and
a lower Q
10
increased average global NPP from
55.1 to 67.6 Pg C/year (Table 2), a value consid-
ered unrealistically high by a synthesized ground
NPP data set (13) (Fig. 2) and flux towers (14).
Our current knowledge of long-term climatic ad-
aptation in respiratory traits of leaves, stems, and
roots among diverse plant taxa remain inadequate,
as well as scaling these physiological processes
to global carbon components (15).
Satellite-based global data sets must be vali-
dated against ground data, although one must be
acutely aware of the scale mismatches of the data
sets. Samanta et al. (2) compared isolated pixels
from our global MODIS NPP data set with 14
small plots of field-measured NPPtaken on various
years from one region, the Amazon, and ques-
tioned the credibility of our results. However,
the authors ignore a number of methodological
differences between NPP measured on the ground
A B
C D
E F
Fig. 1. For the period from 2000 through 2009, the spatial pattern of trends
in the growing season average reliable Collection 5 MODIS vegetation indices
at 0.05°, FPAR (A), NDVI (C), and EVI (E) and the agreement and disagreement
in the change directions between MODIS NPP and FPAR (B), NDVI (D), and EVI
(F). For (B), (D), and (F), the plus sign refers to an upward trend, whereas the
minus sign denotes a downward trend. For the three vegetation indices, we
followed exactly the same method as used by Samanta et al. (2) to screen out
contaminated FPAR, NDVI, and EVI by cloudiness and aerosols without temporal
filling before calculating growing season average (SOM text S1). Results are
listed in Table 1. Spatial pattern of NPP change was shown in Fig. 2 in (1).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1093-e
TECHNICAL COMMENT

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

and NPP measured by satellite. First, the satellite
is radiometrically integrating each entire square
kilometer of landscape, whereas their field data
directly measures only a few fully forested ha at
each study site. The field measures quantified
only growth of trees [more than 10-cm diameter
in many cases; see SOM text in (2)], whereas
the satellite measurement quantifies growth of
all vegetation leaf areas. The field measure-
ments were taken sporadically for various
single years, as shown in Table 1 of Samanta
et al. (2), not annually for the whole period, so
they document no interannual variability [see
SOM text in (2)]. Even the field methodologies
offered by Samanta et al. are very different from
plot to plot, because these were not an organized
project, yet the authors present no discussion of
measurement uncertainty or error.
Beginning in 1996, an exhaustive international
effort was made to synthesize and scale published
ground-measured NPP data to a data set for val-
idating global carbon models called the Global
Primary Production Data Initiative (GPPDI). We
evaluated our MODIS NPP using the GPPDI NPP
data set (13) (Fig. 2). For all 2335 half-degree
cells, MODIS NPP generally agrees well with
the GPPDI NPP (r
2
= 0.8, P= 0.00). Nevertheless,
the GPPDI data set represents a single NPP val-
ue for each cell with no interannual variability,
whereas NPP measures year-to-year variability
of T20% (16). Time series of NPP are best esti-
mated from eddy covariance flux tower data, of
which >400 towers exist worldwide, some with
10 to 15 years of continuous data. However, flux
towers directly measure net ecosystem CO
2
ex-
change (NEE), which then must be disaggre-
gated to NPP from carbon cycle principles (14).
As to the remaining critiques by Samanta et al.,
Phillips et al. (6) stated that “These (biomass)
losses were driven by occasionally large mortal-
ity increases and by widespread but small de-
clines in growth,” a signal more likely quantified
by the satellite data than by isolated plot data.
Based on table S4 of ZR10 (1), the reduction of
NPP in the 2005 drought of the Amazon is 5.56%
of the mean NPP from 2000 to 2004. Our spatio-
temporally complete satellite-based MODIS NPP
data set provides more consistent information on
spatial heterogeneity and interannual variability
in global NPP than 1 year of field measurements,
such as those 14 small plot samples supposedly
representing the entire Amazon Basin. Relative-
ly higher VPD and temperatures, as evidenced
in our meteorological data, were the likely causes
of lower Amazonian NPP in 2006 and 2007
[figure S14 in (1)].
We found previously that as the global land
air temperature increased from 1982 to 1999,
there was a 6% increase in global NPP, and 80%
of that increase came from energy-limited eco-
systems (17). In the past decade, water limitations
appear to have overtaken energy limitations as
the strongest global constraint on NPP. Jung et al.
(18) found that global evapotranspiration has
declined since 1998, with drought impacts con-
Table 1. Percentage (%) of vegetated land areas with consistent or opposite changes in the Collection 5
MODIS annual NPP and growing season average vegetation indices, FPAR, NDVI, and EVI from 2000
through 2009 (SOM text S1). The plus sign refers to an upward trend, whereas the minus sign denotes a
downward trend. Sum shows the total percentage of vegetated land areas with either decreased or
increased trends. The consistency in directions refers to the percentage of total land areas with the same
direction in changes between NPP and the three different vegetation indices. The corresponding spatial
results are shown in Fig. 1.
FPAR – FPAR + NDVI – NDVI + EVI – EVI +
NPP – 29.98 17.24 35.32 11.95 34.11 13.16
NPP + 19.48 33.30 24.92 27.81 24.43 28.30
Sum 49.46 50.54 60.24 39.76 58.54 41.46
Consistency
in the directions
63.28 63.13 62.41
Fig. 2. Comparison of average MODIS NPP from2000 through 2009 at half-degree scale with 2335 NPP
cells fromthe Global Primary Production Data Initiative (13). The gray line is a slope of 1.0, and the black
line is the regression line.
Table 2. Global terrestrial MODIS NPP and its changes (the slope of linear
trends) calculated with the control and four sensitivity tests for the period from
2000 through 2009 (SOMtext S2). Anomalies are shown in Fig. 3. The percentage
of samples with a negative slope for each calculation is also listed base on 10,000
times of bootstrap resample, each time drawing a sample of equal size to the
original dataset and calculating the regression slope fromeach bootstrap sample.
Control
Sensitivity 1
(Q
10
= 2.0)
Sensitivity 2
(Q
10
= 1.4)
Sensitivity 3
(VPD
max
= 5000 Pa,
Q
10
= 1.4)
Sensitivity 4
(VPD
max
= 10,000 Pa,
Q
10
= 1.4)
Average NPP (Pg C/year) 53.39 53.02 55.10 61.07 67.63
NPP trend (Pg C/year) -0.0552 -0.0548 -0.0061 -0.0141 -0.0187
% bootstraps with negative slopes 76.5 75.5 54.9 67.7 65.9
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1093-e
TECHNICAL COMMENT

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

centrated in the Southern Hemisphere, consistent
with our PDSI results. Beer et al. (14) found that
variation in measured global gross primary pro-
duction was best correlated with precipitation.
Recent studies have revealed a persistent drying
trend in the Amazon from2000 to 2005 [Fig. 3 in
(19)], and a severe drought occurred again in
2010 (20). Gobron et al. (21) evaluated a 1998 to
2009 time series of FPAR from the Sea-Viewing
Wide Field-of-View Sensor (SeaWifs) and Me-
dium Resolution Imaging Spectrometer (MERIS)
satellite sensors and confirm independently the
recent negative trend in global vegetation activity
that we report using the MODIS sensor.
The significant positive correlation between
land surface temperature and global CO
2
annual
growth rates shown by Samanta et al. in fact
suggests the negative role of warming tempera-
ture in land carbon sink strength. That is, warming
induces drought and high ecosystem respiration
that weaken the terrestrial carbon sink, consistent
with the conclusion in ZR10 (1). In 2010, severe
droughts occurred again in the Amazon and
Europe (20, 22), leading to concerns of a possible
weakening of terrestrial carbon sinks (20, 1).
In summary, changes in NDVI and EVI and
sensitivity tests of our model, as well as other
studies, all suggest reduced vegetation activity
and NPP from 2000 through 2009. Our contin-
uous monitoring shows that global NPP in 2010
(53.19 Pg C) was lower than that in 2009 (53.84
Pg C), largely due to the two large-scale droughts
in the Amazon and Europe. We expect that the
strongest impacts of changing climate on terres-
trial ecosystem productivity will continue to be
manifested through the hydrologic cycle, but
whether these current trends continue can only
be answered by global monitoring.
References and Notes
1. M. Zhao, S. W. Running, Science 329, 940 (2010).
2. A. Samanta et al., Science 333, 1093 (2011);
www.sciencemag.org/cgi/content/333/6046/1093-c.
3. B. E. Medlyn, Science 333, 1093 (2011);
www.sciencemag.org/cgi/content/333/6046/1093-d.
4. M. Zhao, S. W. Running, R. R. Nemani, J. Geophys. Res.
111, (G1), G01002 (2006).
5. A. Samanta et al., Geophys. Res. Lett. 33, L06405
(2010).
6. O. L. Phillips et al., Science 323, 1344 (2009).
7. P. Jönsson, L. Eklundh, Comput. Geosci. 30, 833
(2004).
8. D. A. Clark, S. C. Piper, C. D. Keeling, D. B. Clark,
Proc. Natl. Acad. Sci. U.S.A. 100, 5852 (2003).
9. S. B. Peng et al., Proc. Natl. Acad. Sci. U.S.A. 101,
9971 (2004).
10. M. J. Tjoelker, J. Oleksyn, P. B. Reich, Glob. Change Biol.
7, 223 (2001).
11. S. Piao et al., Ecology 91, 652 (2010).
12. M. D. Mahecha et al., Science 329, 838 (2010).
13. D. Zheng, S. Prince, R. Wright, Glob. Change Biol. 9,
46 (2003).
14. C. Beer et al., Science 329, 834 (2010).
15. O. Atkin, H. Millar, M. Turnbull, New Phytol. 187,
268 (2010).
16. A. K. Knapp, M. D. Smith, Science 291, 481 (2001).
17. R. R. Nemani et al., Science 300, 1560 (2003).
18. M. Jung et al., Nature 467, 951 (2010).
19. P. M. Brando et al., Proc. Natl. Acad. Sci. U.S.A. 107,
14685 (2010).
20. S. L. Lewis, P. M. Brando, O. L. Phillips, G. M. F. van der Heijden,
D. Nepstad, Science 331, 554 (2011).
21. N. Gobron, A. Belward, B. Pinty, W. Knorr, Geophys. Res. Lett.
37, L15402 (2010).
22. D. Barriopedro, E. M. Fischer, J. Luterbacher, R. M. Trigo,
R. García-Herrera, Science 332, 220 (2011).
Acknowledgments: We thank the anonymous reviewers,
as well as L. Jones, J. Oyler, R. Anderson and Q. Mu
for helpful comments. This work is funded by NASA
Earth Observing System MODIS project (grant
NNH09ZDA001N-TERRA-AQUA).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1093-e/DC1
SOM Text
Figs. S1 to S4
Table S1
References and Notes
17 November 2010; accepted 1 August 2011
10.1126/science.1199169
Fig. 3. Interannual variations from 2000 through 2009 in anomalies of annual total global terrestrial
NPP calculated with control and four sensitivity tests (SOMtext S2): Sensitivity 1 (Q
10
= 2.0), Sensitivity
2 (Q
10
= 1.4), Sensitivity 3 (VPD
max
= 5000 Pa, Q
10
= 1.4), and Sensitivity 4 (VPD
max
= 10,000 Pa, Q
10
=
1.4). Control NPP is NPP estimated without these modifications of parameters.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1093-e
TECHNICAL COMMENT

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

BOOKS ET AL.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1094
T
he Changing Body uses the “plastic-
ity, flexibility and responsiveness” of
the human body to changes in nutri-
tion, disease, work, and warmth to gener-
ate long-term insights on human develop-
ment. The opening chapter lays out a series of
claims that set up the book’s narrative: (i) The
nutritional status of a generation determines
its longevity and ability to work. Individu-
als with a better nutrition status—manifest
as height for age and weight for height; the
result of the net accumulation of energy and
nutrient intake, infection,
care, and activity—have
better brain development,
maintain stronger immune
systems, and are less
likely to succumb to cer-
tain chronic diseases later
in life. (ii) The work of that
generation, when allied
to technology (broadly
defined), determines the
generation’s output. (iii) A
generation’s output, partly
determined by its inheri-
tance from past genera-
tions (i.e., malnourished mothers are more
likely to give birth to malnourished babies),
determines its standard of living through the
enhanced ability to acquire material goods
and to invest in technology. (iv) A genera-
tion’s standard of living determines the nutri-
tion status of the next generation (through the
ability of adolescent girls, pregnant women,
and parents to get access to food, health ser-
vices, and care for themselves and their chil-
dren). (v) And so on…
This circle underpins the book’s cen-
tral theory: “technophysio evolution,” a link
between technological and physiological
change. The authors explain that this differs
from conventional forms of evolution in its
emphasis on the control that humans have
over its speed and their environment. Two
chapters review the evidence behind these
propositions (using equations and lots of
data). The authors then devote three chapters
to examining, in great detail (again provid-
ing much data), technophysio evolution in
England and Wales, continental Europe, and
the United States. Over the past 300 years,
adult heights and life expectancies in all
three regions have responded very rapidly to
improvements in diet, disease prevention, and
sanitation. Today, for example, adult males in
the United Kingdom are, on average, 10 cm
taller than their counterparts in the early 18th
century, and their life expectancies at birth
have doubled.
As my own work centers on the links
among income, food consumption, and nutri-
tion; the links between
nutrition and productivity;
and the distribution of food
and other resources within
households in develop-
ing countries, much of
this territory is familiar
to me. Nevertheless, the
book makes several impor-
tant contributions. First, a
number of the authors are
historians, and they intro-
duce a long view into the
relationships among dif-
ferent variables. Too often,
researchers working on developing countries
neglect these intergenerational effects (from
grandparents to children through to their own
grandchildren). Our need to understand long-
wave phenomena will only increase as we
tackle issues such as aging, chronic disease,
urbanization, and climate change. The authors
offer some insights into how to improve our
grasp on long-term links and the added value
of doing so. For example, are we using short-
run estimates of the responsiveness of calo-
rie consumption to changes in income (vital
in projecting food needs and potential hunger
crises) when we should, as the book argues,
be using time series estimates (which are
much lower)? If we did, our forecasts of the
numbers of hungry people would be substan-
tially reduced, with profound implications for
public policy.
Second, the authors contrast experiences
from the rich and developing worlds. This
does not happen nearly enough. The two
spheres are quite distinct in terms of research
and policy communities. Yet the issues,
methods, and policy prescriptions are very
similar, and so the scope for cross-learning
seems immense. For example, the evidence
on how urbanization in the 19th-century
United States and United Kingdom led to
declines in average male heights (as disease
and overcrowding overwhelmed any higher
wages earned) should serve as a wake-up
call to those who seem to be taking a rather
casual view of the implications of the growth
of cities for well-being.
Third, the book outlines a number of
important questions that could shape the
future research agenda. Two stand out:
Why does it take more than one generation
for changes in environment to manifest as
improved adult height? Given that heights
were not measured at the population level
anywhere before the early 18th century, what
laments will 22nd-century analysts express
over variables that we should be measuring
today but have not even considered?
But the book also frustrates. It fails to
make the most of comparisons between
developed and developing countries. The
authors are not sufficiently familiar with the
literature on developing countries, and their
evidence relies too much on work by a small
number of top U.S. academics. Thus they
miss some opportunities to apply interesting
findings from Europe and the United States
to the developing world (and vice versa). For
example, the insight about the extent to which
heights can improve in one generation might
explain the curious lack of response of nutri-
tion status to sparkling economic growth in
India. (The book mistakenly brackets China
and India’s progress in improving nutrition
status.) Nor do the authors sufficiently chal-
lenge current notions of economic growth.
While effectively critiquing the partiality of
income as a measure of welfare (and argu-
ing that, as a measure, nutritional status is
“analogous to measures of capability”), they
remain sanguine about the capacity of cur-
rent economic growth patterns to generate
“bads” such as carbon emissions, obesity,
and inequality. The long-term perspective
afforded by the intergenerational historical
perspectives is also presumably one reason
that the book is very light on policy impli-
cations, but surely the long view imposes a
greater obligation to think about core policy
mechanisms, unencumbered by electoral
cycles. At the individual level, I have a nag-
ging worry that the authors paid insufficient
attention to people’s ability to influence the
impacts of long-term trends on their bodies.
The body is more than a diagnostic tool, it is
the servant of our agency.
Nevertheless, The Changing Body offers
an authoritative summary of the field of tech-
nophysio evolution. The authors place the
size and shape of the human body at the cen-
Well-Being and Its Consequences
ECONOMICS
Lawrence Haddad
The reviewer is at the Institute of Development Studies,
Library Road, Brighton BN1 9RE, UK. E-mail: l.haddad@
ids.ac.uk
The Changing Body
Health, Nutrition, and Human
Development in the Western
World Since 1700
by Roderick Floud, Robert W. Fogel,
Bernard Harris, and Sok Chul Hong
Cambridge University Press,
Cambridge, 2011. 457 pp. $90, £55.
ISBN 9780521879750. Paper, $32.99,
£19.99. ISBN 9780521705615.
New Approaches to Economic and
Social History.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

BOOKS ET AL.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1095
C
R
E
D
I
T
:

J
O
H
N

H
I
M
M
E
L
M
A
N
/
F
R
O
M

C
R
I
C
K
E
T

R
A
D
I
O

©

T
H
E

B
E
L
K
N
A
P

P
R
E
S
S

O
F

H
A
R
V
A
R
D

U
N
I
V
E
R
S
I
T
Y

P
R
E
S
S
W
hoever decided that the singing of
crickets should be called “stridu-
lation” must have been a cur-
mudgeon. Scientifically, the term describes
the act of creating sound by rubbing body
parts together, but etymologically it derives
from the Latin root strido, “to utter a harsh
sound.” It seems incongruous to call the
sounds made by crickets harsh when they
are among the few noises in nature that
humans perceive as musical.
The musical propensities of the Ensifera—
members of a suborder of Orthoptera com-
prising crickets, bush crickets, and katydids—
provide the theme for John Himmelman’s
Cricket Radio. An entomologist by avoca-
tion, the author has an enthusiasm for
crickets that, by his own description,
borders on obsession. He aims to help
people find and identify night-singing
orthopterans. But the book, more spirit
guide than field guide, is also intended
to raise awareness of these ubiquitous
night sounds, explain what purpose
they serve, and teach how to “tune
in” and appreciate these extraordinary
singers. Because of the centrality of
sound to the lives of ensiferans, Him-
melman can provide abundant infor-
mation on cricket biology while never
straying far from his central theme.
Moreover, his own experiences in
hunting for and learning about crickets
make this potentially abstruse topic far
more engaging.
Three chapters introduce ensiferan
biology: one on morphology, life his-
tories, and classification and a pair that
survey katydids and crickets. Him-
melman then focuses on interactions
Lively Notes on
Nocturnal Singers
ENTOMOLOGY
May Berenbaum
The reviewer is at the Department of Entomol-
ogy, 320 Morrill, University of Illinois, 505 South
Goodwin, Urbana, IL 61801–3795, USA. E-mail:
maybe@uiuc.edu
ter of generational transitions. In doing so,
they generate new insights into contemporary
development processes—even if they under-
state our ability to do something about the
trends that affect our bodies.
between ensiferans and humans. In addition
to disquisitions on a wide range of historical
and cultural references to these songsters, he
offers detailed instructions for finding and
handling crickets in their natural habitats and
for maintaining them at home (“assembling
your cricket radio”). Liberally illustrated with
black-and-white images (photographs and
drawings), the book features a
48-page color insert. The many
close-ups give this section the
look of a (rather peculiar) fam-
ily photo album assembled by
an affectionate uncle.
It may seem curmudgeonly
to criticize the book (clearly
aimed at lay readers) on techni-
cal grounds, but the regrettably
numerous errors are likely to
irritate professional entomolo-
gists. Most annoying are outright mistakes.
For example, Himmelman lumps mayflies
in with dipteran midges and mosquitoes as
belonging to “the world of flies”; as members
of the order Ephemeroptera, they are “flies”
in name only. Editors failed to catch various
misspelled or misused terms, both technical
(e.g., “tachnid” instead of “tachinid”) and
nontechnical (e.g., Himmelman explains that
moths are valuable “because they pollinate a
large percentage of our fauna”). In addition
to outright errors, there are some peculiarities
of style (such as capitalizing common names)
that as an entomologist I found jarring. Most
curmudgeonly, I have a personal aversion to
language in books about insects that refers to
“magic” or soul-stirring. However pleasant
the resulting sound may be, rubbing wings
together is not a magic trick.
But, again, this book is not intended for
hard-bitten entomologists,
so I should be more toler-
ant. In fact, when Him-
melman ranges beyond the
science to explore cultural
aspects of cricket singing,
I found myself entranced
and in awe of his determi-
nation to track down every
relevant detail. He is as
relentless in chasing inter-
esting references in litera-
ture as he is in pursuing crickets themselves
in the greenery near his home. He effectively
embeds his own cricket-watching experi-
ences into often-strange history of associa-
tions between crickets and people. In “The
Mighty Cricket Gladiators,” for example, he
segues from his adventures as an 8-year-old
who received a gift cricket farm inexplicably
equipped with six tiny plastic chariots into
a fascinating history of cricket fighting in
China. (This includes the tale of Ming Xuan-
Zhong, the “Cricket Emperor” of the Song
Dynasty who so prized his fighting
crickets that the accidental death of
a champion due to his wife’s care-
lessness led to a double suicide.)
He also provides a lively account
of tracking down the etymology of
the name katydid: delving into its
onomatopoetic nature and going so
far as to look up the 1751 journal of
botanist John Bartram to read what
may be the first mention of a katydid
(or, as Bartram put it, “catedidist”).
I enjoyed and learned from the bio-
graphical sketches of “bug people,”
which encompass not just ento-
mologists enamored of Orthoptera
but a diverse and colorful assort-
ment of people of many professions
(including a pirate), illustrative of
the widespread appeal of ensiferans.
Reading these stories did make me
realize that not all entomologists are
averse to magic and mystery with
respect to crickets. So I have prom-
ised myself to cut back on curmud-
geonliness (at least in the evenings)
and listen to cricket sounds with a
more sensitive attitude.
Cricket Radio
Tuning In the Night-Singing
Insects
by John Himmelman
Harvard University Press,
Cambridge, MA, 2011. 316 pp.
$22.95, £16.95, €20.70.
ISBN 9780674046900.
Nebraska conehead (Neoconocephalus nebrascensis). A male,
from a West Virginia cornfield. 10.1126/science.1209737
10.1126/science.1210542
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

BOOKS ET AL.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1095
C
R
E
D
I
T
:

J
O
H
N

H
I
M
M
E
L
M
A
N
/
F
R
O
M

C
R
I
C
K
E
T

R
A
D
I
O

©

T
H
E

B
E
L
K
N
A
P

P
R
E
S
S

O
F

H
A
R
V
A
R
D

U
N
I
V
E
R
S
I
T
Y

P
R
E
S
S
W
hoever decided that the singing of
crickets should be called “stridu-
lation” must have been a cur-
mudgeon. Scientifically, the term describes
the act of creating sound by rubbing body
parts together, but etymologically it derives
from the Latin root strido, “to utter a harsh
sound.” It seems incongruous to call the
sounds made by crickets harsh when they
are among the few noises in nature that
humans perceive as musical.
The musical propensities of the Ensifera—
members of a suborder of Orthoptera com-
prising crickets, bush crickets, and katydids—
provide the theme for John Himmelman’s
Cricket Radio. An entomologist by avoca-
tion, the author has an enthusiasm for
crickets that, by his own description,
borders on obsession. He aims to help
people find and identify night-singing
orthopterans. But the book, more spirit
guide than field guide, is also intended
to raise awareness of these ubiquitous
night sounds, explain what purpose
they serve, and teach how to “tune
in” and appreciate these extraordinary
singers. Because of the centrality of
sound to the lives of ensiferans, Him-
melman can provide abundant infor-
mation on cricket biology while never
straying far from his central theme.
Moreover, his own experiences in
hunting for and learning about crickets
make this potentially abstruse topic far
more engaging.
Three chapters introduce ensiferan
biology: one on morphology, life his-
tories, and classification and a pair that
survey katydids and crickets. Him-
melman then focuses on interactions
Lively Notes on
Nocturnal Singers
ENTOMOLOGY
May Berenbaum
The reviewer is at the Department of Entomol-
ogy, 320 Morrill, University of Illinois, 505 South
Goodwin, Urbana, IL 61801–3795, USA. E-mail:
maybe@uiuc.edu
ter of generational transitions. In doing so,
they generate new insights into contemporary
development processes—even if they under-
state our ability to do something about the
trends that affect our bodies.
between ensiferans and humans. In addition
to disquisitions on a wide range of historical
and cultural references to these songsters, he
offers detailed instructions for finding and
handling crickets in their natural habitats and
for maintaining them at home (“assembling
your cricket radio”). Liberally illustrated with
black-and-white images (photographs and
drawings), the book features a
48-page color insert. The many
close-ups give this section the
look of a (rather peculiar) fam-
ily photo album assembled by
an affectionate uncle.
It may seem curmudgeonly
to criticize the book (clearly
aimed at lay readers) on techni-
cal grounds, but the regrettably
numerous errors are likely to
irritate professional entomolo-
gists. Most annoying are outright mistakes.
For example, Himmelman lumps mayflies
in with dipteran midges and mosquitoes as
belonging to “the world of flies”; as members
of the order Ephemeroptera, they are “flies”
in name only. Editors failed to catch various
misspelled or misused terms, both technical
(e.g., “tachnid” instead of “tachinid”) and
nontechnical (e.g., Himmelman explains that
moths are valuable “because they pollinate a
large percentage of our fauna”). In addition
to outright errors, there are some peculiarities
of style (such as capitalizing common names)
that as an entomologist I found jarring. Most
curmudgeonly, I have a personal aversion to
language in books about insects that refers to
“magic” or soul-stirring. However pleasant
the resulting sound may be, rubbing wings
together is not a magic trick.
But, again, this book is not intended for
hard-bitten entomologists,
so I should be more toler-
ant. In fact, when Him-
melman ranges beyond the
science to explore cultural
aspects of cricket singing,
I found myself entranced
and in awe of his determi-
nation to track down every
relevant detail. He is as
relentless in chasing inter-
esting references in litera-
ture as he is in pursuing crickets themselves
in the greenery near his home. He effectively
embeds his own cricket-watching experi-
ences into often-strange history of associa-
tions between crickets and people. In “The
Mighty Cricket Gladiators,” for example, he
segues from his adventures as an 8-year-old
who received a gift cricket farm inexplicably
equipped with six tiny plastic chariots into
a fascinating history of cricket fighting in
China. (This includes the tale of Ming Xuan-
Zhong, the “Cricket Emperor” of the Song
Dynasty who so prized his fighting
crickets that the accidental death of
a champion due to his wife’s care-
lessness led to a double suicide.)
He also provides a lively account
of tracking down the etymology of
the name katydid: delving into its
onomatopoetic nature and going so
far as to look up the 1751 journal of
botanist John Bartram to read what
may be the first mention of a katydid
(or, as Bartram put it, “catedidist”).
I enjoyed and learned from the bio-
graphical sketches of “bug people,”
which encompass not just ento-
mologists enamored of Orthoptera
but a diverse and colorful assort-
ment of people of many professions
(including a pirate), illustrative of
the widespread appeal of ensiferans.
Reading these stories did make me
realize that not all entomologists are
averse to magic and mystery with
respect to crickets. So I have prom-
ised myself to cut back on curmud-
geonliness (at least in the evenings)
and listen to cricket sounds with a
more sensitive attitude.
Cricket Radio
Tuning In the Night-Singing
Insects
by John Himmelman
Harvard University Press,
Cambridge, MA, 2011. 316 pp.
$22.95, £16.95, €20.70.
ISBN 9780674046900.
Nebraska conehead (Neoconocephalus nebrascensis). A male,
from a West Virginia cornfield. 10.1126/science.1209737
10.1126/science.1210542
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1096
EDUCATIONFORUM
Drawing to Learn in Science
SCIENCE EDUCATION
Shaaron Ainsworth
1
*, Vaughan Prain
2
, Russell Tytler
3

Emerging research suggests drawing should
be explicitly recognized as a key element
in science education.
C
R
E
D
I
T
:

I
M
A
G
E

C
O
U
R
T
E
S
Y

O
F

R
I
L
S

(
A
R
C

D
P
0
7
0
9
9
9
)
S
hould science learners be
challenged to draw more?
Certainly making visual-
izations is integral to scientific
thinking. Scientists do not use
words only but rely on diagrams,
graphs, videos, photographs, and
other images to make discoveries,
explain findings, and excite pub-
lic interest. From the notebooks
of Faraday and Maxwell ( 1) to
current professional practices of
chemists ( 2), scientists imagine
new relations, test ideas, and elab-
orate knowledge through visual
representations ( 3– 5).
However, in the science class-
room, learners mainly focus on
interpreting others’ visualiza-
tions; when drawing does occur,
it is rare that learners are system-
atically encouraged to create their
own visual forms to develop and show under-
standing ( 6). Drawing includes constructing
a line graph from a table of values, sketch-
ing cells observed through a microscope, or
inventing a way to show a scientific phenom-
enon (e.g., evaporation). Although interpre-
tation of visualizations and other informa-
tion is clearly critical to learning, becoming
proficient in science also requires learners
to develop many representational skills. We
suggest five reasons why student drawing
should be explicitly recognized alongside
writing, reading, and talking as a key element
in science education. We offer distinct ratio-
nales, although in practice any single draw-
ing activity will likely rest upon multiple jus-
tifications. Both old and new technologies
offer exciting opportunities. We conclude by
highlighting important questions yet to be
answered and key future research to extend
teachers’ and learners’ use of drawing.
Drawing to Enhance Engagement
Many students disengage from school
science because rote learning and tradi-
tional topics reduce them to passive roles
( 7, 8). Reformers advocate more interac-
tive, inquiry-based learning ( 9). Surveys of
teachers and students indicate that, when
students drew to explore, coordinate, and
justify understandings in science, they were
more motivated to learn than from conven-
tional teaching ( 10). The use of drawing
caters to individual learner differences, as
a drawing is shaped by the learner’s current
or emerging ideas and knowledge of visual
conventions.
Drawing to Learn to Represent in Science
Students need to learn how scientists use
multiple literacies of this subject to con-
struct and record knowledge, where reading,
writing, and talk are integrated with visual
modes ( 11– 13). Generating their own rep-
resentations can deepen students’ under-
standing of the specific conventions of rep-
resentations (e.g., “This is how a line graph
works.”) and their purposes (e.g., the effec-
tiveness of line graphs for showing continu-
ous quantitative information), as well as how
representations work more generally (e.g.,
a representation is better when it is coher-
ent, compact, and parsimonious) ( 3, 14, 15).
Teachers can guide students to acquire the
visual literacies of science at the point when
they will see their relevance and appreciate
their explanatory power ( 16).
Drawing to Reason in Science
To show conceptual understanding, students
must learn how to reason with multiple, often
visual, modes ( 9). Understanding sound
waves, for instance, can involve being able to
coordinate a range of wave diagrams, time-
sequenced representations of air particle
movement, and pressure variation. Different
representations have distinctive attributes that
both guide and constrain what learners do and
come to understand ( 17– 19). As they select
specific features to focus on in their draw-
ing, learners reason in various ways, align-
ing their drawing with observation, measure-
ment, and/or emerging ideas ( 6, 20). Practice
in flexible manipulation of representations
has been argued to be central for develop-
ing expertise ( 21). Classroom research shows
how students reason as they generate and
refine models supported by expert teacher
guidance ( 22, 23). This creative reasoning is
distinct from, but complementary to, reason-
ing through argumentation ( 24).
Drawing as a Learning Strategy
Effective learning strategies help learners
overcome limitations in presented material,
organize their knowledge more effectively,
and integrate new and existing understand-
ing; ultimately, they can be transformative
by generating new inferences ( 25, 26). Draw-
ing can be one such effective strategy ( 6, 27).
Revealing understanding. Drawings by two 11-year-olds (left and right) of an evaporating handprint show representa-
tional choices that guide and communicate individual understandings.
*Author for correspondence. E-mail: shaaron.ainsworth@
nottingham.ac.uk
1
School of Psychology, University of Nottingham, University
Park, Nottingham NG7 2RD, UK.
2
Faculty of Education, La
Trobe University, Bendigo 3552, Australia.
3
School of Edu-
cation, Deakin University, Waurn Ponds 3217, Australia.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1097
EDUCATIONFORUM
For example, asking learners to read a text
and draw what they have understood requires
them to make explicit this understanding in
an inspectable form [( 28), see fig. S1 in sup-
porting online material (SOM)]. Unlike other
constructive strategies, such as writing sum-
maries or providing oral self-explanations,
visual representations have distinct attributes
that match the visual-spatial demands of
much of science learning. Moreover, visual
representation has been shown to encourage
further constructive strategies ( 29). Inventing
representations (including drawings) acts as
preparation for future learning, because it can
help students discern key features and chal-
lenges of new tasks ( 30).
Drawing to Communicate
Scientists draw to clarify ideas for colleagues,
students, and the public ( 2, 5). In externalizing
private knowledge more permanently, visual
representation is one way to enable broader
dissemination ( 4). Through drawing, students
make their thinking explicit and specific,
which leads to opportunities to exchange and
clarify meanings between peers ( 31). Where
learners generate and publicly share their rep-
resentations, they learn by critiquing the clar-
ity, coherence, and content of what they and
their peers have drawn ( 32). These windows
into student thinking can serve teachers in
diagnostic, formative, and summative assess-
ment ( 33, 34) (fig. S2).
Current Programs and New Directions
Various programs featuring drawing are now
in progress ( 22, 23, 35). The Role of Repre-
sentation in Learning Science (RiLS) project
( 36) is an exemplar showing how, through
hands-on activities and a variety of multi-
modal representations in which drawing was
central, learners aged 10 to 13 were guided to
generate, justify, and refine representations in
science (fig. S3).
In a unit on water, students produced
representations of particle ideas beyond the
teachers’ experience of previous perfor-
mance. For example, in one task, a class of
students placed their wet hands on paper and
then were challenged to represent what hap-
pens as the handprint diminishes. The draw-
ings reflect learners’ expanding on previous
work to reason about particle distribution
and movement, energy exchange, and time-
sequencing (see the figure). Students’ visual
choices indicate thoughtful engagement with
the task of creating a coherent account of the
phenomenon. Through appraisal and refine-
ment of drawings, teachers and students
established some representational conven-
tions, such as the circles reflecting particles.
Teachers used these diagrams to assess and
then further refine students’ understandings
of particle behavior.
The RiLS approach supported students
to deepen their understanding of the selec-
tive purpose of representational choices.
For example, a student justified the selective
nature of his animation of particles in evap-
oration thus: “I was just focusing on what
they do, not representing other things like
shape and size—they are very, very tiny.”
RiLS teachers have noted that their students
engaged more in class, discussed at a higher
level, and performed better in their work-
books ( 36). Analysis of test results showed
stronger outcomes than in previous stud-
ies using comparable methods ( 37). Further
research is now needed to establish explicit
connections between drawing used in this
way and learning.
Although there is growing evidence of the
benefits of drawing to learn science, many
unanswered questions remain. One active
arena is exploration of how learning with
new technologies can benefit from drawing.
Learners can draw to help them understand
what they are seeing in complex visualization
environments ( 38). Drawing can be the way
learners create models and interact with a sys-
tem ( 39, 40), or their freehand sketches can
be automatically marked to provide timely
feedback ( 41). Technology is also broaden-
ing our concept of drawing as learners cre-
ate animations ( 42) or use cameras and clay
models on drawn backdrops to generate 1-s
stop-frame movies of science processes ( 43).
We also need to research the fundamen-
tal mechanisms of drawing to learn. What
skills do you first need to develop in order to
best take advantage of learning by drawing?
Perhaps some topics are sufficiently difficult
to draw that attempting to do so is counter-
productive. A further important research area
concerns how teachers can best support their
students to use drawing alongside writing
and talking in the classroom. However, what
is clear is the growing interest in drawing as
it reflects new understandings of science as
a multimodal discursive practice, as well as
mounting evidence for its value in supporting
quality learning.
References and Notes
1. D. C. Gooding, J. Cogn. Cult. 4, 551 (2004).
2. R. Kozma, E. Chin, J. Russell, N. Marx, J. Learn. Sci. 9, 105
(2000).
3. J. K. Gilbert, Visualization in Science Education (Springer-
Verlag, New York, 2005).
4. B. Latour, Pandora’s Hope: Essays on the Reality of Science
Studies (Harvard Univ. Press, Cambridge, MA, 1999).
5. N. Nersessian, in Teaching Scientific Inquiry: Recommen-
dations for Research and Implementation, R. Duschl, R.
Grandy, Eds. (Sense Publishers, Rotterdam, Netherlands,
2008), pp. 57–79.
6. P. Van Meter, J. Garner, Educ. Psychol. Rev. 17, 285 (2005).
7. T. Lyons, Int. J. Sci. Educ. 28, 591 (2006).
8. J. Osborne, J. Dillon, Science Education in Europe: Critical
Reflections (Nuffield Foundation, London, 2008).
9. R. A. Duschl, R. E. Grandy, Teaching Scientific Inquiry: Rec-
ommendations for Research and Implementation (Sense
Publishing, Rotterdam, Netherlands, 2005).
10. M. Hackling, V. Prain, Primary Connections: Stage 2 trial
(Australian Academy of Science, Canberra, 2005).
11. J. S. Krajcik, L. M. Sutherland, Science 328, 456 (2010).
12. J. L. Lemke, in Crossing Borders in Literacy and Science
Instruction: Perspectives on Theory and Practice, E. W. Saul,
Ed. (International Reading Association, Newark, DE, 2004),
pp. 33–47.
13. P. D. Pearson, E. Moje, C. Greenleaf, Science 328, 459
(2010).
14. A. A. diSessa, Cogn. Instr. 22, 293 (2004).
15. E. Stern, C. Aprea, H. G. Ebner, Learn. Instr. 13, 191
(2003).
16. N. Enyedy, Cogn. Instr. 23, 427 (2005).
17. S. E. Ainsworth, Learn. Instr. 16, 183 (2006).
18. M. Scaife, Y. Rogers, Int. J. Hum. Comput. Stud. 45, 185
(1996).
19. B. Tversky, Top. Cogn. Sci. 3, 499 (2011).
20. R. Cox, Learn. Instr. 9, 343 (1999).
21. J. G. Greeno, R. P. Hall, Phi Delta Kappan 78, 361 (1997).
22. J. Clement, M. A. Rea-Ramirez, Model Based Learning and
Instruction in Science (Springer Science + Business Media,
Dordrecht, Netherlands, 2008).
23. R. Lehrer, L. Schauble, in The Cambridge Handbook of the
Learning Sciences, K. Sawyer, Ed. (Cambridge Univ. Press,
Cambridge, 2006), pp. 371–388.
24. J. Osborne, Science 328, 463 (2010).
25. M. T. H. Chi, M. Bassok, M. W. Lewis, P. Reimann, R. Glaser,
Cogn. Sci. 13, 145 (1989).
26. U. Kombartzky, R. Ploetzner, S. Schlag, B. Metz, Learn. Instr.
20, 424 (2010).
27. J. D. Gobert, J. J. Clement, J. Res. Sci. Teach. 36, 39 (1999).
28. S. E. Ainsworth, M. J. Nathan, P. van Meter, Proceedings of
the 9th International Conference of the Learning Sciences,
Chicago, 29 June to 2 July 2010 (International Confer-
ence of the Learning Sciences, Chicago, 2010), vol. 2, pp.
164–165.
29. S. E. Ainsworth, A. T. Loizou, Cogn. Sci. 27, 669 (2003).
30. D. L. Schwartz, T. Martin, Cogn. Instr. 22, 129 (2004).
31. D. L. Schwartz, J. Learn. Sci. 4, 321 (1995).
32. M. C. Linn, C. Lewis, I. Tsuchida, N. B. Songer, Educ. Res.
29, 4 (2000).
33. J. E. Dove, L. A. Everett, P. F. W. Preece, Int. J. Sci. Educ. 21,
485 (1999).
34. K. Ehrlén, Int. J. Sci. Educ. 31, 41 (2009).
35. Picturing to Learn, www.picturingtolearn.org.
36. P. Hubber, R. Tytler, F. Haslam, Res. Sci. Educ. 40, 5 (2010).
37. P. Hubber, in Physics Community and Cooperation, D.
Raine, L. Rogers, C. Hurkett, Eds. (Univ. of Leicester, Leices-
ter, UK, 2010), pp. 45–64.
38. H. Z. Zhang, M. Linn, Proceedings of the 9th International
Conference of the Learning Sciences, Chicago, 29 June to
2 July 2010 (International Conference of the Learning Sci-
ences, Chicago, 2010), vol. 2, pp. 165–166.
39. Crayon Physics Deluxe, http://crayonphysics.com/.
40. W. R. van Joolingen, L. Bollen, F. A. J. Leenaars, in Advances
in Intelligent Tutoring Systems, R. Nkambou, J. Bourdeau,
R. Mizoguchi, Eds. (Springer-Verlag, Berlin, 2010), pp.
266–282.
41. K. Forbus, J. Usher, A. Lovett, K. Lockwood, J. Wet-
zel, Top. Cogn. Sci., published online 11 April 2011
().10.1111/j.1756-8765.2011.01149.x
42. H. Y. Chang, C. Quintana, J. S. Krajcik, Sci. Educ. 94, 73
(2010).
43. D. Macdonald, G. Hoban, Int. J. Learn. 16, 319 (2009).
44. The authors are affiliated with RiLS.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1096/DC1
10.1126/science.1204153
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1098
PERSPECTIVES
M
eteorites are the oldest rocks in our
solar system, and therefore record
the earliest stages of its evolution.
On the basis of their mineralogy, petrogra-
phy, bulk chemistry, and oxygen-isotope
compositions, meteorites are classified into
more than 70 groups. Each group is believed
to represent samples of one or a few larger
bodies that formed early in the solar system
and may still be represented by asteroids ( 1)
in the main asteroid belt, which lies between
the orbits of Mars and Jupiter. Various types
of asteroids have been identified based on
their spectral properties. Linking the mete-
orite groups to these asteroid types is critical
for assessing the meteorite data and using it
as a basis for inferring, for example, the com-
position and growth of Earth and other plan-
ets. Because we have lacked direct samples
of specific asteroids, we have had to rely on
remote observations of asteroids and relate
them to the meteorite groups. This has been
problematic because the surfaces of asteroids
have been heavily modified by space weather-
ing, which changes their spectral properties.
To establish such a connection and understand
the role of space weathering, the JAXA (Japan
Aerospace Exploration Agency) spacecraft
Hayabusa was sent to near-Earth asteroid
25143 Itokawa, which is 0.5 by 0.3 by 0.2 km
in size to collect and bring back to Earth sam-
ples of its surface (see the figure). Itokawa is
an S-type asteroid, the most abundant spectral
type of asteroid in the inner asteroid belt.
On 25 November 2005, Hayabusa gen-
tly landed on Itokawa. Because of several
glitches, it wasn’t clear whether the spacecraft
had successfully sampled the asteroid, and the
return to Earth became longer and more har-
rowing than planned. However, on 13 June
2010, Hayabusa successfully returned to Earth
along with more than 1500 rocky particles, up
to 180 µm in size, from Itokawa. These par-
ticles are the first samples returned from an
asteroid and the second set of samples of
extraterrestrial surface rock material returned
to Earth. (The first were the lunar samples col-
lected by the NASA Apollo and Soviet Luna
missions.) The detailed laboratory studies
of the mineralogy, petrography, chemistry,
and noble gas and oxygen-isotope composi-
tions of the Itokawa particles reported in this
issue ( 2– 7) provide unequivocal evidence that
S-type asteroids like Itokawa were the parent
bodies of ordinary chondrites, the most abun-
dant type of meteorites found on Earth. Three
groups of ordinary chondrites are currently
recognized—H, L, and LL ( 1)—with slightly
different chemistry and thermal histories.
Itokawa is composed of thermally metamor-
phosed LL4, LL5, and LL6 materials.
Despite the small sizes of the samples
returned (typically 10 to 50 µm), they revealed
that Itokawa had a complex formation his-
tory and that its surface is dynamic today. The
asteroid was initially much larger, more than
20 km in diameter, and experienced inten-
sive thermal metamorphism at ~800°C, pos-
sibly by decay of a short-lived radionuclide
26
Al, during the first 10 million years of solar
system evolution. It was subsequently cata-
strophically disaggregated by impacts into
small pieces, some of which reaccreted into
the present rubble-pile asteroid ( 2).
High-resolution x-ray microtomography
combined with scanning and TEM ( 5, 6) and
noble gas analyses ( 7) of Itokawa particles
provide insights into the formation of the
loose material on the surface of the asteroid
(regolith), the history of irradiation by cosmic
rays and solar wind, and the nature of space
weathering on asteroid-sized bodies. The Ito-
kawa particles resulted from mechanical dis-
aggregation by meteoroids and grain abra-
sion due to seismic-induced grain migration,
but escaped in situ melting and formation of
agglutinates, commonly observed in lunar
regolith. Space weathering was also differ-
ent from the lunar-style space weathering and
resulted in vapor deposition of sulfur-bear-
ing Fe-rich nanoparticles and formation of
an amorphous surface layer accompanied by
minor in situ reduction of Fe
2+
to nanophase
Fe metal by deeply implanted solar wind ions
( 6). The grains experienced multiple implan-
tations of solar wind particles and on average
resided for less than 8 million years on the
surface. Itokawa is thus continuously losing
surface materials into space at a rate of tens of
centimeters per million years. At this rate, it
will disappear in less than 1 billion years ( 7).
The results from the JAXA Hayabusa ( 2–
7) and NASA Stardust (returned samples of
Jupiter Family Comet Wild2) ( 8, 9) and Gen-
esis (returned samples of solar wind) mis-
sions ( 10, 11) have demonstrated that solar
system samples returned for study in terres-
trial laboratories are crucial in understanding
the origin and evolution of the solar system.
These missions, however, have not delivered
any water-bearing mineral species, which
are important for understanding one of the
most outstanding questions in planetary sci-
ence—the origin of Earth’s water ( 12). Sev-
eral potential sources of that water are being
discussed, including cometary, asteroidal,
and water adsorbed on the surface of fine-
grained dust during the initial stages of plan-
etary accretion ( 12– 14). Hydrogen isotopic
measurements of water-bearing minerals in
terrestrially uncontaminated samples from
the hydrated B-, C-, or D-type asteroids,
possible sources of hydrated carbonaceous
chondrites, can potentially provide a key
for understanding a role of asteroidal water
in the origin of Earth’s oceans. Two sample-
returning missions to asteroids rich in organ-
Bringing Part of an Asteroid
Back Home
PLANETARY SCIENCE
Alexander N. Krot
A space mission that took samples of
an asteroid has provided insights into
the evolution of the solar system.
Sampling an asteroid. Illustration of the release of
the target marker by Hayabusa before landing on
the surface of the Itokawa asteroid.
C
R
E
D
I
T
:

A
K
I
H
I
R
O

I
K
E
S
H
I
T
A
/
J
A
X
A
Hawai`i Institute of Geophysics and Planetology, Univer-
sity of Hawai`i at Manoa, Honolulu, HI 96822, USA. E-mail:
sasha@higp.hawaii.edu
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1099
PERSPECTIVES
ics and hydrated minerals have been recently
approved. The JAXA Hayabusa-2 spacecraft
will be launched in 2014 to a C-type asteroid,
1999 JU3. The NASA mission OSIRIS-Rex
(standing for Origins, Spectral Interpreta-
tion, Resource Identification, and Security–
Regolith Explorer) scheduled for 2016 will
explore a B-type asteroid, 1999 RQ6.
These voyages of exploration will
undoubtedly expand our knowledge of near-
Earth space as it currently exists and contrib-
ute significantly to humanity’s understanding
of Earth’s most distant past.
References
1. A. N. Krot, K. Keil, C. A. Goodrich, M. K. Weisberg, E. R. D.
Scott, in Meteorites, Comets and Planets, A. M. Davis, Ed.
(Elsevier, Amsterdam, 2005), pp. 83–129.
2. T. Nakamura et al., Science 333, 1113 (2011).
3. H. Yurimoto et al., Science 333, 1116 (2011).
4. M. Ebihara et al., Science 333, 1119 (2011).
5. A. Tsuchiyama et al., Science 333, 1125 (2011).

6. T. Noguchi et al., Science 333, 1121 (2011).
7. K. Nagao et al., Science 333, 1128 (2011).
8. D. Brownlee et al., Science 314, 1711 (2006).
9. M. E. Zolensky et al., Science 314, 1735 (2006).
10. K. D. McKeegan et al., Science 332, 1528 (2011).
11. B. Marty, M. Chaussidon, R. C. Wiens, A. J. Jurewicz,
D. S. Burnett, Science 332, 1533 (2011).
12. M. Mottl, B. Glazer, R. Kaiser, K. Meech, Chem. Erde 67,
253 (2007).
13. A. Morbidelli et al., Meteorit. Planet. Sci. 35, 1309 (2000).
14. M. J. Drake, Meteorit. Planet. Sci. 40, 519 (2005).
10.1126/science.1212145
Arranging a Cellular Checkerboard
CELL BIOLOGY
Wangsun Choi
1
and Mark Peifer
1, 2

Heterotypic cell adhesion mediated by nectins
directs the complex cellular architecture of the
inner ear.
C
R
E
D
I
T
:

B
.

S
T
R
A
U
C
H
/
S
C
I
E
N
C
E
B
iological patterns are all around us,
ranging from the complex patterns on
butterfly wings to the baroque archi-
tecture of kidney tubules. These patterns
reflect the organization of millions of cell types
in intricate yet reproducible ways to build tis-
sues and organs. Even more remarkable, these
patterns often arise through the self-assem-
bly of undifferentiated cells. Understanding
how this occurs is a key challenge for biolo-
gists. On page 1144 in this issue,
Togashi et al. ( 1) implicate the
interactions between nectin adhe-
sion molecules in this process.
More than 50 years ago,
experiments revealed that cells
from different tissues sort from
one another in reproducible
ways ( 2). We now know that this
also occurs during development,
modulating events as diverse as
segmentation of the vertebrate
hindbrain and separation of the
front and back halves of fruit fly
wings ( 3). Large-scale cell sort-
ing was hypothesized to occur by
differential cell adhesion ( 4) and
to underlie tissue segregation, in
which the attachment of an adhe-
sion molecule on one cell binds to
an identical (homophilic) adhe-
sion molecule on an adjacent cell.
Cell surface membrane proteins
called cadherins ( 5) are expressed
in patterns consistent with roles
in tissue sorting. However, there
are only a few clear examples of
cadherin-based tissue segregation—the posi-
tioning of oocytes by E-cadherin during ovary
development in the fly Drosophila melano-
gaster; boundary formation between the cere-
bral cortex and striatum by classic cadherins
in the developing mouse brain; and the segre-
gation of groups of motor neurons by type II
cadherins in the chick spinal cord ( 3).
Tissue sorting shapes the body plan but
does not explain more intricate patterns seen at
the single-cell level. The human inner ear pro-
vides a striking example ( 6). We hear because
sound generates fluid waves within the inner
ear, which are detected when they deflect ste-
reocilia projecting from inner-ear hair cells.
Outer and inner hair cells are arranged in a
“checkerboard” pattern that is interspersed
with different types of supporting cells, but
the cellular mechanisms that produce this pat-
tern have not been known.
Togashi et al. provide evi-
dence for cell sorting at the sin-
gle-cell level that relies on cell
surface proteins called nectins.
Like cadherins, nectins medi-
ate cell-cell adhesion ( 7). They
are part of the immunoglobulin
superfamily, with extracellu-
lar domains distantly related to
antibodies. Mammals have four
distinct nectins that mediate
both homophilic adhesion and
binding to nonidentical (hetero-
philic) nectins. The latter inter-
action is preferred, with a range
of binding strengths: Binding
of nectin-1 to nectin-3 is stron-
gest, followed by nectin-2–nec-
tin-3, and then either nectin-1–
1
Department of Biology, University of North
Carolina, Chapel Hill, NC 27599, USA.
2
Lineberger Comprehensive Cancer Center,
University of North Carolina, Chapel Hill,
NC 27599, USA. E-mail: peifer@unc.edu
Outer
hair cell
Supporting
cell
Organ of Corti
Stereocilia
Coiled duct
Cochlea
Adhesion and patterning. The organ
of Corti in the cochlea of the mamma-
lian inner ear contains sensory hair
cells interspersed with supporting cells.
In the mouse ear, heterophilic inter-
action between nectin-1 expressed on
hair cells and nectin-3 on supporting
cells generates a checkerboard pattern
of hair cells and supporting cells dur-
ing development (shown). Homophilic
interactions between cadherins (one or
more types) may also regulate cell-cell
adhesion between these cell types.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1100
PERSPECTIVES
nectin-1, nectin-2–nectin-2, or nectin-3–
nectin-3 binding.
In the mouse inner ear, nectins are
expressed in a checkerboard pattern: nectin-1
in hair cells, nectin-3 in supporting cells, and
nectin-2 in both. Because each hair cell is sur-
rounded by many supporting cells, heterotypic
adhesion between nectin-1 and nectin-3 joins
hair cells and supporting cells (see the figure).
To test the biological function of differential
nectin expression, Togashi et al. examined
cell arrangements in mice lacking each nec-
tin. Without heterophilic adhesion between
nectin-1 and nectin-3, the checkerboard pat-
tern of hair cells was disrupted, leading to
homophilic contacts between the same type of
cells—for example, the authors observed nec-
tin-1–mediated contact between hair cells in
nectin-3 mutants. It remains unclear whether
this affects hearing. Togashi et al. also dem-
onstrated that nectins are sufficient for cel-
lular patterning in vitro. When they created
an artificial boundary between cultured cells
expressing nectin-1 or nectin-3, the cells rear-
ranged at the border to form a mosaic pattern
resembling the checkerboard pattern of cells
in the inner ear.
Genetic analysis in the mouse is compli-
cated by the presence of four nectins, which
may have redundant functions. Thus, mice
lacking nectin-1, nectin-2, or nectin-3 indi-
vidually survive. However, as observed by
Togashi et al. in the ear, their absence causes
defects in cell-cell adhesion in other tissues.
In the mouse eye, pigment and nonpigment
cells both express nectin-1, -2, and -3. Nec-
tin-1 and nectin-3 mediate heterophilic adhe-
sion between these cell layers, and deleting
the gene for either of these nectins causes the
cell layers to detach from each other, leading
to severe eye defects ( 8). Likewise, in the tes-
tis, developing spermatids express nectin-3,
whereas nectin-2 is restricted to supporting
Sertoli cells. The nectins form heterophilic
cell-cell junctions required for spermatid–
Sertoli cell adhesion and proper spermatogen-
esis ( 9). In the developing mouse brain, syn-
apses between mossy fiber terminal axons and
CA3 pyramidal cell dendrites require asym-
metric distribution of nectin-1 and nectin-3,
respectively ( 10). While these defects do not
affect neural transmission in the hippocam-
pus, mutations in human nectin-1 sometimes
lead to mental retardation.
The findings of Togashi et al. are reminis-
cent of events in fruit fly eyes, another complex
sensory structure with a specific arrangement
of photoreceptors and supporting cells ( 11).
As in mammalian ears, differential cell-cell
adhesion shapes this intricate cellular array.
Differential expression of E- and N-cadherin
regulates cone cell shape. Notably, differen-
tial expression of cell surface proteins belong-
ing to another subset of the immunoglobulin
superfamily regulates precise arrangement
of different types of supporting cells. Family
members Kirre and Roughest are expressed in
one supporting cell type, whereas Hibris and
Sns are present in the other ( 12, 13). As with
nectins, heterophilic protein (and thus cellu-
lar) interactions are preferred to maintain the
intricate arrangement of cell types.
This family of adhesion proteins also regu-
lates other processes that rely on heterophilic
cell interactions in the fly ( 14), including myo-
blast fusion, axon guidance, and formation of
a slit diaphragm–like structure in Drosoph-
ila nephrocytes. This latter function is strik-
ing because the closest human relatives of
fly Sns and Kirre are nephrins. Nephrins are
mutated in patients with congenital nephrotic
syndrome and are required for proper slit dia-
phragm formation during kidney develop-
ment. Although nephrins and nectins are not
orthologs, these parallels suggest that these
and other families of immunoglobulin super-
family proteins may modulate cell sorting and
maintain intricate cell patterning in a wide
variety of situations.
References
1. H. Togashi et al., Science 333, 1144 (2011) 10.1126/
science.1208467.
2. P. L. Townes, J. Holtfreter, J. Exp. Zool. 128, 53 (1955).
3. U. Tepass et al., Curr. Opin. Genet. Dev. 12, 572 (2002).
4. M. S. Steinberg, Science 141, 401 (1963).
5. M. Takeichi, Science 251, 1451 (1991).
6. M. Schwander et al., J. Cell Biol. 190, 9 (2010).
7. Y. Takai et al., Annu. Rev. Cell Dev. Biol. 24, 309 (2008).
8. M. Inagaki et al., Development 132, 1525 (2005).
9. K. Ozaki-Kuroda et al., Curr. Biol. 12, 1145 (2002).
10. T. Honda et al., Mol. Cell. Neurosci. 31, 315 (2006).
11. U. Tepass, K. P. Harris, Trends Cell Biol. 17, 26 (2007).
12. S. Bao, R. Cagan, Dev. Cell 8, 925 (2005).
13. S. Bao et al., Dev. Biol. 344, 948 (2010).
14. K. F. Fischbach et al., J. Neurogenet. 23, 48 (2009).
10.1126/science.1211158
I
n 1859, English naturalist Henry Wal-
ter Bates left Brazil after 11 strenuous,
danger-filled, but blissful years of explor-
ing and collecting in the Amazon. Despite all
of the privations he had suffered, the self-
taught amateur dreaded exchanging a land
of “perpetual summer,” “endless streams,”
and “boundless forests” for the “gloomy
winters,” “murky atmosphere,” and “factory
chimneys” of England ( 1). But Bates’s return
home could not have been timed better. Just
as he began to sort out his vast collections,
Darwin’s On the Origin of Species appeared
and gave him a framework for everything
that he had seen in the jungle. Well, almost
everything; Bates soon realized that he had
noticed some things that had escaped the
great Darwin’s attention but that could lend
support to Darwin’s controversial new theory
of natural selection.
What Bates discovered were harmless,
edible insects, particularly butterflies, whose
appearance resembled distasteful or noxious
species. Bates reasoned that such striking
“analogous resemblances” could not be coin-
cidences but must have come about because
the imitators gained an advantage by mim-
icking well-defended species. Bates’s hard-
earned, intimate familiarity with the variation
within species led him to realize that individ-
uals that were better mimics would fare better
than poorer imitators that would be culled by
predators. Bates’s explanation of mimicry as
“a most beautiful proof of the theory of natu-
ral selection” ( 2, 3) was quickly embraced by
Darwin and his advocates, and the phenome-
non has remained of intense interest to ecolo-
gists and evolutionary biologists ever since.
During the 150 years since Bates’s descrip-
tions, however, researchers have learned noth-
ing about the mechanisms underlying the gen-
eration of mimetic patterns. Until now. On
page 1137 of this issue, Reed et al. ( 4) iden-
tify the optix gene, a homolog of a Drosoph-
ila transcription factor involved in eye devel-
opment ( 5), as the key regulator of the vivid
red color patterns that are characteristic of the
classic mimetic wing patterns of Heliconius
butterflies. The discovery is a major advance
because it opens the way toward solving the
long-standing mystery of how similar patterns
can arise in both closely related and largely
unrelated species. The findings also represent
a major technological breakthrough in gene
How Great Wings Can Look Alike
EVOLUTION
Sean B. Carroll
The optix gene controls mimetic wing pattern evolution in Heliconius butterflies.
Howard Hughes Medical Institute, Laboratory of Molecu-
lar Biology, University of Wisconsin–Madison, 201 Bock
Laboratories, Madison, WI 53706, USA. E-mail: sbcarrol@
facstaff.wisc.edu
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1101
PERSPECTIVES
mapping and identification in these experi-
mentally challenging, nonmodel species.
The strikingly beautiful Heliconius wing
patterns actually present a twist on those first
cases of mimicry deciphered by Bates, in that
both species of any pair of mimics are dis-
tasteful and display warning coloration. It was
Fritz Müller, a German-born naturalist who
emigrated to Brazil in 1852, who showed that
each member of a pair of unpalatable mimics
benefited from mimicry and, moreover, that
the rarer species gained a greater advantage
( 3, 6). Heliconius butterflies are thus referred
to as “Müllerian mimics” to distinguish that
phenomenon from “Batesian mimicry,” in
which a palatable species resembles an unpal-
atable form.
Heliconius butterflies have attracted
extensive interest because of the well-doc-
umented co-mimicry that occurs among
closely related species in different geo-
graphic areas. In the best-studied example,
the two species Heliconius melpomene and
Heliconius erato each exhibit great intra-
specific variation between regions, but their
wing patterns converge upon one another in
any given area (see the figure). Such exten-
sive intraspecific morphological diversity,
and the convergent evolution of wing pat-
terns in different species under natural
selection, has raised the questions of how
such diversity is generated during devel-
opment and how similar patterns evolve in
different species ( 7). To answer these ques-
tions, investigators must identify the genes
involved in wing pattern formation. Crossing
experiments performed several decades ago
revealed that, despite the apparent complex-
ity of the diverse wing patterns, major differ-
ences among geographical races are due to
largely Mendelian factors ( 8), but gene iden-
tification was technically unapproachable.
Reed et al.’s breakthrough required a com-
bination of genomic approaches and some
intrepid, long-term fieldwork in Central and
South America by members of this extraor-
dinary international research team. Recent
studies had narrowed the locus responsible
for the red pattern elements on the H. erato
wing to a 380-kb genomic interval ( 9, 10).
After a systematic survey of all of the genes
in the interval revealed a striking correlation
between optix expression and red patterns
in different H. erato races and other Helico-
nius species (see the figure), the research-
ers genotyped wild-caught specimens from
hybrid zones in Peru and Costa Rica. These
individuals were drawn from populations
that had been subjected to thousands of gen-
erations of recombination between butterflies
with different wing patterns; this enabled the
researchers to determine that butterfly wing
phenotypes were strongly associated with the
genotype at the optix locus.
The link between the optix gene and the
highly variable red patterns in these butter-
flies raised the question of the nature of the
functional variation at the optix locus. The
researchers found no coding differences in
the Optix protein sequences between highly
divergent races of the same species, between
divergent phenotypes of closely related spe-
cies, or between similar phenotypes of dis-
tantly related species. Instead, they attributed
the functional differences at the optix locus
to noncoding cis-regulatory sequences. This
makes perfect sense in light of the observed
differences in the spatial regulation of optix
messenger RNA (mRNA) expression in dif-
ferent races and species, and what we know
about the central role of cis-regulatory muta-
tions in morphological evolution, particu-
larly for changes involving genes with mul-
tiple roles (pleiotropic genes), such as optix.
Research has shown that mutations in the dis-
crete, modular enhancers of pleiotropic genes
alter the spatial expression of a gene and the
morphology of one body part without alter-
ing gene expression or morphology in other
tissues, or altering the biochemical activity
of the gene product; in general, both kinds of
changes would be deleterious ( 11– 13).
The precise identification of the func-
tionally relevant mutations at the optix locus
remains to be determined. In addition, we do
not yet know the origins of the different optix
alleles in different races and species. It is of
fundamental interest to understand whether
similar patterns in different races or species
are due to the presence of the same alleles or
the consequence of independently derived
functional mutations. Thanks to the identifi-
cation of optix, the resolution of this question
is now within reach.
In The Naturalist on the River Amazons,
Bates’s riveting narrative of his journey, the
explorer waxed poetic about the meaning of
the great diversity of butterfly wing patterns
he had observed: “It may be said, therefore,
that on these expanded membranes nature
writes, as on a tablet, the story of the modifi-
cation of species” ( 1). Those tablets and sto-
ries are now being decrypted at their most
fundamental level. Bates would be thrilled
beyond words.
References
1. H. W. Bates, The Naturalist on the River Amazons
(Penguin Books, New York, 1988).
2. H. W. Bates, Trans. Linn. Soc. Lond. 23, 495 (1862).
3. T. N. Sherratt, Naturwissenschaften 95, 681 (2008).
4. R. D. Reed et al., Science 333, 1137 (2011).
5. M. Seimiya, W. J. Gehring, Development 127, 1879 (2000).
6. F. Müller, Proc. Entomol. Soc. Lond. 1879, 20 (1879).
7. H. Frederik Nijhout, The Development and Evolution of
Butterfly Wing Patterns (Smithsonian Press, Washington,
1991).
8. P. M. Sheppard, J. R. G. Turner, K. S. Brown, W. W. Ben-
son, M. C. Singer, Philos. T. R. Soc. B 308, 433 (1985).
9. R. Papa et al., BMC Genomics 9, 345 (2008).
10. S. W. Baxter et al., Genetics 180, 1567 (2008).
11. S. B. Carroll, PLoS Biol. 3, e245 (2005).
12. G. A. Wray, Nat. Rev. Genet. 8, 206 (2007).
13. S. B. Carroll, Cell 134, 25 (2008).
H. erato
H. melpomene
H. erato hindwing H. melpomene hindwing optix expression optix expression
The optix gene and butterfly mimicry. (Top) H. erato and H. melpomene butterflies exhibit great variation
in their wing patterns but converge on similar patterns in the same geographic areas (butterflies from the
same area are aligned vertically). (Bottom) The red patterns on the wings of each species are controlled by the
Optix protein. Schematics of the convergent red ray patterns found on the hindwings of one race of H. erato
and H. melpomene are shown next to schematics of optix mRNA expression (purple) in developing hindwings.
Reed et al. ( 4) have shown that the pattern of optix expression corresponds to the adult red color patterns.
10.1126/science.1211025
C
R
E
D
I
T
:

R
O
B
E
R
T

R
E
E
D

A
N
D

A
R
N
A
U
D

M
A
R
T
I
N
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1102
PERSPECTIVES
S
quamous cell carcinomas of the head
and neck (HNSCCs) that arise in the
mucosal linings of the upper respira-
tory and digestive tracts are the sixth lead-
ing cancer by incidence worldwide, with
~600,000 new cases each year ( 1). The most
important risk factors for these cancers are
tobacco use and alcohol consumption, while
a subgroup is caused by infection with human
papillomaviruses that also cause cervical
cancer ( 2). Patients with early-stage disease
are treated by either surgery or radiotherapy,
whereas patients in advanced stages receive
a combination of these modalities or concur-
rent chemotherapy and localized radiation.
Only 40 to 50% of patients will survive for 5
years after treatment ( 2). Antibodies directed
against the epidermal growth factor recep-
tor have been the only recent new therapy for
this life-threatening disease ( 3). Two papers in
this issue—by Agrawal et al. ( 4) on page 1154
and Stransky et al. ( 5) on page 1157—provide
new insight into the genetic changes causing
HNSCC that may guide the development of
alternative treatment strategies.
Agrawal et al. and Stransky et al. used
a high-throughput technique called mas-
sively parallel sequencing or next-genera-
tion sequencing to analyze the genomes of
head and neck cancers in great detail. Both
groups sequenced the exons of all known
human genes in tumor DNA and compared
the sequence to that of the corresponding
normal DNA of the same patient. In total,
the genomic landscapes of 32 ( 4) and 74 ( 5)
tumors were examined, including tumors
that were positive or negative for the human
papillomavirus. Agrawal et al. also pro-
vided genetic profiling data on chromosomal
changes, verified the mutations by classical
Sanger sequencing, and validated some muta-
tions in an additional panel of tumor and nor-
mal tissues. Mutations were found in many
of the genes already known to play a role in
HNSCC, such as TP53, CDKN2A, PIK3CA,
PTEN, and HRAS, but at least one new can-
cer gene previously not known to be involved
in HNSCC, NOTCH1, was identified. In both
studies, inactivating mutations of NOTCH1
were found in 10 to 15% of the head and neck
tumors, and it was the second most frequently
mutated gene after TP53 (which is mutated in
50 to 80% of the tumors). In several tumors,
both alleles harbored mutations in NOTCH1.
Why was NOTCH1 not found before in
this type of cancer or even in other malig-
nancies ( 6) as an important tumor suppres-
sor? Functional studies had identified a role
for NOTCH1 in squamous cell carcinogene-
sis, at least in the skin ( 7), but robust muta-
tional data in clinical samples were missing.
NOTCH1 is a very large gene consisting of
34 coding exons, which hampers classical
(Sanger) DNA sequencing, thus demonstrat-
ing the major improvement afforded by next-
generation sequencing platforms.
The finding of numerous inactivat-
ing mutations in NOTCH1 in HNSCCs and
the observation that mice with a disrupted
NOTCH1 gene in the skin show a skin can-
cer phenotype ( 8, 9) provide strong evidence
that NOTCH1 is an important tumor sup-
pressor gene in HNSCC. NOTCH1 encodes
a transmembrane receptor that functions in
cell-to-cell communication ( 10) and is in the
skin typically located in the cilia of the squa-
mous cells, the dermal keratinocytes ( 11).
After ligand binding, the cytoplasmic tail of
NOTCH is cleaved by a secretase enzyme,
translocates to the nucleus, and functions as
a transcription factor, driving the expression
of numerous genes (see the figure). All four
NOTCH receptors encoded in the human
genome are important for cell differentia-
tion. Stransky et al. also found mutations in
other cell differentiation-related genes, such
as NOTCH2, NOTCH3, and TP63, suggest-
ing that deregulation of the terminal differ-
entiation program of mucosal keratinocytes
is critical for squamous cancer development.
This is not unexpected because terminal dif-
ferentiation of normal keratinocytes in skin
and mucosal epithelia is characterized by loss
of cell organelles and even the nucleus during
cornification—events that support the bar-
rier function of squamous epithelia but which
would inhibit malignant transformation.
However, some questions remain. A high-
throughput sequencing approach can reveal
many mutations in a large number of genes,
but this does not necessarily imply that these
are all “driver mutations” causally related
to the malignant transformation process.
Tumors are genetically unstable and acquire
many mutations including so-called “pas-
senger mutations” ( 12, 13) that are a result
of malignant transformation and not the
cause. Functional studies in animal models
are required to elucidate the exact role of the
NOTCH receptors and the other genes that are
mutated in HNSCC. As an example, Agrawal
et al. indicated that they also found mutations
in FBXW7 in tumors that lack inactivating
NOTCH1 mutations. The FBXW7 protein is a
component of a ubiquitin ligase complex that
Another NOTCH for Cancer
CANCER
Ruud H. Brakenhoff
Next-generation sequencing reveals NOTCH1
as an important tumor suppressor gene in head
and neck cancer.
NOTCH and squamous cells. Upon its interaction with
a ligand, NOTCH is cleaved and its intracellular domain
(NICD) translocates to the nucleus and activates tar-
get genes. In mouse skin, this pathway decreases
WNT signaling and initiates differentiation. It is likely,
although not proven, that the same mechanism oper-
ates in squamous cells of the mucosal linings of the
throat, oral cavity, and tongue. P, phosphorylation.
C
R
E
D
I
T
:

B
.

S
T
R
A
U
C
H
/
S
C
I
E
N
C
E
Department of Otolaryngology/Head-Neck Surgery, VU
University Medical Center, 1007 MB Amsterdam, Nether-
lands. E-mail: rh.brakenhoff@vumc.nl
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1103
PERSPECTIVES
F
lowering plants possess the remark-
able capacity to selectively move
large, functional proteins and RNA
molecules between cells through regulated
channels called plasmodesmata ( 1). For
example, within a meristem (a group of stem
cells) plants can steer macromolecular traf-
fic in particular directions along these “high-
ways” to orchestrate stem cell division and
the development of specialized tissues. How
they control this cell-to-cell traffic, however,
has been a question of major importance. On
page 1141 of this issue, Xu et al. ( 2) reach a
major milestone in the quest for an answer.
They show that selective trafficking of a
functional protein, the maize transcription
factor KNOTTED1 (KN1), requires KN1 to
unfold and then be refolded in the destina-
tion cell by a group of proteins known as the
chaperonin complex.
Plasmodesmata create highways along
which proteins move from synthesis sites
on the endoplasmic reticulum (ER) within
one cell, through a plasmodesmal channel,
and into a neighboring cell. When a protein
arrives at a plasmodesma, there are two main
possibilities for how it might get through
the narrow channel. One is that the channel
dilates to permit passage; the other is that the
protein is unfolded to allow passage and then
refolded when it reaches its destination.
Xu et al. used a clever research design,
involving genes that control the growth of
leaf hairs (trichomes) in Arabidopsis plants
engineered to express maize KN1, to inves-
tigate which of these two mechanisms was
involved in KN1 trafficking between leaf
cells. It enabled the researchers to show that
KN1 trafficking in the shoot apical meristem
(the growing tip of a stem) requires CCT8,
a subunit of the type II chaperonin complex.
In their experimental system, they found that
trafficking required the presence of the chap-
eronin complex in destination cells, which
were located on leaf surfaces, but not in the
cells where KN1 was produced, which were
located in the inner leaf mesophyll, the site
of photosynthesis. This suggests that KN1 is
unfolded, moves through the plasmodesmal
channel, and then is refolded at its destination
to restore function (see the figure).
Xu et al.’s findings have broad impli-
cations for understanding other plant pro-
cesses, such as the induction of flowering.
Nearly 50 years ago, the late plant physiol-
ogist Jan Zeevaart remarked that the most
urgent problem in flowering physiology was
identifying the signaling molecule known
as “florigen” ( 3). Today, we know that flori-
gen is the protein FLOWERING LOCUS T
(FT). It initiates flowering by moving from
leaves, where it is produced in vascular tis-
sues in response to environmental stimulus,
into the phloem stream and up the stem to
the vegetative apex, where it reprograms the
shoot apical meristem, causing the plant to
produce flowers rather than leaves ( 4). Now,
however, plant biologists can shift from just
identifying “what” is moving between plant
cells to understanding how the plant con-
trols the movements of these molecules and
how it “chooses” to move a specific protein
to orchestrate complex transformations, such
as from stem cells into organs, from epider-
mal cells into hair cells, and from vegetative
development into reproductive development.
This is not the only window that Xu et al.
have helped open into the “how” of selective
A Window on the Sophistication
of Plants
PLANT SCIENCE
Richard A. Jorgensen
Chaperonin proteins are necessary for
the movement of functional transcription
factor KN1 between cells.
Langebio, Cinvestav Irapuato, Guanajuato 36821 Mexico.
E-mail: rajorgensen@langebio.cinvestav.mx
targets NOTCH receptors for degradation by
the proteasome, the protein degradation sys-
tem of the cell, and this could be considered
an inhibitory regulatory system of NOTCH1.
Surprisingly, these FBXW7 mutations were
also inactivating. One would have expected
activating mutations in this inhibitory down-
stream pathway, assuming that NOTCH1 is
the target. Hence, this requires more detailed
investigation. Relating mutations to phe-
notypic consequences is a challenge for all
potential cancer genes identified by these
high-throughput methods. Even nonsynony-
mous mutations in established cancer genes
may not always be driving, unless supported
by functional testing in relevant models.
An issue even more relevant to clinical
application is that identification of a can-
cer gene does not mean that it is druggable.
As Agrawal et al. note, proteins encoded by
oncogenes (genes that, when activated, cause
a normal cell to become cancerous) are most
suited for treatments because inhibitory drugs
will result in a reduction of cellular prolifera-
tion. However, in the case of inactivated or lost
tumor suppressor proteins, inhibitors are of no
use, and reactivation is complex or impossible.
Instead, one has to make use of the principal
of synthetic lethality—finding another path-
way that compensates the effect of, for exam-
ple, NOTCH pathway inactivation ( 14). Can-
cer-associated signaling pathways are often
so critical for cellular homeostasis that there
are mechanisms of redundancy to compen-
sate inactivation, and these take over in tumor
cells. Blocking this compensating pathway is
then lethal for tumor cells, whereas in normal
cells this has less effect as both pathways are
active. This principle of synthetic lethality is
a highly successful strategy, as shown by the
application of poly(ADP-ribose) polymerase
inhibitors in BRCA1- and BRCA2-deficient
breast cancers ( 15). However, the presence of
such compensating pathways and their syn-
thetic lethal character need to be identified.
Hence, there is more work to be done, but the
studies by Agrawal et al. and Stransky et al.
indicate that there are more candidate can-
cer genes to be identified and we should keep
searching for them.
References
1. F. Kamangar, G. M. Dores, W. F. Anderson, J. Clin. Oncol.
24, 2137 (2006).
2. C. R. Leemans, B. J. M. Braakhuis, R. H. Brakenhoff, Nat.
Rev. Cancer 11, 9 (2011).
3. J. A. Bonner et al., N. Engl. J. Med. 354, 567 (2006).
4. N. Agrawal et al., Science 333, 1154 (2011); 10.1126/
science.1206923.
5. N. Stransky et al., Science 333, 1157 (2011); 10.1126/
science.1208130.
6. A. Klinakis et al., Nature 473, 230 (2011).
7. G. P. Dotto, Oncogene 27, 5115 (2008).
8. M. Nicolas et al., Nat. Genet. 33, 416 (2003).
9. A. Proweller et al., Cancer Res. 66, 7438 (2006).
10. P. Ranganathan, K. L. Weaver, A. J. Capobianco, Nat. Rev.
Cancer 11, 338 (2011).
11. E. J. Ezratty et al., Cell 145, 1129 (2011).
12. T. Sjöblom et al., Science 314, 268 (2006).
13. L. D. Wood et al., Science 318, 1108 (2007).
14. J. D. Iglehart, D. P. Silver, N. Engl. J. Med. 361, 189 (2009).
15. P. C. Fong et al., N. Engl. J. Med. 361, 123 (2009).
10.1126/science.1210986
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1104
PERSPECTIVES
trafficking. Plants not only move functional
proteins; they also traffic full-length, func-
tional mRNA molecules, both between cells
and via the phloem long-distance transport
stream ( 1). It will be fascinating to learn how
plants accomplish this, whether by means of
RNA chaperones or other means. In addition,
plants traffic short interfering RNA (siRNA)
molecules that mediate RNA interference in
the cytoplasm and epigenetic alteration of
chromatin states; this movement occurs both
between cells and throughout the plant via the
phloem stream ( 5). Clearly, the “how” ques-
tions are likely to be extensive, because the
“what” targets are so diverse. In addition,
plant biologists must continue to uncover the
full repertoire of plant processes that depend
on selective trafficking.
This brings us to the most difficult, but
most important, question: Why? What evo-
lutionary advantage have plants derived from
this remarkable ability? And in what ways
might this capability make them unique and
different from animals and fungi? For exam-
ple, plants may look placid and unintelligent,
but perhaps we need to reconsider the sophis-
tication of their information processing sys-
tems, especially given the diverse types of
plant informational proteins and RNA mole-
cules that can move about in a regulated fash-
ion, respond to the environment, and enter the
nuclei of distant cells to orchestrate intricate
symphonies of responses (as in the case of
florigen). The computational capabilities of
such a system can only be imagined, but they
are potentially enormous.
Researchers have already shown that
flowering plants can “remember” newly
induced states for long periods by altering
the form of chromatin states at particular
genetic loci, through self-reinforcing pro-
tein modifications, or through stable meta-
bolic or developmental states—and con-
ceivably there might be thousands of such
states. This suggests that we should at least
consider the possibility that flowering plants
possess sophisticated abilities to integrate,
process, store, and retrieve information—
and that they possess an advanced mode of
intelligence based on novel mechanisms that
we are only beginning to understand, and
that this intelligence is quite different from
animal or artificial intelligence. Thus, the
real importance of understanding selective
trafficking mechanisms is highlighted by the
question: What is the unique nature of intel-
ligence in plants?
References
1. W. J. Lucas, B. K. Ham, J. Y. Kim, Trends Cell Biol. 19, 495
(2009).
2. X. M. Xu et al., Science 333, 1141 (2011).
3. J. A. D. Zeevaart, Science 137, 723 (1962).
4. F. Turck, F. Fornara, G. Coupland, Annu. Rev. Plant Biol.
59, 573 (2008).
5. C. A. Brosnan et al., Proc. Natl. Acad. Sci. U.S.A. 104,
14741 (2007).
T
he properties of polymers depend
not only on their composition—the
types of monomers used to synthe-
size them—but also on their topology. Dif-
ferences in how polymer chains are con-
nected within the molecule can lead to
materials properties that vary for polymers
made from the same monomer. For exam-
ple, high-density polyethylene made with
mostly linear chains and few branches is
stiff and strong, and can be used for pipes.
When made with many branching chains,
it is more flexible and can be used in shop-
ping bags. More complex structures can be
created with copolymers containing two
or more monomers that allow variations in
both composition and chemical functional-
ity. In this way, advanced materials used in
health and beauty products, optoelectronic
and microelectronic materials, and struc-
tural applications have been developed ( 1,
2). The latest challenge is to combine all of
these elements—composition, topology, and
functionality—into one material, and to do
so in ways that reduce the complexity and
cost of synthesis.
There are many ways in which two or
more different types of monomer units,
which can be chosen from thousands of
available monomers (3, 4), can be distrib-
uted along a polymer chain. Monomers can
be distributed statistically, grouped into
segments or blocks, repeated periodically,
or changed progressively in composition
to create a gradient. Topologies of poly-
mers include (but are not limited to) linear,
branched, hyperbranched, and even cyclic
chains, as well as structures that resemble
combs, stars, and lattice networks (see the
figure, panel A). Various functional groups
can be distributed randomly along the chain,
or they can be placed in specific locations;
for example, in telechelic polymers, reac-
tive groups are placed at the chain ends.
The functional groups can allow further
reactions to occur—they can be polymeriz-
able groups, so the chain acts as a “macro”
monomer, or they can react with small mole-
Architecturally Complex Polymers
with Controlled Heterogeneity
CHEMISTRY
Krzysztof Matyjaszewski
New materials result by creating polymers that
combine the many ways in which chains are
connected and composition and functionality
are varied.
Department of Chemistry, Carnegie Mellon University, Pitts-
burgh, PA 15213, USA. E-mail: km3b@andrew.cmu.edu
ER
Origin cell
(mesophyll)
KN1
CELL WALL PLASMA
MEMBRANE
Destination cell
(epidermis)
PLASMODESMAL
CHANNEL
KN1 unfolding
KN1
CCT
Unfolding story. Plants move regulatory proteins
from cell to cell, often over long distances, to con-
trol a wide range of activities, including growth and
flowering. The mechanisms, however, have been
unclear. Now, Xu et al. ( 2) show that the movement
of the maize transcriptional factor KN1 through
channels known as plasmodema requires the pres-
ence of a CCT in the destination cell. This suggests
that KN1 unfolds to gain passage, and is then
refolded in the destination cell by CCT.
10.1126/science.1211194
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1105
PERSPECTIVES
cules. When multiple architectural elements
are combined into one material (see the fig-
ure, panel B), complex macromolecules
can form, such as molecular brushes with
grafted side chains in the form of gradient
or block copolymers, pom-pom–like struc-
tures, or stars with mixed (“mikto”) arms
with different functionalities.
A recently reported example combined
block and gradient copolymers in a circular
shape ( 5). The synthesis of cyclic polymers
is among the most difficult tasks for poly-
mer chemists, as the ends of long chains
must somehow be brought together before
they react with other chains ( 6). Reactions
of various telechelic materials can proceed
intramolecularly to form a cyclic product
but also compete with intermolecular reac-
tions that lead to longer polymer chains. To
prepare sufficiently pure cyclic polymers, it
is necessary to work under extreme dilution
conditions or to use special templating pro-
cesses to avoid intermolecular reactions ( 7).
Alternatively, ring expansion can also lead
to cyclic polymers, and such processes can
also proceed catalytically ( 8).
The composition of the reported cyclic
copolymer illustrates the challenges of cre-
ating both block and gradient copolymers
( 5). Pure block copolymers can be prepared
in several ways, including coupling of two
already synthesized blocks (a quite difficult
process), using initiators that can grow two
segments by different mechanisms ( 9), and
switching from one monomer to another
monomer during chain growth. Such reac-
tions require very high selectivities for add-
ing the correct monomer. Synthesis of a typ-
ical symmetrical diblock copolymer with a
molar mass of ~200,000 g/mol consisting of
~1000 units of each monomer must proceed
with selectivity of a chain growth over any
chain-breaking reactions exceeding 99.9%
for each block, and requires clever chemis-
try to switch efficiently from one monomer
to the other.
However, these efforts are worthwhile
in that they can lead to new nanostructured
materials. Most polymers cannot mix with
each other but instead separate like oil and
water, so block copolymers often adopt
morphologies in which one polymer forms
nanoscale components within the other
polymer, which acts as a matrix. Different
morphologies can result by changing how
the blocks are connected, as well as the vol-
ume fraction and the relative solubility of the
different blocks. Typically, diblock copoly-
mers separate into spherical, cylindrical, and
lamellar morphology, but triblock copoly-
mers can lead to more than 30 different and
more complex structures ( 10). Block copo-
lymers are used as thermoplastic elastomers,
dispersants, surfactants, additives, and many
other advanced-technology products ( 1).
Gradient copolymers are in a sense
easier to prepare than pure block copoly-
mers because the synthesis can be switched
gradually from one monomer to the other.
They can be considered to be “fuzzy” block
copolymers; the phase boundary in nano-
structured morphologies is more diffuse
than that found in pure block copolymers
( 11). However, they have many new excit-
ing properties, including an unusually broad
temperature range for vitrification and sev-
eral energy-relaxation processes that con-
fer excellent vibration- and noise-damp-
ing properties. Gradient copolymers have
much higher critical micelle concentration
(the maximum concentration of free surfac-
tants in solution), which leads to very good
surfactant properties. Thus, these gradient
copolymers, which can be viewed as a “poor
man’s” block copolymer, may be even more
interesting than pure block copolymers.
Precisely controlled size distribution of
polymer chains—that is, a low-molecular
weight distribution (MWD)—is usually the
target for most polymer synthetic efforts,
but gradient copolymers can sometimes out-
perform pure block copolymers despite a
broader MWD. For example, they can lead
to materials with new morphologies and
much larger nanodomains than block copo-
lymers with very low dispersity ( 12– 14).
Thus, control of MWD and designed shape
of MWD—which is accessible by using
exchange reactions between macromolecu-
lar growing species of different activity ( 15),
some deliberately chosen chain-breaking
reactions, or even slow initiation—can open
new avenues to nanostructured materials.
The cyclic polymer with both blocky
and gradient structures and relatively broad
MWD reported in ( 5) illustrates how several
architectural features can be incorporated
into one macromolecule. It will be interest-
ing to precisely characterize such polymers
and correlate their properties with molecu-
lar structural parameters and to compare
cyclic and linear analogs. New polymers
with complex architecture and synergisti-
cally combined controlled heterogeneity,
including MWD, topology, functionality,
and composition, can lead to new advanced
nanostructured functional materials.
References and Notes
1. K. Matyjaszewski, Y. Gnanou, L. Leibler, Eds., Macro-
molecular Engineering: From Precise Macromolecular
Synthesis to Macroscopic Materials Properties and Appli-
cations (Wiley-VCH, Weinheim, Germany, 2007).
2. K. Matyjaszewski, N. V. Tsarevsky, Nat. Chem. 1, 276
(2009).
3. N. Hadjichristidis, H. Iatrou, M. Pitsikalis, J. Mays, Prog.
Polym. Sci. 31, 1068 (2006).
4. N. Hadjichristidis, M. Pitsikalis, S. Pispas, H. Iatrou,
Chem. Rev. 101, 3747 (2001).
5. E. J. Shin et al., Angew. Chem. Int. Ed. 50, 6388 (2011).
6. J. A. Semlyen, Pure Appl. Chem. 53, 1797 (1981).
7. M. Glassner, J. P. Blinco, C. Barner-Kowollik, Macromol.
Rapid Commun. 32, 724 (2011).
8. C. W. Bielawski, D. Benitez, R. H. Grubbs, Science 297,
2041 (2002).
9. Y. Yagci, M. A. Tasdelen, Prog. Polym. Sci. 31, 1133
(2006).
10. F. S. Bates, G. H. Fredrickson, Phys. Today 52, 32 (1999).
11. K. Matyjaszewski, M. J. Ziegler, S. V. Arehart, D. Greszta,
T. Pakula, J. Phys. Org. Chem. 13, 775 (2000).
12. J. Listak et al., Macromolecules 41, 5919 (2008).
13. N. A. Lynd et al., Prog. Polym. Sci. 33, 875 (2008).
14. S. Li, R. A. Register, B. G. Landes, P. D. Hustad, J. D.
Weinhold, Macromolecules 43, 4761 (2010).
15. N. V. Tsarevsky, K. Matyjaszewski, Chem. Rev. 107, 2270
(2007).
16. I thank A. M. Elsen for helpful discussion. Supported by
NSF grant DMR 09-69301.
10.1126/science.1209660
Composition
Composition, topology, and functionality
A B
Topology
Functionality
Homopolymer
Block copolymer
Linear
Cyclic Star
Graft
Macromonomer End functional
Telechelic Side functional Multifunctional
Network Hyperbranched
Periodic Gradient
Gradient brush
Side chain,
block copolymer brush
Cyclic gradient
or block copolymer
Pom-pom,
grafted block copolymer
Multifunctional
(miktoarm) star
Statistical
Macromolecular architectural guide. The three main aspects of polymer architecture—composition, topol-
ogy, and function—are shown separately in (A) and in combination in (B).
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1106
ESSAY
Astronomical Perspectives
for Young Children
SPORE* SERIES WINNER
Carolina J. Ödman-Govender
1,2,3
and Deirdre Kelleghan
4

Crowd-sourced efforts and online educational
resources are able to multiply the impact
of astronomy outreach and education.
M
any of us have experienced a
moment when, mesmerized by the
beauty of a night sky, we reflected
upon life, here and elsewhere. The realization
of our place in the universe is humbling, but it
is also inspiring.
Universe Awareness (UNAWE) is an
astronomy education and outreach program
targeting young children, from 4 to 10 years
old, in underprivileged environments. At this
age, children are developing their cultural
landscape and value systems, which serve
as references as they grow and learn. The
rationale behind UNAWE is that the unique
perspective brought by astronomy, e.g., see-
ing Earth from space, without real borders
between countries, helps young children
develop the skills and values that will help
them create a better future for themselves.
Since 2005, UNAWE has grown to involve
more than 500 volunteers, organizations, and
governments in 40 countries. UNAWE grew
most during the International Year of Astron-
omy (IYA) in 2009 and is now an important
ingredient of the International Astronomical
Union’s decadal development strategy (1).
Educational resources, generously con-
tributed by the network of volunteers to
UNAWE, have been collected and published
(www.unawe.org/). Many UNAWE mem-
bers have contributed translations, which
are then featured on the site. Quality con-
trol comes through an automatic peer adop-
tion process. The resources are checked for
scientific accuracy by qualified astronomers
before they are posted online, where the
most useful resources naturally become the
most used resources. One of the most suc-
cessful activities UNAWE features is the
“Deadly Moons” workshop.
Deadly Moons is a 1-hour interactive
workshop that teaches students about our
Moon and the other exotic moons in our solar
system. Deadly Moons is immediately made
relevant by its title, as many children in Ire-
land use the street phrase “That’s deadly!”
in everyday life about things that impress
them totally. On UNAWE, the local nature
of this expression is described, and educators
in other countries either teach their students
about the Irish expression or use the equiv-
alent colloquial expression in their own lan-
guage. Students learn about moon phases,
active moons, quiet moons, Galilean moons,
moon exploration, and moon terminology.
When they have absorbed enough informa-
tion about these moons they are invited to
vote in a vocal way to indicate which moons
they like the best. The group then produces
drawings of their favorite moons.
Thousands of observational drawings
have been produced by participants of
Deadly Moons (see the figures). Some of
these drawings were exhibited as part of
the IYA, both at the opening celebrations in
Paris and at the international astronomical
sketching exhibition in Ireland. Addition-
ally, many schools and libraries have held
exhibitions of the work produced for the
local communities to see.
Each year, an increasing number of
schools in Ireland have used Deadly Moons
toward their Discover Primary Science
award. This award acknowledges the efforts
of primary school children and teachers who
have shown an increased knowledge of sci-
ence and math. Evaluations revealed reten-
tion of information; many of the children
were unaware of other moons before the
workshop but could remember the names of
several after their participation.
Deadly Moons has spawned several other
successful workshops using the same format,
including The Suns Massive, about Sun ter-
minology and robotic solar exploration, and
Rapid Rockets and Wicked Robots, which is
a history, through drawing, of space explo-
ration. Additionally, Deadly Moons now has
a follow-up moon drawing workshop that
uses richly colored images from the U.S.
National Aeronautics and Space Adminis-
tration (NASA) lunar explorers. The new
workshop was the inspiration behind the
paintings that some Irish children produced
for OPTICKS 2011, a live science and art
audiovisual performance using radio waves
between Earth and the Moon (www.ustream.
tv/recorded/13911581).
In general, UNAWE activities come from
places where financial means are often scarce
and are therefore full of good ideas at low cost
or no cost (2). The science is not shrouded
in high-tech imagery that may distract chil-
dren, and the hands-on nature of the activities
ensures real learning. They are also locally
relevant, as ideas are contributed by educators
close to the children, the resources convey
the right message, children identify with the
characters featured in the stories, and they are
familiar with the real-life examples shown.
UNAWE is not just about sharing
resources. Astronomical events present the
perfect opportunity to connect children and
educators worldwide. Two classrooms, sepa-
rated by the Atlantic Ocean, are preparing for
a lunar eclipse. They understand that Earth is
round and that while it is day in some places,
it is night in others. In rural South Africa, it
is early morning. Pupils have spent an excit-
ing sleepover at the school with an astronomy
workshop and star-gazing the evening before.
Now, at 4 a.m., they are awake and ready to C
R
E
D
I
T
:

D
R
A
W
I
N
G

B
Y

E
D
E
N

M
U
N
R
O
E
;

P
H
O
T
O

D
E
I
R
D
R
E

K
E
L
L
E
G
H
A
N
Illuminating understanding. A drawing of Mimas,
one of Saturn’s moons, by Eden Munroe, age 12, of
Ashbourne Irish Girl Guides, Dublin, during Discover
Science and Engineering’s Science Week 2010.
1
Leiden University, NL-2300 Leiden, Netherlands.
2
South
African Astronomical Observatory, Observatory 7935,
South Africa.
3
African Institute for Mathematical Sciences,
Muizenberg 7945, South Africa. E-mail: carolina@odman.
org (C.J.O.-G.).
4
Independent educator, UNAWE Ireland.
E-mail: skysketcher@gmail.com (D.K.)
*SPORE, Science Prize for Online Resources in Education;
www.sciencemag.org/site/special/spore/
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1107
ESSAY
view our eclipsed Moon. Across the globe in
Bermuda, it is evening after an unusual school
day in which UNAWE volunteers have come
to talk about the Sun, the stars, and the Moon.
As the eclipse draws near, an Internet con-
nection is established between the schools.
“Good morning!” wish the South African
learners. “Good evening!” retort the Bermu-
dans, and the children are left to smile at the
time difference and chat among themselves.
Children strike up an immediate friendship
and start realizing the differences in their
lives: “Here, water is a problem, we do not
always have enough,” says one South African
pupil. “Water is a problem here too, but we
are trying to keep our houses dry” says his
Bermudan friend, and both laugh at the seem-
ingly absurd situation. Suddenly, one child
runs into the Bermudan classroom. “The
Moon has gone red!” she shouts. All abandon
the computers and run outside, in both Ber-
muda and South Africa.
Events like this one
would not be possible
without a strong net-
work of dedicated volun-
teers—the true strength
of UNAWE. The idea
is simple and powerful:
Teachers, students, par-
ents, children, and other
people involved in the
education of the young-
est generation under-
stand easily how astron-
omy inspires children
and take ownership of the
program. UNAWE is a
bottom-up program rely-
ing on local ownership of
its philosophy. Living in a poor rural area of
the world with a privileged access to the night
sky but no one to explain it is a
different challenge from grow-
ing up in a city where teach-
ers struggle to impart concepts
only seen through television
screens or advertising. Rural
settings are rich in natural
beauty. Urban environments
may have more infrastructure,
but light pollution is a real hur-
dle for astronomy. Only those
in touch with the children
know what specific conditions
they are learning and living in.
UNAWE also connects its
volunteers through regular
workshops, enabling them to
exchange ideas and best prac-
tices. At one such meeting,
primary school teachers from
Spain solved the problem of
an Indian educational nongov-
ernmental organization try-
ing to explain why days grow
shorter or longer at different
times of the year with a model
easy enough for a 6-year-old
to understand. Another teacher
had the simple idea of cover-
ing a ball with aluminum foil
to demonstrate that the Moon
reflects sunlight and does not
emit its own. Venezuelan col-
leagues showed how to make a
spiral galaxy magically appear
by using sand and water in a
flat-bottomed glass bowl on
a projector. Such exchanges
help the teachers build confi-
dence in their ability to teach
scientific topics.
Until 2010, the coordination of UNAWE
was funded by a seed grant from the Dutch
Ministry of Education, Science, and Cul-
ture. Since then, UNAWE has been awarded
a European grant of almost €2 million
(U.S. $2.87 million), allocated to create
and implement a professional develop-
ment program for primary school teachers
in six countries. The resource Web site is
currently being redeveloped as part of the
European Union–funded program. Profes-
sional expertise will be involved in evaluat-
ing the resources, and although the site will
continue to welcome educational resources
from all our partners, it will also offer new
resources, developed professionally.
Children around the world are different,
and yet they share curiosity and an appetite
for learning and for understanding the world
around them and their place in it. UNAWE
volunteers bring the unique perspective of
astronomy to young children, giving them
a chance to appreciate the scale and beauty
of the universe, empowering them to think
independently, and bringing them closer to
each other.
References and Notes
1. International Astronomical Union, Astronomy for the
Developing World: Strategic Plan 2010–2020; http://iau.
org/static/education/strategicplan_091001.pdf.
2. See also http://www.arvindguptatoys.com/ for examples
of excellent hands-on low- or no-cost science toys.
3. This work would not have been possible without the sup-
port of the Ministry of Education, Science, and Culture
of the Netherlands and the visionary initiative of George
Miley, founder of UNAWE. Most of all, C.J.O.-G. is grate-
ful to the network of 500+ volunteers worldwide whose
generous efforts and unique contributions enabled
UNAWE to grow into a reputable community of practice.
D.K. thanks the Dublin Institute of Advanced Studies,
Blackrock Castle Observatory, and Dublin City Libraries.
About the Authors
Carolina Ödman-Govender
studied physics engineering at
the Swiss Federal Institute of
Technology in Lausanne before
completing a Ph.D. in theoreti-
cal cosmology at the Univer-
sity of Cambridge in the United
Kingdom. She has consulted for UNESCO and tutored at
the African Institute for Mathematical Sciences (AIMS)
in South Africa. After working as a post-doc, she joined
Leiden Observatory in the Netherlands where she devel-
oped “Universe Awareness.” In 2010, she became a
research fellow at the South African Astronomical Obser-
vatory and at AIMS. In 2011, she became the director of
academic development of the AIMS Next Einstein Initia-
tive. She remains a member of the Universe Awareness
Project Board and of the AIMS Cosmology Group.
Deirdre Kelleghan lives in Ire-
land where she is both an ama-
teur astronomer and an artist
with a passion for experimenta-
tion in lunar, solar, and deep-sky
sketching. She is an avid informal
educator who develops drawing
workshops and talks for children
and adults. As an amateur astron-
omer, Deirdre is vice chair of the
Irish Federation of Astronomical Societies, Irish repre-
sentative of Astronomers without Borders, and a mem-
ber of Jet Propulsion Laboratory–NASA’s Saturn observa-
tion campaign. She is coauthor of Sketching the Moon:
An Astronomical Artists’ Guide (R. Handy, D. Kelleghan,
T. McCagne, E. Rix, S. Russell, Springer, New York, forth-
coming 2011).
Deadly Moons Workshop at Wicklow Arts Festival.
10.1126/science.1196982 C
R
E
D
I
T
S
:

(
T
O
P

T
O

B
O
T
T
O
M
)

D
E
I
R
D
R
E

K
E
L
L
E
G
H
A
N
;

(
F
O
R

C
.
O
.
-
G
.
)

H
A
N
S

H
O
R
D
I
J
K
;

(
F
O
R

D
.
K
.

W
I
T
H

T
E
L
E
S
C
O
P
E
)

B
E
R
N
A
R
D

K
E
L
L
E
G
H
A
N
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1108
AAASNEWS&NOTES EDITED BY EDWARD W. LEMPINEN
She suffered a stroke at the age of 42, and
for nearly a decade after, she was unable to
move or communicate beyond shaking her
head. Such patients usually experience little
improvement, but today, thanks to a neural
implant that links her brain to a computer,
she has used her mind to control a keyboard
and move a robotic arm.
Leigh R. Hochberg, a neuroengineer and
doctor who specializes in brain-computer
interface systems, told her story at a AAAS
Capitol Hill briefing to illustrate how advances
in neuroscience may transform once-futuris-
tic ideas into better lives for soldiers, accident
victims, and others. “We’re hoping to develop
technologies that will restore the ability to
communicate and restore the ability to move,”
Hochberg said.
The briefing focused on the military
applications of neuroscience, and the pre-
sentations reflected the sense that today’s
advances could have far broader future
impact. But with the benefits of enhanced
brain function will come critical social and
ethical issues.
“These issues are emerging now and will
only get more prominent over time,” said
moderator Alan I. Leshner, the chief execu-
tive officer of AAAS and executive publisher
of Science.
About 100 congressional staff and
others attended the 90-min briefing, held
26 July in cooperation with the House Armed
Services Committee and with financial sup-
port from The Dana Foundation. The AAAS
Office of Government Relations has scheduled
a second neuroscience briefing—on possible
links between cell phones and brain tumors—
for 7 September. The final briefing in
the series, on traumatic brain injury,
will be held in October.
In areas such as “shell shock” and
brainwashing, neuroscience has
long been a military interest. Today,
the military is exploring new realms
of neuroscience to make soldiers
safer and more effective. Jonathan
D. Moreno, an historian and ethi-
cist at the University of Pennsylvania,
detailed the Pentagon’s 2011 neuroscience
investments: U.S. Army, $55 million; Navy,
$34 million; Air Force, $24 million; and the
Defense Advanced Research Projects Agency
(DARPA), more than $240 million.
Martha J. Farah, director of Penn’s Cen-
ter for Neuroscience & Society, said a key
military interest is cognitive enhancement.
Research is assessing newer drugs to keep sol-
diers alert or neutralize the effects of wartime
fatigue, stress, and trauma.
Another area of interest, Farah said, is non-
invasive brain stimulation in which a weak
current is directed into a targeted area of the
brain. Such techniques have been found to
affect attention, learning, memory, decision-
making, and mood, and some believe that the
effects could extend to visual perception, reac-
tion time, and even social behavior.
Given the legion of war casualties from
Iraq and Afghanistan, research into prosthet-
ics is acutely important. Hochberg, who has
appointments at Brown and Harvard universi-
ties and the Providence VA Medical Center, is
focused on technology that may allow people
with tetraplegia to move robotic limbs.
When a spinal injury
leaves someone paralyzed, he
explained, the brain still generates a signal
to move the arm, but the signal never arrives.
At scientific meetings, he has described early
research in which tiny arrays are implanted
in the brain’s motor cortex; the array picks
up that signal, conveys it through a thin wire
into a computer, and then to external devices.
Almost inevitably, however, neuroscience
research raises social and ethical questions.
Newer drugs developed for sleep disorders
and attention deficit hyperactivity disorder
are increasingly misused now by travelers,
students, and others battling fatigue, Farah
reported. Meanwhile, said Moreno, military
research has developed a “robo-rat” that can
be controlled through electrodes in its brain—
and other creatures could be used as living
robots too, he said.
For now, though, the potential bene-
fits provide powerful motivation for further
research. Hochberg said the stroke victim
with whom he’s worked sent a video mes-
sage last year to a Society for Neuroscience
conference. Using an on-screen keyboard,
her intentions guided the cursor, letter by
letter, to type a simple message:
“There is hope.”
SCIENCE AND SOCIETY
Advances in Neuroscience Raise
Medical Hopes, Social Questions
C
R
E
D
I
T
:

P
H
O
T
O

C
O
U
R
T
E
S
Y

O
F

D
E
K
A

R
E
S
E
A
R
C
H

&

D
E
V
E
L
O
P
M
E
N
T
AAAS and 139 other scientific societies and
universities urged U.S. lawmakers to refrain
from eliminating or substantially defunding
specific areas of science as they consider the
fiscal year 2012 budget.
In an 11 July letter sent to the House Appropria-
tions Committee, the societies acknowledged the
nation’s fiscal challenges, but strongly opposed
the elimination of entire interdisciplinary pro-
grams such as the National Science Foundation
Directorate for Social, Behavioral, and Economic
Sciences. Proposed legislation that would under-
mine the scientific peer-review process, they cau-
tioned, could set “a dangerous precedent that, in
the end, will inhibit scientific progress and our
international competitiveness.” (Read the full
letter at www.aaas.org/go/nsf.)
The bipartisan agreement to raise the U.S.
debt ceiling could have a negative impact on
research and development investment, said
Joanne Carney, director of AAAS’s Office of
Government Relations. In a statement after
President Barack Obama and congressional
leaders reached agreement on 31 July, she
said long-term spending caps, when combined
with inflation, “could result in a real decrease
in R&D investment for the foreseeable future.”
—Becky Ham
BUDGET POLICY
Science Societies Warn of Risks to R&D Funding
Better than bionic. DARPA , the VA, and
others are exploring whether brain power
can drive a new generation of prosthetics
like the DEKA robotic arm.
n
,
nd
n
i-
nia,
oscience
n; Navy,
; and the
with tetraplegia to move robotic limbs.
When a spinal injury
leaves someone paralyzed he
Better than bionic. DARPA , the VA, and
others are exploring whether brain power
can drive a new generation of prosthetics
like the DEKA robotic arm.
Published by AAAS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Mitochondria and the
Autophagy–Inflammation–Cell
Death Axis in Organismal Aging
Douglas R. Green,
1
* Lorenzo Galluzzi,
2,3,4
Guido Kroemer
2,5,6,7,8
*
Alterations of mitochondrial functions are linked to multiple degenerative or acute diseases.
As mitochondria age in our cells, they become progressively inefficient and potentially toxic,
and acute damage can trigger the permeabilization of mitochondrial membranes to initiate apoptosis
or necrosis. Moreover, mitochondria have an important role in pro-inflammatory signaling.
Autophagic turnover of cellular constituents, be it general or specific for mitochondria (mitophagy),
eliminates dysfunctional or damaged mitochondria, thus counteracting degeneration, dampening
inflammation, and preventing unwarranted cell loss. Decreased expression of genes that regulate
autophagy or mitophagy can cause degenerative diseases in which deficient quality control results in
inflammation and the death of cell populations. Thus, a combination of mitochondrial dysfunction
and insufficient autophagy may contribute to multiple aging-associated pathologies.
Aging seems to be the only available way
to live a long life.
Daniel-François-Esprit Auber (1)
T
here is a simple arithmetic of multicellu-
lar life that holds deep insights into health
and disease—the rate of cell generation
minus the rate of cell loss determines the growth
or degeneration of a tissue. Somatic cells fall into
two classes: those that are constantly renewed
from proliferating stem cells, such as epithelial
cells and leukocytes, and those that are rarely
renewed after birth, such as neurons and cardio-
myocytes. Although the former are programmed
to undergo cell death and are replaced through-
out life, the latter must endure until we expire.
Excessive cell death in postmitotic tissues pre-
cipitates degenerative states (2, 3), whereas the
failure to execute timely programmed death in
renovating tissues contributes to hyperplasia and
cancer (4). As we age, cells accumulate errors
in their nuclear and mitochondrial genomes and
damage to organelles and macromolecules, and
therefore quality control mechanisms determine
the consequences of such accumulation for the
cell and the organism. Recent evidence indicates
that the interplay between mitochondria and au-
tophagy links aging to health or disease.
Mitochondria are the evolutionary relics of
aerobic bacteria that invaded the proto-eukaryotic
cell about a billion years ago. As such, they
have a separate genome and provide the oxygen
consumption–driven synthesis of adenosine tri-
phosphate (ATP) (via oxidative phosphorylation,
OXPHOS). However, as a side product of nor-
mal respiration, mitochondria produce reactive
oxygen species (ROS) that must be detoxified.
Moreover, as they replicate, mitochondrial ge-
nomes accumulate mutations that eventually
compromise the efficiency of OXPHOS (5). Mito-
chondrial deficiency, excessive ROS, or both
appear to be driving forces in aging because
they reduce cellular fitness, inflict damage to
other organelles, and/or cause mutations of the
nuclear genome (5). Beyond their roles in the
chronic process of cellular and organismal aging,
mitochondria also mediate acute cell death. In
many instances, developmental, homeostatic, and
pathological cell death involves a critical step
in which mitochondria release proteins that trig-
ger the self-destructive enzymatic cascade that
causes apoptosis (6).
Macroautophagy (herein “autophagy”) occurs
by the formation of autophagosomes, double-
membraned vesicles that sequester organelles,
proteins, or portions of the cytoplasm, which then
fuse with lysosomes. As a result of this process,
the sequestered contents are degraded by lyso-
somal enzymes and recycled as a source of en-
ergy (7). Autophagy may occur either as a general
phenomenon, for instance, when cells lack nu-
trients and mobilize their energy reserves, or it
can specifically target distinct cellular structures
such as damaged mitochondria (“mitophagy”).
Autophagy has cardinal roles in the cellular adap-
tation to stress, in innate immune responses, and
as a quality-control mechanism (8). Thus, autoph-
agy represents an essential cytoprotective path-
way and a potential anti-aging mechanism (9).
Inflammation is a critical organismal response
to tissue damage and infection, in which various
secreted mediators such as cytokines, chemokines,
and eicosanoids coordinate defense and repair,
but its long-term consequences are a major cause
of disease associated with aging (10). Both au-
tophagy and apoptotic cell death tend to dampen
inflammation, whereas necrotic cell death can
promote it (11, 12). A key mechanism of inflam-
mation is the activation of the “inflammasome,”
a molecular platform that activates caspase-1,
which in turn processes the inflammatory cyto-
kines interleukin-1 and interleukin-18 and facil-
itates secretion of these and other inflammatory
mediators. Mitochondria play a central role in
the initiation of inflammasomes and other in-
flammatory pathways (discussed below) and as
such integrate autophagy, cell death, and inflam-
mation (Fig. 1). Here, we discuss the functions
of these organelles in aging as a consequence of
the interplay of these three processes.
Damaging Mitochondrial Signals
in Inflammation and Cell Death
There are at least three major mechanisms through
which mitochondria can damage or kill their
host, namely through production of ROS or pro-
inflammatory signals or through mitochondrial
membrane permeabilization.
About 1 to 3% of the molecular oxygen is in-
completely reduced during OXPHOS, hence gen-
erating the ROS superoxide anion. This fraction
REVIEW
1
Department of Immunology, St. Jude Children’s Research
Hospital, Memphis, TN 38105, USA.
2
INSERM, U848, F-94805
Villejuif, France.
3
Institut Gustave Roussy, F-94805 Villejuif,
France.
4
Université Paris-Sud, Paris 11, F-94805 Villejuif, France.
5
Metabolomics Platform, Institut Gustave Roussy, F-94805
Villejuif, France.
6
Centre de Recherche des Cordeliers, F-75005
Paris, France.
7
Pôle de Biologie, Hôpital Européen Georges
Pompidou, Assistance Publique–Hôpitaux de Paris, F-75908
Paris, France.
8
Université Paris Descartes, Sorbonne Paris Cité,
F-75270 Paris, France.
*To whom correspondence should be addressed. E-mail:
douglas.green@stjude.org (D.R.G.); kroemer@orange.fr (G.K.)
Young individuals
AA
Mitochondria
Inflammation
Apoptosis Autophagy
Aging
?
Inflammation
Apoptosis Autophagy
mma
mmat
B
A
?
s A
Mitoch Mit M o ndria dria
Infl tion a
Mitochondria
Fig. 1. Interactions between mitochondria, cell
death, autophagy, and inflammation in young indi-
viduals (A) and during aging (B). In this scenario,
loss of autophagy with age leads to accumulation
of damaged mitochondria, which promotes cell
death and inflammation, both of which are oth-
erwise limited by autophagy.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1109

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

can increase when mitochondria are
compromised by damage or muta-
tion. ROS cause mutations in mito-
chondrial DNA (mtDNA), in turn
compromising OXPHOS and initiat-
ing a vicious cycle of mitochondrial
collapse. Mice in which the poly-
merase that replicates mtDNA has
been mutated to increase the frequen-
cy of mtDNA mutations exhibit signs
of premature aging, and this has
been attributed to apoptotic death
of stem cells (5).
Mitochondria participate in the
detection of infectious microorga-
nisms and cellular damage to acti-
vate innate immune responses. The
pattern recognition receptors RIG-I
and MDA-5 recognize viral RNA
in the cytosol and interact with an
adapter on the mitochondrial mem-
brane, MAVS, to trigger a signal
transduction cascade that drives the
production of type I interferon (13).
Similarly, IRGM, a protein that is
required for the control of intracel-
lular mycobacteria, translocates to
mitochondria. Through the induction
of mitochondrial fragmentation and
mitochondrial outer membrane per-
meabilization (MOMP, see below),
distinct IRGMisoforms can either ac-
tivate autophagy (similarly required
for defense against intracellular my-
cobacteria) or cell death (14).
ROS produced by mitochondria
can activate an inflammasome com-
posed of NLRP3, the adapter pro-
tein ASC, and caspase-1, a process
that occurs at the interface between
mitochondria and the endoplas-
mic reticulum (ER), the so-called
mitochondria-associated ER mem-
branes (MAMs) (15), possibly con-
comitant with the cytosolic release
of mtDNA (11). Mitochondria are
essential for NLRP3 inflammasome
activation in response to various
noninfectious agents, including uric
acid and silica, and, further, ROS-
triggering OXPHOS inhibitors also
activate this inflammasome. It is
therefore likely that accumulation
of damaged mitochondria is an im-
portant cause of inflammation.
In the intrinsic pathway leading
to apoptotic death, MOMP results
in the release of soluble mitochon-
drial intermembrane proteins that cause cell death
by apoptosis (6). MOMP results from the inter-
actions of proteins of the B cell lymphoma 2
(BCL-2) family that protect or disrupt the outer
mitochondrial membrane. Alternatively, cell death
can be triggered by the so-called mitochondrial
permeability transition (MPT), which depends on
the mitochondrial matrix protein cyclophilin D.
MPT results in the instantaneous dissipation of
the mitochondrial transmembrane potential (Dy
m
)
and cessation of OXPHOS, thus triggering rapid
necrotic cell death (16).
Mitochondria are highly dynamic organelles
that can change their morphology, fragment by
fission, or undergo fusion to gen-
erate highly interconnected tubular
networks. These dynamics are also
regulated by the BCL-2 proteins and
undoubtedly affect the respiratory,
ROS-generating, pro-inflammatory,
and lethal signaling functions of
mitochondria in ways that are only
partially understood.
General Versus
Mitochondrion-Specific Autophagy
In response to multiple forms of cel-
lular stress including the shortage of
growth factors, nutrients, or oxygen
and excessive ROS or DNA dam-
age, general autophagy is stimulated
through the coordinated activation
of several multiprotein complexes
[such as the complexes organized
around the pro-autophagic protein
kinase ULK1/2 and its upstream in-
hibitory kinase, mammalian target
of rapamycin (MTOR), as well as
the lipid kinase PIK3C3/HVPS34
and its obligatory allosteric activator
Beclin 1], a conjugation system that
transfers the ubiquitin-like proteins
ATG12 and MAP1LC3A/LC3 to
their substrates (usually ATG5 and
phosphatidylethanolamine, respec-
tively), a dynein-dependent transport
system that moves autophagosomes
to lysosomes, and a subsequent fu-
sion machinery (7). During general
autophagy, cytoplasmic cargo, includ-
ing portions of the cytosol, mitochon-
dria, and other organelles, can be
sequestered and digested. Nonethe-
less, mitochondria appear to have
a major role in general autophagy
because they supply membranes
for the biogenesis of autophago-
somes during starvation (17). Dis-
ruption of MAMs by knockdown
of MFN2 (an essential component
of the mitochondrial fusion machin-
ery) abolishes starvation-induced
autophagy in human cancer cell
lines (17), implying that MAMs
(or perhaps MAM-unrelated func-
tions of MFN2) are essential for
phagophore formation. Knockout
or inhibition of cyclophilin D also
prevents starvation-induced autoph-
agy in some cell types (18). However,
it remains a matter of controversy
[supporting online material (SOM)
text] whether MPT constitutes an essential step
toward autophagy. Notably, cyclophilin D defi-
ciency has not been associated with premature
aging or degeneration.
Low ATP production or enhanced ROS gen-
eration by mitochondria induces general autoph-
agy (7, 8). Cells also have several distinct systems
OM proteins
PARL
Parkin
p97
BNIP3
NIX
Ubiquitin
Beclin 1
Active PINK1
import and
degradation
by PARL
Inhibition
of Beclin 1 by
BCL-2/BCL-X
L Degraded
PINK1
Full length
PINK1
BCL-2/BCL-X
L
Healthy mitochondrion
Hypoxia
Erythroblastic
differentiation
Derepression
of Beclin 1
Other mechanisms?
Accumulation of PINK1
Recruitment of Parkin
Ubiquitination of OM proteins
Recruitment of p97
AMPK-mediated
phosphorylation
of ULK1
Activation of an
ATG12-ATG3-
dependent
mechanism
Mitochondrial
turnover
MPT, uncoupling,
damage
MITOPHAGY
A
B
ATP
AMP
Fig. 2. Mechanisms of mitophagy. In healthy mitochondria (A), PINK1 is ac-
tively imported by a Dy
m
-dependent mechanism and degraded by the inner
mitochondrial membrane protease PARL. BCL-2 and BCL-X
L
bind to and inhibit
Beclin 1. Different triggers can stimulate distinct pathways to mitophagy (B). The
BH3-only proteins NIX and BNIP3 are activated during erythroblast differ-
entiation and under hypoxia, respectively, and may cause mitophagy by dis-
placing Beclin 1 frominhibitory interactions with BCL-2 and BCL-X
L
. In response
to uncoupling, mitochondrial damage, or the MPT, the Dy
m
is dissipated, and
full-length PINK1 accumulates at the outer mitochondrial membrane (OM). This
allows for the recruitment of the AAA ATPase p97 and of Parkin, which together
render mitochondria a palatable substrate for the autophagic machinery. Mito-
chondrial turnover can also be mediated by accumulation of AMP, leading to the
phosphorylation of ULK1 by AMPK and possibly involving a ATG12-ATG3
conjugate that functions specifically in mitophagy.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1110
REVIEW

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

to specifically target mitochondria to mitophagy
(Fig. 2). One such system comes into action in
red blood cell precursors, which eliminate mito-
chondria by overexpressing a BCL-2 family protein,
BNIP3L/NIX. NIX associates with mitochon-
drial membranes to engage direct molecular in-
teractions with LC3 (19) and/or causes Dy
m
dissipation (20) (which can suffice to target mito-
chondria for mitophagy, see below) (21). Sim-
ilarly, BNIP3, a hypoxia-inducible BCL-2 family
protein, is suggested to trigger mitophagy by
competitively disrupting the inhibitory interaction
between BCL-2 and Beclin 1 (8). Mitochondria
can divide asymmetrically into functional prog-
eny (with a high Dy
m
), which can reintegrate the
mitochondrial network by fusion, and dysfunc-
tional organelles (with a low Dy
m
), which are
specifically destined for mitophagy (21).
The kinase PINK1 is imported into healthy
mitochondria, dependent on Dy
m
, where it is
degraded by the protease PARL (21). On the
surface of mitochondria with low Dy
m
, PINK1
accumulates, leading to the recruitment of the
ubiquitin ligase Parkin, which ubiquitinylates outer
membrane proteins including BCL-2 (22), VDAC1
(23), MFN1, and MFN2 (24). This may de-
repress the VPS34/Beclin 1 complex (22) (with
which Parkin may also directly interact) (25). By
favoring the proteasomal degradation of MFN1
and MFN2 through a mech-
anism that requires the AAA-
type ATPase p97/VCM, Parkin
suppresses mitochondrial fusion
and promotes mitophagy (24).
This can be further stimulated
by histone deacetylase 6, which
is recruited to mitochondria by
ubiquitinylated proteins and cat-
alyzes pro-autophagic cytoplas-
mic deacetylation reactions (26).
In cells that lack Parkin, Dy
m
dissipation does not induce rap-
id removal of mitochondria,
although other mechanisms al-
most certainly ensure a mito-
chondrial quality control by
mitophagy. Another, possibly
redundant, mechanism of mito-
phagy involves the activation
of adenosine monophosphate–
activated protein kinase (AMPK)
under conditions where AMP
is produced (such as when ATP
concentrations decline). AMPK
phosphorylates and thereby acti-
vates ULK1, one of the initiators
of autophagy, and this engages
autophagic removal of mito-
chondria (27). This appears to
require the formation of conju-
gates between ATG3 and ATG12,
because cells carrying a mutant
ATG3 that binds LC3 but not
ATG12 can activate autophagy
but do not remove mitochondria
(28). How the different mechanisms of mito-
phagy (Fig. 2) interact is not known, although
defects in mitophagy (for example, that caused
by loss of PINK1) can lead to increased gen-
eral autophagy (23), possibly as a compensatory
mechanism triggered by ROS-mediated damage.
Autophagy for the Avoidance of Cell Death
Although autophagy often precedes apoptosis or
necrosis, it rarely constitutes a suicidal mech-
anism, and it probably reflects failed attempts of
cells to adapt to stress (29). Usually, autophagy
has a cytoprotective rather than cytocidal func-
tion (although there are exceptions). When MOMP
is induced and postmitochondrial caspase ac-
tivation is blocked or disrupted, permeabilized
mitochondria are removed by autophagy (30, 31),
and inhibition of mitophagy can accelerate cell
death (31). Mitochondria that have not under-
gone MOMP can repopulate the mitochondrial
pool and rescue the cell (30, 32).
Induction of general autophagy before cells
are stressed with otherwise lethal stimuli can pro-
tect them against cell death. This has been cor-
related with the relative resistance of residual
mitochondria to MOMP or MPTand may be ex-
plained by the removal of mitochondria that have
a rather lowthreshold for permeabilization (mito-
chondrial “purging”) (21). Alternatively, occur-
rence of MOMP or MPT in a fraction of
mitochondria may activate repair or recycling
pathways that activate autophagic sequestration
of depolarized mitochondria (3, 21). In this case,
the intensity of the autophagic flow might set the
threshold beyond which partial MOMP becomes
lethal. However, evidence that forced induction
of mitophagy would confer cytoprotection in
these circumstances is scarce, and it is possible
that general autophagy (as opposed to specific
mitophagy) may reduce the cell’s propensity to
engage in lethal signal transduction cascades.
How can general autophagy confer cytopro-
tection and interrupt signals that lead to MOMP?
Autophagy facilitates the maintenance of high
concentrations of ATP (which reduces the pro-
pensity of mitochondria to undergo MPT) and
furnishes basic building blocks for the adaptive
synthesis of proteins, including potential apo-
ptosis inhibitors (7, 8). It may also lead to the
elimination of potentially toxic protein aggre-
gates and help limit the accumulation of ubiqui-
tinylated proteins that otherwise would inhibit
proteasome function (33). Induction of autoph-
agy affects the progression of the cell cycle
(34) (and vice versa), suggesting that it can im-
pinge on the propensity of cells to succumb to
cell cycle–dependent cell death inducers. More-
over, as a correlate of autophagy induction, anti-
apoptotic proteins like BCL-2
and the caspase-8 inhibitor FLIP
may be liberated from inhibitory
interaction with autophagic ef-
fectors (such as Beclin 1 and
ATG3, respectively) (8). Multi-
ple mechanisms exist through
which autophagy can intercept
lethal signaling before or at the
level of mitochondria (Fig. 3).
Thus, induction of autophagy
might affect the circuitry through
which lethal signals are relayed
at mitochondria.
Autophagy and Aging
Autophagy appears to decline
with age, and several key players
in the autophagic pathway (for
example, ATG5 and ATG7) show
decreased expression in the brains
of aging individuals (35). Con-
ditions that promote autophagy,
such as caloric restriction and
exercise, delay aging-associated
degeneration (3), suggesting that
autophagy counteracts the aging
process. Stimulation of autophagy
can increase the healthy life span
in multiple model organisms, in-
cluding mice and primates (9)
(table S1).
Experimental inactivation of
genes required for the execution
of autophagy is lethal at the whole-
body level, whereas tissue-specific
MITOPHAGY
CELL DEATH INFLAMMATION
MITOCHONDRIAL
DYSFUNCTION
MOMP
MPT
MPT
BAK
BAX
RIG-I
MDA5
PTPC
MAVS
Viral
RNA
Release of
mtDNA
Pro-inflammatory signals
Activation of
NF-kB and IRFs
Activation of
caspases
Activation of
caspase-independent
cell death effectors
Activation of
the NLRP3
inflammasome
Excess
ROS
Release of IMS proteins
Uncouplers
mtDNA
Respiratory
chain
Fig. 3. Mitophagy exerts cytoprotective effects by intercepting lethal signals before or
at the level of mitochondria. In response to lethal stimuli, mitochondria can undergo
BAX- or BAK-mediated MOMP or activate the permeability transition pore complex
(PTPC), driving the MPT. In both instances, intermembrane space (IMS) proteins are
released into the cytosol, where they activate caspase-dependent and -independent
mechanisms that mediate cell death. One MPT trigger is represented by ROS, which can
be generated upon respiratory chain uncoupling. Production of mitochondrial ROS can
trigger the NALP3 inflammasome via an unknown mechanism. In some settings, ROS-
mediated MPT may favor the release of mitochondrial DNA (mtDNA), which can
stimulate pro-inflammatory signaling via RIG-I and MDA5, both of which function as viral
RNA sensors and interact with mitochondria through the adaptor MAVS. Activated RIG-I
and MDA5 promote the activation of NF-kB and interferon regulatory factors (IRFs).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1111
REVIEW

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

knockouts induce organ-specific degenerative
changes (29). Major neurodegenerative diseases
affecting humans have been linked to defects in
mitochondria and autophagy. Parkinson’s disease
(PD) is caused by the selective loss of dopaminergic
neurons, and such cell loss can be experimentally
inducedbymitochondrial toxins, including1-methyl-
4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and
the complex I inhibitor rotenone. Several muta-
tions that cause hereditary PD affect genes, in-
cluding those encoding PINK1, Parkin, or DJ1,
whose products function in mitophagy (21, 23–25),
suggesting that deficient mitochondrial quality
control may contribute to PD. In both sporadic
and familial PD, intraneuronal inclusions called
Lewy bodies have been detected. Lewy bodies
contain a-synuclein, a protein that impairs auto-
phagy when overexpressed in cells or mice (36).
Huntington’s disease is a monogenic degen-
eration affecting striatal neurons that is caused
by an extended polyglutamine stretch in the
Huntingtin protein. In mice, a Huntington-like dis-
ease can be provoked by systemic injection of
mitochondrial uncouplers. Moreover, Huntingtin
can interact with mitochondria and patients’ lym-
phocytes manifest mitochondrial dysfunction,
while whole-body energy expenditure is affected
by signs of inefficient OXPHOS (37). Although
Huntingtin itself may be an autophagic substrate,
it also regulates autophagy. Thus, complete dele-
tion of the polyglutamine tract within Huntingtin
increases baseline autophagy (and augments lon-
gevity) in mice (38). Moreover, expression of
mutant Huntingtin in mice results in deficient
sequestration of autophagic cargoes (39).
In hereditary Alzheimer’s disease, deficient
presenilin-1 activity has been thought to induce
the accumulation of the b-amyloid peptide, which
can mediate mitochondrial toxicity (2). However,
presenilin-1 also acts as a chaperone for one sub-
unit of the lysosomal proton pump, and its loss-
of-function mutation results in defective lysosomal
acidification and impaired autophagosome clear-
ance, which are reflected in increased autophagic
vacuolization (40). The decline in the expression
of autophagy genes with age is exacerbated in
the brains of Alzheimer’s patients (35).
The aforementioned examples (and others,
table S2) follow a common leitmotif. After the
identification of pathogenic mutations, the disease
etiology was initially ascribed to primary mito-
chondrial effects of the mutant proteins. More
recently, however, it has been discovered that
the disease-causing gene product subverts autoph-
agy. It remains to be determined whether the
mitochondrial alterations simply result from de-
ficient quality control or whether the genetic de-
fects act as “dual hits” and simultaneously disrupt
mitochondrial and autophagic functions. It is al-
so possible that environmental factors (such as
mitochondrial toxins) and hereditary perturbations
in mitophagy cooperate in disease pathogenesis.
Irrespective of these incognita, it appears that phar-
macological induction of autophagy can postpone
the manifestation of neurodegenerative diseases
at least in some animal models of hereditary neu-
rodegeneration, including Huntington’s disease (41).
Autophagy may also have a major cardio-
protective role. Ischemic preconditioning, the tran-
sient reduction of coronary blood flow, protects
the heart against subsequent ischemic necrosis
through a massive induction of autophagy (3).
This correlates with an increased mitochondrial
resistance to MPT, perhaps because mitochon-
dria with an elevated threshold for MPT induc-
tion have been selected. Exercise and caloric
restriction stimulate autophagy in most tissues,
including the myocardium (3), and it is possible
that autophagy constitutes (one of) the mecha-
nisms through which physical activity and lean-
ness confer cardioprotection.
Autophagy can mitigate inflammatory reac-
tions through several mechanisms. Autophagy in
dying cells is required for optimal macrophage-
mediated clearing of apoptotic corpses, thus
reducing inflammatory reactions (29). Beyond
its contribution to the control of intracellular mi-
croorganisms, autophagy can inhibit signaling
via RIG-I–like receptors by directly conjugating
the receptors to ATG5-ATG12 complexes and
through elimination of dysfunctional mitochon-
dria (13). Autophagy can also inhibit NLRP3
activation by removing permeabilized or ROS-
producing mitochondria (11, 15). Because neuro-
degenerative processes and pathological aging
are accompanied by chronic inflammation, these
anti-inflammatory effects of autophagy may me-
diate additional health benefits.
Open Questions and Perspectives
Undoubtedly, inhibition of autophagy can partic-
ipate in the pathogenesis of major diseases, in-
cluding neurodegeneration, and stimulation of
autophagy may mediate cytoprotective and anti-
inflammatory effects that at least partially can
be ascribed to the removal of dysfunctional mito-
chondria. It remains to be seen whether the
mechanisms of autophagy-driven longevity are
the inverse of those accounting for age-related
disease—for example, declining autophagy, pro-
gressive mitochondrial dysfunction, and ensuing
cell death and inflammation—and whether behav-
ioral or pharmacological measures destined to in-
duce general autophagy can be broadly used to
improve human health. In rodents, 1 day of nutrient
deprivation reportedly suffices to halve mito-
chondrial mass in various tissues (3). Therefore
intermittent fasting may invoke drastic recycling
of the mitochondrial pool while improving longev-
ity to the same extent as does continuous caloric
restriction (2). Nonetheless, it not clear whether
the longevity-extending and health-protective ef-
fects of specific induction of mitophagy would
be as positive as those of general autophagy. For
the development of specific mitophagy-inducing
gene therapies and drugs, it will be important to
resolve questions as to how mitochondria and
autophagy cross-talk in molecular terms: Does
the deconstruction of mitochondrial membranes
induced by BCL-2 family proteins directly relate to
the construction of autophagic membranes? How
does the mitochondrial fusion-fission cycle impact
on autophagy? How is the autophagic membrane
generated at MAMs, and how does this interface
with metabolic signaling, ion fluxes and lethal sig-
nals, and, in turn, inflammatory disease?
References and Notes
1. E. R. Mackenzie, B. Rakel, Eds., Complementary and Alternative
Medicine for Older Adults: A Guide to Holistic Approaches to
Healthy Aging (Springer, New York, 2006), p. 136.
2. M. P. Mattson, M. Gleichmann, A. Cheng, Neuron 60,
748 (2008).
3. R. A. Gottlieb, R. M. Mentzer Jr., Annu. Rev. Physiol.
72, 45 (2010).
4. R. S. Hotchkiss, A. Strasser, J. E. McDunn, P. E. Swanson,
N. Engl. J. Med. 361, 1570 (2009).
5. D. C. Wallace, W. Fan, V. Procaccio, Annu. Rev. Pathol. 5,
297 (2010).
6. S. W. Tait, D. R. Green, Nat. Rev. Mol. Cell Biol. 11, 621 (2010).
7. C. He, D. J. Klionsky, Annu. Rev. Genet. 43, 67 (2009).
8. G. Kroemer, G. Mariño, B. Levine, Mol. Cell 40, 280 (2010).
9. F. Madeo, N. Tavernarakis, G. Kroemer, Nat. Cell Biol.
12, 842 (2010).
10. V. H. Perry, C. Cunningham, C. Holmes, Nat. Rev.
Immunol. 7, 161 (2007).
11. K. Nakahira et al., Nat. Immunol. 12, 222 (2011).
12. H. Kono, K. L. Rock, Nat. Rev. Immunol. 8, 279 (2008).
13. T. Saitoh, S. Akira, J. Cell Biol. 189, 925 (2010).
14. S. B. Singh et al., Nat. Cell Biol. 12, 1154 (2010).
15. R. Zhou, A. S. Yazdi, P. Menu, J. Tschopp, Nature 469,
221 (2011).
16. G. Kroemer, L. Galluzzi, C. Brenner, Physiol. Rev. 87, 99
(2007).
17. D. W. Hailey et al., Cell 141, 656 (2010).
18. R. S. Carreira, Y. Lee, M. Ghochani, A. B. Gustafsson,
R. A. Gottlieb, Autophagy 6, 462 (2010).
19. I. Novak et al., EMBO Rep. 11, 45 (2010).
20. H. Sandoval et al., Nature 454, 232 (2008).
21. R. J. Youle, D. P. Narendra, Nat. Rev. Mol. Cell Biol. 12,
9 (2011).
22. D. Chen et al., J. Biol. Chem. 285, 38214 (2010).
23. S. Geisler et al., Nat. Cell Biol. 12, 119 (2010).
24. A. Tanaka et al., J. Cell Biol. 191, 1367 (2010).
25. S. Michiorri et al., Cell Death Differ. 17, 962 (2010).
26. J. Y. Lee, Y. Nagano, J. P. Taylor, K. L. Lim, T. P. Yao,
J. Cell Biol. 189, 671 (2010).
27. D. F. Egan et al., Science 331, 456 (2011).
28. L. Radoshevich et al., Cell 142, 590 (2010).
29. B. Levine, G. Kroemer, Cell 132, 27 (2008).
30. A. Colell et al., Cell 129, 983 (2007).
31. L. Xue, G. C. Fletcher, A. M. Tolkovsky, Curr. Biol. 11, 361 (2001).
32. S. W. Tait et al., Dev. Cell 18, 802 (2010).
33. V. I. Korolchuk, A. Mansilla, F. M. Menzies,
D. C. Rubinsztein, Mol. Cell 33, 517 (2009).
34. M. M. Lipinski et al., Dev. Cell 18, 1041 (2010).
35. M. M. Lipinski et al., Proc. Natl. Acad. Sci. U.S.A. 107,
14164 (2010).
36. A. R. Winslow et al., J. Cell Biol. 190, 1023 (2010).
37. E. Bossy-Wetzel, A. Petrilli, A. B. Knott, Trends Neurosci.
31, 609 (2008).
38. S. Zheng et al., PLoS Genet. 6, e1000838 (2010).
39. M. Martinez-Vicente et al., Nat. Neurosci. 13, 567 (2010).
40. J. H. Lee et al., Cell 141, 1146 (2010).
41. K. Moreau, S. Luo, D. C. Rubinsztein, Curr. Opin. Cell
Biol. 22, 206 (2010).
Acknowledgments: We apologize to all colleagues whose
work we could not cite owing to space limitations. D.R.G.
receives grants from the NIH and from the American
Lebanese Syrian Associated Charities. L.G. is financed by the
European Union (EU) (APO-SYS). G.K. is supported by Agence
Nationale de la Recherche, Association pour la Recherche sur
le Cancer, AXA Chair for Longevity Research, Fondation
pour la Recherche Médicale, Institut National du Cancer, EU,
and Ligue Nationale Contre le Cancer (équipe labelisée).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1109/DC1
SOM Text
Tables S1 and S2
10.1126/science.1201940
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1112
REVIEW

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Itokawa Dust Particles: A Direct Link
Between S-Type Asteroids and
Ordinary Chondrites
Tomoki Nakamura,
1
* Takaaki Noguchi,
2
Masahiko Tanaka,
3
Michael E. Zolensky,
4
Makoto Kimura,
2
Akira Tsuchiyama,
5
Aiko Nakato,
1
Toshihiro Ogami,
1
Hatsumi Ishida,
1
Masayuki Uesugi,
6
Toru Yada,
6
Kei Shirai,
6
Akio Fujimura,
6
Ryuji Okazaki,
7
Scott A. Sandford,
8
Yukihiro Ishibashi,
6
Masanao Abe,
6
Tatsuaki Okada,
6
Munetaka Ueno,
6
Toshifumi Mukai,
6
Makoto Yoshikawa,
6
Junichiro Kawaguchi
6
The Hayabusa spacecraft successfully recovered dust particles from the surface of near-Earth asteroid 25143
Itokawa. Synchrotron-radiation x-ray diffraction and transmission and scanning electron microscope analyses
indicate that the mineralogy and mineral chemistry of the Itokawa dust particles are identical to those of
thermally metamorphosed LL chondrites, consistent with spectroscopic observations made from Earth and by
the Hayabusa spacecraft. Our results directly demonstrate that ordinary chondrites, the most abundant
meteorites found on Earth, come from S-type asteroids. Mineral chemistry indicates that the majority of
regolith surface particles suffered long-termthermal annealing and subsequent impact shock, suggesting that
Itokawa is an asteroid made of reassembled pieces of the interior portions of a once larger asteroid.
T
he Hayabusa spacecraft arrived at S(IV)-
type asteroid 25143 Itokawa (former-
ly 1998 SF36) in September 2005 (1).
Remote-sensing measurements from the space-
craft suggest that Itokawa consists of rocks sim-
ilar to LL5 and LL6 ordinary chondrites (2, 3),
confirming ground-based spectral characterization
(4). On 20 and 26 November 2005, the spacecraft
descended to touchdown and capture dust par-
ticles from MUSES-C Regio. This area consists
of dust and gravel deposits dominated by grains
up to 1 cm in diameter (5). Although the sam-
pler did not operate as planned, an elastic sampling
horn impacted onto the asteroid surface, directing
dust particles into the spacecraft’s sample catcher
device (5). The Hayabusa sample capsule suc-
cessfully landed in the Woomera Prohibited Area
in South Australia on 13 June 2010. Dust par-
ticles collected at the second touchdown were
recovered by two methods. In one method, we
used a Teflon spatula to sweep particles from
about 10% of the surface of a sample catcher. In
the other method, we gently tapped on the ex-
terior of the sample catcher, causing particles to
drop onto a pure silica glass slide (6).
On the Teflon spatula, we identified 1534
rocky particles by means of a field-emission
scanning electron microscope. The particles have
diameters ranging from 3 to 40 mm but are
mostly smaller than 10 mm (7). Most Itokawa
particles are angular and are probably broken
pieces of larger rocks. Among the 1534 harvested
rocky particles, 1087 are monomineralic, includ-
ing 580 olivine particles, 126 low-Ca pyroxenes,
56 high-Ca pyroxenes, 186 feldspars (172 plagi-
oclase and 14 K-feldspar), 113 troilites, 13 chro-
mites, 10 Ca phosphates, and 3 Fe-Ni metal
grains. The remaining 447 particles are poly-
mineralic mixtures, mainly silicates. Several other
particles are silica minerals and K-bearing halite,
all of uncertain origin.
Of the 40 particles removed by tapping
(diameters ranging from 30 to 180 mm) that
were analyzed by x-ray computed microtomog-
raphy (7) and x-ray diffraction, 38 were sub-
jected to more detailed mineralogic analysis.
Backscattered electron images of selected par-
ticles are shown in Fig. 1, A to D. RA-QD02-
0030 (Fig. 1A) and RA-QD02-0024 (Fig. 1B)
have a platy morphology, are polymineralic, and
have many mineral grains 1 to 10 mm in diam-
eter adhering to their surfaces. Their appearance
is typical of most Itokawa particles. Two par-
ticles show different morphologies. RA-QD02-
0013 (Fig. 1C) has a smoother soccer-ball shape,
whereas RA-QD02-0027 (Fig. 1D) consists of a
large troilite crystal and smaller silicates. Parti-
cles that contain troilite or taenite as major com-
ponents like RA-QD02-0027 are rare.
Mineralogical analysis of individual “tapped”
particles indicates that they consist mainly of coarse
[typically 10 to 50 mm in diameter (7)] crystalline
silicates, the most abundant being olivine. The next
most abundant minerals are low- and high-Ca
pyroxene and plagioclase (fig. S6A). Low-Ca py-
roxene is exclusively composed of orthopyroxene,
except for RA-QD02-0060, which is dominated
by low-Ca clinopyroxene (monoclinic structure
was confirmed by x-ray diffraction). The degree
of crystallinity of silicates differs between and
within particles, particularly for plagioclase. Some
particles contain chromite, chlorapatite, merrillite,
and troilite up to 25 mm in size. Small inclusions
(up to 10 mm) of taenite, kamacite, troilite, and
REPORTS
1
Department of Earth and Planetary Material Sciences, Faculty
of Science, Tohoku University, Aoba, Sendai, Miyagi 980-8578,
Japan.
2
College of Science, Ibaraki University, 2-1-1 Bunkyo,
Mito, Ibaraki 310-8512, Japan.
3
Synchrotron X-ray Station at
SPring-8, National Institute for Materials Science, Sayo, Hyogo
679-5198, Japan.
4
ARES, NASA Johnson Space Center, Houston,
TX 77058, USA.
5
Department of Earth and Space Science,
Graduate School of Science, Osaka University, Toyonaka 560-0043,
Japan.
6
JAXA-ISAS, 3-1-1 Yoshinodai, Sagamihara, Kanagawa
229-8510, Japan.
7
Department of Earth and Planetary Science,
Faculty of Science, Kyushu University, Hakozaki, Fukuoka 812-
8581, Japan.
8
NASA Ames Research Center, Moffett Field, CA
94035, USA.
*To whom correspondence should be addressed. E-mail:
tomoki@m.tohoku.ac.jp
Fig. 1. (A to D) Backscattered electron (BSE) images of RA-QD02-0030 (A), RA-QD02-0024 (B),
RA-QD02-0013 (C), and RA-QD02-0027 (D).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1113

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

chromite occur within coarse silicates in many
particles. The occurrence of triple junctions at the
boundary between coarse silicates suggests that
most particles have been thermally annealed.
Internal, frequently elongate, pores in silicates de-
fining curved planes are interpreted as partially
healed, impact-generated fractures (7).
Silicates in 38 of the “tapped” Itokawa par-
ticles show a limited compositional range. Of
these, olivine was found in 29, with an average
compositional range of Fa
28.6T1.1
(Fig. 2, A and
B). Low-Ca pyroxene (present in 15 particles) is
Fs
23.1T2.2
Wo
1.8T1.7
(Fig. 2, A and C); high-Ca py-
roxene (present in 14 particles) is Fs
8.9T1.6
Wo
43.5T4.5
(Fig. 2C); and plagioclase (present in 23 parti-
cles) is Ab
83.9T1.3
Or
5.5T1.2
. FeNi metallic inclu-
sions within silicates showa narrowcompositional
range: Ni and Co concentrations for kamacite
are 3.8 to 4.2 weight percent (wt%) and 9.4 to
9.9 wt%, respectively, and those for taenite are
42 to 52 wt% and 2.0 to 2.5 wt%, respectively.
Compositional ranges of olivine, low-Ca pyrox-
ene, and Co and Ni concentrations in kamacite
for equilibrated members (petrologic types 4 to 6)
of the main ordinary chondrite groups are as fol-
lows: H (Fa
16–20
, Fs
14.5–18
, 0.44 to 0.51 wt% Co,
~6.9 wt%Ni), L(Fa
22–26
, Fs
19–22
, 0.70 to 0.95 wt%
Co, ~6.54 wt%Ni), and LL(Fa
26–32
, Fs
22–26
, 1.42
to 37.0 wt% Co, ~4.98 wt% Ni) (8, 9). Olivine,
low-Ca pyroxene, and kamacite compositions
of most Itokawa particles fall within the range
of LLchondrites, indicating that MUSES-CRegio
is covered with LL-chondrite particles.
Close examination of the mineral chemistries
of these 38 Itokawa particles revealed that they
belong to two different populations: six poorly
equilibrated particles and 32 highly equilibrated
particles. The poorly equilibrated particles
contain olivine and low-Ca pyroxene with com-
positional ranges Fa
24.4–28.9
and Fs
11.2–23.8
,
respectively (Fig. 2, B and C); pyroxene is more
heterogeneous because of slower cation diffusion
(10). RA-QD02-0060 contains a large low-Ca
clinopyroxene crystal with lamellar-like Fe-Mg
zoning (Fig. 3A) and Mg-rich host (Fs
11.3
Wo
0.5
;
table S1) with many Mg-poor lamella (~Fs
20.4
Wo
0.6
;
table S1). Three particles contain mesostasis con-
sisting of tiny diopside and troilite crystals em-
bedded in albitic glass (Fig. 3B). The texture and
composition of the mesostasis is similar to that of
chondrule mesostasis, which suggests that these
particles are pieces of chondrules.
The highly equilibrated Itokawa particles
(Fig. 1, A to D) have narrow olivine and low-Ca
pyroxene compositional ranges of Fa
27.1–30.7
and
Fs
22.5–25.7
, respectively (Fig. 2, B and C). Meso-
stasis is absent, and coarse-grained diopside and
plagioclase (Fig. 3C)—some of which include
exsolved K-feldspar (Fig. 3D)—are abundant.
Major and minor element concentrations in sili-
cates in these particles are nearly constant within
and between particles. The mean compositions
of olivine and low- and high-Ca pyroxene are
Fa
29.0T0.7
, Fs
24.0T0.6
Wo
1.4T0.3
, and Fs
8.6T0.8
Wo
44.8T0.9
,
respectively (table S2). Virtually constant Fs and
Wo contents in both low- and high-Ca pyroxenes
indicate that the highly equilibrated particles ex-
perienced intense thermal metamorphism (11);
Mg, Fe, and Ca partitioning between pyroxenes
has almost progressed to completion.
Ordinary chondrite meteorites exhibit a com-
plete thermal metamorphic sequence from
unequilibrated type 3, through types 4 and 5, to
totally equilibrated type 6 (11). Mg-rich clino-
pyroxene in type-3 ordinary chondrites is pro-
Fig. 2. (A) Mean Fa and Fs contents of all tapped Itokawa particles in
comparison with those of H, L, and LL chondrites. (B) Fa distribution of
olivine crystals from 29 Itokawa particles. Approximately 10 analyses were
made for each olivine crystal. (C) Pyroxene compositions from 5 poorly equil-
ibrated particles (left) and from 18 highly equilibrated particles (right); 5 to 10
analyses were made for each pyroxene crystal. (D) Chromite compositions of
three highly equilibrated particles: RA-QD02-0030, -0031, and -0047. Chro-
mite data of meteorites are from (16).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1114
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

gressively transformed to orthopyroxene with
increasing metamorphism. Devitrified chondrule
mesostasis glass (table S1), clinopyroxene, and
the characteristic lamellar-like Fe-Mg zoning
(like that in the poorly equilibrated Itokawa par-
ticles) is present in type 4 ordinary chondrites
but absent in type 5 (12, 13). Thus, the poorly
equilibrated Itokawa particles are weakly me-
tamorphosed material, classified as petrologic
type 4.
The narrow compositional variations of the
highly equilibrated Itokawa pyroxenes (Fig. 2C)
are similar to LL5 and LL6 chondrites (10, 14).
The characteristics of high-temperature meta-
morphism of grades 5 and 6 (14) that we ob-
serve in highly equilibrated Itokawa particles
include (i) very low CaO and Cr
2
O
3
contents
(<0.02 wt%) in olivine, (ii) predominance of
low-Ca orthopyroxene over clinopyroxene (Fig.
3E), (iii) preferential occurrence of diopside and
absence of augite (Fig. 2C), and (iv) presence of
coarsened plagioclase (typically 20 to 50 mm in
diameter; Fig. 3C) and diopside (typically 10 to
30 mm). Similarly, chromite in the highly equil-
ibrated particles has compositions (Fig. 2D and
table S3) similar to those of LL5 and LL6 chon-
drites (15, 16). Thus, the highly equilibrated
particles formed during intense thermal meta-
morphism. Itokawa is a breccia of poorly equil-
ibrated LL4 and highly equilibrated LL5 and
LL6 materials. The conclusion is confirmed by
independent results of a companion paper (7).
With our mineralogic data of the highly equil-
ibrated particles, it is possible to follow three crit-
ical phases of Itokawa’s thermal evolution. The
prograde peak crystallization temperature is pro-
vided by plagioclase triclinicity (6, 17, 18) from
five particles: 570°C for RA-QD02-0010, 560°C
for RA-QD02-0025-01, 570°C for RA-QD02-
0055, 575°C for RA-QD02-0067, and 820°C
for RA-QD02-0013. These plagioclases crystal-
lized at different points as temperatures rose to
a peak of 820°C, with each crystal locking in
a different triclinicity (18, 19). The peak meta-
morphic temperature (i.e., the temperature of last
equilibration) is revealed by two-pyroxene geo-
thermometry (20). Low- and high-Ca pyroxenes
with homogeneous compositions coexist in three
Itokawa particles, and peak metamorphic temper-
atures at 0.1 MPa were calculated using QUILF
95 software: 783° T 12°C for RA-QD02-0013,
814° T 21°C for RA-QD02-0024, and 837° T
10°C for RA-QD02-0010. These temperatures
are slightly lower than the range reported for LL6
chondrite meteorites [875° to 945°C (19)]. To
characterize the subsequent cooling period, we
applied the olivine-spinel geothermometer (21)
to three particles containing olivine-chromite pairs:
RA-QD02-0030 (Figs. 1A and 3C), -0031, and
-0047, in which yielded equilibration tempera-
tures of 636°C, 625°C, and 595°C, respectively.
These temperatures are lower than those obtained
by two-pyroxene thermometry, reflecting contin-
ued Fe-Mg exchange between chromite and oli-
vine during cooling, which occurred more quickly
than Ca diffusion in pyroxene (10). The olivine-
spinel geospeedometry (22) applied to mean
chromite size (~20 mm) and the equilibrated tem-
peratures and the results indicate slow cooling at
a rate of ~0.5 K per 1000 years near 600°C. The
combined results of these temperature estimates
suggest that the highly equilibrated particles ex-
perienced a peak metamorphic temperature of
~800°C and cooled slowly to 600°C.
During thermal metamorphism in asteroids,
temperatures increase with depth and thus higher
petrologic-type materials form deeper than the
lower types, assuming internal heating due to
decay of short-lived radioisotopes such as
26
Al
(23–25). The slow cooling of the highly equili-
brated Itokawa particles suggests that they
formed at considerable depth. The temperature
of 800°C experienced by many Itokawa particles
requires a diameter of the original asteroid larger
than 20 km (25). The current size of Itokawa
(0.5 × 0.3 × 0.2 km) (1) is much smaller than
this required size. Therefore, the Itokawa parent
S-class asteroid was originally much larger, expe-
rienced intense thermal metamorphism, and was
then catastrophically disaggregated by one or more
impacts into many small pieces, some of which re-
accreted into the present greatly diminished,
rubble-pile asteroid, consistent with previous sug-
gestions (1). Impact processes are responsible for
cracks in crystals (7), dislocations in some olivine
crystals (Fig. 3F), and local variations in the degree
of silicate crystallinity. These impact effects vary
Fig. 3. (A to F) Poorly equilibrated particles RA-QD02-0060 (A) and RA-QD02-0011-1 (B). Highly
equilibrated particles RA-QD02-0030 (C), RA-QD02-0013 (D), RA-QD02-0024 (E), and RA-QD02-0032
(F). (A) Thin, FeO-rich light lamellae repeatedly occur in Low-Ca clinopyroxene. (B) Mesostasis glass-like
portions occur between coarse silicate crystals. Inset: An EBSD map (red box) indicates that mesostasis
shows no diffraction and thus it is glass. (C) A variety of minerals are included in this particle. (D) Three
parallel exsolution lamella of K-feldspar sanidine occur in albite. (E) Low-Ca pyroxene consists of
orthopyroxene with many stacking disorders parallel to (100). Inset: Diffraction spots show streaks
(indicated by arrows) due to disturbance of the 1.8-nm repetition (indicated by white lines). (F) The
dislocation density in this olivine is 1.6 × 10
8
cm
−2
, suggestive of moderate shock. (A) to (D), BSE
images; (E) and (F), bright-field transmission electron microscope images. Ol, olivine; LPx, low-Ca
pyroxene; LCPx, low-Ca clinopyroxene; Di, diopside; Chr, chromite; Tr, troilite; K, kamacite; Tae, taenite;
Pl, plagioclase; Kfs, K feldspar; Mes, mesostasis.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1115
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

greatly between particles (fig. S7), which is typ-
ical of moderately shocked astromaterial cor-
responding shock stages up to S4 (6, 26).
MUSES-C Regio probably formed by segre-
gation and accumulation of fine gravel into areas
close to the gravitational center of Itokawa due to
global-scale electrostatic grain levitation, vibration-
induced granular migration, and deposition of
slowmoving ejecta launched fromsurface impacts
(27–29). Therefore, particles in MUSES-C Regio
originally derived from diverse regions of Itokawa.
Fortunately, despite the small mass of the recovered
Itokawa samples, they record the critical steps in
the history of this asteroid. Itokawa was classified
as an S-type asteroid from terrestrial remote sen-
sing, and it has been commonly suggested that
S-type asteroids, the most abundant asteroids
in the inner asteroid belt, are the parent bodies of
ordinary chondrites. Our petrologic data from
MUSES-C Regio confirm that Itokawa is indeed
an ordinary chondrite (LL4 to LL6), thereby finally
linking these asteroids and meteorites.
References and Notes
1. A. Fujiwara et al., Science 312, 1330 (2006).
2. M. Abe et al., Science 312, 1334 (2006).
3. T. Okada et al., Science 312, 1338 (2006).
4. R. Binzel, A. S. Rivkin, S. J. Bus, J. M. Sunshine,
T. H. Burbine, Meteorit. Planet. Sci. 36, 1167 (2001).
5. H. Yano et al., Science 312, 1350 (2006).
6. See supporting material on Science Online.
7. A. Tsuchiyama et al., Science 333, 1125 (2011).
8. A. J. Brearley, R. H. Jones, Rev. Mineral. 36, 3-1 (1998).
9. A. E. Rubin, Geochim. Cosmochim. Acta 54, 1217 (1990).
10. J. B. Brady, D. J. Cherniak, In Diffusion in Minerals
and Melts, Y. Zhang, D. J. Cherniak, Eds. (Mineralogical
Society of America, Chantilly, VA, 2010), pp. 899–920.
11. W. R. Van Schmus, J. A. Wood, Geochim. Cosmochim.
Acta 31, 747 (1967).
12. E. A. Jobbins et al., Mineral. Mag. 35, 881 (1966).
13. T. J. McCoy, E. R. D. Scott, R. H. Jones, K. Keil,
G. J. Taylor, Geochim. Cosmochim. Acta 55, 601 (1991).
14. G. R. Huss, A. E. Rubin, J. N. Grossman, in Meteorites and
the Early Solar System II, D. S. Lauretta, H. Y. McSween
Jr., Eds. (University of Arizona Press, Tucson, AZ, 2006),
pp. 567–586.
15. T. E. Bunch, K. Keil, K. G. Snetsinger, Geochim.
Cosmochim. Acta 31, 1569 (1967).
16. M. Kimura, H. Nakajima, H. Hiyagon, M. K. Weisberg,
Geochim. Cosmochim. Acta 70, 5634 (2006).
17. J. V. Smith, J. Geol. 80, 505 (1972).
18. Y. Nakamuta, Y. Motomura, Meteorit. Planet. Sci. 34,
763 (1999).
19. V. Slater-Reynolds, H. Y. McSween Jr., Meteorit. Planet.
Sci. 40, 745 (2005).
20. D. H. Lindsley, Am. Mineral. 68, 477 (1983).
21. J. Fabriés, Contrib. Mineral. Petrol. 69, 329 (1979).
22. K. Ozawa, Geochim. Cosmochim. Acta 48, 2597
(1984).
23. M. Miyamoto, N. Fujii, H. Takeda, Proc. Lunar Planet. Sci.
12B, 1145 (1981).
24. M. Trieloff et al., Nature 422, 502 (2003).
25. H. Y. McSween et al., in Asteroids III, W. Bottke et al.,
Eds. (Univ. of Arizona Press, Tucson, AZ, 2002),
pp. 559–571.
26. D. Stöffler, K. Keil, E. R. D. Scott, Geochim. Cosmochim.
Acta 55, 3845 (1991).
27. H. Miyamoto et al., Science 316, 1011 (2007).
28. P. Lee, Icarus 124, 181 (1996).
29. D. G. Korycansky, E. Asphaug, Icarus 171, 110
(2004).
Acknowledgments: We thank the Hayabusa project team
for sample return; KEK for synchrotron experiments;
H. Nakano, Y. Yamazaki, K. Shimada, Y. Kakazu,
T. Hashimoto, M. Konno, Y. Katsuya, and Y. Matsushita,
for technical support; and J. Grossman, T. Ikeda,
T. Hokada, K. Ozawa, Y. Nakamuta, and S. Wakita for
helpful discussions. Supported by NASA grant
769583.07.03 (M.E.Z. and S.A.S.).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1113/DC1
Figs. S1 to S8
Tables S1 to S5
References (30–40)
2 May 2011; accepted 2 August 2011
10.1126/science.1207758
Oxygen Isotopic Compositions of
Asteroidal Materials Returned from
Itokawa by the Hayabusa Mission
Hisayoshi Yurimoto,
1
* Ken-ichi Abe,
1
Masanao Abe,
2
Mitsuru Ebihara,
3
Akio Fujimura,
2
Minako Hashiguchi,
1
Ko Hashizume,
4
Trevor R. Ireland,
5
Shoichi Itoh,
1
Juri Katayama,
1
Chizu Kato,
1
Junichiro Kawaguchi,
2
Noriyuki Kawasaki,
1
Fumio Kitajima,
6
Sachio Kobayashi,
1
Tatsuji Meike,
1
Toshifumi Mukai,
2
Keisuke Nagao,
7
Tomoki Nakamura,
8
Hiroshi Naraoka,
6
Takaaki Noguchi,
9
Ryuji Okazaki,
6
Changkun Park,
1
Naoya Sakamoto,
1
Yusuke Seto,
10
Masashi Takei,
1
Akira Tsuchiyama,
4
Masayuki Uesugi,
2
Shigeyuki Wakaki,
1
Toru Yada,
2
Kosuke Yamamoto,
1
Makoto Yoshikawa,
2
Michael E. Zolensky
11
Meteorite studies suggest that each solar system object has a unique oxygen isotopic composition.
Chondrites, the most primitive of meteorites, have been believed to be derived from asteroids, but oxygen
isotopic compositions of asteroids themselves have not been established. We measured, using secondary ion
mass spectrometry, oxygen isotopic compositions of rock particles from asteroid 25143 Itokawa returned
by the Hayabusa spacecraft. Compositions of the particles are depleted in
16
Orelative to terrestrial materials
and indicate that Itokawa, an S-type asteroid, is one of the sources of the LL or L group of equilibrated
ordinary chondrites. This is a direct oxygen-isotope link between chondrites and their parent asteroid.
M
ineral compositions of asteroids are
inferred from visible and near-infrared
reflectance spectroscopy. The spectro-
scopic similarity between some asteroids and
meteorites suggests that meteorites come from
asteroids and allows indirect assessments of
asteroid-meteorite connections and inferences
regarding chemical compositions of asteroids
(1). Of the ~40,000 meteorites we know of, only
14 have had their pre-impact orbits ascertained
(2). The aphelia of these 14 orbits are located
within the Main Asteroid Belt between Martian
and Jovian orbits, which is consistent with an
asteroidal origin. However, even the parent
asteroids of these 14 meteorites have not been
identified.
The taxonomy of meteorites largely has been
based on the whole-rock chemical and oxygen
isotopic compositions. Each meteorite group, and
probably each planet, has a characteristic chem-
ical composition and a unique oxygen isotopic
composition (3, 4). The origin of oxygen isotopic
variations in the solar system is thought to be
an isotope-selective photodissociation of carbon
monoxide that occurred before planet formation
(5–7). The unique oxygen isotopic composition
of a planet is thought to be produced by a com-
bination of gas-dust chemistry and accretion
physics in the solar nebula (6, 8). The Earth and
the Moon—the only bodies for which we have
measurements—have similar oxygen isotopic
compositions within an uncertainty of T0.016 per
mil (‰) [2 SD (2s)] (9, 10). The determination
of an oxygen isotopic composition of an asteroid
or a planet therefore would provide an indisputable
means to clarify mechanisms of planet formation
in the solar nebula and to connect an asteroid or
a planet to a specific meteorite group.
The Hayabusa spacecraft made two touch-
downs on the surface of asteroid 25143 Itokawa
on 20 and 26 November 2005 JSTand success-
fully collected grain particles from the surface
of the asteroid. Itokawa is classified as an S-type
asteroid. As inferred from reflectance spectrom-
etry, it consists of materials similar to primitive
1
Natural History Sciences, Hokkaido University, Sapporo 060-
0810, Japan.
2
Japan Aerospace Exploration Agency–Institute
of Space and Astronautical Science, 3-1-1 Yoshinodai, Chuo
Sagamihara, Kanagawa, 252-5210, Japan.
3
Graduate School
of Science and Engineering, Tokyo Metropolitan University,
1-1 Minami-Osawa, Hachioji, Tokyo 192-0397, Japan.
4
De-
partment of Earth and Space Science, Osaka University, 1-1
Machikaneyama, Toyonaka 560-0043, Osaka, Japan.
5
Research
School of Earth Sciences, College of Physical and Mathematical
Sciences, Australian National University, Canberra, ACT 0200,
Australia.
6
Department of Earth and Planetary Sciences, Kyushu
University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581,
Japan.
7
Geochemical Research Center, University of Tokyo,
Hongo, Tokyo 113-0033, Japan.
8
Department of Earth and
Planetary Material Sciences, Tohoku University, Aramaki, Aoba,
Sendai, Miyagi 980-8578, Japan.
9
Collage of Science, Ibaraki
University, 2-1-1 Bunkyo, Mito, Ibaraki 310-8512, Japan.
10
De-
partment of Earth and Planetary Sciences, Kobe University,
Kobe 657-8501, Japan.
11
Astromaterials Research and Explo-
ration Science, KT, NASA Johnson Space Center, Houston, TX
77058, USA.
*To whom correspondence should be addressed. E-mail:
yuri@ep.sci.hokudai.ac.jp
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1116
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

achondrites or ordinary chondrites (11), which
can be distinguished by their oxygen isotopic
compositions (4). Previous near-infrared reflec-
tance spectroscopy by Hayabusa suggests that
the asteroid’s surface has an olivine-rich mineral
assemblage that is potentially similar to that of
LL5 or LL6 chondrites, with different degrees
of space weathering (12). The major mineral
assemblage of the sample grains collected by
Haybusa is olivine, pyroxene, plagioclase, iron
sulfide, and iron-nickel metal (13). The grain
sizes are less than 150 mm (mostly less than sev-
eral tens of micrometers), and crystal sizes in
the grain are less than 80 mm (mostly less than
20 mm) (14).
We used the Hokudai isotope microscope
system (15) to determine oxygen isotopic com-
positions of minerals in 28 of these grains, cor-
responding to measurements of 19 olivine crystals,
7 orthopyroxene crystals, and 7 plagioclase crys-
tals (table S4). The results include multiphase
measurements within a grain: analysis for the co-
existing olivine-orthopyroxene-plagioclase system
in grains RA-QD02-0010 and RA-QD02-0030
and analysis for the coexisting olivine-plagioclase
system in grain RA-QD02-0031 (Fig. 1).
The analytical uncertainty was determined
from oxygen isotope measurements of an ordi-
nary chondrite, Ensisheim LL6. It is T0.7‰
(2s) for d
17
O
SMOW
, T1.5‰ (2s) for d
18
O
SMOW
for olivine and orthopyroxene, and twice that
for plagioclase, where SMOW is standard mean
ocean water. The precision of D
17
O
SMOW
is
~T0.5‰ (2s) for all analyses [Fig. 2 and sup-
porting online material (SOM) text]. This preci-
sion is sufficient to distinguish most meteoritic
materials known to date from terrestrial mate-
rials. However, uncertainties of d
17
O
SMOW
and
d
18
O
SMOW
are too high to allow a precise de-
termination of metamorphic temperatures by
means of the isotopic fractionation among the
minerals. Nevertheless, the mineralogical order
of isotopic equilibration by thermal metamor-
phism on the parent body could be recognized
within the measurement uncertainties. Thus, the
analytical uncertainties could be applied to the
Itokawa grains.
The variations of D
17
O
SMOW
for Itokawa
minerals are about T0.5‰ (2s) (table S4), which
are equivalent to the dispersion expected from
measurement uncertainties. All oxygen isotopic
compositions of the minerals from Hayabusa
sample return capsule plot on the upper side of
terrestrial standards on a three-isotope oxygen
diagram and are distributed parallel to the ter-
restrial mass fractionation line (Fig. 3). This in-
dicates that the grains returned by Hayabusa are
not terrestrial materials and further demonstrates
that the spacecraft retrieved asteroid Itokawa’s
surface materials during touchdown.
Isotopic compositions of meteorites occupy
distinct regions of the oxygen three-isotope dia-
gram according to meteorite group. The region
of the Itokawa grains overlaps with those of the
ordinary chondrites (16). Ordinary chondrites are
subdivided into H, L, and LL chondrites. These
groups also have distinct ranges of whole-rock
oxygen isotopic compositions, with magnitudes
of departure from the terrestrial fractionation
Fig. 2. Oxygen isotopic compositions of Ensisheim minerals (A and B) com-
pared with those of a forsterite crystal from San Carlos, Arizona, USA and an
anorthite crystal from Miyake-jima, Japan. Instrumental mass fractionation for
each mineral is corrected by use of the reference value shown in table S3.
Isotope variation defined by 2s for each mineral phase is shown by a rec-
tangle with a color of the corresponding symbol. Open circles on the ECL
(equilibrated chondrite line) correspond to average O isotopic compositions
of ordinary chondrites, LL, L, H, from top to bottom. TF, terrestrial fraction-
ation line; Ol, olivine; Opx, orthopyroxene; Pl, plagioclase; An, anorthite;
Miyake, Miyake-jima. Data are from tables S2 and S3. D
17
O
SMOW
= d
17
O
SMOW

0.52 d
18
O
SMOW
. A mass fractionation line of the average O isotopic com-
position of LL chondrite group is shown as a reference. Variations (2 s) of
whole-rock D
17
O
SMOW
values for H, L, and LL chondrite groups are shown to
the right of (B).
20µm
Opx
Ol
Pl
Tr
Chr
Pl
RA-QD02-0030
Fig. 1. Measurement spots for oxygen isotope
analysis. An optical microscope image after mea-
surements is superimposed on the backscattered
electron image before measurements. Primary ion
beam craters are indicated with dashed circles.
The spatial resolution (~10 mm) was sufficient to
measure an object with coexisting minerals and to
allow analyses free of contamination from the re-
spective minerals. Ol, olivine; Opx, orthopyroxene;
Pl, plagioclase; Chr, chromite; Tr, troilite.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1117
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

line, D
17
O
SMOW
, being 0.73 T 0.18‰ (2s) for
H-chondrite group, 1.07 T 0.18‰ (2s) for L-
chondrite group, and 1.26 T 0.24‰ (2s) for
LL-chondrite group (17). The ranges of D
17
O
SMOW
of L and LL group overlap each other, but com-
positions from the H group are distinct from the
other groups.
Unequilibrated chondrites consist of min-
erals having highly variable D
17
O
SMOW
values.
Two mechanisms can homogenize the D
17
O
SMOW
among minerals: metamorphism and melting
(18). Minerals of equilibrated chondrites be-
come homogenized to a D
17
O
SMOW
value by
metamorphism toward the whole-rock oxygen
isotopic composition, with variability decreas-
ing in the order of metamorphic grades from
type 4 to 6.
The D
17
O
SMOW
values for Itokawa are 1.46 T
0.41‰ (2s) for olivine, 1.57 T 0.62‰ (2s) for
orthopyroxene, and 1.15 T 0.51‰ (2s) for plagi-
oclase (table S4). The observed variations among
the minerals are within analytical uncertainties
of our measurements. The mean D
17
O
SMOW
for
minerals from Itokawa, 1.39 T 0.36‰ (2s), co-
incides with that of LL or L chondrite groups
but is clearly distinguished from H chondrites
(Fig. 3B). The small variation of D
17
O
SMOW
dem-
onstrates that the Itokawa minerals were equil-
ibrated during metamorphism.
The variations of d
18
O
SMOW
of orthopyrox-
ene and plagioclase from Itokawa are similar
to those measured fromthe EnsisheimLL6 chon-
drite. The range of variation in d
18
O
SMOW
of
Itokawa olivine is greater than that of Ensisheim
olivine and is as large as those of Itokawa pla-
gioclase. The large variation for Itokawa olivine
could be attributed to instrumental mass frac-
tionation relating to irregularities of the sample
surface owing to the small size of the grains.
Nevertheless, the isotopic relationship among
olivine, orthopyroxene, and plagioclase shows
that the oxygen isotopes fractionated under equi-
libriumbetween coexisting phases. Degrees of the
isotopic fractionation among minerals are slightly
larger in Itokawa materials than in Ensisheim.
The larger isotopic fractionation among the min-
erals may indicate that the metamorphic tem-
perature was lower in Itokawa material than in
Ensisheim.
The metamorphic temperature would be
determined by means of the oxygen isotopic
fractionation among minerals. The plagioclase-
olivine, orthopyroxene-olivine, and plagioclase-
orthopyroxene temperatures are calculated to
be 600, 650, and 720°C, respectively, through
application of an oxygen isotope thermometer
(19). The estimated temperatures from 600 to
720°C for Itokawa are lower than those for LL6,
L6, and L5 chondrites and higher than for a L4
chondrite (16).
On the basis of this equilibration and the
small variation of D
17
O
SMOW
, the petrographic
type of Itokawa is equivalent to type 4-6 in the
LL or L chondrite group. The Itokawa material
is compatible with an LL4-6 chondrite classifi-
cation if we combine the oxygen isotope data
with the results of the chemical compositions
of minerals (13).
The oxygen isotopic composition of asteroid
Itokawa thus provides unequivocal evidence that
ordinary chondrites come from S-type asteroids.
References and Notes
1. D. J. Tholen, M. A. Barucci, in Asteroids II, W. F. Bottke,
A. Cellino, P. Paolicchi, R. P. Binzel, Eds. (University of
Arizona Press, Tucson, AZ, 1989), pp. 298–315.
2. P. Brown et al., Meteorit. Planet. Sci. 46, 339 (2011).
3. A. N. Krot, K. Keil, C. A. Goodrich, E. R. D. Scott,
M. K. Weisberg, in Meteorites, Comets, and Planets,
A. M. Davis, Ed. (Elsevier, Amsterdam, 2005), pp. 83–128.
4. R. N. Clayton, in Meteorites, Comets, and Planets,
A. M. Davis, Ed. (Elsevier, Amsterdam, 2005), pp. 129–142.
5. R. N. Clayton, Nature 415, 860 (2002).
6. H. Yurimoto, K. Kuramoto, Science 305, 1763 (2004).
7. J. R. Lyons, E. D. Young, Nature 435, 317 (2005).
8. K. Kuramoto, H. Yurimoto, in Chondrites and the
Protoplanetary Disk, A. N. Krot, E. R. D. Scott,
B. Reipurth, Eds. (ASP Conference Series, 2005),
vol. 341, pp. 181–192.
9. Martian meteorites are thought to be martian rocks, but
definitive proof will require a direct measurement of
the oxygen isotopic composition of Mars.
10. U. Wiechert et al., Science 294, 345 (2001).
11. P. A. Abell, F. Vilas, K. S. Jarvis, M. J. Gaffey, M. S. Kelley,
Meteorit. Planet. Sci. 42, 2165 (2007).
12. M. Abe et al., Science 312, 1334 (2006).
13. T. Nakamura et al., Science 333, 1113 (2011).
14. All samples collected by the Hayabusa spacecraft and
analyzed here have been characterized by means of
x-ray microtomography, x-ray diffraction analysis,
x-ray fluorescence analysis, scanning electron microscopy,
and electron probe microanalysis before isotope
measurements were made (13). For our analysis,
chemically equilibrated grains were prepared. Chemically
less equilibrated grains described in (13) have not
been included because a precise chemical characterization
was still in process, and they were not ready for this
study. The less equilibrated grains were not common in
Fig. 3. Oxygen isotopic compositions of Itokawa minerals (Aand B) compared
to those of a forsterite crystal from San Carlos, Arizona, USA and an anorthite
crystal fromMiyake-jima, Japan. Isotope variation defined by 2s for each mineral
is shown by a rectangle with a color of the corresponding symbol. Open circles on
the ECL correspond to average Oisotopic compositions of ordinary chondrites, LL,
L, H, from top to bottom. TF, terrestrial fractionation line; Ol, olivine; Opx, or-
thopyroxene; Pl, plagioclase; An, anorthite; Miyake, Miyake-jima. Data are from
tables S2 and S4. A mass fractionation line of the average O isotopic composition
of LL chondrite group is shown as a reference. Variations (2s) of whole-rock
D
17
O
SMOW
values for H, L, and LL chondrite groups are shown to the right of (B).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1118
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

the Hayabusa sample return capsule (13). We mounted
each grain at the center of an epoxy disk and polished
the surface according to the processes established for
the preliminary examination. We coated a thin layer
of gold with a thickness of 60 nm on the samples for
secondary ion mass spectrometry.
15. H. Yurimoto, K. Nagashima, T. Kunihiro, Appl. Surf. Sci.
203-204, 793 (2003).
16. R. N. Clayton, Annu. Rev. Earth Planet. Sci. 21, 115 (1993).
17. R. N. Clayton, T. K. Mayeda, J. N. Goswami, E. J. Olsen,
Geochim. Cosmochim. Acta 55, 2317 (1991).
18. R. C. Greenwood, I. A. Franchi, A. Jambon, P. C. Buchanan,
Nature 435, 916 (2005).
19. R. N. Clayton, S. W. Kieffer, in Stable Isotope
Geochemistry: A Tribute to Samuel Epstein, H. P. Taylor Jr,
J. R. O'Neil, I. R. Kaplan, Eds. (Geochem. Soc. Spec. Pub.
No. 3, 1991), pp. 3–10.
Acknowledgments: We thank the Hayabusa sample
curation team and the Hayabusa project team for close
cooperation. This study was supported by the Monka-sho
grant (H.Y.) and by the NASA Muses-CN/Hayabusa
Program (M.E.Z.).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1116/DC1
Materials and Methods
SOM Text
Figs. S1 to S3
Tables S1 to S4
Reference (20)
2 May 2011; accepted 1 August 2011
10.1126/science.1207776
Neutron Activation Analysis of a
Particle Returned from Asteroid Itokawa
M. Ebihara,
1
* S. Sekimoto,
2
N. Shirai,
1
Y. Hamajima,
3
M. Yamamoto,
3
K. Kumagai,
1
Y. Oura,
1
T. R. Ireland,
4
F. Kitajima,
5
K. Nagao,
6
T. Nakamura,
7
H. Naraoka,
5
T. Noguchi,
8
R. Okazaki,
5
A. Tsuchiyama,
9
M. Uesugi,
10
H. Yurimoto,
11
M. E. Zolensky,
12
M. Abe,
10
A. Fujimura,
10
T. Mukai,
10
Y. Yada
10
A single grain (~3 micrograms) returned by the Hayabusa spacecraft was analyzed by neutron
activation analysis. This grain is mainly composed of olivine with minor amounts of plagioclase,
troilite, and metal. Our results establish that the Itokawa sample has similar chemical characteristics
(iron/scandium and nickel/cobalt ratios) to chondrites, confirming that this grain is extraterrestrial
in origin and has primitive chemical compositions. Estimated iridium/nickel and iridium/cobalt
ratios for metal in the Itokawa samples are about five times lower than CI carbonaceous chondrite
values. A similar depletion of iridium was observed in chondrule metals of ordinary chondrites.
These metals must have condensed from the nebular where refractory siderophile elements already
condensed and were segregated.
T
he Hayabusa spacecraft was launched on
9 May 2003 and reached asteroid 25143
Itokawa in September 2005 (1). After ac-
complishing numerous scientific observations
(2, 3), the spacecraft tried to collect surface ma-
terial from Itokawa by touching down to the as-
teroid in November 2005 (4). The spacecraft then
navigated back to Earth. Despite encountering
several difficulties, Hayabusa finally returned to
Earth on 12 June 2010, and its entry capsule was
successfully recovered. Although the sample col-
lection was not nominally performed, it was hoped
that some extraterrestrial material was stored in
the capsule. After careful and extensive exami-
nation, more than 1500 particles were recognized
by microscopes, most of which were eventually
judged to be extraterrestrial, probably originating
from Itokawa (5).
We analyzed one of the largest grains returned
by Hayabusa (RA-QD02-0049) through instru-
mental neutron activation analysis (INAA). We
used a scanning electron microprobe (SEM) to
perform the initial characterization of this grain
at the receiving room at Institute of Space and
Astronautical Science, Japan Aerospace Explo-
ration Agency. The SEM results indicate that
the particle is a large crystal of olivine (Fig. 1A),
and small inclusions of troilite are contained in
this olivine (Fig. 1B). In addition, small pieces of
silicates are attached on the surface of olivine
(Fig. 1B). Before assaying to INAA, the sample
was rinsed with ethanol for the inspection of or-
ganic materials at Kyushu University, Fukuoka,
Japan, where the rinsed sample was also analyzed
by a Raman spectrometer for the characteriza-
tion of carbonaceous compounds. The sample
was then carefully placed into a quartz sample
holder for neutron irradiation at Kyushu and
brought to the Kyoto University Research Re-
actor Institute (KURRI), Kumatori, Osaka, Japan.
Along with reference standards, the Itokawa
sample was irradiated with neutrons at a thermal
neutron flux of 8.2 × 10
13
cm
−2
s
−1
for 19 hours.
After irradiation, the quartz holder was replaced
with a new(nonirradiated) one to reduce the back-
ground radioactivity during gamma-ray counting.
During this procedure, the sample was split into
five small grains, the largest of which was named
RA-QD02-0049-1. The remaining four smaller
grains were placed together into one sample
holder and named RA-QD02-0049-2. Together
with reference standards, both samples were
measured for their radioactivity at KURRI for
the first three weeks and then at the Low Level
Radioactivity Laboratory of Kanazawa Univer-
sity, Tatsunokuchi, Kanazawa, Japan. The de-
tailed procedure for INAA is described in the
supporting online material (SOM).
Both samples have similar chemical compo-
sitions, suggesting that the grain is fairly homo-
geneous in its chemical composition (Table 1).
According to the surface observation by SEM,
grain RA-QD02-0049 is composed mainly of
olivine, with plagioclase and troilite as trace
components. A slightly high Na content for
RA-QD02-0049-1 suggests that it is somewhat
plagioclase-rich relative to RA-QD02-0049-2.
1
Department of Chemistry, Tokyo Metropolitan University,
Hachioji, Tokyo 192-0397, Japan.
2
Kyoto University Research
Reactor Institute, Kumatori, Osaka, Japan.
3
Low Level Radio-
activity Lab, Kanazawa University, Tatsunokuchi, Japan.
4
Re-
search School of Earth Sciences, The Australian National
University, Canberra, Australia.
5
Department of Earth and
Planetary Sciences, Kyushu University, Fukuoka, Japan.
6
Geo-
chemical Research Center, The University of Tokyo, Tokyo, Japan.
7
Department of Earth and Planetary Material Sciences, Tohoku
University, Sendai, Japan.
8
College of Science, Ibaraki Uni-
versity, Mito, Japan.
9
Department of Earth and Space Science,
Osaka University, Toyonaka, Japan.
10
Institute of Space and
Astronautical Science Japan Aerospace Exploration Agency,
Sagamihara, Kanagawa, Japan.
11
Natural History Sciences,
Hokkaido University, Sapporo, Japan.
12
NASA Johnson Space
Center, Houston, TX 77058, USA.
*To whom correspondence should be addressed. E-mail:
ebihara-mitsuru@tmu.ac.jp
Fig. 1. Back-scattered electron image of a whole
view (A) and a partial view (B) of Itokawa particle
RA-QD02-0049. (A) The particle is almost exclusive-
ly made of olivine with minor amounts of what are
probably plagioclase and opaque inclusions mostly
of troilite. (B) Enlarged view of the boxed area in (A)
where some troilite (Tr) inclusions are observed.
Many small pieces of silicates deposit on the surface.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1119
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

There are few data for chondritic olivine with
which our data can be compared. We used chon-
dritic olivines separated from St. Severin (LL6)
and Modoc (L6) (6, 7) for comparison. There
seems to be an apparent discrepancy in elemen-
tal abundances between the Itokawa samples and
chondritic olivine. We will discuss such a discrep-
ancy in the following text and in the SOM.
In INAA of the Itokawa samples, Fe and Sc
could be determined very reliably, because their
neutron-capturing radionuclides (
59
Fe and
46
Sc)
have relatively long half-lives, permitting repeated
measurements for long periods, and because they
emit duplicate high-intensity gamma rays. Dif-
ferentiated planetary bodies such as Earth have
undergone their own differentiation histories and
thus have different element compositions in their
silicate layers from those of chondrites. In par-
ticular, Fe/Sc ratios in silicates for differentiated
planetary bodies (Earth, the Moon, Mars, and
4Vesta) are lower than those of chondrites due
to the partitioning of some Fe into cores. Ac-
cording to partition coefficients of Fe and Sc for
silicates (8), Fe is more compatible with olivine
than Sc. Therefore, the Fe/Sc ratio for olivine
should be even higher than the whole-rock
value. Fe/Sc ratios for the Itokawa samples
(86,000 T 2700, 84,000 T 3200) are higher than
those for terrestrial olivine (19,000) (9) and
martian olivine (38,000) (10) and are similar to
those for olivine separates from ordinary
chondrites (24,000 to 85,000) (SOM). Therefore,
the Itokawa samples analyzed here are not
terrestrial, but rather extraterrestrial in origin.
Cobalt and Ni are siderophile elements and
behave similarly during cosmochemical and geo-
chemical processes. According to condensation
calculations, Co and Ni condensed from gas of
solar composition at 1351 and 1354 K, respec-
tively, under a total pressure of 10
−4
atm, which is
close to that of Fe (1336 K) (11). Because these
elements condensed into solid material at fairly
high temperatures, they are assumed not to have
fractionated from each other among chondrite
groups. In fact, bulk chondrites (12) do plot on
the CI carbonaceous chondrite (CI) line (Fig. 2)
defined by CI chondrites and on which ordinary
chondrite metal and magmatic iron meteorites
also fall. In ordinary chondrites and iron mete-
orites, siderophile elements are largely distributed
into metal grains. This can be explained in terms
of preferential partitioning or diffusion of sider-
ophile elements into a metal phase. In partition-
ing, Ni and Co are concentrated into metal phases
without fractionation from each other. Because
ordinary chondrites have the same Ni/Co ratio as
the CI value, Ni/Co ratios of ordinary chondrite
metals are similar to those for bulk samples of
ordinary chondrites (Fig. 2). On the other hand,
olivine separates from ordinary chondrites, bulk
eucrites, and Earth’s continental crust are rela-
tively depleted in Ni and Co and fall off the CI
line (Fig. 2), lying in the region where Ni is more
depleted than Co. Such a spreading of Ni and
Co can be primarily explained in terms of more
preferential partitioning of Ni into metal com-
pared with silicate than Co (8). The Itokawa sam-
ples do not plot in the region of these samples,
but instead fall on the CI line primarily defined
by the solar nebula composition (Fig. 2). Although
Co and Ni contents are variable, most chondrules
in equilibrated ordinary chondrites have a CI
chondritic Ni/Co ratio (13, 14). Grossman and
Wasson (13, 14) suggested that siderophiles such
as Ni and Co are mostly concentrated in metal
phases dispersed in chondrules. Ni and Co abun-
dances for the Itokawa samples fall within the
range for chondrules in ordinary chondrites. As
they are similar to chondrules in ordinary chon-
drites, our Itokawa samples contain 1 to 2%chon-
dritic metal if we assume that the Ni and Co
Table 1. Major, minor, and trace element contents of Itokawa, upper continental crust, and ordinary
chondrite (SOM text). Uncertainties cited inclue only counting statistics (1s) in gamma counting. The
upper half of the table is for the absolute mass of each element. In calculating the elemental abun-
dances given in the lower half of the table, we assume that grain is composed of olivine with plagioclase
and metal, and we obtain 1.66 and 1.68 mg for RA-QD02-0049-1 and RA-QD02-0049-2, respectively
(SOM text). ppm, parts per million; N/A, not applilcable.
Element
Itokawa Upper
continental
crust*
Olivine in
ordinary
chondrite†
No. 49-1 No. 49-2
Na (ng) 2.14 T 0.07 1.56 T 0.08 N/A N/A
Sc (ng) 0.0039 T 0.0001 0.0040 T 0.0001 N/A N/A
Cr (ng) 0.061 T 0.003 0.076 T 0.002 N/A N/A
Fe (ng) 347 T 6 350 T 6 N/A N/A
Co (ng) 0.29 T 0.01 0.23 T 0.01 N/A N/A
Ni (ng) 6.65 T 0.16 5.33 T 0.15 N/A N/A
Zn (ng) 0.07 T 0.02 0.067 T 0.017 N/A N/A
Se (ng) <0.012 <0.013 N/A N/A
Sm (ng) <26 <48 N/A N/A
Eu (ng) <410 <410 N/A N/A
Hf (ng) <0.0022 <0.0025 N/A N/A
Ir (pg) 0.031 T 0.007 0.037 T 0.012 N/A N/A
Estimated mass
(mg)
1.66 1.67 N/A N/A
Na (ppm) 1290 T 50 931 T 48 24300 N/A
Sc (ppm) 2.3 T 0.1 2.4 T 0.1 14.0 2.1–8.7
Cr (ppm) 37 T 2 45 T 1 92 160–600
Fe (%) 20.9 T 0.4 20.9 T 0.4 3.92 14.17–20.79
Co (ppm) 176 T 3 140 T 3 17.3 7–47
Ni (%) 0.40 T 0.01 0.32 T 0.01 0.0047 0.010–0.029
Zn (ppm) 45 T 12 40 T 10 67 19–20
Se (ppm) <7.2 <7.8 0.09 N/A
Sm (ppm) <16 <29 4700 0.20–0.26
Eu (ppm) <250 <250 1.0 0.03
Hf (ppm) <1.3 <1.5 5.3 N/A
Ir (ppb) 19 T 4 22 T 7 0.022 N/A
Ni/Co 23 T 1 23 T 1 2.7 5.5–14.3
Fe/Sc 86000 T 2700 84000 T 3200 2800 24000–87000
*Data from (16). †Data from (6, 7, 17)
Fig. 2. Correlation of Co and Ni contents
in several astromaterials. The black solid
line indicates the Co/Ni ratio defined by CI
chondrites. Data for magmatic iron mete-
orite, metal in ordinary chondrite, olivine in
ordinary chondrite, chondrule fromordinary
chondrite, bulk chondrite, continental crust,
and eucrite are plotted for comparison.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1120
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

contents of their olivine and metal phases are
equal to those in olivine separates from St.
Severin (LL6) and Modoc (L6) (6, 7) and to
those in metal separates from H, L, and LL
chondrites (15), respectively. Ni/Co ratios for the
Itokawa samples clearly indicate that they
preserve the primitive elemental abundances of
the early solar system.
Samples RA-QD02-0049-1 and RA-QD02-
0049-2 contain 31 T 7 and 37 T 12 fg (10
−15
g) of
Ir, respectively (Table 1), their Ir abundances
are 19 T 4 and 22 T 7 parts per billion ( ppb),
respectively. Because Ir is more siderophile than
Ni and Co, it must be totally present in the metal
phase. Bulk samples of ordinary chondrites have
compositions close to CI, showing no apparent
fractionation among the three elements in chon-
dritic meteorites (Fig. 3). Slightly lower values
of Ir/Co and Ir/Ni ratios for metal separates from
ordinary chondrites (bulk metals) suggest prefer-
ential partitioning of Ni and Co into nonmetal
phases compared with Ir. Even though Ir in the
Itokawa samples is assumed to be in the metal
phase, the Itokawa metals are relatively depleted
in Ir compared with Ni and Co (Fig. 3). Such a
depletion of Ir relative to Co and Ni was observed
in chondrule separates from ordinary chondrites
(13, 14); chondrule data disperse along the line
tying the Itokawa metal and bulk chondrites (and
their metal separates) (Fig. 3). Because element
abundances of siderophile elements (Ir, Co, and
Ni) in the two Itokawa samples (RA-QD02-0049-1
and RA-QD02-0049-2) are similar to each other,
such metal phases must be small in size and dis-
persed. However, it cannot be ruled out that metal
grains are adhering to the surface of the grains
rather than contained within the grains, although
this is unlikely because the grains were washed
twice with ethanol, and no adhering metal grains
were observed in the preliminary SEM analysis
of the sample. Regardless, it is clear that Ir-depleted
metals are present in the returned Itokawa samples.
The elemental compositions of metal sepa-
rates fromordinary chondrites were systematically
studied, and the relative abundances of sidero-
phile elements such as Co, Ni, and Ir were ob-
served to be near CI values (13, 14), as shown in
Figs. 2 and 3. Based on elemental compositions of
metals in ordinary chondrites, Kong and Ebihara
(15) proposed that chondritic metals and chon-
drules were possibly produced by melting of their
precursor materials. Coexistence of liquid metal
and silicate and subsequent segregation of these
two materials produce slight enrichments of Ni
over Co, Ir over Co, and of Ni in chondritic metals
compared with CI values, which are clearly visi-
ble in Figs. 2 and 3. Although this scenario can
explain most of the metals in ordinary chondrites,
it cannot explain them all. For example, small
metal grains in chondrules in ordinary chondrites,
whose data are dispersed in Fig. 3, are one case
in point, and so are the Itokawa metals. Iridium
is a highly refractory element, as implied by its
condensation temperature of 1610 K at 10
−4
atm
(11). Considering that the metal phase in the
Itokawa samples analyzed in this study is highly
depleted in Ir compared with Ni and Co, it must
have condensed from a fractionated nebula gas
where refractory siderophiles such as Ir had al-
ready condensed and been removed.
References and Notes
1. A. Fujiwara et al., Science 312, 1330 (2006).
2. M. Abe et al., Science 312, 1334 (2006).
3. T. Okada et al., Science 312, 1338 (2006).
4. H. Yano et al., Science 312, 1350 (2006).
5. T. Nakamura et al., Science 333, 1113 (2011).
6. B. Mason, A. L. Graham, Smithson. Contrib. Earth Sci. 3,
1 (1970).
7. R. O. Allen Jr., B. Mason, Geochim. Cosmochim. Acta 37,
1435 (1973).
8. J. H. Jones, in Experimental Trace Element Partitioning,
Rock Physics and Phase Relations: A Handbook of
Physical Constants, T. J. Ahrens, Ed. (American
Geophysical Union, Washington, DC, 1995), pp. 73–104.
9. S. M. Eggins, R. L. Rudnick, W. F. McDonough, Earth
Planet. Sci. Lett. 154, 53 (1998).
10. C. Meyer Jr., The Mars Meteorite Compendium,
(Johnson Space Center #27672 revision, NASA Johnson
Space Center, Houston, TX, 2008);
http://curator.jsc.nasa.gov/antmet/mmc/index.cfm.
11. J. T. Wasson, in Meteorites-Their Record of Early
Solar-System History (W. H Freeman and Company,
New York, 1985), pp. 1–267.
12. J. T. Wasson, G. W. Kallemeyn, Philos. Trans. R. Sco.
London Ser. A 325, 535 (1988).
13. J. N. Grossman, J. T. Wasson, Geochim. Cosmochim. Acta
46, 1081 (1982).
14. J. N. Grossman, J. T. Wasson, Geochim. Cosmochim. Acta
49, 925 (1985).
15. P. Kong, M. Ebihara, Geochim. Cosmochim. Acta 61,
2317 (1997).
16. R. L. Rudnick, S. Gao, in Composition of the Continental
Crust, Treatise on Geochemistry, vol. 3, H. D. Hooland,
K. K. Turekian, Eds. (Elsevier, Amsterdam, 2003), pp. 1–64.
17. A. E. Rubin, Geochim. Cosmochim. Acta 54, 1217
(1990).
Acknowledgments: This work was supported in part by a
grant-in-aid defrayed by the Ministry of Education,
Culture, Science and Technology of Japan. M.E.Z. was
supported by NASA.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1119/DC1
Materials and Methods
SOM Text
Figs. S1 to S4
Tables S1 to S3
References (18–23)
3 May 2011; accepted 2 August 2011
10.1126/science.1207865
Incipient Space Weathering Observed
on the Surface of Itokawa Dust Particles
T. Noguchi,
1
* T. Nakamura,
2
M. Kimura,
1
M. E. Zolensky,
3
M. Tanaka,
4
T. Hashimoto,
5
M. Konno,
5
A. Nakato,
2
T. Ogami,
2
A. Fujimura,
6
M. Abe,
6
T. Yada,
6
T. Mukai,
6
M. Ueno,
6
T. Okada,
6
K. Shirai,
6
Y. Ishibashi,
6
R. Okazaki
7
The reflectance spectra of the most abundant meteorites, ordinary chondrites, are different from those
of the abundant S-type (mnemonic for siliceous) asteroids. This discrepancy has been thought to be
due to space weathering, which is an alteration of the surfaces of airless bodies exposed to the space
environment. Here we report evidence of space weathering on particles returned from the S-type asteroid
25143 Itokawa by the Hayabusa spacecraft. Surface modification was found in 5 out of 10 particles,
which varies depending on mineral species. Sulfur-bearing Fe-rich nanoparticles exist in a thin (5 to 15
nanometers) surface layer on olivine, low-Ca pyroxene, and plagioclase, which is suggestive of vapor
deposition. Sulfur-free Fe-rich nanoparticles exist deeper inside (<60 nanometers) ferromagnesian
silicates. Their texture suggests formation by metamictization and in situ reduction of Fe
2+
.
T
he surfaces of airless bodies exposed to
interplanetary space gradually have their
structures, optical properties, chemical com-
positions, and mineralogy changed by solar wind
implantation and sputtering, irradiation by galac-
tic and solar cosmic rays, and micrometeorite
Fig. 3. Relation among Ir, Co, and Ni in
bulk chondrites, metal separates fromchon-
drites (bulk metal), and chondrules from
ordinary chondrites. A broad red band for
Itokawa metal designates the range of es-
timated Ni and Co contents in metal, which
depend on the Ni and Co contents in oli-
vine separates and bulk metals in L and LL
chondrites.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1121
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

bombardment. These alteration processes and the
resultant optical changes are known as space
weathering (1–3). Our knowledge of it depends
almost entirely on studies of the surface materials
returned from the Moon (1, 4), with the addition
of recent analyses of regolith breccia meteorites
(5). Space weathering reduces (darkens) the al-
bedo of lunar soil and regolith (6), steepens the
slopes of their reflectance spectra (reddening),
and attenuates the characteristic absorption bands
of their reflectance spectra (1, 7). It is caused by
vapor deposition of small (<40 nm) metallic Fe
nanoparticles within the grain rims of lunar soils
(5, 7–11) and agglutinates (12).
Spectra of near-Earth asteroids suggest that
ordinary chondrite-like spectra change to S-type
(mnemonic for siliceous) ones by space weather-
ing (13, 14). The Galileo spacecraft observed
ongoing space weathering on the S-type aste-
roids Ida and its satellite Dactyl (15). Laboratory
studies apparently succeeded in simulating space
weathering observed on asteroids (2). Labora-
tory and astronomical studies indicate that
asteroid spectra are modified in their early history
by solar wind irradiation operating on unexpect-
edly short time scales of 10
4
to 10
6
years, and
later by micrometeoroid bombardment operating
on longer time scales of 10
8
to 10
9
years (16–19).
The Hayabusa spacecraft touched down at
and lifted off from the smooth terrain of the
MUSES-C Regio on asteroid 25143 Itokawa on
19 and 25 November 2005 UTC (20). Although
pebbles (debris ranging from 4 mm to 6.4 cm in
diameter) that cover the regio uniformly (20, 21)
appeared to lack a distinct powdery covering,
Hayabusa successfully collected submillimeter
particles from the regio (22). Because no im-
pactors were fired by the sampling mechanism
(21), the collected samples are unlikely to be frag-
ments formed during sampling and are thus likely
to retain their original surfaces. In addition, Itokawa
shows space-weathered S-type spectra (23, 24),
and the Itokawa particles have LL chondrite-like
mineralogy (22). Therefore, the particles are ideal
for investigation of asteroidal space weathering.
Ten Itokawa particles (average diameter
52 mm) were embedded in epoxy resin and ultra-
microtomed into ~100-nm-thick sections. We used
high-angle annular dark-field scanning transmis-
sion electron microscope (HAADF–STEM) im-
1
College of Science, Ibaraki University, 2-1-1 Bunkyo, Mito,
Ibaraki 310-8512, Japan.
2
Department of Earth Science, Grad-
uate School of Science, Tohoku University, 6-3 Aoba, Aramaki,
Aoba-ku, Sendai 980-8578, Japan.
3
Astromaterials Research
and Exploration Science Directorate, NASA, Johnson Space
Center, 2101 NASA Parkway, Houston, TX 77058, USA.
4
Syn-
chrotron X-ray Station at SPring-8, National Institute for
Materials Science, 1-1-1 Kouto, Sayo, Hyogo 679-5148, Japan.
5
Hitachi High-Technologies, 882 Ichige, Hitachinaka, Ibaraki
312-8504 Japan.
6
Institute of Space and Astronautical Science,
Japan Aerospace Exploration Agency, 3-1-1 Yoshinodai, Chuo-
ku, Sagamihara, Kanagawa 252-5210, Japan.
7
Department of
Earth and Planetary Science, Kyushu University, Hakozaki,
Fukuoka 812-8581, Japan.
*To whom correspondence should be addressed. E-mail:
tngc@mx.ibaraki.ac.jp
Fig. 1. (A) Backscattered electron
images of a space-weathered par-
ticle (RA-QD02-0035) and (B) a
space-weathered lunar soil par-
ticle (15004, 194) for compari-
son. Abbreviations: olivine, Ol;
plagioclase, Pl; high-Ca pyroxene,
HPx; vesicle, V. (C) HAADF-STEM
images of nanoparticle-bearing
rims on the surface of olivine in
RA-QD02-0041and(D) aninclusion-
richrimon the surface of glass in
15004, 194. Because the surface
of the ultrathin section in (C) is
not edge-on, a slanted surface is
shown. Sigmoidal cracks in (C)
and parallel cracks in the inset
of (D) are artifacts formed during
ultramicrotomy.
Fig. 2. Edge-on BF-
STEMand HAADF-STEM
images of (A and B) ol-
ivine in RA-QD02-0041,
(C and D) low-Ca pyrox-
ene in RA-QD02-0042,
and (E and F) plagio-
clase in RA-QD02-0042.
The rims of these min-
erals are divided into
three zones based on
their texture: zone I,
amorphous surface lay-
er containing npFe; zone
II, partially amorphized
area; zone III, crystal-
line substrate minerals.
Zone II of olivine and
pyroxene contains abun-
dant npFe. The bound-
aries between zones
are indicated by dotted
curves.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1122
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

ages, in which Fe-rich nanoparticles appear as
bright spots (25). Figure 1, A and B, show back-
scattered electron images of an Itokawa particle
and a lunar soil grain, both of which have Fe-rich
nanoparticle–bearing (npFe-bearing) rims. The
npFe-bearing rims are found on five of the
Itokawa particles (Fig. 1C). The thickness of
the rims ranges from 30 to 60 nm, which over-
laps with the typical thickness of those on lunar
soil grains (9). The average grain size of the
npFe’s in the Itokawa particles (~2 nm; fig. S3) is
similar to that of lunar soils (~3 nm) (7, 9, 12),
although the lunar soil contains larger npFe’s
(5 to 10 nm across) (Figs. 1D; 3S).
Cross sections of olivine, low-Ca pyroxene,
and plagioclase showthat they have a 5- to 15-nm-
thick amorphous surface layer containing a dense-
ly arranged layer of npFe’s (1 to 2 nm) within or
at the bottom of the layer (zone I in Fig. 2).
Below zone I, the texture of the rim depends on
whether host minerals contain abundant Fe
2+
or
not. Ferromagnesian silicates have a npFe-rich
zone with variable thickness from 20 to 50 nm
(zone II in Fig. 2, A to D). In contrast to lunar
npFe-bearing rims, in which npFe’s are em-
bedded in amorphous material (5, 7, 9, 12), in
Itokawa grains the host minerals in zone II show
various degrees of amorphization (Fig. 3, A
and B). Zone II gives way to the well-crystalline
minerals in zone III. Although lattice fringes of
the substrate minerals hamper the identification
of npFe’s in bright-field (BF)–STEM images,
HAADF-STEM images suggest that the bright
npFe’s are enclosed in darker material (composed
of lighter elements), which can be recognized in
Fig. 2D. On the contrary, in zone II of plagioclase,
no npFe’s were observed (Fig. 2, Eand F). There-
fore, when plagioclase is adjacent to olivine,
only the olivine rim has the npFe-rich zone II
(Fig. 3, C and D). Because plagioclase was rap-
idly damaged during analysis by the intensely
focused electron beam, we could not observe the
texture of amorphization of plagioclase in zone
II and the boundary between zones II and III
(Fig. 2, E and F).
Elemental distribution maps clearly showthat
the zone I areas of olivine and low-Ca pyroxene
are enriched in Fe, S, and Mg, and depleted in Si
(Fig. 4). As seen by comparing Figs. 2 and 4, the
npFe’s in zone I are enriched in S, Fe, and prob-
ably Mg. In this zone, small amounts of elements
that are not included in the substrate minerals
were ubiquitously detected. Sulfur, Al, Na, and K
were detected on ferromagnesian silicates; Mg,
Fe, and S on plagioclase; and Na, Mg, Si, K on
troilite. On the other hand, abundant npFe’s (1 to
3 nm) in zone II do not contain sulfur. With the
exception of zone I, the bulk chemical composi-
tions of the rims are similar to those of their
substrate minerals (Figs. 2 and 4, and fig. S4).
Sulfur-bearing npFe’s in zone I show 0.22- to
0.23-nm lattice fringes (Fig. 3E), which is not
consistent with those of troilite and pyrrhotite.
This discrepancy may be related to the presence
of Mg accompanied by the npFe’s in zone I. On
the other hand, S-free npFe’s in zone II have
0.20-nm lattice fringes, which is consistent with
the spacing of (110) of a-Fe (d
110
= 0.203 nm)
(Fig. 2D). The surface of the rim contains npFe
sulfide (npFeS), and the deep (>10 to 15 nm) in-
terior of the rim contains npFe metal (npFe
0
).
Studies of lunar space weathering have sug-
gested that abundant soil particles in the 10- to
100-mm range are required to form sufficiently
large numbers of npFe
0
’s to affect the spectra (7).
Because abundant fine-grained regolith did not
exist on the MUSES-C Regio, the major lunar
process operating to formabundant npFe’s on the
Moon is ineffective on Itokawa. The high albedo
of Itokawa (0.23), within the range of that of
S-type asteroids (0.11 to 0.22) (26), suggests poor
development of npFe’s. However, Itokawa is as
red as large main-belt S-type asteroids (26). Be-
cause small npFe’s (<10 nm) redden the spectra,
and only large ones (especially >50 nm) darken
Fig. 3. High resolution BF-STEM images of three zones in (A) olivine and (B) low-Ca pyroxene shown in
Fig. 2. They were obtained at the boundary between zones I and II, within II, and at the transitional area
fromzone II to III. (C and D) BF-STEMand HAADF-STEMimages showing the boundary between olivine and
plagioclase in RA-QD02-0042. The boundary is indicated by a dotted curve. High-resolution BF-STEMimages
of (E) an S-bearing npFe in zone I of RA-QD02-0042 and (F) an S-free npFe in zone II of RA-QD02-054. The
former shows 0.22- to 0.23-nm lattice fringes and the latter 0.20-nm lattice fringes.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1123
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

the spectra (10, 11, 27, 28), it is likely that the
small npFe’s (~2 nm) in the Itokawa particles
efficiently redden the spectra.
Asteroid studies indicate that their surfaces
do not necessarily experience lunar-style space
weathering (17, 29, 30). This is attributable to
variations in surface composition, differences in
micrometeoroid flux and velocities, solar wind
particle flux, and the porosity and grain size of
the regolith (12). Elements found in the outer-
most surface (zone I) can be derived from major
minerals of Itokawa particles (ferromagnesian
silicates, plagioclase, and troilite) (22). Zone I is
likely to represent vapor recondensation products
attributed to neighboring minerals. Laboratory
laser irradiation of ordinary chondrites, which
simulate micrometeorite bombardment, forms
an S-rich surface layer (31). A thin npFeS-
bearing layer could have been formed by both
vapor deposition attributable to micrometeorite
bombardment (31) and solar wind irradiation and
sputter deposition (1). However, it cannot pres-
ently be determined which process dominates.
Below zone I (15 to 60 nm from the surface),
the structure of the rims clearly depends on wheth-
er the substrate minerals contain abundant Fe
2+
.
The texture of partially amorphized ferromagnesian
silicates in zone II is similar to that of radiation-
damaged zircon (32). Although radiation-induced
amorphization has been proposed as the mech-
anism of asteroidal space weathering (30), our
observations suggest that not only amorphization
but also contemporaneous reduction of Fe
2+
to
Fe
0
by deeply implanted solar wind ions are re-
quired to form npFe
0
’s in zone II.
Multilayered rims with structure similar to
that of Itokawa ferromagnesian silicates have been
identified on ilmenite in lunar soils (33). Because
abundant solar wind protons that have ~1 keV/
atomic mass unit kinetic energy (34) penetrate
only the very thin surface to cause sputtering and
deposition (1), heavy ions and/or solar flare pro-
tons may contribute to the formation of npFe
0
’s
deeper in the rims of the lunar ilmenite (~50 nm)
(33) and of zone II of the Itokawa ferromagnesian
silicates (~60 nm). Noble gas study of the Itokawa
particles shows that the degree of noble gas im-
plantation is variable between particles (35), pos-
sibly related to the presence or absence of the
npFe-bearing rims. In conclusion, (i) the optical
properties of asteroids are altered by space weath-
ering that produces npFe
0
’s and npFeS’s, and (ii)
the structure of the rims on ferromagnesian sili-
cates suggests that npFeS’s near the surfaces are
vapor deposits, and npFe
0
’s existing deep in-
side the rims were formed by radiation-induced
amorphization and in situ reduction of Fe at-
tribute to solar wind irradiation.
References and Notes
1. B. Hapke, J. Geophys. Res. 106, 10039 (2001).
2. B. E. Clark et al., in Asteroids III, W. F. Bottke Jr.,
A. Cellino, P. Paolicchi, R. P. Binzel, Eds. (Univ. of Arizona
Press, Tucson, AZ, 2002), pp. 255–271.
3. C. R. Chapman, Annu. Rev. Earth Planet. Sci. 32, 539
(2004).
4. L. A. Taylor, C. M. Pieters, L. P. Keller, R. V. Morris,
D. S. McKay, J. Geophys. Res. Planets 106, 985 (2001).
5. S. K. Noble, L. P. Keller, C. M. Pieters, Meteorit. Planet.
Sci. 45, 2007 (2011).
6. The surface of the Moon is covered by unconsolidated
debris referred to as the lunar regolith. The fine-grained
fraction of the regolith is called lunar soil. Irregularly
shaped clusters composed of a mixture of lithic fragments
and interstitial glass in lunar soil are called agglutinates.
7. C. M. Pieters et al., Meteorit. Planet. Sci. 35, 1101
(2000).
8. B. Hapke, W. Cassidy, E. Wells, Moon 13, 339 (1975).
9. L. P. Keller, D. S. McKay, Geochim. Cosmochim. Acta 61,
2331 (1997).
10. L. P. Keller, S. J. Wentworth, D. S. McKay, Lunar Planet. Sci.
XXXIX, 1762 (1998).
11. S. K. Noble, C. M. Pieters, L. P. Keller, Icarus 192, 629
(2007).
12. S. K. Noble, L. P. Keller, C. M. Pieters, Meteorit. Planet.
Sci. 40, 397 (2005).
13. S-type (or S-class) asteroids are characterized by
moderate albedos and reddish-sloped spectra with
moderate absorption features at 1 and 2 mm, which
correspond to olivine and pyroxene absorption bands.
They are most abundant among both the inner asteroid
belt and the near-Earth asteroids.
14. R. P. Binzel, S. J. Bus, T. H. Burbine, J. M. Sunshine,
Science 273, 946 (1996).
15. C. R. Chapman, Meteorit. Planet. Sci. 31, 699 (1996).
16. S. Sasaki, K. Nakamura, Y. Hamabe, E. Kurahashi,
T. Hiroi, Nature 410, 555 (2001).
17. P. Vernazza, R. P. Binzel, A. Rossi, M. Fulchignoni,
M. Birlan, Nature 458, 993 (2009).
18. M. Willman et al., Icarus 195, 663 (2008).
19. D. Nesvorny, D. Vokrouhlicky, W. F. Bottke, M. Sykes,
Icarus 181, 107 (2006).
20. H. Yano et al., Science 312, 1350 (2006).
21. H. Miyamoto et al., Science 316, 1011 (2007).
22. T. Nakamura et al., Science 333, 1113 (2011).
23. M. Abe et al., Science 312, 1334 (2006).
24. T. Hiroi et al., Nature 443, 56 (2006).
25. Materials and methods are available as supporting
material on Science Online.
26. J. E. Thomas-Osip et al., Earth Planets Space 60, 39
(2008).
27. L. P. Keller, S. J. Clemett, Lunar Planet. Sci. XXXII, 2097
(2001).
28. P. G. Lucey, M. A. Riner, Icarus 212, 451 (2011).
29. S. M. Murchie et al., Icarus 155, 145 (2002).
30. G. Strazzulla et al., Icarus 174, 31 (2005).
31. S. K. Noble et al., Lunar Planet. Sci. 42, 1382 (2011).
32. R. C. Ewing, A. Meldrum, L. Wang, W. J. Weber,
L. R. Corrales, in Zircon, J. M. Hanchar, P. W. O. Hoskin,
Eds. (Mineralogical Society of America, Chantilly, VA,
2003), pp. 387–425.
33. S. Zhang, L. P. Keller, Lunar Planet. Sci. 41, 1432 (2010).
34. M. J. Loeffler, C. A. Dukes, R. A. Baragiola, J. Geophys.
Res. Planets 114, E3003 (2009).
35. K. Nagao et al., Science 333, 1128 (2011).
Acknowledgments: Special thanks to the Hayabusa project
team for sample return. We are grateful to Y. Suzuki,
S. Tsujimoto, R. Sagae, R. Hinoki, and M. Kawamoto for
Fig. 4. Elemental distribution maps of nanoparticle-bearing rims on (A) olivine and (B) low-Ca pyroxene
shown in Fig. 2. In both minerals, the very surface (<10 to 15 nmthick) of these minerals is enriched in S.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1124
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

supporting N
2
purge sample preparation at the Institute
of Space and Astronautical Sciences of the Japan
Aerospace Exploration Agency and Ibaraki University,
and for STEM observation at Hitachi High-Technologies.
Thanks to H. Hidaka, we could compare the rims of
the Itokawa particles and those of space-weathered
lunar soil. M. Zolensky was supported by NASA’s
Muses-C/Hayabusa Program.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1121/DC1
Materials and Methods
Figs. S1 to S4
References (36–39)
2 May 2011; accepted 2 August 2011
10.1126/science.1207794
Three-Dimensional Structure of
Hayabusa Samples: Origin and
Evolution of Itokawa Regolith
Akira Tsuchiyama,
1
* Masayuki Uesugi,
2
Takashi Matsushima,
3
Tatsuhiro Michikami,
4
Toshihiko Kadono,
5
Tomoki Nakamura,
6
Kentaro Uesugi,
7
Tsukasa Nakano,
8
Scott A. Sandford,
9
Ryo Noguchi,
1
Toru Matsumoto,
1
Junya Matsuno,
1
Takashi Nagano,
1
Yuta Imai,
1
Akihisa Takeuchi,
7
Yoshio Suzuki,
7
Toshihiro Ogami,
6
Jun Katagiri,
3
Mitsuru Ebihara,
10
Trevor R. Ireland,
11
Fumio Kitajima,
12
Keisuke Nagao,
13
Hiroshi Naraoka,
12
Takaaki Noguchi,
14
Ryuji Okazaki,
12
Hisayoshi Yurimoto,
15
Michael E. Zolensky,
16
Toshifumi Mukai,
2
Masanao Abe,
2
Toru Yada,
2
Akio Fujimura,
2
Makoto Yoshikawa,
2
Junichiro Kawaguchi
2
Regolith particles on the asteroid Itokawa were recovered by the Hayabusa mission. Their
three-dimensional (3D) structure and other properties, revealed by x-ray microtomography, provide
information on regolith formation. Modal abundances of minerals, bulk density (3.4 grams per cubic
centimeter), and the 3D textures indicate that the particles represent a mixture of equilibrated and
less-equilibrated LL chondrite materials. Evidence for melting was not seen on any of the particles.
Some particles have rounded edges. Overall, the particles’ size and shape are different from those seen
in particles from the lunar regolith. These features suggest that meteoroid impacts on the asteroid surface
primarily form much of the regolith particle, and that seismic-induced grain motion in the smooth
terrain abrades them over time.
T
he Hayabusa mission recovered at least
1534 particles from the smooth terrain
of MUSES-C Regio on asteroid 25143
Itokawa (1, 2). These grains are less than a few
hundred micrometers in diameter (3). These sam-
ples can be compared with the other extraterres-
trial regolith to have been sampled, that of the
Moon, which was sampled by the Apollo and
Luna missions (4).
It is accepted that most meteorites originate
from asteroids, as, for example, demonstrated
by orbital determination from observed meteorite
falls. Ground-based telescope observation (5) and
remote-sensing observation by the Hayabusa
spacecraft (6) indicate that the materials on S
(IV)–type asteroid Itokawa are similar to LL5 or
LL6 chondrites. Itokawa samples allow a direct
validation of the relation between asteroidal and
meteorite mineralogy. In addition, the properties
of Itokawa particles allow studies of regolith
formation on an asteroid. Here, we describe the
three-dimensional (3D) structures of Itokawa par-
ticles by using x-ray microtomography to under-
stand their textures, mineralogy, and shapes in
comparison with those seen in meteorites and
in lunar samples and use the results to infer how
Itokawa’s regolith formed.
Our imaging tomography experiments were
made at beamline BL47XU of SPring-8 (7). We
obtained 40 particles ranging from 30 to 180 mm
in size from sample catcher A by tapping it
(tapping samples). These particles were collected
during the spacecraft’s second touchdown on the
asteroid (2). These particles were imaged with
effective spatial resolutions of ~200 or ~500 nm,
which is sufficient for comparisons with the tex-
tures of ordinary chondrites. Successive 3D com-
puted tomography (CT) images (fig. S1), which
show quantitative 3D mineral distribution, were
obtained. Mineralogy of the particle was derived
by comparing a set of CT images taken at dual
x-ray energies (7 and 8 keV) (fig. S2). In addi-
tion to the tapping samples, we swept 1534 par-
ticles from sample catcher A surface by using a
Teflon (Dupont) spatula (spatula samples). We used
a scanning electron microscope (SEM) to measure
the size of 1469 particles larger than 0.5 mm (3).
The CT images of different particles show
substantial textural variations (Fig. 1). None of
the textures is consistent with in situ melting
caused by the impact of meteoroids, such as
seen in the lunar agglutinates. The textures are
also different from those of porous interplanetary
dust particles and Stardust particles of cometary
origin. Eighteen of the 40 particles are polyminer-
alic, mainly composed of olivine, low-Ca pyroxene,
high-Ca pyroxene, plagioclase, and/or troilite
(e.g., Fig. 1A), whereas 22 particles are almost
[>~80 volume (vol.) %] monomineralic and are
dominated by olivine (Fig. 1B), low-Ca pyroxene
(Fig. 1C), or plagioclase. Most of the particles
seem to have equilibrated chondritic textures, sug-
gesting thermal metamorphism ( petrologic type
of 5 and/or 6), whereas a few of them have less-
equilibrated textures, such as a chondrule frag-
ment where mesostasis and pyroxene with Ca
zoning are seen (Fig. 1D). Some less-equilibrated
materials are also present (3), suggesting that the
Itokawa material is representative of a breccia.
Voids, both spherical and elongated, are common
in 15 of the particles (Fig. 1B). Seven particles
contain cracks showing partially healed impact-
generated fractures (Fig. 1A).
The total volume of the 40 particles obtained
from CT image analysis (7) is 4.23 × 10
6
mm
3
.
This corresponds to a sphere 201 mm in diameter
(typical chondrule diameter in LL chondrites is
~900 mm). The modal mineral abundances (vol.
%) of the entire 40-particle sample were obtained
from the relative volumes of crystalline minerals
in the 3Ddata of individual particles (64%olivine,
19% low-Ca pyroxene, 3% high-Ca pyroxene,
11% plagioclase, 2% troilite, ~0.02% kamacite,
~0.2%taenite, ~0.1%chromite, and ~0.01% Ca
phosphates). These are similar to those of LL
chondrites, although the abundance of troilite and
metals is ~2% lower than those in the average
LL chondrite (8) (table S1). The chemical and
oxygen isotope data of Itokawa materials (3, 9)
also show a similarity to LL chondrites.
The porosities of individual particles range
from0 to 11%. The mean porosity of 1.4%is less
than the average porosity of LL chondrites (8.2 T
5.5%) (Tstandard deviation) (10). This is because
most porosity in LL chondrites is in cracks
between grains, and these cracks are not repre-
sented in the Itokawa samples because of a size
effect: LL chondrites are bigger than the particles
1
Department of Earth and Space Science, Osaka University,
Toyonaka, 560-0043, Japan.
2
Japan Aerospace Exploration
Agency, Sagamihara, 252-5210, Japan.
3
Department of En-
gineering Mechanics and Energy, Graduate School of Systems
and Information Engineering, University of Tsukuba, 1-1-1
Tennodai, Tsukuba, Ibaraki, 305-8573, Japan.
4
Fukushima
National College of Technology, Iwaki, Fukushima, 970-8034,
Japan.
5
Institute of Laser Engineering, Osaka University, 2-6
Yamadagaoka, Suita, 565-0871, Japan.
6
Department of Earth
and Planetary Material Sciences, Tohoku University, Aoba-ku,
Sendai, 980-8578, Japan.
7
Japan Synchrotron Radiation Re-
search Institute ( JASRI)/SPring-8, Sayo, Hyogo, 679-5198, Japan.
8
Geological Survey of Japan, National Institute of Advanced
Industrial Science and Technology, Tsukuba, 305-8568, Japan.
9
NASA Ames Research Center, Moffett Field, CA 94035, USA.
10
Graduate School of Science and Engineering, Tokyo Metro-
politan University, Hachioji, 192-0397 Japan.
11
Research School
of Earth Sciences, Australian National University, Canberra, ACT
0200, Australia.
12
Department of Earth and Planetary Sciences,
Kyushu University, Higashi-ku, Fukuoka, 812-8581, Japan.
13
Geochemical ResearchCenter, University of Tokyo, Bunkyo-ku,
Tokyo, 113-0033, Japan.
14
College of Science at Ibaraki Uni-
versity, Mito, 310-0056, Japan.
15
Natural History Sciences,
Hokkaido University, Kita-ku, Sapporo, 060-0808, Japan.
16
NASA
Johnson Space Center, Houston, TX 77058, USA.
*To whom correspondence should be addressed. E-mail:
akira@ess.sci.osaka-u.ac.jp
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1125
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

analyzed here. The mass of each mineral (table
S1) was obtained from the modal abundances,
and the density calculated from its mean chem-
ical compositions (3). The total mass of our ex-
amined particles is 14.5 mg. From the mineral
mass and the porosity, we obtained an average
density of 3.4 g/cm
3
. This corresponds to grain
density and is comparable to the measured grain
density of LL chondrites (3.54 T 0.13 g/cm
3
) (10).
If the collected sample is representative of Itokawa
and has the average porosity of LL chondrites, its
bulk density would be 3.1 T 0.2 g/cm
3
. The
macroporosity of Itokawa would then be 39 T 6%
on the basis of the bulk density of Itokawa (1.9 T
0.13 g/cm
3
) (1). This is consistent with a rubble-
pile asteroid model of Itokawa (1).
The sphere-equivalent diameters of tapping
sample particles calculated from their volumes
range from 14 to 114 mm (median 36.8 mm),
whereas the diameters of spatula sample particles
range from 0.5 to 32 mm (median 3.5 mm). These
particles are smaller than the size-sorted, mm- to
cm-sized particles observed in close-up images of
MUSES-C Regio (2). The mm- to cm-sized parti-
cles were not comminuted by the pressure of the
spacecraft during touchdown [~0.02 MPa (7)] if
they are coherent (11), and the collected small
particles should be original regolith particles from
the smooth terrain. Three sampling mechanisms are
possible (7): (i) impact by the sampler horn, (ii)
electrostatic interaction between charged particles
and possibly charged sampler horn, and (iii) le-
vitation by thruster jets fromthe ascending space-
craft. Some mechanisms may have caused some
size sorting. However, because details concerning
the touchdown conditions are not known, we
cannot specify the mechanism(s) and such effect.
The cumulative size distribution of the
tapping samples has a log slope of about –2 in
the range of 30 to 100 mm (Fig. 2). Large par-
ticles might have been selectively picked up
from the tapping samples. The spatula samples
have a log slope of –2.8 in the range of 5 to 20 mm
(Fig. 2). However, sweeping by the spatula
would have pulverized some of the particles,
and the slope is an upper limit to the original
slope. Thus, the slope for the fine particles (~5
to 100 mm) in the smooth terrain should be
shallower than –2.8 and probably around –2. This
slope is shallower than that of Itokawa boulders
of 5 to 30 m (–3.1 T 0.1) (12). If transition of the
slope from about –3 to –2 occurs at the mm- to
cm-sized region, then we can explain the obser-
vation of abundant mm- to cm-sized regolith
(2). The lack of mm-sized particle in the Hayabusa
samples might be explained by a small probability
of collecting these small particles. However, the
possibility of size selection biases during sampling
from Itokawa or agglomerations of small particles
(11) cannot be excluded. In contrast, abundant
sub-mm regolith powder was observed on the
Moon, and the size distribution from lunar sam-
ples has a steep slope (–3.1 to –3.3 in the range
of 20 to 500 mm) (4), suggesting repeated frag-
mentation on this relatively large celestial body.
The lower abundance of ~10- to 100-mm par-
ticles in the smooth terrain can potentially be
explained by (i) smaller grains having higher
ejection velocity and therefore higher loss rates
from Itokawa after impacts (13), (ii) selective
electrostatic levitation of smaller grains (13), and/or
(iii) size-dependent segregation by vibration [the
Brazil-nut effect (14)].
Figure 3A shows the shape distributions of the
tapping samples. The mean b/a and c/a ratios are
0.71 T 0.13 and 0.43 T 0.14, respectively [a, b,
and c are longest, middle, and shortest axial
diameters, respectively, of a best-fit ellipsoid (7)].
The distribution among polymineralic and mono-
mineralic particles does not have any significant
difference based on the Kolmogorov-Smirnov
Fig. 1. Slice images of Itokawa particles obtained by microtomography with a gray scale showing the
linear attenuation coefficient (LAC) of objects (from0 to X cm
−1
), where X is the maximumLAC value in the
CT images. (A) Sample RA-QD02-0063 (7 keV, X = 431 cm
−1
). (B) RA-QD02-0014 (7 keV, X = 287 cm
−1
).
Some voids define a 3D plane (arrows). (C) RA-QD02-0042 (7 keV, X = 575 cm
−1
). (D) RA-QD02-0048
(7 keV, X = 431 cm
−1
). Concentric structure is a ring artifact. Bright edges of particles and voids are
artifacts resulting from refraction contrast. Ol indicates olivine; LPx, low-Ca pyroxene; HPx, high-Ca
pyroxene; Pl, plagioclase; CP, Ca phosphate; Tr, troilite; and Meso, mesostasis.
Fig. 2. Cumulative size distribu-
tion of Itokawa particles. Sphere-
equivalent diameters of the tap-
pingsamples anddiameters of the
spatula samples are shown.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1126
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

(K-S) test (probability, P=0.84) probably because
the monomineralic particles are polycrystalline
and not affected by anisotropy, such as cleavage.
The mean axial ratio (a:b:c) of fragments gen-
erated in laboratory impact experiments is ~2:√2:1
(b/a = 0.71 and c/a = 0.5) over a broad size range
(~0.4 to ~10 cm) (15, 16) (Fig. 3). The mean
ratios, b/a, for boulders on Itokawa (0.68, range
from 0.1 to 5 m) and asteroid (433) Eros (0.71 to
0.73, range from0.1 to 150 m) have similar values
(17). The K-S test indicates that the shape distrib-
ution of the Itokawa particles is not significantly
different from that of the laboratory experimental
fragments (16) (P = 0.17). The Itokawa particles
are probably the results of mechanical disaggre-
gation, primarily as a response to impacts. How-
ever, other processes, such as disaggregation of
larger particles by thermal cycling (18), cannot be
excluded. The edges of 30 of the tapping par-
ticles are angular (Figs. 1, A, B, and D, and 4, A
and B, and movie S1), suggesting that they are
fragments of mechanically crushed precursors,
whereas some edges of remaining particles are
rounded (Figs. 1C, 4, C and D, and movie S2).
Their 3D shapes are also more spherical than the
particles with angular edges (Fig. 3A). In contrast
to the Itokawa particles, the shapes of lunar re-
golith particles are more spherical (4, 19) than
experimental impact fragments (Fig. 3B) (P=0.00
in the K-S test).
Two types of terrains are observed on the
surface of Itokawa: boulder-rich rough terrain
and smooth terrain (2). It has been proposed that
Itokawa is a rubble-pile asteroid that was formed
by an early collisional breakup of a preexisting
large parent body followed by a re-agglomeration
of some of the original fragments (1). It has been
also proposed that mm to cm particles formed by
impact processing and selectively migrated into
the smooth terrains of gravitational potential lows
by granular processes (14) induced by seismic
vibration (13). Smaller Itokawa particles would
have formed in situ by impact processing on
Itokawa’s surface, although a possibility of di-
rect re-accumulation from the catastrophic im-
pact that formed Itokawa cannot be excluded.
Even though Itokawa has a low escape velocity
Fig. 3. The 3D shape distributions of (A) Itokawa particles and (B) lunar regolith (19, 26). Fragments
of impact experiments (16) are also shown in both graphs. Large circles in (A) shows particles with
rounded edges.
Fig. 4. The 3D external shapes of Itokawa particles. (A) Stereogram (box size 232 µm by 232 µm by 203 mm) and (B) SEM micrograph of RA-QD02-0023.
(C) Stereogram (box size, 112 µm by 112 µm by 93 mm) and (D) SEM micrograph of RA-QD02-0042.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1127
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

(~0.2 m/s), some amount of small particles
(<cm) should have low enough impact ejection
velocities (20) to allow them to reaccumulate
onto the surface. An in situ origin seems to be
consistent with the residence time of Itokawa
particles in the regolith deduced from galactic
cosmic ray noble gas analyses (<10 million years)
(21), which is younger than the lower limit of
Itokawa’s age (>~75 million years) (22).
The lack of in situ melting textures in
Itokawa particles can be explained by relatively
low-impact velocities of the type expected among
asteroids (~5 km/s) (23). Greater quantities of
melt-containing ejecta would be expected with
impact velocities of >10 km/s, of the type that
produce agglutinates on the Moon.
Particles with rounded edges were probably
formed from particles that were originally more
angular. Sputtering by solar-wind particles is un-
likely to explain the rounded morphology be-
cause of the short residence time on the uppermost
regolith layer deduced from solar-wind noble gas
analyses (~150 years) (21). The rounded particles
may be a result of abrasion as grains migrate
during impacts. The spherical shapes of lunar
regolith particles (Fig. 3B) are due to their longer
residence time in the regolith, which allows more
thorough sputtering by solar-wind particles.
The size and 3D shape of collected 10- to
100-mm-sized Itokawa particles suggest that they
were primarily formed on Itokawa’s surface by
impact and suffered abrasion by seismic-induced
grain motion in the terrain together with minor
repeated solar-wind particle implantation (21)
and space weathering (24). Because these pro-
cesses are mechanical and substantial melting did
not occur during impacts, the particles collected
by the Hayabusa spacecraft may not have suf-
fered a large degree of chemical fractionation and
are thus largely representative of the surface ma-
terials of Itokawa.
References and Notes
1. A. Fujiwara et al., Science 312, 1330 (2006).
2. H. Yano et al., Science 312, 1350 (2006).
3. T. Nakamura et al., Science 333, 1113 (2011).
4. G. H. Heiken et al., Eds., Lunar Sourcebook (Cambridge
Univ. Press, Cambridge, 1991).
5. R. P. Binzel, A. S. Rivkin, S. J. Bus, J. M. Sunshine,
T. H. Burbine, Meteorit. Planet. Sci. 36, 1167 (2001).
6. M. Abe et al., Science 312, 1334 (2006).
7. Materials and methods are available as supporting
material on Science Online.
8. R. Hutchison, Meteorites: A Petrologic, Chemical and Isotopic
Synthesis (Cambridge Univ. Press, Cambridge, 2004).
9. H. Yurimoto et al., Science 333, 1116 (2011).
10. G. J. Consolmagno, D. T. Britt, R. J. Macke, Chem. Erde
68, 1 (2008).
11. Some of the mm- to cm-sized particles might be
agglomerations of smaller particles weakly bounded by
cohesive forces (25). We cannot exclude a possibility that
the collected particles are these smaller particles broken
apart by the sampling mechanism(s).
12. T. Michikami et al., Earth Planets Space 60, 13 (2008).
13. H. Miyamoto et al., Science 316, 1011 (2007); 10.1126/
science.1134390.
14. J. Duran et al., An Introduction to the Physics of Granular
Materials (Springer, New York, 1999).
15. A. Fujiwara, G. Kamimoto, A. Tsukamoto, Nature 272,
602 (1978).
16. F. Capaccioni et al., Nature 308, 832 (1984).
17. T. Michikami, A. Nakamura, N. Hirata, Icarus 207, 277
(2010).
18. M. S. Robinson, P. C. Thomas, J. Veverka, S. Murchie,
B. Carcich, Nature 413, 396 (2001).
19. The 3D structures of lunar regolith particles (105 to
250 mm) at Apollo 16 landing site were measured by
microtomography (b/a = 0.79 T 0.10 and c/a =
0.61 T 0.10) (26).
20. T. Michikami, K. Moriguchi, R. Nakamura, Lunar Planet.
Sci. XXXVI, 1729 (2005).
21. M. Ebihara et al., Science 333, 1119 (2011).
22. P. Michel, D. P. O’Brien, S. Abe, N. Hiratad, Icarus 200,
503 (2009).
23. Few amounts of melting products were reported in ejecta
in laboratory impact experiments with silicate rocks at
impact velocities of ~4 to 5 km/s (27).
24. T. Noguchi et al., Science 333, 1121 (2011).
25. D. J. Scheeres, C. M. Hartzell, P. Sánchez, M. Swift, Icarus
210, 968 (2010).
26. J. Katagiri, T. Matsushima, Y. Yamada, “Statistics on 3D
particle shapes of lunar soil (No. 60501) obtained by
micro x-ray CT and its image-based DEM simulation,” in
Earth and Space 2010: Engineering, Science,
Construction, and Operations in Challenging
Environments, G. Song, R. M. Malla, Eds. (Proceedings
of the 12th International Conference on Engineering,
Science, Construction, and Operations in Challenging
Environments, American Society of Civil Engineers,
2010), pp. 254–259.
27. T. Kadono et al., J. Geophys. Res. 115, E04003 (2010).
Acknowledgments: We thank the Hayabusa sample curation team
and the Hayabusa project team. A. Tsuchiyama was
supported by a grant-in-aid of the Japan Ministry of Education,
Culture, Sports, Science, and Technology (19104012). M.E.Z.
and S.A.S. were supported by NASA’s Muses-C/Hayabusa
Program. The tomography experiment was performed under
the approval of the SPring-8 Proposal Review Committee
(2010B1531).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1125/DC1
Materials and Methods
SOM Text
Figs. S1 to S4
Table S1
References (28–38)
Movies S1 and S2
2 May 2011; accepted 5 August 2011
10.1126/science.1207807
Irradiation History of Itokawa Regolith
Material Deduced from Noble Gases
in the Hayabusa Samples
Keisuke Nagao,
1
* Ryuji Okazaki,
2
Tomoki Nakamura,
3
Yayoi N. Miura,
4
Takahito Osawa,
5
Ken-ichi Bajo,
1
Shintaro Matsuda,
1
Mitsuru Ebihara,
6
Trevor R. Ireland,
7
Fumio Kitajima,
2
Hiroshi Naraoka,
2
Takaaki Noguchi,
8
Akira Tsuchiyama,
9
Hisayoshi Yurimoto,
10
Michael E. Zolensky,
11
Masayuki Uesugi,
12
Kei Shirai,
12
Masanao Abe,
12
Toru Yada,
12
Yukihiro Ishibashi,
12
Akio Fujimura,
12
Toshifumi Mukai,
12
Munetaka Ueno,
12
Tatsuaki Okada,
12
Makoto Yoshikawa,
12
Junichiro Kawaguchi
12
Noble gas isotopes were measured in three rocky grains from asteroid Itokawa to elucidate a
history of irradiation from cosmic rays and solar wind on its surface. Large amounts of solar helium
(He), neon (Ne), and argon (Ar) trapped in various depths in the grains were observed, which
can be explained by multiple implantations of solar wind particles into the grains, combined
with preferential He loss caused by frictional wear of space-weathered rims on the grains.
Short residence time of less than 8 million years was implied for the grains by an estimate on
cosmic-ray–produced
21
Ne. Our results suggest that Itokawa is continuously losing its surface
materials into space at a rate of tens of centimeters per million years. The lifetime of Itokawa
should be much shorter than the age of our solar system.
T
he Hayabusa spacecraft arrived at asteroid
25143 Itokawa in November 2005. Itokawa
is a small (535 by 294 by 209 m) (1) S-type
asteroid with the appearance of a rubble pile.
Global remote-sensing observations revealed that
there are two geological settings, boulder-rich
rough terrains and smooth terrains (2–4). Hayabusa
carried out two touchdowns on a smooth terrain,
MUSES-CRegio, and collected regolith particles
disturbed by the touching down of the sampler
horn (5). After the sampling, Hayabusa returned
to Earth, and the sample capsule was successfully
recovered on 13 June 2010. The sample container
was opened at the curation facility of the Japan
Aerospace Exploration Agency (JAXA), and
a large number of small particles were found.
Scanning electron microscope energy-dispersive
1
Geochemical Research Center, The University of Tokyo, Bunkyo-ku,
Tokyo 113-0033, Japan.
2
Department of Earth and Planetary
Sciences, Kyushu University, Higashi-ku, Fukuoka, Japan.
3
De-
partment of Earth and Planetary Material Sciences, Tohoku
University, Aoba-ku, Sendai, Japan.
4
Earthquake Research In-
stitute, The University of Tokyo, Bunkyo-ku, Tokyo, Japan.
5
Japan Atomic Energy Agency, Tokai-mura, Ibaraki, Japan.
6
Grad-
uate School of Science and Engineering, Tokyo Metropolitan
University, Hachioji, Japan.
7
Research School of Earth Sciences,
The Australian National University, Canberra, Australia.
8
The
College of Science at Ibaraki University, Mito, Ibaraki, Japan.
9
Department of Earth and Space Science, Osaka University,
Toyonaka, Japan.
10
Natural History Sciences, Hokkaido Uni-
versity, Kita-ku, Sapporo, Japan.
11
National Aeronautics andSpace
Administration (NASA) Johnson Space Center, Houston, TX, USA.
12
Institute of Space and Astronautical Sciences (ISAS) of the
Japan Aerospace Exploration Agency ( JAXA), Sagamihara, Japan.
*To whom correspondence should be addressed. E-mail:
nagao@eqchem.s.u-tokyo.ac.jp
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1128
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

x-ray analyses revealed that at least 1500 grains
were of extraterrestrial origin and were definitely
from asteroid Itokawa (6).
The Hayabusa samples are pristine undam-
aged grains collected fromthe surface of Itokawa
and are essentially different from other extrater-
restrial materials: Micrometeorites recovered on
Earth have experienced frictional heating and ab-
lation of the surface layer during passage through
the atmosphere and have then suffered from ex-
posure to terrestrial atmosphere. Even the Star-
dust samples fromcomet 81P/Wild 2 have suffered
from frictional heating and decomposition in the
aerogel collector (7). Although interplanetary dust
particles (IDPs) recovered in the stratosphere
contain high concentrations of solar noble gases
and are not observed to have been heated much
upon entry into Earth’s atmosphere, they have been
contaminated by atmosphere. Samples returned by
Hayabusa were simply recovered fromthe sample
return capsule and handled in the clean chambers
of the JAXA curation facility under vacuum or
clean nitrogen gas with low concentrations of no-
ble gases, which minimized contamination of atmo-
spheric noble gases, oxygen, and water vapor (8).
Regolith on the surface of small asteroids is
exposed to various energetic particles, including
solar wind (SW), solar cosmic rays (SCRs), and
galactic cosmic rays (GCRs). SW(with energy of
~keV/nucleon), which includes noble gases, is
implanted within 1 mmof the grain surfaces. SCRs
are composed of more energetic solar particles
(1 to 100 MeV/nucleon), and GCRs have even
higher energies of >0.1 GeV. The high-energy
protons fromSCRs penetrate several centimeters,
whereas GCRs penetrate up to 1 m or more be-
neath the surface. Nuclear reactions caused by
these cosmic rays can produce noble gases with
characteristic isotopic compositions that are dif-
ferent from those associated with SW. Unlike the
regolith breccia represented by meteorites and on
the Moon, which formed on relatively large par-
ent bodies, the regolith on Itokawa, which is small
and thus has little gravity, is unconsolidated. Here,
we use noble gas concentrations and isotope ra-
tios in the Hayabusa samples to assess the ex-
posure histories and regolith processes on this
microgravity asteroid.
We measured three grains: RA-QD02-0015
(#0015), RA-QD02-0053 (#0053), and RA-
QD02-0065 (#0065). These grains are blocky
and translucent olivine with dimensions of ~40,
40, and 60 mm(6), and their estimated masses are
0.06, 0.06, and 0.2 mg for #0015, #0053, and
#0065, respectively [see supporting online ma-
terial (SOM)]. Noble gases were extracted from
each grain by means of stepwise heating at 200°
and 300°Cand finally melted for complete extrac-
tion (CE) using a yttrium-aluminum-garnet–Nd
laser. The extracted gases were measured with a
modified-VG5400(MS-III) at the Geochemical
Research Center, University of Tokyo.
Signals from the extracted heavy noble gas
isotopes, such as
40
Ar,
84
Kr, and
132
Xe, are indis-
tinguishable from the blank levels because of the
small amounts of gases released. This is reason-
able in that the concentrations of these trapped
and radiogenic gases should be low in chondritic
olivine. In contrast, released amounts of
4
He,
20
Ne, and
36
Ar from the samples are clearly dis-
tinguished from the respective blank levels. Iso-
topic ratios of Ne obtained (table S1) are close to
the SW composition (9) as depicted in Fig. 1.
Argon isotopic ratios,
38
Ar/
36
Ar, also are close to
that of SW (table S1).
3
He/
4
He ratios (0.00035 to
0.00038) are similar to those observed for IDPs
and Antarctic micrometeorites (10–12) and lunar
soils (13–15) but lower than the Genesis-SW
value of 0.000464 (9) and also below the lowest
value of ~0.00041 in a bulk metallic glass (BMG)
target measured by closed system stepwise etch-
ing (CSSE) (16). The concentrations of
4
He of
the Hayabusa samples are (1.8 to 3.5) × 10
−2
Fig. 1. Isotopic ratios of Ne released from the three Hayabusa samples by stepped laser heating. Bulk
compositions of the three samples plot close to the SW values. The inset shows data of each temperature
step of the samples varying between SW and fractionated SW (FS), but not being shifted toward
cosmogenic Ne produced via nuclear reaction by cosmic rays. CE, complete noble gas extraction.
Fig. 2. Elemental ratios of
4
He/
20
Ne and
4
He concentrations for the Hayabusa samples, along with those
for olivine grains in lunar soil with dimensions of 300 to 700 mm(13) and 150 to 200 mm(14, 15), IDPs of
8 to 40 mm (10), unmelted Antarctic micrometeorites (AMMs) of ~100 mm (11, 12), and cosmic spherules
of 50 to 250 mm(26). The
4
He concentrations in the Hayabusa samples are as high as those for lunar soils
and IDPs.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1129
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

cm
3
STP/g (table S1), which are as high as those
for IDPs and olivines in lunar soils (Fig. 2). These
results indicate that the three grains had been
exposed directly to corpuscular SW particles
on the surface of Itokawa. Gas release profiles
of
4
He,
20
Ne, and
36
Ar, and respective
3
He/
4
He,
20
Ne/
22
Ne, and
40
Ar/
36
Ar ratios (Fig. 3) are dif-
ferent among the samples: Most of He, Ne, and
Ar in #0015 was released at the last CE step; in
#0053, most of the He was released at 200° and
300°C, but
36
Ar was released at the CE step; in
#0065,
4
He,
20
Ne, and
36
Ar were released at 200°C.
All the isotopic ratios at the peak releases are
similar to those of SW except for
40
Ar/
36
Ar ra-
tios, for which exact correction for contaminating
atmospheric
40
Ar (
40
Ar/
36
Ar = 296) was difficult.
The resulting values as bulk
40
Ar/
36
Ar ratios for
each sample (3 to 22) (table S1) are much higher
than those for the SW (~10
−4
), albeit with large
experimental uncertainties.
The
4
He/
20
Ne ratios depictedinFig. 4(110−170)
(table S1) for the Hayabusa samples are distinctly
lower than those of Genesis SW value (656) (9),
whereas the
36
Ar/
20
Ne ratios are close to the SW
value (0.024) (9). Most measurements from gas-
rich meteorites (17) plot between the Hayabusa
and SW data, suggesting that there is a common
mechanism for He depletion in regolith materials
on the asteroidal surface. Asimilar noble gas com-
position was found in the Genesis BMG target,
where
4
He/
20
Ne ratios decreased to 100 to 400
after ~30 nm was removed from the target sur-
face (16). At deeper levels, the He and Ne isotopic
ratios, as well as the elemental ratio
20
Ne/
36
Ar,
were almost constant. Most He was found at
shallowlevels in the BMGtarget (<30 nm), where-
as Ne was retained at depths of 30 to 70 nm, with
only small isotopic fractionation. The deeper lay-
er (>70 nm) contains noble gases fractionated
in respect of both elemental and isotopic ratios,
which is commonly observed in lunar soils and
IDPs (18, 19), but not in the case of Hayabusa
samples. Thus, we infer that the noble gas com-
positions observed for the Hayabusa samples re-
sult fromrepeated SWimplantation in addition to
a preferential loss of SW-He, which is implanted
at shallower depths. One process that may re-
move the rims of grains is mechanical erosion
caused by friction among the grains or sputtering
by other SW particles. Some He is probably also
lost by diffusion; this should reasonably lower
the
3
He/
4
He ratios below the SW value (9, 16).
Noguchi et al. (20) describe space-weathered rims
containing metallic nanoparticles in the Itokawa
grain surfaces, which would be fragile and easily
removed mechanically from the grain surface.
SW implantation occurs only when a grain is
exposed directly to SWon Itokawa’s surface, be-
cause the low velocity of SW particles (mostly
300 to 800 km/s) results in a penetration depth of
20 to 100 nm in olivine (21, 22). We estimated
the time needed for
20
Ne to accumulate to the
observed concentration by assuming that the ir-
radiation occurred in a single stage and ignoring
erosion of grain surface and backscatter effects of
Ne ions (SOM). The resulting time spans are
410, 150, and 550 years for #0015, #0053, and
#0065, respectively. The estimated time spans,
however, should be lower limits of total periods
for SW implantation, and the actual SW ir-
radiation ages might be several times longer
because the release profiles of noble gases among
the samples (Fig. 4) indicate rather complex SW
implantation and loss histories for these grains.
The more energetic particles (i.e., SCRs and
GCRs) can reach much deeper levels in the as-
teroid and can produce cosmogenic nuclides
through nuclear reactions (table S2). Neon data
for the Hayabusa samples, however, do not
show a noticeable contribution of the cosmic-
ray–produced
21
Ne in excess of the SW-Ne iso-
topic ratios beyond experimental errors (Fig. 1).
We estimated the upper limit of cosmic-ray ex-
posure ages for the Hayabusa samples based on
the Ne isotopic compositions (SOM). The best
constraint obtained from#0065 is ≈3 million years
(My) if the grain remained in the uppermost
regolith layer, and at most 8 My if the grain re-
mained several tens of centimeters below the sur-
face. These are much shorter than the nominal
exposure ages of over ~400 My for mature lunar
regolith samples (21).
The cosmic-ray exposure and SW-implantation
ages for the Hayabusa samples suggest that the
regolith materials have been continuously escap-
ing from the Itokawa surface at a relatively high
rate, several tens of cm/My. This estimate supports
observations of Itokawa surface: Only a small por-
tion (~20%) of the Itokawa surface is covered with
regolith, and where it is present the regolith is
estimated to be several meters thick (23). The weak
gravity on Itokawa (escape velocity ~0.2 m/s) (24)
may allow its grains to easily escape.
The release profiles of the Hayabusa samples
(Fig. 3) and the difference in diffusion rates of
noble gases between the space-weathered rim
and crystalline part (25) allow the history of each
Hayabusa grain in regolith on Itokawa to be
explained: Grain #0015 contained the highest
amounts of
4
He,
20
Ne, and
36
Ar among the three
samples, and these gases were released at tem-
perature (CE) higher than 300°C, which suggests
that noble gases were trapped deep in the crystals.
Therefore, this grain must have been exposed to
SW particles for a time long enough to accumu-
late large amounts of solar particles. After im-
plantation, the grain was buried in the regolith
layer and shielded from SW. Solar particles that
remained in the uppermost surface layer of the
grain were completely lost by removal of the
fragile rim during mechanical contact and by dif-
fusion. After that, the grain was picked up by
Hayabusa. For sample #0053,
4
He was released
at low temperatures of 200° and 300°C, mainly
from the rim (25) but scarcely at melting. The
profile suggests that grain #0053 appeared re-
Fig. 3. Release profiles of
4
He,
20
Ne, and
36
Ar through stepped laser heating
extractions, together with respective
3
He/
4
He,
20
Ne/
22
Ne, and
40
Ar/
36
Ar. The
release profiles are different among the samples, although isotopic ratios of
He and Ne in the three samples are similar to those of SW.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1130
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

cently on the uppermost regolith layer and expe-
rienced SWimplantation for ~150 years, thereby
accumulating only a lowamount of SWparticles.
The release pattern of #0065 reflects a combina-
tion of two histories for #0015 and #0053. At
first, #0065 was exposed to SW and the upper-
most rim (<30 nm) was weathered, after which
the rim was eroded. The grain again experienced
SWirradiation and space weathering. Consequent-
ly, the gases in the present rim consists of at least
two different SWgases, one implanted during the
latest irradiation and another implanted deeper
(>30 nm) during the earlier irradiation.
Itokawa regolith materials have formed from
surface rocks by impacts of meteoroids after its
formation and then migrated to low gravitational
potential areas forming smooth terrains (23). The
short period for cosmic-ray exposure and SW
implantation obtained for the regolith grains indi-
cates that Itokawa is continuously shrinking by
losing its surface materials into space at a loss
rate of several tens of cm/ My. The lifetime of
Itokawa should be 100 to 1000 My.
References and Notes
1. A. Fujiwara et al., Science 312, 1330 (2006).
2. J. Saito et al., Science 312, 1341 (2006).
3. M. Abe et al., Science 312, 1334 (2006).
4. T. Okada et al., Science 312, 1338 (2006).
5. H. Yano et al., Science 312, 1350 (2006).
6. T. Nakamura et al., Science 333, 1113 (2011).
7. D. Brownlee et al., Science 314, 1711 (2006).
8. R. Okazaki et al., 42nd Lunar and Planetary
Science Conference, The Woodlands, TX, 7 to
11 March 2011, abstract no. 1653 (2011).
9. V. S. Heber et al., Geochim. Cosmochim. Acta 73,
7414 (2009).
10. A. O. Nier, D. J. Schlutter, Meteoritics 25, 263 (1990).
11. T. Osawa, K. Nagao, Antarct. Meteorite Res. 15, 165
(2002).
12. T. Osawa, T. Nakamura, K. Nagao, Meteorit. Planet. Sci.
38, 1627 (2003).
13. J. Kiko, T. Kirsten, D. Ries, Proc. Lunar Planet. Sci. 9,
1655 (1978).
14. P. Signer et al., Proc. Lunar Planet. Sci. 8, 3657 (1977).
15. R. Wieler, Ph. Etique, P. Signer, G. Poupeau, Proc. Lunar
Planet. Sci. 11, 1369 (1980).
16. A. Grimberg, H. Baur, F. Bühler, P. Bochsler, R. Wieler,
Geochim. Cosmochim. Acta 72, 626 (2008).
17. L. Schultz, L. Franke, Meteorit. Planet. Sci. 39, 1889
(2004).
18. R. Wieler, H. Baur, P. Signer, Geochim. Cosmochim. Acta
50, 1997 (1986).
19. R. O. Pepin, R. L. Palma, D. J. Schlutter, Meteorit. Planet.
Sci. 35, 495 (2000).
20. T. Noguchi et al., Science 333, 1121 (2011).
21. R. Wieler, Rev. Mineral. Geochem. 47, 21 (2002).
22. E. N. Parker, in Cosmic Winds and the Heliosphere,
J. R. Jokipii, C. P. Sonett, M. S. Giampapa, Eds.
(Univ. of Arizona Press, Tucson, AZ, 1997), pp. 3–27.
23. H. Miyamoto et al., Regolith on a tiny asteroid: Granular
materials partly cover the surface of Itokawa. 37th Lunar
and Planetary Science Conference, League City, TX,
13 to 17 March 2006, abstract no. 1686 (2006).
24. H. Miyamoto et al., Science 316, 1011 (2007).
25. H. Ducati, S. Kalbitzer, J. Kiko, T. Kirsten, H. W. Müller,
Moon 8, 210 (1973).
26. T. Osawa, K. Nagao, T. Noguchi, A. Nakazawa, J. Mikada,
Antarct. Meteorite Res. 16, 196 (2003).
Acknowledgments: We thank the Hayabusa project team
for their major effort in the success in the mission
returning the samples to Earth. The National Institute for
Materials Science is acknowledged for the special thin
thermocouple used for the calibration of laser heating.
We acknowledge constructive comments from reviewers.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1128/DC1
Fig. S1
Tables S1 and S2
References (27–35)
2 May 2011; accepted 2 August 2011
10.1126/science.1207785
Synthesis and Structure Determination
of the Hierarchical Meso-Microporous
Zeolite ITQ-43
Jiuxing Jiang,
1,3
Jose L. Jorda,
1
Jihong Yu,
3
Laurent A. Baumes,
1
Enrico Mugnaioli,
2
Maria J. Diaz-Cabanas,
1
Ute Kolb,
2
Avelino Corma
1
*
The formation of mesopores in microporous zeolites is generally performed by postsynthesis acid,
basic, and steamtreatments. The hierarchical pore systems thus formed allow better adsorption, diffusion,
and reactivity of these materials. By combining organic and inorganic structure-directing agents and
high-throughput methodologies, we were able to synthesize a zeolite with a hierarchical system
of micropores and mesopores, with channel openings delimited by 28 tetrahedral atoms. Its complex
crystalline structure was solved with the use of automated diffraction tomography.
Z
eolites are crystalline microporous mate-
rials, primarily made from silica and alu-
mina, that have found wide application in
industry as adsorbents, ion exchangers, and cat-
alysts. In the field of catalysis, the introduction
of zeolites has represented a revolution for oil
and gas processing, and their use has been ex-
tended to the manufacture of chemicals (1, 2).
However, zeolites are ineffective as catalysts for
processing large molecules that cannot diffuse
inside the pores. Several ways to overcome this
difficulty have been considered, such as increas-
ing the pore dimensions (3–8) and increasing
the external to internal surface ratio by decreas-
ing the crystallite size or by preparing monolay-
ered zeolites (9, 10). Another approach is to
generate mesopores within the zeolite crystallites,
yielding a network of connected mesopores and
Fig. 4. Elemental ratios of
36
Ar/
20
Ne and
4
He/
20
Ne for Hayabusa samples and gas-rich meteorites. The
4
He/
20
Ne ratios for the Hayabusa samples are distinctly lower than those of SW (9), whereas
36
Ar/
20
Ne
ratios keep the SW value.
4
He/
20
Ne released from Genesis BMG by CSSE analysis with estimated depths
from the surface is shown (16).
1
Instituto de Tecnología Química, Universidad Politécnica de
Valencia–Consejo Superior de Investigaciones Científicas, Avenida
de los Naranjos s/n, 46022 Valencia, Spain.
2
Institute of Phys-
ical Chemistry, Johannes Gutenberg University Mainz, Welderweg
11, 55099 Mainz, Germany.
3
State Key Laboratory of Inorganic
Synthesis and Preparative Chemistry, College of Chemistry, Jilin
University, Changchun 130012, China.
*To whom correspondence should be addressed. E-mail:
acorma@itq.upv.es
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1131
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

micropores. Such networks can improve the activ-
ity and selectivity of catalysts for cracking and
hydrocracking of vacuum gasoil. The mesopores
can be formed by different postsynthesis treat-
ments that involve steaming, acid or base leaching,
or some combination of these (11). Alternatively,
mesoporous materials can be synthesized in which
walls are formed or contain zeolite nanocrystals
(12). However, none of these techniques yields a
regular distribution or, ideally, a unique channel
system of mesopores structurally connected with
the regular micropores of the zeolite.
We synthesized the zeolite ITQ-43, which
features a structurally hierarchical system of con-
nected mesopores and micropores, and solved
its structure with the use of automated electron
diffraction tomography (ADT) (13, 14). ITQ-43
was synthesized as a silicogermanate using
(2´R,6´S)-2´,6´-dimethylspiro[isoindoline-2,1´-
piperidin]-1´-ium hydroxide (fig. S1) as the
organic structure-directing agent (SDA), colloi-
dal silica (Ludox AS-40) as the silicon source,
and GeO
2
as the germanium source. The boro-
silicogermanate and aluminosilicogermanate
analogs were also synthesized by modifying
the gel composition and incorporating the cor-
responding boron or aluminum sources (table
S1) (15).
13
C magic-angle spinning nuclear magnetic
resonance (MAS-NMR) (fig. S3) and chemical
analysis indicated that the SDA remained mainly
intact after the synthesis, and the
19
F MAS-NMR
spectrum (fig. S4) suggested the presence of F

in Ge-rich D4R (double four-ring) units. In the
case of the aluminosilicogermanate analog,
27
Al
MAS-NMR (fig. S5) indicated that Al was tet-
rahedrally coordinated in the framework (15).
N
2
and Ar adsorption isotherms (fig. S2) indi-
cated a very open structure with large pore di-
ameters (15).
The x-ray powder diffraction (XRPD) pat-
tern of ITQ-43 calcined in situ (15) was suc-
cessfully indexed with an orthorhombic unit
cell, with a = 26.090 Å, b = 41.866 Å, and c =
12.836 Å. The systematic extinctions correspond
to the space groups C222, Cmm2, Cm2m, C2mm,
or Cmmm. Because of the large cell parame-
ters, data resolution was reduced by severe peak
overlap even at low 2q values (d
min
> 1.8 Å).
In combination with the low stability of the
calcined sample, this precluded the extraction
of an appropriate set of intensities for struc-
tural determination using powder data.
High-resolution transmission electron micros-
copy (HRTEM) can provide direct space infor-
mation at resolutions of 0.8 to 0.5 Å (16), but this
technique requires high-intensity illumination con-
ditions and a large electron dose on the sample,
which result in rapid amorphization. Also, the
use of oriented “in-zone” electron diffraction pat-
terns, which demands long exposure times while
orienting ITQ-43 crystals for data acquisition,
gives rise to dynamical effects. Therefore, we used
a recently developed method of ADT (13, 14)
to investigate the ITQ-43 structure. Because the
XRPD data indicated low stability of the cal-
cined sample when exposed to air, the ADT data
acquisition was performed on a noncalcined, as-
prepared sample.
A first attempt to collect ADT data at room
temperature resulted in low-quality data, suffi-
cient for cell parameter confirmation but not for
reliable reflection intensity integration. The fast
degradation of the crystal structure during the
acquisition led to a progressive loss of resolution
and weak reflections. Moreover, several acquired
patterns coming from degraded areas of the crys-
tals showed very weak or even amorphous
scattering.
A second attempt was performed by cooling
the sample down to liquid N
2
temperature (~100 K)
and resulted in a substantial improvement of crys-
tal stability under the beam, and consequently in
a more reliable reflection intensity integration.
Two crystals (dimensions 200 nm × 90 nm and
250 nm× 140 nm) were examined. Projections of
the reconstructed reciprocal space for the sec-
ond crystal are shown in fig. S6. The two data
sets delivered cell parameters consistent with
those obtained by XRPD within the ADT ex-
perimental error. The three-dimensional recon-
structed volume revealed only extinctions due to
C-centering. A combined intensity data set was
obtained by merging the two experimental ac-
quisitions, using a scale factor of 1.92 cal-
culated from the strongest reflections of each
crystal. Reflection intensity symmetry was con-
sistent with Laue class Cmmm, confirming the
orthorhombic setting; the related residual R
sym
was 24%.
Structure solution was performed ab initio
by direct methods implemented in SIR2008
software (17) using a fully kinematic approach
intensity proportional to the square of the struc-
ture factor, and the best results were obtained
when the merged data set was used. Among the
allowed space groups, only the noncentrosym-
metric C222 provided a consistent solution. From
the structure solution, 20 Si-Ge positions were
recognized alongside 24 O positions (from a total
of 42 in the complete structure). Strong electron
densities were present inside the cavities, and in
particular inside the large four-leaf clover–like
pore. These potentials derive from the diffuse
scattering produced by the remaining SDA used
in the synthesis.
Subsequent Fourier map refinement per-
formed by SHELX97 (18) identified 13 further
O positions, and the remaining 5 O positions
were placed on the basis of geometric consid-
erations. The refined structure showed additional
symmetry elements consistent with the centrosym-
metric space group Cmmm. Because the Fourier
map refinement proved to be inconclusive, space
group Cmmm was assumed as correct on the
basis of geometrical considerations. The fact that
SIR2008 delivered a structure solution only in
space group C222 was attributed to the lack of
constraints on the phases in the noncentrosym-
metric case, which made it easier to get a set of
approximately right phases good enough to com-
plete the structure, and to the template diffuse
scattering contribution that spoiled the accurate
phase attribution. The SIR2008 output residual R
(34%) was considerably smaller than the equiv-
alent SHELX97 residual R
1
of the final model
(>50%) because of the suppression of the intrapore
electron densities.
Fig. 1. Rietveld refinement
of the x-ray diffraction pat-
tern of ITQ-43 calcined at
973 K. Data points show
the observed XRPD pattern;
the line along these points
is the calculated XRPD pat-
tern, with the difference
profile at the bottom. The
vertical tick marks below
the pattern give the posi-
tions of the Bragg reflections.
Fig. 2. The 28-ring channel of zeolite ITQ-43
(oxygens omitted for clarity, except those in the
silanol groups). The dimensions of the pore open-
ing are calculated using an oxygen radius of 1.35 Å.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1132
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

The presence of the SDA precluded an ap-
propriate refinement of the structure on the basis
of ADT data; therefore, a Rietveld refinement in
Cmmm symmetry was performed using the XRPD
data of the calcined sample with the program
FULLPROF (19), obtaining a good match be-
tween experimental and calculated patterns (Fig.
1). The R
exp
, R
wp
, R
F
, and R
B
residuals of the
refinement were 0.042, 0.160, 0.096, and 0.075,
respectively. Note, however, that the low resolu-
tion of the data introduced an uncertainty in the
precise atomic coordinates of the O atoms (table
S2). Rietveld refinements in lower-symmetry
space groups did not produce an improvement in
the refinement. Also, the structure of this material
has been predicted with an in-house simulation
program based on an evolutionary algorithm using
GPU hardware (15, 20), appearing as one of the
50 most probable hypothetical frameworks gen-
erated for those given cell parameters.
ITQ-43 has a very open structure, with a frame-
work density as low as 11.4 tetrahedral atoms
per 1000 Å
3
. The most notable characteristic of
ITQ-43 is the presence of cloverleaf-like chan-
nels formed by 28-rings (units of 28 tetrahedral
atoms) along the c axis, with a shape similar to
that found in the zeolite cloverite (21) but with
pore diameters of 21.9 Å × 19.6 Å in their
longest axis (Fig. 2), in the range of the mesopore.
This mesoporous system is also connected by
12-ring channels of 6.8 Å × 6.1 Å along the a
axis, and two additional sets of 12-ring chan-
nels along (a + b) and (a – b), related by sym-
metry, with apertures of 7.8 Å× 5.7 Å(Fig. 3 and
fig. S7).
The framework of ITQ-43 can be described
as a basic tertiary building unit formed by the
combination of three mel cages [45
2
6
4
] (Fig. 4A),
a building unit also present in zeolite IM-20 (In-
ternational Zeolite Association code UWY) (22).
These tertiary units connect to each other ei-
ther by a double four-ring d4r [4
6
] or by lau
[4
2
6
4
] cages and distorted d4r units that are
connected to the hydroxyl groups to produce
a network in layers (Fig. 4B). Finally, those
sheets connect to each other, forming additional
d4r units to give the complete three-dimensional
framework.
Regarding the preferential location of Ge in
different tetrahedral sites (T-sites), the results ob-
tained are consistent with those for previous
Ge-containing zeolites (3, 4, 23–25). ITQ-43
presents three different D4R units: (i) in the side
of the 28-ring, formed by T11 (31% Ge); (ii)
between the sheets of the structure, formed by
T5 (47% Ge) and T7 (82% Ge), with an av-
erage value of 65% Ge; and (iii) distorted so
as to connect to the OH groups, formed by T2
(0% Ge), T6 (0% Ge), T8 (48% Ge), and T10
(92% Ge, directly connected to the OH), for an
average of 35% Ge. The remaining positions, cor-
responding to a 4-ring and the two first neighbors
of the D4R, remain pure silica. The preferential
occupation of Ge in T-sites directly connected to
the hydroxyl groups has been found in the zeo-
litic silicogermanate ITQ-37 (5).
Zeolite ITQ-43 has a Si/Ge ratio too low to
preserve the framework stability after removal
of the SDA in a moisture-containing atmosphere,
and consequently too low for use in catalytic
applications. However, the Si/Ge ratio of other
silicogermanates has been increased with the
use of improved direct-synthesis methods, a more
efficient SDA (26, 27), or postsynthesis treat-
ments (28) and the same may be possible for
ITQ-43.
We also believe that the use of ADT data
offers new possibilities for the solution of zeo-
litic structures. Even though the combination of
HRTEM, electron diffraction, and XRPD data
provides an important step toward the solution
of complicated zeolitic structures (29), ab initio
structure solution by direct methods based on
pure electron diffraction data may be a faster
and more reproducible path applicable to any
class of catalytic porous materials.
References and Notes
1. A. Corma, J. Catal. 216, 298 (2003).
2. M. E. Davis, Nature 417, 813 (2002).
3. A. Corma, M. J. Díaz-Cabañas, J. L. Jordá, C. Martínez,
M. Moliner, Nature 443, 842 (2006).
4. J. Jiang, J. L. Jorda, M. J. Diaz-Cabanas, J. Yu, A. Corma,
Angew. Chem. Int. Ed. 49, 4986 (2010).
5. J. Sun et al., Nature 458, 1154 (2009).
6. K. G. Strohmaier, D. E. W. Vaughan, J. Am. Chem. Soc.
125, 16035 (2003).
7. C. C. Freyhardt, M. Tsapatsis, R. F. Lobo, K. J. Balkus,
M. E. Davis, Nature 381, 295 (1996).
8. A. Burton et al., Chem. Eur. J. 9, 5737 (2003).
9. A. Corma, V. Fornes, S. B. Pergher, Th. M. L. Maesen,
J. G. Buglass, Nature 396, 353 (1998).
10. M. Choi et al., Nature 461, 246 (2009).
11. J. Čejka, S. Mintova, Catal. Rev. 49, 457 (2007).
12. Y. Liu, T. J. Pinnavaia, J. Mater. Chem. 12, 3179
(2002).
13. U. Kolb, T. Gorelik, C. Kübel, M. T. Otten, D. Hubert,
Ultramicroscopy 107, 507 (2007).
14. U. Kolb, T. Gorelik, M. T. Otten, Ultramicroscopy 108,
763 (2008).
15. See supporting material on Science Online.
16. C. Kisielowski et al., Microsc. Microanal. 14, 469
(2008).
17. M. C. Burla et al., J. Appl. Cryst. 40, 609 (2007).
18. G. M. Sheldrick, Acta Crystallogr. A 64, 112
(2008).
19. J. Rodriguez-Carvajal, IUCr Newsletter 26, 12 (2001).
20. L. A. Baumes, F. Kruger, S. Jimenez, P. Collet, A. Corma,
Phys. Chem. Chem. Phys. 13, 4674 (2011).
21. M. Estermann, L. B. McCusker, Ch. Baerlocher,
A. Merrouche, H. Kessler, Nature 352, 320 (1991).
22. M. Dodin et al., J. Am. Chem. Soc. 132, 10221
(2010).
23. G. Sastre et al., Angew. Chem. Int. Ed. 41,
4722 (2002).
24. A. Corma, F. Rey, S. Valencia, J. L. Jordá, J. Rius,
Nat. Mater. 2, 493 (2003).
25. A. Corma et al., J. Am. Chem. Soc. 130, 16482
(2008).
26. A. Cantín et al., Angew. Chem. Int. Ed. 45, 8013
(2006).
27. J. L. Jorda et al., Z. Kristallogr. Suppl. 26, 393 (2007).
28. F. Gao et al., J. Am. Chem. Soc. 131, 16580
(2009).
29. Ch. Baerlocher et al., Science 315, 1113 (2007).
Acknowledgments: We thank the Spanish government
(projects MAT2009-14528-C02-01, PLE2009-0054, and
CONSOLIDER INGENIO 2010) and Generalitat Valenciana
(Project Prometeo) for financial support, and the
Deutsche Forschungsgemeinschaft SFB 625 Projekt B8
for supporting the development of automated diffraction
tomography. J.J. thanks the China Scholarship Council for
Fig. 3. Structure of ITQ-43 viewed along the three main crystallographic axes (oxygens omitted for
clarity, except those in the silanol groups).
Fig. 4. Building units of ITQ-43. (A) Composition
of the tertiary building unit. (B) Combination of
the building units in layers. (C) Bonding between
layers to form the complete structure (oxygens
omitted for clarity, except those in the silanol
groups).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1133
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

support and Instituto de Tecnologia Quimica (UPV-CSIC)
for a Ph.D. scholarship. J.Y. thanks the National Basic
Research Program of China (grants 2011CB808703 and
2007CB936402) and the Major International Collaboration
Project of China. Further details on the crystal structure
may be obtained from the Fachinformationszentrum
Karlsruhe (www.fiz-karlsruhe.de), depository
no. CDS-423044.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1131/DC1
Materials and Methods
Figs. S1 to S7
Tables S1 and S2
References (30–42)
20 May 2011; accepted 14 July 2011
10.1126/science.1208652
Unraveling the Perplexing Structure
of the Zeolite SSZ-57
Christian Baerlocher,
1
Thomas Weber,
1
Lynne B. McCusker,
1
* Lukas Palatinus,
2
Stacey I. Zones
3
Previous high-resolution x-ray powder diffraction and transmission electron microscopy studies
of the zeolite SSZ-57 could not fully elucidate the structural basis for its puzzling adsorption
behavior, which appears to be intermediate between that of a medium- (10-ring) and a
large-pore (12-ring) zeolite. Now by applying advanced crystallographic techniques (structure
solution in four-dimensional space and interpretation of three-dimensional diffuse scattering by
Monte Carlo simulation) and crystal chemistry considerations to high-quality single-crystal
x-ray diffraction data collected on a microcrystal (about 2 micrometers by 2 micrometers by
8 micrometers), we have been able to derive a comprehensive description of its silicate framework
structure. The framework is related to that of ZSM-11 but is commensurately modulated along
the c axis (P4m2, a = b = 20.091 Å, c = 110.056 Å) to yield a structure with a 12-ring:10-ring
ratio of 1:15. Disorder of the 12-rings results in a three-dimensional 10-ring channel system
with large isolated pockets. The structure helps to clarify the material’s catalytic activity.
S
SZ-57 is a high-silica zeolite catalyst that
is synthesized in hydroxide media by using
the bulky N-butyl-N-cyclohexylpyrrolidinium
cation as a structure-directing agent (SDA) (1).
In fluoride media, the same SDA produces the
medium-pore zeolite ZSM-11 [framework type
MEL (2)] (3). Adsorption and catalytic studies
show SSZ-57 to behave curiously different
from related zeolites (4, 5). In particular, the 2,2-
dimethylbutane (~6.2 Å cross section) uptake
shows that the initial pore filling is less hindered
than that of a series of zeolites (including ZSM-
11) with multidimensional 10-ring (6) channels.
At the same time, the material does not have as
steep a pore-filling curve as known 12-ring zeo-
lites or reach the same filling capacity. Similarly,
xylene isomerization tests for para selectivity
show SSZ-57 to be slightly less selective than
the medium-pore ZSM-5 (MFI), which is the
commercial standard (5), but more selective than
large-pore zeolites such as beta (*BEA). That is,
the pore-filling profile shows a unique discrimi-
nation behavior that can be exploited in catalytic
applications (7).
Electron microscopy investigations indicate
that the structure is both commensurately mod-
ulated (i.e., the structure can be described in
terms of a smaller basis structure that changes in
a well-defined manner over an integral number
of repeat units to create a crystalline superstruc-
ture) and disordered (i.e., contains faults that
disturb the long-range order) (5), so it is not
surprising that, despite considerable effort, its
structure has not been resolved. In the diffrac-
tion data, the modulation manifests itself in sat-
ellite reflections next to the main Bragg peaks
and the disorder produces regions of diffuse scat-
tering under and/or between the reflections. Dis-
order is not uncommon in zeolite framework
structures [e.g., zeolite beta, ZSM-48, or SSZ-31
(*BEA, *MRE, and *STO, respectively) (8)],
but its characterization is difficult and usually re-
stricted to model building based on high-resolution
transmission electron microscopy (HRTEM) im-
ages and powder diffraction pattern simulations (9).
Now, by collecting high-quality single-crystal
diffraction data on a microcrystal of SSZ-57
(~2 µm by 2 µm by 8 mm) using a PILATUS 6M
pixel detector (10) and by combining advanced
crystallographic techniques [structure solution in
four-dimensional (4D) space and interpretation
of 3D diffuse scattering by Monte Carlo simu-
lation] with both physical and in silico model
building, we have been able to construct a com-
prehensive description of its structure. The sili-
cate framework is related to that of ZSM-11
(I 4m2, a = b = 20.12 Å, c = 13.44 Å) (11) but is
modulated along the c axis to yield a unit cell more
than eight times larger (c = 110 Å). This mod-
ulation produces a structure with a 12-ring:10-ring
ratio of 1:15. Disorder of the 12-rings results in
a 10-ring channel system with large isolated pock-
ets, which are an intrinsic and essential feature
of the structure. In a conventional 3D descrip-
tion of the idealized structure (Fig. 1A), there
are 99 Si atoms in the asymmetric unit. For com-
parison, the two most complex zeolite framework
structures known to date, TNU-9 (TUN) (12) and
IM-5 (IMF) (13), each have 24. The structure
clarifies the previously published HRTEM images
(5) and explains the material’s catalytic activity.
In view of the fact that a closely related syn-
thesis procedure produces pure ZSM-11, it is not
surprising that both powder diffraction (3) and
electron microscopy (5) data indicate a qualified
resemblance between SSZ-57 and ZSM-11. Their
powder diffraction patterns have some similarities
but are clearly not identical. In particular, the first
strong peak in SSZ-57 would require a c lattice
parameter of ~13.8 Å (versus 13.4 Å for ZSM-
11). The HRTEM images of SSZ-57 show re-
gions reminiscent of a ZSM-11 [100] projection
(pentasil layer), separated by regions with ill-
defined boundaries that are distinctly different
(5). Not only do these ladderlike regions appear
at fairly regular intervals in the image (~100 Å
apart), but satellite reflections are also observed
in the electron diffraction (ED) patterns. Both
phenomena indicate that the structure has a much
longer repeat distance than that of the simple
ZSM-11 structure (i.e., it is modulated). The ED
patterns also show regions of diffuse scattering,
implying the presence of disorder in the structure.
X-ray diffraction data were collected on one
of the larger crystallites (~2 µm by 2 µm by
8 mm) in the SSZ-57 sample, synthesized ac-
cording to the procedure outlined in (3), on
beamline X06SA at the Swiss Light Source. The
high dynamic range, low background, and ex-
treme speed of the PILATUS 6M pixel detector
(10) on this beamline allowed a full data set (1800
frames) of excellent quality to be measured in
just 15 min. Indeed, it is the quality of these data
that has made the necessarily elaborate structure
analysis possible.
The synchrotron data were reduced, and the
reflections indexed [6374 unique reflections with
I > 2s(I), where I is intensity; see (14)]. Conven-
tional structure analysis [SHELX (15)] produced
a solution in the space group P4m2 (a = b =
20.091 Å, c = 110.056 Å) with many overlapping
atomic positions. Although these positions could
be interpreted and a disordered structural model
derived, it proved to be very difficult to verify this
structure.
Therefore, we tried a different approach. The
positions of the satellite reflections indicate the
presence of an eightfold modulation along the c
axis, so they can be interpreted to stem from a
commensurately modulated structure with a te-
tragonal unit cell (a = 20.091 Å and c = 13.756
Å) and a modulation vector q = 0.125c*. All re-
flections were assigned four-integer indices
according to the superspace formalism [for an
overview see (16)]. Then the superspace electron
density was reconstructed by using the charge-
1
Laboratory of Crystallography, Eidgenössische Technische
Hochschule (ETH) Zurich, CH-8093 Zurich, Switzerland.
2
De-
partment of Structure Analysis, Institute of Physics of the
Academy of Sciences of the Czech Republic, 182 21 Prague,
Czechia.
3
Chevron Energy Technology Company, Richmond,
CA 94802, USA.
*To whom correspondence should be addressed. E-mail:
mccusker@mat.ethz.ch
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1134
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

flipping algorithm (17–19) implemented in the
computer program Superflip (20), which is spe-
cifically designed to deal with higher-dimensional
structures. The solution was performed in the
space group P1 (i.e., translational symmetry only),
and the superspace group was determined unam-
biguously after structure solution from the phased
structure factors (21) to be I4
1
/amd(00g)s00s.
This superspace model better describes the true
symmetry of the structure and accounts for more
of the systematic absences observed in the data
than does the 3D space group P4m2. The output
of the structure solution procedure is a 4D super-
space electron density that can be analyzed in
terms of atomic domains (atomic positions and
associated modulation functions). By artificially
shortening each atomic domain, we derived an
ordered variant of the structure found in three
dimensions (space group P4m2) (Fig. 1A). In
the 4D description, there are just 7 independent
Si and 15 independent O atomic domains. The
structure was then refined in 4D space by using
the computer program Jana2006 (14, 22).
Although the structure of SSZ-57 is enor-
mously complex, the idealized model shown in
Fig. 1A can be described relatively easily as a
disturbed ZSM-11 structure. The disturbances
are 12-ring channels that substitute for 1 out of
16 10-ring channels in an ordered manner. At
each 12-ring interruption, the ZSM-11 structure
is rotated by 90° around the c axis. Consequent-
ly, and consistent with the tetragonal symmetry
of SSZ-57, the next 12-ring is also rotated by
90°, so there are two mutually perpendicular 12-
ring channels per unit cell. However, this ide-
alized structure tells only part of the story.
It is readily apparent from Fig. 1B that the
simple idealized model does not explain all of
the peaks in the electron density map. The extra
peaks can be accounted for by moving the 12-
ring/8-ring column (running parallel to b) from
one 10-ring/6-ring column to the next in a zig-
zag fashion following the numbers in Fig. 1A
(Fig. 1C). The electron density (Fig. 1B) indi-
cates that all possible positions of this column in
the asymmetric unit are equally populated. Ex-
amination of the connection between adjacent
layers (pentasil layers stacked along the a axis)
reveals that when the 12-ring in the upper layer
is either directly above the lower one (position 0
in Fig. 1A) or above either of the two 10-rings
to the left or right of the lower 12-ring (positions
1 and 1', Fig. 2A), the straight 10-ring channels
of the MEL sections of the structure are pre-
served. However, when the upper 12-ring is
moved further away (positions 2 to 8 or 2' to 7'),
all 10-rings lying between the two 12-rings are
out of register with one another, and kinks such
as those in ZSM-5 (MFI) are built into the
straight 10-ring channels. Distance least-squares
calculations (23) show that no strain is involved
in connecting layers with the 12-ring in posi-
tions 0 and 1 (or 1') (Fig. 2A and table S2).
The idealized model has a column of 12- and
8-rings within a layer (Fig. 2B left). From a
Fig. 1. (A) The [100] projection of the idealized structure of SSZ-57 with the different parts of the
structure highlighted and the rings numbered. The 12-ring channels (yellow) separate the two ZSM-11
structure sections (black and blue), which are related to one another by a 90° rotation around z. The
dividing line is shown in red. Bridging O atoms have been omitted for clarity. (B) The [100] projection
of the electron density map generated by Superflip with the idealized structure shown in (A) overlaid.
The larger densities not described by the idealized model are circled in red, and the 12-/8-ring column
is highlighted in yellow. (C) A schematic diagram (12-/8-ring columns shown as ellipses) of (B) showing
how the extra peaks, depicted as red dots, arise from the different placements of the 12-/8-ring column
in different layers.
Fig. 2. The disorder in SSZ-57. (A) How layers
with displaced 12-rings (yellow) fit together
along the [100] (left) and [010] (right) pro-
jections. (B) A simple schematic drawing (12- and
8-rings as yellow elipses and 10- and 6-rings
as circles) of a single layer showing how the
positions of the 12-rings perpendicular to the
layer (yellow cylinders) could be arranged
(magnitude of each shift from top to bottom
indicated on the right). The two different ori-
entations of the ZSM-11 sections are distin-
guished with black and blue coloring and the
direction of the arrows in the 6-rings. (C) The
layer in (B) viewed from above.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1135
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

structural point of view, it is only possible to
disturb this 12-/8-ring column by replacing the
8-ring with a somewhat strained bre unit (8) and
introducing a displaced 12-ring (fig. S3), so this
fault is not expected to occur very often. In
other words, the layer has longer range order
along the 12-/8-ring column, but the 12-rings
perpendicular to the layer are disordered. The
extended 12-/8-ring column acts as a barrier for
the possible distribution of the 12-rings perpen-
dicular to it and thereby ensures that the staggered
12-rings remain within certain bounds. In the
layers perpendicular to the one shown in Fig.
2B, the extended column is on the right-hand
side of the layer running parallel to the a axis,
and the staggered 12-rings are on the left (Fig.
2C). The disorder is thus 3D, resulting in diffuse
scattering in all three dimensions. This 12-ring
disorder explains the shape of the diffuse crosses
in Fig. 3, which are sharp along the extended
direction (highly ordered) and broad along the
staggered direction (disordered).
Models of this disorder were constructed and
used as input to a Monte Carlo algorithm spe-
cifically tailored to this problem to simulate the
observed diffuse scattering. The periodic dis-
ordered structure was first disentangled so that it
could be understood as a superposition of sev-
eral chemically reasonable ordered fragments.
We could identify 26 equally probable 12-ring
positions within the unit cell, half oriented along
a and half along b. The 12-rings form a se-
quence of alternating orientations along the c
axis, that is, a 12-ring must always be followed
by a second one rotated by 90°. To reduce the
modeling to a manageable complexity, only the
12-ring positions with 0 ≤ z ≤ 0.5 were con-
sidered for one orientation and only those with
0.5 ≤ z ≤ 1.0 for the other (perpendicular) ori-
entation. This simplification reduced the total num-
ber of possible 12-ring positions from 13 to 9 for
each orientation (Fig. 1, B and C). Model pa-
rameter values were found by trial and error. The
simulated crystal covered 160, 160, and 32 unit
cells along a, b and c, respectively, representing
a total of more than 2.06 × 10
9
atoms (14).
These simulations reproduce the diffuse
scattering extremely well (Fig. 3) and allow us
to draw several conclusions regarding the dis-
order. The 12-rings in adjacent layers clearly
tend to adopt positions close to one another.
That is, given the layer shown in Fig. 1A, the
12-rings in the next layer are most likely to be
found in positions 0, 1, or 1' (i.e., shifts of 0 or
T1) with equal probability. This simple rule proved
to be sufficient to generate the positions and
basic shapes of the observed diffuse scattering.
Optimization of the simulation to better repro-
duce the diffuse scattering profile showed that
shifts of 0 and T1 account for roughly three-
quarters of all shifts, with the remaining 25%
distributed about equally over the shifts beyond
T1. The simulations cannot distinguish between
equal or decreasing probability with 12-ring sep-
aration for these larger shifts but do show that
Fig. 3. The hk0 (top) and 13kl (bottom) layers of the diffraction pattern as (A) measured and (B)
calculated from the disordered model simulation. Note the satellite reflections indicative of the
eightfold modulation in the 13kl layer and the diffuse scattering indicative of disorder in both layers.
The lower left-hand corner is the position of the reciprocal space origin in the hk0 layer and of the 13 0 0
reflection in the 13kl layer. Disagreements are mainly a consequence of the simplifications in our model
(fig. S4).
Fig. 4. (A) A possible arrangement of larger cavities in a portion of the channel system of SSZ-57,
shown in the form of a structure envelope (27). The third and fourth vertical channels contain the larger
cavities that are associated with the disordered 12-rings, whereas the others are the simple 10-ring
channels of ZSM-11. (B) A closeup view of a 12-ring pocket in a 10-ring channel (top) and a simple 10-
ring channel (bottom). The 12-rings have effective pore widths of ~8.2 by 5.3 Å, whereas those of the
10-rings range from 5.4 Å by 5.4 Å to 5.5 Å by 5.6 Å (versus 5.3 Å by 5.4 Å for ZSM-11).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1136
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

allowing only the next nearest position (i.e., shifts
of T2) did not reproduce the measured diffuse
scattering as well. The disorder shown schemat-
ically in Fig. 2B reflects the approximate fre-
quency found for the different shifts. Further
fine-tuning revealed that there is no correlation
between the positions of the 12-rings in the two
halves of the unit cell along the long c axis.
The disordered structure described above is ful-
ly consistent with the HRTEM images (5), which
show predominantly single-step displacements
from layer to layer. The fact that none of the
images shows larger 12-ring openings is also
consistent with the lack of open 12-ring chan-
nels. The ladderlike regions between the ZSM-11
regions are those in which 12-rings are partially
blocked by 10-rings (Fig. 2). Although the basic
channel systemof SSZ-57 is the same as that of
ZSM-11, there are additional bulges in the chan-
nels where the 12-rings occur (Fig. 4). That is,
SSZ-57 has a 3D10-ring channel systemwith a
dilute population of larger isolated cavities. These
loosely ordered cavities must be the reason for
the different catalytic properties observed for
this material.
It has been shown for the zeolites SSZ-25
(MWW) and SSZ-35 (STF), which have 10-
ring apertures leading to larger cavities, that their
catalytic selectivities reflect the presence of these
less-restricted regions (24). The situation with SSZ-
57 is different in that the larger cavities are more
highly dispersed in the 10-ring channel system.
In the idealized structure (Fig. 1A), which does
not actually exist as an extended structure, there
would be just single 12-ring channels embedded
in the 10-ring ZSM-11 framework. Only as a
result of the disorder are the 12-ring pockets dis-
tributed throughout the structure, thereby pro-
viding SSZ-57 with its unique properties. The
12-ring pockets provide space for larger inter-
mediates (e.g., for bimolecular transition states)
in a catalytic reaction and/or faster diffusion,
whereas the 10-ring channel system preserves
the shape-selective sorption and desorption prop-
erties of a classical pentasil framework structure.
This could be particularly valuable in forming
products for fine chemicals, where modification
of the substituents of single rings, either by acid-
catalyzed or partial oxidation reactions, is often
required.
The crystallization mechanism of SSZ-57
remains to be clarified, but the synthesis is quite
reproducible. This structure could be viewed as
a first step along the way to a hierarchical ma-
terial (microporous system with mesopores to fa-
cilitate diffusion) with well-defined mesopores.
Other attempts in this direction have involved
carving mesoporosity into zeolite crystals, co-
templating the formation of both types of pores
(25), and producing a material with arrested crys-
tallization (26). SSZ-57 offers a fourth option.
References and Notes
1. S. Elomari, U.S. Patent 6,544,495 (2003).
2. Framework type codes for all zeolites mentioned in the
text are given in parentheses.
3. A. W. Burton et al., Stud. Surf. Sci. Catal. 170, 690 (2007).
4. S. I. Zones et al., J. Catal. 250, 41 (2007).
5. S. I. Zones et al., Solid State Sci. 13, 706 (2011).
6. An n-ring is a ring of n Si atoms alternating with n O
atoms that defines the dimensions of a pore opening in a
zeolite framework structure. A 10-ring is usually
referred to as a medium, and a 12-ring as a large,
pore opening.
7. S. Elomari, U.S. Patent 6,616,830 (2003).
8. C. Baerlocher, L. B. McCusker, D. H. Olson, Atlas of
Zeolite Framework Types (Elsevier, Amsterdam, 2007).
9. H. Gies, H. van Koningsveld, Catalog of Disorder in
Zeolite Frameworks, www.iza-structure.org/databases.
10. C. Broennimann et al., J. Synchrotron Radiat. 13,
120 (2006).
11. G. T. Kokotailo, P. Chu, S. L. Lawton, W. M. Meier,
Nature 275, 119 (1978).
12. F. Gramm et al., Nature 444, 79 (2006).
13. C. Baerlocher et al., Science 315, 1113 (2007).
14. Materials and methods are available as supporting
material on Science Online.
15. G. M. Sheldrick, Acta Crystallogr. A 64, 112 (2008).
16. S. van Smaalen, Incommensurate Crystallography
(Oxford Univ. Press, Oxford, 2007).
17. G. Oszlányi, A. Sütö, Acta Crystallogr. A 60, 134 (2004).
18. G. Oszlányi, A. Sütö, Acta Crystallogr. A 64, 123 (2008).
19. L. Palatinus, Acta Crystallogr. A 60, 604 (2004).
20. L. Palatinus, G. Chapuis, J. Appl. Cryst. 40, 786 (2007).
21. L. Palatinus, A. van der Lee, J. Appl. Cryst. 41, 975 (2008).
22. V. Petricek, M. Dusek, L. Palatinus, Jana2006 (Institute of
Physics, Prague, Czech Republic, 2006).
23. C. Baerlocher, A. Hepp, W. M. Meier, DLS-76, Distance
least squares refinement program (Institut für
Kristallographie, ETH Zürich, Switzerland, 1977).
24. J. R. Carpenter, S. Yeh, S. I. Zones, M. E. Davis, J. Catal.
269, 64 (2010).
25. J.-B. Koo et al., J. Catal. 276, 327 (2010).
26. J. Kim, W. Park, R. Ryoo, ACS Catal 1, 337 (2011).
27. S. Brenner, L. B. McCusker, C. Baerlocher, J. Appl. Cryst.
30, 1167 (1997).
Acknowledgments: We thank S. Elomari for synthesizing
the SSZ-57 sample. Crystallographic experiments were
conducted at the X06SA beamline at the Swiss Light
Source, Paul Scherrer Institute, Villigen, Switzerland. We
thank the staff there for assistance with the data
collection. This work was funded in part by the Swiss
National Science Foundation. Additional information
on the indexing, the structure solution, the model
construction and the Monte Carlo simulations is
available in the supporting online material.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1134/DC1
Materials and Methods
SOM Text
Figs. S1 to S4
Tables S1 and S2
References
25 April 2011; accepted 25 July 2011
10.1126/science.1207466
optix Drives the Repeated
Convergent Evolution of Butterfly
Wing Pattern Mimicry
Robert D. Reed,
1
*† Riccardo Papa,
1,2
*† Arnaud Martin,
1
Heather M. Hines,
3
Brian A. Counterman,
4
Carolina Pardo-Diaz,
5
Chris D. Jiggins,
5
Nicola L. Chamberlain,
6
Marcus R. Kronforst,
6
Rui Chen,
7
Georg Halder,
8
H. Frederik Nijhout,
9
W. Owen McMillan
10,3
Mimicry—whereby warning signals in different species evolve to look similar—has long served
as a paradigm of convergent evolution. Little is known, however, about the genes that underlie
the evolution of mimetic phenotypes or to what extent the same or different genes drive such
convergence. Here, we characterize one of the major genes responsible for mimetic wing pattern
evolution in Heliconius butterflies. Mapping, gene expression, and population genetic work all
identify a single gene, optix, that controls extreme red wing pattern variation across multiple
species of Heliconius. Our results show that the cis-regulatory evolution of a single transcription
factor can repeatedly drive the convergent evolution of complex color patterns in distantly
related species, thus blurring the distinction between convergence and homology.
S
ome of the most dramatic examples of nat-
ural selection involve Müllerian mimicry—
a phenomenon in which two or more spe-
cies share predator avoidance signals. This type
of mimicry is one of the main forces driving the
evolution of warning coloration in Heliconius
(1), a neotropical butterfly genus of ~40 spe-
cies famous for its extensive intraspecific wing
pattern variation. The two best-studied Heliconius
species are the comimics Heliconius erato and
Heliconius melpomene (Fig. 1); each has more
than 25 named geographic races, most of which
mimic co-occurring Heliconius species, other un-
related butterflies, and day-flying moths. Mimetic
wing pattern radiations like those seen in H. erato
1
Department of Ecology and Evolutionary Biology, University
of California, Irvine, CA 92697, USA.
2
Department of Biology
and Center for Applied Tropical Ecology and Conservation,
University of Puerto Rico, Rio Piedras, Puerto Rico.
3
Depart-
ment of Genetics, North Carolina State University, Raleigh, NC
27695, USA.
4
Department of Biological Sciences, Mississippi
State University, Mississippi State, MS 39762, USA.
5
Department
of Zoology, University of Cambridge, Cambridge CB2 3EJ, UK.
6
FAS Center for Systems Biology, Harvard University, Cam-
bridge, MA 02138, USA.
7
Baylor Human Genome Sequencing
Center, Houston, TX 77030, USA.
8
Department of Biochemistry
and Molecular Biology, M.D. Anderson Cancer Center, University
of Texas, Houston, TX 77030, USA.
9
Department of Biology,
Duke University, Durham, NC 22708, USA.
10
Smithsonian Trop-
ical Research Institute, Panama City, Panama.
*These authors contributed equally to this work.
†To whom correspondence should be addressed. E-mail:
rreed@uci.edu (R.D.R.); rpapa.lab@gmail.com (R.P.)
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1137
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

and H. melpomene have resulted in the evolution
of hundreds of wing pattern races within the
genus (2). Despite increasing evidence that the
same genomic regions contribute to wing pat-
tern variation in different Heliconius species, the
specific genes that underlie the repeated conver-
gent evolution of wing patterns have remained
unknown (3).
Three genomic regions control most wing
pattern variation in Heliconius—in both co-
mimetic species like H. erato and H. melpomene
and also in highly dissimilar species such as
Heliconius numata and Heliconius cydno (3).
The region that controls the major red color
patterns (the hindwing rays, forewing band,
and basal forewing patch) maps to a 380-kb
genomic interval corresponding to the D locus
in H. erato and H. melpomene (4, 5) and the
G locus in H. cydno (6). To pinpoint transcrip-
tion units in this interval involved in wing pat-
tern variation we screened for differential color
pattern–related RNA transcription across the en-
tire region. We examined two H. erato forewing
phenotypes: one with a red midwing band, and
one with a yellow midwing band and a red basal
patch (Fig. 2A). For both phenotypes, we com-
pared transcription in basal, midwing, and apical
wing sections from five different developmental
stages (Fig. 2A). This strategy allowed us to
perform separate comparisons of each distinct
red color pattern element, while also including
comparisons between invariant apical forewing
tips to control for population-associated expres-
sion differences unrelated to phenotype. A glob-
al analysis of all stages and tissue sections from
the two phenotypes identified a single major
cluster of red color pattern–associated probes
at the position of optix (Fig. 2A), an intronless
homeobox transcription factor gene with an 801–
base pair coding region. Further analysis of the
data revealed that this pattern-specific differ-
ential expression of optix begins between 12 and
60 hours after pupation and persists throughout
pupal development (fig. S1). No other genes
in the map interval besides optix were consist-
ently expressed in association with color pat-
tern variation.
The finding that optix is differentially ex-
pressed in association with red color patterns in
two H. erato phenotypes raised the question of
whether optix expression is also associated with
red patterns in other H. erato races and other
Heliconius species. We thus examined in situ
spatial expression patterns of optix mRNA in
72-hour pupal forewings and hindwings across
a diversity of phenotypes and species. Across
three geographic races of H. erato, three races
of H. melpomene, and one race of H. cydno, we
found that all red wing patterns, including red
hindwing rays in both H. erato and H. melpomene,
were perfectly prefigured by optix expression
(Fig. 2, B and C). optix also prefigured forewing
bands in H. melpomene plesseni that consist of
both red and white scales (Fig. 2B), consistent
with the activity of an unlinked modifier locus
that modulates the amount of red in forewing
bands (7). Overall, optix expression predicted
color pattern shapes with such precision that it
was possible to identify samples to species and
race by optix expression alone.
The concurrence of mapping and expression
data led us to hypothesize that optix is the major
gene that controls red color pattern variation in
H. erato, H. melpomene, and H. cydno. To test
this, we examined optix genotype-by-phenotype
associations in hybrid zones where there is nat-
ural variation for optix-associated red color patterns
(Fig. 3). These hybrid zones consist of populations
that represent thousands of generations of re-
combination between wing pattern types; there-
fore, DNA sequence associations are due to
linkage with functional variation and not other
phenomena like population structure (8, 9). In a
Peruvian H. erato hybrid zone where linkage
disequilibrium is reported to decay to background
levels within 1 kb (8), we examined DNA se-
quence variation in a series of red color pattern
interval genes from 73 individuals (Fig. 3B).
optix showed by far the strongest association,
with 13 of 30 variable nucleotides having sig-
nificant associations of P < 1 × 10
−15
, with the
strongest being P < 1 × 10
−32
(Fig. 3B and table
S2). Similarly, we sampled 62 individuals from
the parallel Peruvian H. melpomene hybrid zone
and found that optix showed the most significant
associations of any sampled marker, with 5 of
29 variable nucleotides having significant asso-
ciations of P < 1 × 10
−10
, with the strongest
being P < 1 × 10
−32
(Fig. 3C and table S3). In
H. melpomene, a cluster of genes (slu7, kinesin,
GPCR) ~150 kb away from optix showed a total
of six nucleotide associations with P < 1 × 10
−10
(Fig. 3C), though none of these genes displayed
color pattern–related spatial expression and they
showed only marginal associations in H. erato.
This broad spread of associated nucleotides in
H. melpomene is consistent with the observation
that linkage disequilibrium in this hybrid zone
can extend 100 kb or more at wing pattern–
associated sites (9). Lastly, we assessed optix
genotype-by-phenotype associations in 32 indi-
viduals collected from the Costa Rican H. cydno/
H. pachinus hybrid zone wherein, as with H. erato,
highly significant associations are only expected
within 1 kb of functional variation (6). Of all the
sampled genes, optix had the most significant
association with a nucleotide that showed P< 1 ×
A B
H. hewitsoni H. sapho
candidus
H. sapho
leuce
H. eleuchia
eleuchia
H. melpomene
melpomene
200 0 km
100 0 mi
Panama
Costa Rica
Brazil
Venezuela
Colombia
Guyana
Surinama
Guyane
Bolivia
Nicaragua
Ecuador
Peru
H. melpomene
xenoclea
H. melpomene
plesseni
H. melpomene
malleti
H. melpomene
meriana
H. erato
hydara
H. erato
microclea
H. erato
notabilis
H. erato
lativitta
H. erato
amalfreda
H. pachinus H. cydno
galanthus
H. cydno
alithea
H. cydno
alithea
Fig. 1. Mimicry rings show convergent and divergent evolution within
Heliconius. (A) Comimetic Heliconius species are typically distantly related as
shown in this phylogenetic tree of selected Heliconius species. Illustrated are
two major mimicry groups: the blue mimics represented by the H. cydno/
H. pachinus complex and the H. sapho/H. hewitsonii/H. eleuchia complex,
and the red/yellow mimics represented by H. erato and H. melpomene. In
addition to local convergence, wing patterns often vary markedly across the
geographic range of a species to match local mimicry rings. For H. cydno, we
show three of more than a dozen color pattern races. For the red/yellow
mimicry ring, we show five of more than 25 different geographic races of
H. erato and H. melpomene. (B) Simplified geographic distribution of the
H. erato and H. melpomene color pattern forms shown in (A), as color-
coded by the dots. The entire range of H. erato and H. melpomene is much
larger (gray shaded region).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1138
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

10
−12
(Fig. 3E and table S4). Together, these three
independent population genetic analyses all pin-
point optix as the major gene that controls red
color pattern variation.
Because optix alleles from different geo-
graphic races of Heliconius show clear differ-
ences in spatial expression patterns (Fig. 2B), it
follows that there are cis-regulatory differences
between alleles. To rule out allelic variation in
optix protein amino acid sequences, we sequenced
the entire coding region of optix from H. erato
emma, H. erato favorinus, H. erato petiverana,
H. melpomene rosina, H. melpomene melpomene,
H. pachinus, and H. cydno galanthus. All varia-
tion that we observed was synonymous, meaning
that despite 12 to 25 million years of evolution
(10) the optix amino acid sequence has remained
invariable. This was true between highly divergent
races of the same species (within H. erato), be-
tween divergent phenotypes of closely related
species (H. cydno versus H. melpomene), and
between convergent phenotypes of distantly re-
lated species (H. erato versus H. melpomene).
Thus, when considered in the context of the
mapping work, we deduce that the functional
variation between optix alleles is cis-regulatory.
To make an initial assessment of whether
optix is associated with color pattern determina-
tion in non-Heliconius butterflies and moths, we
examined in situ optix mRNA expression in 72-
hour pupal wings of the nymphalid butterflies
Vanessa cardui (Fig. 2D) and Agraulis vanillae
(fig. S2), and the pyralid moth Ephestia kuehniella
H. erato hydara Vanessa cardui
B
C
12
10
8
6
4
2
0
-
l
o
g
1
0

p
-
v
a
l
u
e
FDR 0.01
hybrid B vs. petiverana B
NADUF
TM2 domain
LRR-1
LRR-1
RVT
optix
hydrolase like
GPCR
Kinesin
sorting-nexin
RVT RVT RVT
RVT
DNA H
DALR bves
DNA J
Rte-1
RPS13
TRAPP
Rab 39
RVT? RVT?
Van Gogh
SCY1 like
PDEase-1 Slu 7
hybrid M vs. petiverana M
hybrid A vs. petiverana A
A B
M A
H. erato petiverana H. erato hybrid
B
M
A
D
H. melpomene rosina H. melpomene malleti H. melpomene plesseni
H. erato petiverana H. erato erato H. cydno galanthus
Fig. 2. optix expression prefigures red wing patterns in Heliconius. (A) H. erato
genomic expression tiling microarrays spanning the entire red color pattern
genomic map interval pinpoint optix as the only gene strongly differentially
expressed in association with red forewing pattern elements. Details are pro-
vided in the SOM and fig. S1. (B) optix mRNA expression in 72-hour pupal
wings of different species and races of Heliconius shows spatial association with
all red wing patterns and (C) discrete patches of acute scales. (D) In V. cardui,
optix expression is associated with acute scale patches but not color patterns.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1139
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

(fig. S2). None of these species showed color
pattern–related expression of optix, even though
both butterflies have red color patterns and
A. vanillae is from a heliconiine genus closely
related to Heliconius. In all lepidopteran spe-
cies sampled, however, optix precisely prefig-
ured small, discrete patches of specialized acute
(pointed) scales (Fig. 2, C and D). We have
identified several H. melpomene mutants that
show homeotic transformations of normal red
scales into acute scales (fig. S2), suggesting a
simple regulatory switch between “red” and
“acute” identities and demonstrating how optix
might serve a dual function in determining these
two scale types. In sum, our observations sug-
gest that optix is ancestrally associated with acute
scale morphology in lepidopteran pupal wings
and that its color patterning function in Helico-
nius may be derived.
optix provides a compelling example of a
gene that drives adaptation because its various
alleles are regulatory variants that have pro-
nounced effects on complex large-scale patterns.
Because red color pattern variation is primarily
achieved through optix cis-regulatory variation,
we hypothesize that the different races within a
given Heliconius species share a common wing
prepattern that is interpreted by different optix
alleles in different ways. Although the nature of
this prepattern remains a mystery, the ability of
optix to produce a complex variety of readouts
from a common regulatory background suggests
that it acts as a signal-integrating “input-output”
regulatory gene (11). On the basis of this model,
we speculate that Heliconius color pattern var-
iation evolves primarily through the gain and
loss of pattern-specific cis-regulatory elements,
as do Drosophila wing melanin patterns (12, 13),
although we have yet to identify any specific
optix cis-regulatory elements. That a single gene
drives the evolution of both convergent and di-
vergent color patterns across a range of species
also presents a conceptual conundrum: Conver-
gence and homology are often presented as con-
trasting explanations of similarity, yet the color
patterns of comimetic Heliconius species could
be considered both convergent and homologous
under many modern definitions (14, 15). The
challenge now is to elucidate the functional role
of optix in color pattern formation in order to
understand how it was co-opted and why it
served as such an efficient catalyst for the evo-
lutionary radiation of Heliconius wing patterns.
References and Notes
1. J. Mallet, L. E. Gilbert, Biol. J. Linn. Soc. London 55,
159 (1995).
2. K. S. J. Brown, thesis, Universidade Estadual de Campinas,
Brazil (1979).
3. R. Papa, A. Martin, R. D. Reed, Curr. Opin. Genet. Dev.
18, 559 (2008).
4. R. Papa et al., BMC Genomics 9, 345 (2008).
5. S. W. Baxter et al., Genetics 180, 1567 (2008).
6. N. L. Chamberlain, R. I. Hill, S. W. Baxter, C. D. Jiggins,
M. R. Kronforst, Heredity, published online 9 February
2011 (10.1038/hdy.2011.3).
7. P. M. Sheppard, J. R. G. Turner, K. S. Brown, W. W. Benson,
M. C. Singer, Philos. Trans. R. Soc. London Ser. B 308,
433 (1985).
8. B. A. Counterman et al., PLoS Genet. 6, e1000796
(2010).
9. S. W. Baxter et al., PLoS Genet. 6, e1000794 (2010).
10. N. Pohl, M. P. Sison-Mangus, E. N. Yee, S. W. Liswi,
A. D. Briscoe, BMC Evol. Biol. 9, 99 (2009).
11. D. L. Stern, V. Orgogozo, Science 323, 746 (2009).
12. B. Prud’homme et al., Nature 440, 1050 (2006).
13. T. Werner, S. Koshikawa, T. M. Williams, S. B. Carroll,
Nature 464, 1143 (2010).
14. J. Arendt, D. Reznick, Trends Ecol. Evol. 23, 26
(2008).
15. G. P. Wagner, Nat. Rev. Genet. 8, 473 (2007).
Acknowledgments: We thank M. A. Flores and A. Portugal for
help rearing butterflies, M.-J. Sung for help with electron
microscopy, J. Miller for help with microarray analysis,
and J. Mallet for sharing samples. Supported by funding
from NSF (R.D.R., W.O.M., M.R.K.), NIH (H.M.H., M.R.K.),
and Centre National de la Recherche Scientifique–
Guyane (B.A.C.) Sequences are available on GenBank
( JN102349–JN102354). Microarray data are available on
Gene Expression Omnibus (GSE30221). Permission to
collect and export butterfly tissue was provided by the
Peruvian Ministerio de Agricultura and Instituto Nacional
De Recursos Naturales (004-2008-INRENA-IFFS-DCB and
011756-AG-INRENA); the French Guiana Ministere de
L’Ecologie, de L’Energie, du Developpemet Durable et de
la Mar (BIODAD-2010-0433); the Ecuadorian Ministerio
del Ambiente Ecuadorian (013-09 IC-FAU-DNB/MA); and
the Panamanian Autoridad Nacional del Ambiente
A B
C
D
E
optix
NADUF
TM2 domain
LRR-1
LRR-1
RVT
hydrolase like
GPCR
Kinesin
sorting-nexin
RVT RVT RVT
RVT
DNA H
DALRbves
DNA J
Rte-1
RPS13
TRAPP
Rab 39
RVT?
Van Gogh
SCY1 like
PDEase-1 Slu 7
H. erato
H. melpomene
H. pachinus / H. cydno
-
l
o
g
1
0

p
-
v
a
l
u
e
-
l
o
g
1
0

p
-
v
a
l
u
e
-
l
o
g
1
0

p
-
v
a
l
u
e
H. melpomene H. melpomene
Peru
H. cydno galanthus
Costa Rica
H. pachinus
H. erato H. erato
emma favorinus
amaryllis aglaope
Fig. 3. optix shows highly significant genotype-by-phenotype
associations in three independent Heliconius hybrid zones. (A)
Overlapping H. erato and H. melpomene hybrid zones in Peru,
where optix shows the strongest genotype-by-phenotype association across all the genes surveyed in (B) H. erato and (C) H. melpomene. (D) A Costa
Rican H. cydno/H. pachinus hybrid zone where (E) optix shows the most significant genotype-by-phenotype association out of all the surveyed genes.
The horizontal lines mark a false discovery rate of 0.001.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1140
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

(SC/A-7-11). W.O.M., F.N, and R.D.R. designed experiments
and conducted initial work. R.P. produced butterfly crosses
and generated and analyzed the microarray data with
H.M.H. A.M. generated in situ hybridization data. B.A.C.,
C.P.-D., C.D.J., N.L.C, and M.R.K. contributed nucleotide
association data. R.C. and G.H. assisted with positional
cloning. R.P., A.M., B.A.C., and R.D.R. prepared figures and
R.D.R. wrote the manuscript with support from R.P., A.M.,
and W.O.M. The authors declare that they have no
competing financial interests.
Supporting Online Material
www.sciencemag.org/cgi/content/full/science.1208227/DC1
Materials and Methods
Figs. S1 and S2
Tables S1 to S5
References
11 May 2011; accepted 5 July 2011
Published online 21 July 2011;
10.1126/science.1208227
Chaperonins Facilitate KNOTTED1
Cell-to-Cell Trafficking and Stem
Cell Function
Xianfeng Morgan Xu, Jing Wang, Zhenyu Xuan,* Alexander Goldshmidt, Philippa G. M. Borrill,†
Nisha Hariharan,‡ Jae Yean Kim,§ David Jackson∥
Cell-to-cell communication in plants includes the selective trafficking of transcription factors and
other signals through plasmodesmata. The KNOTTED1 (KN1) homeobox (KNOX) family transcription
factors, which use this pathway, are essential for stem cell establishment and/or maintenance. Here
we show that KN1 trafficking requires the chaperonin complex, which belongs to a group of cytosolic
chaperones that fold specific substrate proteins. Genetic and physical interaction data show a
functional relevance for chaperonins in KNOX family-dependent stem cell maintenance. Furthermore,
tissue-specific complementation assays indicate a mechanistic basis for chaperonin function during
the posttranslocational refolding process. Our study shows that chaperonins are essential for the
cell-to-cell trafficking of a subset of mobile transcription factors and demonstrates the importance of
chaperonin-dependent protein trafficking for plant stem cell function.
C
ell-to-cell communication plays critical
roles in specifying cell fate and coordi-
nating development in all multicellu-
lar organisms. One way in which plant cells
communicate is by the selective trafficking of
transcription factors and other signals through
plasmodesmata (PD), channels that traverse the
cell wall and connect neighboring cells (1–5).
The KNOTTED1 (KN1) homeodomain transcrip-
tion factor was the first plant protein found to
traffic selectively through PD. Subsequent dis-
coveries identified a growing array of mobile
signaling factors, including transcriptional reg-
ulators and small RNAs (3, 4, 6–8). Despite our
increasing recognition of the importance of this
pathway for transmitting signals between cells,
the molecular factors that contribute to selec-
tive trafficking are largely unknown. Here, we
identify chaperonins as essential for the cell-
to-cell trafficking of transcription factors and
demonstrate that chaperonin-dependent pro-
tein trafficking is required for plant stem cell
function.
Previously, we established a trichome res-
cue system in Arabidopsis to follow KN1 cell-
to-cell trafficking in vivo (9). This system
relies on the observation that maize KN1 can
functionally replace its Arabidopsis homolog
SHOOTMERISTEMLESS (STM), and STM pro-
tein has also been shown to traffic (9, 10). Tri-
chomes are leaf hairs that develop from the
epidermis, and require the cell-autonomous activ-
ity of a MYB transcription factor, GLABROUS1
(GL1) (11). In our system, trichome development
is rescued in the gl1 mutant by expressing a GL1
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 11724,
USA.
*Present address: Department of Molecular and Cell Biology,
Center for Systems Biology, University of Texas at Dallas, 800
West Campbell Road, Richardson, TX 75080, USA.
†Present address: John Innes Centre, Norwich Research Park,
Norwich, Norfolk NR4 7UH, UK.
‡Present address: Department of Molecular and Cell Biology,
University of California, Berkeley, CA 94720–3200, USA.
§Present address: Division of Applied Life Science (BK21/
World Class University program), Plant Molecular Biology and
Biotechnology Research Center, Gyeongsang National Uni-
versity, Jinju 660-701, Korea.
||To whom correspondence should be addressed. E-mail:
jacksond@cshl.edu
Fig. 1. Identification of a chaperonin, CCT8, essential for KN1 trafficking. (Aand
B) The starting line for mutagenesis (A) (gl1; pRbcS::GFP~GL1~KN1
C
) showing
trichome rescue (arrowhead), whereas the 25-277 mutant (B) lost trichome rescue
(arrows) and had upwardly curled leaves (asterisk). (C and D) Scanning electron
microscopic images of leaf epidermis fromthe starting line (C) or mutant (D).
(E and F) Confocal image of a leaf cross section fromthe starting line (E) shows
accumulation of GFP~GL1~KN1
C
in both the mesophyll and the epidermis,
whereas the 25-177 mutant (F) has normal accumulation of GFP~GL1~KN1
C
in
the mesophyll but a severe reduction in the epidermis. Dashed lines denote the
epidermis; arrows point to nuclei with fusion protein. Chlorophyll autofluorescence
is shown in red. Corresponding genotypes are marked in each image. (G) Western
blotting shows a similar expression of the pRbcS::GFP~GL1~KN1
C
transgene in
both the starting line (control) and the 25-177 mutant. Ponceau staining was used
as the loading control. (H) CCT8 gene structure, showing different mutant alleles.
Black bars indicate exons; lines and gray bars represent introns and untranslated
regions, respectively. T-DNA insertions are shown as open triangles. (I) Control
experiments: (1) Mesophyll-expressed GL1~KN1
C
could rescue trichomes in gl1
but failed to rescue in gl1; cct8-1. (2) Expression of the GFP~GL1~KN1
C
fusion in
the gl1; cct8-1 epidermis was sufficient to rescue trichomes. Trichome rescue
efficiency was scored as the percentage of T1 transformants (n >40) with trichome
rescue. (J) A pfd6-1; gl1; pRbcS::GFP~GL1~KN1
C
plant showing normal trichome
rescue. Scale bars: 0.5 cm (A and B), 400 mm (C and D), 25 mm (E and F).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1141
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

fusion to KN1 in the mesophyll cell layers, with
the use of the Rubisco small subunit 2b promoter
( pRbcS::GL1~KN1). Trichome rescue depends
on trafficking of the GL1~KN1 fusion protein
from the mesophyll to the epidermal cell layer.
Using this system, we found that the homeo-
domain of KN1 is both necessary and sufficient
for KN1 trafficking (9).
To identify factors required for KN1 traf-
ficking, a trichome-rescued line ( gl1; pRbcS::
GFP~GL1~KN1
C
, where GFP is green fluo-
rescent protein and KN1
C
represents the KN1
C-terminal trafficking domain) was mutagenized,
and mutants lacking trichomes were isolated
(12). Asecondary screen, using confocal imaging,
was employed to confirm absence or reduction
of the GFP~GL1~KN1
C
fusion protein from the
epidermis. After screening ~2500 M2 lines, we
identified five putative trafficking mutants; here,
we present the characterization of one such mu-
tant, initially referred to as “25-177.”
Compared with the starting line, the 25-177
mutant lost trichome rescue in the first four
leaves and showed reduced trichome rescue in
later leaves (Fig. 1, Ato D). Consistent with the
loss of trichomes, the epidermal GFP~GL1~KN1
C
signal was severely reduced in the mutant (Fig. 1,
E and F); however, the overall expression level
of the pRbcS::GFP~GL1~KN1
C
transgene was
unchanged (Fig. 1G). Hence, this mutant ap-
peared to have impaired KN1 trafficking. The
mutant also displayed pleiotropic developmental
phenotypes, including a reduced size and up-
wardly curling leaves with polarity defects (Fig.
1B and fig. S1).
The 25-177 mutant behaved as a single re-
cessive locus, which we mapped to a ~170 kilo–
base pair interval on chromosome 3, containing
~25 genes. This interval was polymerase chain
reaction (PCR)–amplified fromthe mutant as over-
lapping ~5-kb amplicons, which were pooled and
sequenced to >20× coverage. We compared the
sequence to the Columbia reference genome and
identified potentially disruptive point mutations.
We found one such mutation (G to A, causing an
amino acid substitution from Ala
512
to Val
512
)
within the At3g03960 gene, which encodes a
chaperonin subunit, CCT8 (for chaperonin con-
taining TCP1, also called TRiC, for TCP1-ring
complex) (Fig. 1H). We complemented the tri-
chome rescue defect, as well as other devel-
opmental abnormalities of the mutant, with a
GFP-tagged CCT8 genomic clone, confirming
the lesion in CCT8 as the causal mutation (see
below). Thus, we renamed the mutant as cct8-1,
and the mutation (Ala
512
→Val
512
) was at a residue
that is highly conserved in different organisms,
supporting the idea that it is critical for function.
We isolated two additional alleles (SALK082168C;
cct8-2 and SALK032358; cct8-3) fromthe public
collection of transfer DNA (T-DNA) insertional
mutants (13). Both had T-DNA insertions in the
5′ untranslated region of CCT8 (Fig. 1H), leading
to reduced transcript levels (fig. S2A). Single
mutants of cct8-2 or cct8-3 had upwardly curled
leaves like cct8-1, but they developed trichomes
normally (fig. S2B), indicating that CCT8 is not
required for trichome development. However,
when cct8-2 was introduced into the trichome-
rescue background (cct8-2; gl1; pRbcS::
GFP~GL1~KN1
C
), trichome rescue was lost
(fig. S2C), confirming the requirement for
CCT8 in the KN1 trafficking-dependent trichome
rescue. All three cct8 alleles were weak alleles,
because chaperonin null mutants are lethal in all
organisms that have been tested [reviewed in
(14)], including Arabidopsis (fig. S3).
CCT8 is a subunit of the type II chaperonin
complex, which forms oligomeric double-ring
assemblies that assist in protein folding [reviewed
in (14)]. Type II chaperonins are found in
Archaea and in the eukaryotic cytosol, and they
are composed of eight or nine related subunits.
Chaperonins have a more restricted substrate
spectrum compared with other chaperones (e.g.,
HSP70s) and have been best characterized in
folding actin and tubulin in yeast and mammalian
cells (14). In Arabidopsis, nine genes encode
eight different chaperonin subunits (CCT1-8),
and CCT8 is expressed widely (fig. S4), which is
consistent with its predicted role in assisting the
folding of multiple proteins. Because CCT8 was
required for KN1 trafficking, we tested whether
trafficking might require the whole chaperonin
complex. Using artificial microRNAs, we reduced
CCT1 or CCT5 expression in the trichome rescue
system, and in both cases, trichome rescue was
lost (fig. S5), suggesting that the whole chaperonin
complex is involved in cell-to-cell trafficking.
Given that chaperonins facilitate protein fold-
ing, we asked if the trafficking defects in cct8
mutants might be indirect. First, because GFP
is a chaperonin substrate (15), we asked whether
the GFP tag in our trichome rescue system
might be responsible for the loss of trafficking
in cct8 mutants. Mesophyll-specific expression
of GL1~KN1
C
rescued trichomes in the gl1 mu-
tant but failed to do so in gl1; cct8-1 (Fig. 1I),
showing that the trafficking defects in cct8-1 were
independent of the GFP tag. A second indirect
cause for impaired trafficking in cct8-1 could
be instability of the GFP~GL1~KN1
C
fusion pro-
tein, due to a general folding defect. This pos-
sibility is unlikely, given that the fusion protein
accumulated and was targeted to the nuclei in
mesophyll cells of the cct8-1 mutant (Fig. 1F).
Nevertheless, we also expressed GFP~GL1~KN1
C
directly in the epidermis of gl1; cct8-1 plants,
using the GL1 promoter. This expression led to
trichome rescue, indicating that a general folding
defect was not responsible for the reduced tri-
chome rescue in cct8-1 (Fig. 1I). A third indirect
effect could arise from the established functions
of chaperonins in folding actins and tubulins [re-
viewed in (14)], because the cytoskeleton has been
implicated in trafficking through PD (16, 17).
Therefore, we tested whether mutations in CCT8
Fig. 2. Physical and genetic interaction between CCT8 and KN1/STM. (A) CCT8 coimmunoprecipitated with
KN1 or STM (arrows), but not with MP
CMV
(movement protein of the cucumber mosaic virus). (B and C)
GFP~KN1, when expressed in the L3 of the SAM using pWUS, moved and accumulated in the L2 layer
[arrow in (B)]. However, in the cct8-1 mutant, this movement was severely attenuated [arrow in (C)].
Dashed lines denote the L2 layer. WT, wild type. (D to H) Meristem defects in stm-10 were enhanced by
cct8-1. The stm-10 mutant meristemterminated after two to six rosette leaves [four shown in (D)], whereas
stm-10; cct8-1 double mutants only developed zero to three true leaves [one shown in (F)]. For comparison,
a cct8-1 mutant (E) and a null stm-11 mutant (G) are shown. Plants are 15 days old. (H) Quantification of
the phenotypic enhancement in the stm-10; cct8-1 double mutants. The y axis shows the percentage of
plants; the number of rosette leaves is shown on the x axis. Scale bars: 20 mm (B and C), 0.5 cm (D to G).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1142
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

might affect KN1 trafficking by compromising
the cytoskeleton. First, when grown with drugs to
disrupt microtubules (oryzalin) or microfilaments
(latrunculin B), the cct8-1 mutants exhibited wild-
type levels of root growth inhibition (fig. S6),
suggesting that they did not have gross cytoskel-
etal defects. We then introduced an Arabidopsis
prefoldin6 (PFD6) mutant, pfd6-1, into our tri-
chome rescue system. Prefoldin acts together with
chaperonins during actin and tubulin folding (18),
and the pfd6-1 mutant has cytoskeletal defects and
increased sensitivity to oryzalin and latrunculin B
(19) (fig. S6). Normal trichome rescue was seen
in the presence of the pfd6-1 mutation (Fig. 1J),
suggesting that defects in KN1 trafficking were
not due to cytoskeletal defects.
We next asked if chaperonins might directly
facilitate KN1 trafficking; thus, we tested for a
physical association. Using in vivo coimmuno-
precipitation assays, we found that both KN1 and
STM interacted with CCT8 (Fig. 2A). Further-
more, a second chaperonin subunit, CCT7, also
coimmunoprecipitated with KN1 (fig. S7), sug-
gesting that KN1 associated with the chaperonin
complex and that chaperonins might directly fa-
cilitate KN1 trafficking.
The function of several plant proteins depends
on their ability to traffic between cells. For ex-
ample, TRANSPARENT TESTA GLABROUS1
(TTG1) is a WD40-repeat protein whose move-
ment is involved in trichome spacing (20), and
SHORTROOT(SHR) is a plant-specific transcrip-
tion factor whose movement is essential in spec-
ifying root endodermal identity (21). CCT8 was
required for trafficking of TTG1, but not of SHR
(fig. S8, A to D). Consistent with its dependency
on CCT8 for trafficking, TTG1 also physically in-
teracted with CCT8 (fig. S8E). The requirement
for CCT8 in the trafficking of the structurally
unrelated KN1 and TTG1 proteins, but not for
SHR, suggests that distinct pathways exist for
proteins to traffic between cells.
Our trichome rescue systemreports KN1 traf-
ficking between leaf cells; this process is devel-
opmentally relevant in maize in the context of
dominant Kn1 mutations (22). However, KN1
and STM normally function in the shoot apical
meristem (SAM) to initiate and/or maintain the
shoot stem cell population (23, 24). Therefore,
we asked whether chaperonins might also be
required for KN1 trafficking in the SAM. When
driven by a WUSCHEL promoter ( pWUS), which
drives expression within the L3 layer of the SAM
(9), trafficking of GFP~KN1 into the L2 layer was
visible (Fig. 2B). However, movement into the
L2 layer was reduced in cct8-1 mutants (Fig. 2C),
indicating that KN1 trafficking in the SAM also
required CCT8.
We then assayed to determine if the reduced
KN1 homeobox family (KNOX) protein traffick-
ing in the SAM of cct8-1 mutants had functional
importance. Given that KN1 can functionally com-
plement Arabidopsis stm mutants (10) and that
STM also traffics (9, 10), we genetically tested
whether CCT8 was required for STMfunction. A
weak allele of stm, stm-10, develops ~two to six
rosette leaves before the SAM terminates (Fig.
2D) (25). The cct8-1 mutant had no obvious de-
fects in the number of leaves, rate of their initia-
tion, or in SAM function (Fig. 2E); however,
stm-10; cct8-1 double mutants had an enhanced
SAM termination phenotype (Fig. 2, F and H)
and sometimes resembled the null stm-11 allele
(Fig. 2G). The enhancement of stem cell main-
tenance defects in stm-10 by cct8-1 is probably
due to impaired STM trafficking. To support this
hypothesis, we made a more severe knockdown
of CCT8 using cosuppression, which resulted
in obvious meristem defects in the shoot, but
not in the root, resembling an stm phenotype
(fig. S9). Together with the physical and gen-
etic interactions between CCT8 and STM, these
findings support our hypothesis that CCT8
contributes to stem cell maintenance through its
impact on STM movement. Alternatively, the role
of CCT8 in stem cell maintenance might result
fromimpaired trafficking of other factors required
for this function.
Chaperone activity was previously suggested
to be important for cell-to-cell trafficking through
PD (26). In addition, the beet yellows virus en-
codes a homolog of the HSP70 chaperone to
enable viral spread between cells (27); HSP70
homologs are found in PD-enriched cell wall
fractions, as well as in the phloem, and they can
traffic from cell to cell (28). However, our report
identifies a chaperone that facilitates the traf-
ficking of specific plant signaling proteins, and
we therefore investigated the underlying mech-
anism. Like all chaperones, chaperonins carry
out folding using energy fromcycled adenosine
triphosphate (ATP) hydrolysis (14). Because
ATPase activity has been observed in PD (29)
and CCTsubunits are present in a PD-enriched
proteome (30), we asked if CCT8 was enriched
at PD. A functional, internally tagged CCT8
(CCT8-iGFP) localized throughout the cyto-
plasm (Fig. 3A and fig. S10); however, upon
plasmolysis we sawno evidence of enrichment at
PD (fig. S10), indicating that CCT8 was un-
likely to be specifically enriched at PD. We also
considered the possibility that CCT8 might func-
tion as a mobile transport chaperone, analogous
to a mechanism for nuclear transport (31). How-
ever, imaging of CCT8-iGFP demonstrated that
CCT8 remains within its cell of origin, making
this possibility unlikely (Fig. 3, B and C).
We then considered additional models for
how chaperonins might facilitate KN1 trafficking.
One possibility is that they are required for guiding
Fig. 3. Epidermal expression of CCT8 can restore the trafficking-dependent trichome rescue in
cct8-1. (A) Root cells from complemented cct8-1; pCCT8::CCT8-iGFP plants, showing CCT8-iGFP
localization in the cytoplasm. N denotes nuclei. (B and C) CCT8-iGFP, driven by the ML1 promoter,
accumulated specifically in the epidermis, both in the SAM (B) and in leaves [(C), where chlo-
rophyll autofluorescence is shown in magenta], suggesting its cell autonomy. (D and E) Ex-
pression of HA~CCT8 in the mesophyll failed to restore trichome rescue [(D), pRbcS::HA~CCT8;
cct8-1; gl1; pRbcS::GFP~GL1~KN1
C
], whereas epidermal expression of HA~CCT8 complemented
the trichome rescue defect in cct8-1 [(E), pML1::HA~CCT8; cct8-1; gl1; pRbcS::GFP~GL1~KN1
C
].
(F) Quantification of the trichome rescue efficiency shown as the percentage of plants with
trichomes rescued in the third and fourth leaves. Averages from multiple independent lines are
shown. Error bar indicating SE; double asterisks indicate P < 0.01, in an unpaired t test. Scale
bars: 10 mm (A to C), 1 mm (D and E) (33–39).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1143
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

or sorting KN1 to the PD or for the partial un-
folding of KN1 into a transport-competent state
(26), similar to their role in posttranslational ER
translocation (32). Alternatively, chaperonins might
function in the refolding of KN1 in destination
cells, so that the protein becomes functional after
its passage through the PDchannel. To distinguish
between these possibilities, we expressed CCT8,
as functional HA~CCT8 or CCT8~iGFP fusions,
in either the epidermis or the mesophyll cell lay-
ers of the cct8-1 mutant and asked if expression
was sufficient to complement KN1 trafficking.
We found that trichome rescue in cct8-1 could be
restored by the epidermal expression of CCT8,
but not by mesophyll expression (Fig. 3, D to F),
indicating that CCT8 was required in destination
cells for proper KN1 trafficking. Hence, we pro-
pose that CCT8, and by inference the chaperonin
complex, facilitates successful KN1 trafficking
through refolding translocated, partially unfolded
KN1 proteins (fig. S11).
In summary, we demonstrated a direct in-
volvement of the chaperonin complex in me-
diating cell-to-cell signaling by Arabidopsis and
maize KNOX proteins, suggesting that this mech-
anismis conserved in diverse plant species. More
important, our studies provide evidence that
chaperonin-dependent protein trafficking is re-
quired for stem cell maintenance, demonstrating
the critical importance of this signaling pathway.
Our results also raise an important question about
what exact molecular functions are provided by
mobility of KNOXproteins. We hypothesize that
trafficking could serve to establish protein con-
centration gradients with distinct developmental
readouts or could act as a fail-safe mechanism
for maintaining stemcells; and future studies will
address these possibilities.
References and Notes
1. A. J. Maule, Curr. Opin. Plant Biol. 11, 680 (2008).
2. W. J. Lucas, B. K. Ham, J. Y. Kim, Trends Cell Biol.
19, 495 (2009).
3. X. M. Xu, D. Jackson, Curr. Opin. Plant Biol. 13, 684 (2010).
4. T. Kurata, K. Okada, T. Wada, Curr. Opin. Plant Biol.
8, 600 (2005).
5. I. Kim, P. C. Zambryski, Curr. Opin. Plant Biol. 8, 593
(2005).
6. W. J. Lucas et al., Science 270, 1980 (1995).
7. A. Carlsbecker et al., Nature 465, 316 (2010).
8. P. Dunoyer et al., Science 328, 912 (2010);
10.1126/science.1185880.
9. J. Y. Kim, Y. Rim, J. Wang, D. Jackson, Genes Dev. 19,
788 (2005).
10. J. Y. Kim, Z. Yuan, D. Jackson, Development 130,
4351 (2003).
11. D. G. Oppenheimer, P. L. Herman, S. Sivakumaran,
J. Esch, M. D. Marks, Cell 67, 483 (1991).
12. Materials and methods are available as supporting
material on Science Online.
13. J. M. Alonso et al., Science 301, 653 (2003).
14. A. L. Horwich, W. A. Fenton, E. Chapman, G. W. Farr,
Annu. Rev. Cell Dev. Biol. 23, 115 (2007).
15. J. S. Weissman, H. S. Rye, W. A. Fenton, J. M. Beechem,
A. L. Horwich, Cell 84, 481 (1996).
16. K. M. Wright et al., Traffic 8, 21 (2007).
17. N. Winter, G. Kollwig, S. Zhang, F. Kragler, Plant Cell
19, 3001 (2007).
18. J. Martín-Benito et al., EMBO J. 21, 6377 (2002).
19. Y. Gu et al., Proc. Natl. Acad. Sci. U.S.A. 105, 18064
(2008).
20. D. Bouyer et al., PLoS Biol. 6, e141 (2008).
21. K. Nakajima, G. Sena, T. Nawy, P. N. Benfey, Nature
413, 307 (2001).
22. S. Hake, M. Freeling, Nature 320, 621 (1986).
23. E. Vollbrecht, L. Reiser, S. Hake, Development 127,
3161 (2000).
24. J. A. Long, E. I. Moan, J. I. Medford, M. K. Barton, Nature
379, 66 (1996).
25. S. Kanrar, O. Onguka, H. M. Smith, Planta 224, 1163
(2006).
26. F. Kragler, J. Monzer, K. Shash, B. Xoconostle-Cazares,
W. J. Lucas, Plant J. 15, 367 (1998).
27. V. V. Peremyslov, Y. Hagiwara, V. V. Dolja, Proc. Natl.
Acad. Sci. U.S.A. 96, 14771 (1999).
28. K. Aoki, F. Kragler, B. Xoconostle-Cázares, W. J. Lucas,
Proc. Natl. Acad. Sci. U.S.A. 99, 16342 (2002).
29. J. Cronshaw, J. Histochem. Cytochem. 28, 375 (1980).
30. L. Fernandez-Calvino et al., PLoS ONE 6, e18880
(2011).
31. S. Jäkel, J. M. Mingot, P. Schwarzmaier, E. Hartmann,
D. Görlich, EMBO J. 21, 377 (2002).
32. K. Plath, T. A. Rapoport, J. Cell Biol. 151, 167
(2000).
Acknowledgments: We thank the Arabidopsis Biological
Resource Center, K. Barton, P. Benfey, Y. Gu,
J.-Y. Lee, and D. Chitwood for seeds and plasmids. We
also thank M. Regulski for assisting with the Illumina
sequencing library; T. Mulligan for plant care; and
R. Martienssen, Z. Lippman, F Kragler, and members of
the Jackson lab for discussion and comments on the
manuscript. X.M.X. is a LSRF Fellow of the Life Sciences
Research Foundation and a Chapman Fellow of the
Chapman Foundation. A.G. is supported by the
Vaadia–Binational Agricultural Research and Develop-
ment Postdoctoral Fellowship Award FI-413. P.G.M.B. and
N.H. were supported by the Cold Spring Harbor
Laboratory’s Undergraduate Research Program. J.Y.K. is
currently supported by the Next-Generation BioGreen
21 Program (SSAC grant PJ008109), Rural Development
Administration, Korea. This research is supported by
the NSF (award 1027003). Author contributions: X.M.X.
performed the imaging, RT-PCRs, immunoprecipitations
(Ips), and genetic and transgenic analyses. J.W. started
the screen and performed mutant cloning. Z.X.
processed the Illumina sequencing data. P.G.M.B.
assisted with the IP assays. N.H. analyzed the
cct3-1 embryos. A.G. took the scanning electron
microscopy images. J.Y.K. generated the starting line
for mutagenesis. X.M.X. and D.J. analyzed the data and
wrote the manuscript. All of the authors read and
approved the manuscript.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1141/DC1
Materials and Methods
Figs. S1 to S11
References (33–39)
17 March 2011; accepted 8 July 2011
10.1126/science.1205727
Nectins Establish a Checkerboard-Like
Cellular Pattern in the Auditory Epithelium
Hideru Togashi,
1
* Kanoko Kominami,
1
* Masazumi Waseda,
1
† Hitomi Komura,
1
‡ Jun Miyoshi,
2
Masatoshi Takeichi,
3
Yoshimi Takai
1
§
In the auditory epithelium of the cochlea, the sensory hair cells and supporting cells are
arranged in a checkerboard-like fashion, but the mechanism underlying this cellular patterning
is unclear. We found that mouse hair cells and supporting cells express the immunoglobulin-like
adhesion molecules nectin-1 and -3, respectively, and that their interaction mediates the
heterotypic adhesion between these two cell types. Genetic removal of nectin-1 or -3 disrupted
the checkerboard-like pattern, inducing aberrant attachment between hair cells. When cells
expressing either nectin-1 or -3 were cocultured, they arranged themselves into a mosaic pattern. Thus,
nectin-1 and -3 promote the formation of the checkerboard-like pattern of the auditory epithelia.
A
nimal tissues comprise multiple cell types,
which are arranged in complex, elabo-
rate patterns. One such pattern is the
checkerboard-like cell arrangement, which is ob-
served in certain tissues such as the oviduct and
the auditory epithelia (1, 2). The organ of Corti in
the inner ear consists of mechanosensory hair
cells (subgrouped into inner and outer hair cells),
which are interdigitated with various types of
supporting cells (2); these cells are arranged in
a checkerboard-like pattern (fig. S1). Anumber
of genes or molecules, including those known to
regulate planar cell polarity, have been implicated
in the generation of the highly ordered structures
in the cochlea (3–7). A mathematical model pre-
dicted that the checkerboard-like pattern could
be generated by a mixture of two cell types,
when their heterotypic cell-cell adhesions domi-
nated over their homotypic ones (8). However,
how the checkerboard-like pattern of cells is es-
tablished is unclear.
Epithelial cell-cell adhesions are mediated via
the adherens junctions (AJs), where two classes
of molecules, cadherins and nectins, cooperate.
The major role of cadherins is to connect cells
expressing the same cadherins through their
1
Division of Molecular and Cellular Biology, Department of
Biochemistry and Molecular Biology, Kobe University Graduate
School of Medicine, Kobe 650-0017, Japan.
2
Department of
Molecular Biology, Osaka Medical Center for Cancer and
Cardiovascular Disease, Osaka 537-8511, Japan.
3
Laboratory
for Cell Adhesion and Tissue Patterning, RIKEN Center for
Developmental Biology, Kobe 650-0047, Japan.
*These authors contributed equally to this work.
†Present address: Division of Genetics and Department of
Cancer Biology, the Institute of Medical Science, the Uni-
versity of Tokyo, Tokyo 108-8639, Japan.
‡Present address: Department of Pathology and Tumor Biol-
ogy, Graduate School of Medicine, Kyoto University, Kyoto
606-8501, Japan.
§To whom correspondence should be addressed. E-mail:
ytakai@med.kobe-u.ac.jp
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1144
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

homophilic interactions (9, 10). In contrast, nectins,
a family of immunoglobulin-like molecules, com-
prising four members (nectin-1, -2, -3, and -4),
promote both homophilic and heterophilic inter-
actions between the members (11–13). Their
heterophilic interactions are stronger than their
homophilic interactions in the following order:
nectin 1-3 > nectin 2-3 > nectin 1-1, 2-2, and 3-3,
which might contribute to the mosaic patterning
of two cell populations (11–16). We investigated
the potential role of nectins in the formation of
the checkerboard-like pattern of the auditory epi-
thelial cells in the mouse.
We first examined the localization of nectin-
1, -2, and -3 in the auditory epithelia at different
stages of development. At embryonic day 14
(E14), the three nectins were already localized
at the junctions between the auditory epithelia,
although the checkerboard-like cellular pattern
was not clearly visible; this pattern became de-
tectable at E16 and even clearer at postnatal day
1 (P1). In the middle turn of the cochlea at P1,
nectin-1 was localized only at the heterotypic
boundaries between hair (myosin VIIa–positive)
and supporting (myosin VIIa–negative) cells
(Fig. 1, arrows), not at the junctions between
supporting cells. The distribution of nectin-3
was similar to that of nectin-1; that is, nectin-3
was localized along the heterotypic junctions
between hair cells and supporting cells (Fig. 1,
arrows). However, unlike nectin-1, nectin-3 was
also weakly detected at the junctions between
supporting cells (Fig. 1, arrowheads). On the
other hand, nectin-2 was uniformly distributed at
the boundaries between hair cells and supporting
cells, as well as between supporting cells (Fig. 1).
The localization patterns of nectins in the audi-
tory epithelia at E16 were essentially identical to
those at P1 (fig. S2). All of these nectin mole-
cules were located on the apical side of the
lateral membranes of hair cells and supporting
cells, where the AJs are detected (2, 17) (fig.
S3). To identify the cells expressing nectin-1 and
-3, we performed in situ hybridization for each at
P1. Nectin-1 mRNA was expressed in hair cells,
whereas nectin-3 mRNAwas detected in support-
ing cells (fig. S4). Thus, nectin-1 expressed in hair
cells and nectin-3 expressed in supporting cells
are concentrated at their heterotypic boundaries
(fig. S1E).
We next analyzed the cochleas of mice in
which the genes encoding these nectins were
knocked out (KO) (Fig. 2). The numbers of dif-
ferentiated hair cells and supporting cells were
not altered in these mutant mice (Fig. 2F). How-
ever, the checkerboard-like pattern of these cells
was dramatically affected by the deficiency of
nectin-3. In nectin-3 KO mice, two or three hair
cells aberrantly attached to each other, as revealed
by staining for F-actin and the tight-junction
marker ZO-1 (Fig. 2D, arrows). The signals for
ZO-1 and F-actin appeared as a single line at the
borders between neighboring hair cells, indicating
that these cells came into direct contact with each
other. Most (62.4%, n = 15) outer hair cells aggre-
gated in this manner, disrupting the checkerboard-
like pattern. Inner hair cells were also aberrantly
attached to one another, although less frequent-
ly than outer hair cells. In addition, these inner
hair cells lost their contacts with inner pillar
cells, resulting in the elongation of the contact
Fig. 1. Localization of nectins in the auditory epithelia of wild-type mice at P1. Nectins and myosin VIIa
(MyoVIIa) were doubly stained at apical surfaces of the auditory epithelia in the middle turn of cochleae. Hair
cells were identified on the basis of MyoVIIa expression. OHCs, outer hair cells; IHCs, inner hair cells. Arrows
point to examples of nectin-1 or -3 signals concentrated at the borders between hair cells and supporting
cells. Arrowheads indicate examples of nectin-3 signals concentrated at junctions between supporting cells.
Fig. 2. Cellular patterning in the auditory epithelia of wild-type (WT) and nectin KO mice at P1. (A to
D) F-actin and ZO-1 were doubly stained in the middle turn of the cochlea. Arrows point to examples
of contact sites aberrantly formed between hair cells. Arrowheads indicate abnormally formed
junctions between inner phalangeal cells. (E and F) Statistical analysis of nectin KO phenotypes. (E)
Number of hair cells showing aberrant attachment to other cells in each photographic field, the size
of which is equivalent to those in (A) to (D). Left, attached OHCs. Middle, attached IHCs. Right, IHCs
detached from inner pillar cells (PC). (F) Numbers of supporting cells (SC), OHCs, IHCs, and total hair
cells (HC) per photographic field. Results shown are the means T SEM. *P < 0.01; **P < 0.001. Data
were collected from three individuals and were quantified using five fields in each individual.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1145
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

sites between inner phalangeal cells (depicted
in gray in fig. S1D) (Fig. 2D, arrowheads). In
nectin-1 KO mice, similar phenotypes were ob-
served, but the effects of nectin-1 KO were much
milder than those of nectin-3 KO in mice (Fig.
2, B and E). On the other hand, nectin-2 KO
mice did not show any detectable phenotypes
(Fig. 2C).
To understand howthe above phenotypes were
generated, we examined the distribution patterns
of the nectins remaining in these KOmice (Fig. 3).
In nectin-1 KO mice, the uniform distribution of
nectin-2 was not altered (Fig. 3B), whereas nectin-3
became localized more intensely at the junctions
between supporting cells (Fig. 3B, arrowheads).
In nectin-2 KO mice, the localization patterns of
nectin-1 and -3 remained unchanged (Fig. 3C).
In nectin-3 KO mice, nectin-1 was lost from the
heterotypic junctions between hair cells and sup-
porting cells; instead, it became concentrated at
the boundaries between aberrantly contacting hair
cells (Fig. 3D, arrows, and fig. S5), suggesting
that the nectin-1 in these mice was undergoing
only homophilic interactions at the interfaces be-
tween hair cells. Thus, the loss of nectin-1 and -3
caused the redistribution of their heterophilic
partners to ectopic sites.
To verify the roles of nectin-1 and -3 in cel-
lular patterning, we reexamined the behavior of
cells expressing these nectins in cultures. We pre-
pared HEK293 cells stably expressing nectin-1 or
Fig. 3. (A to D) Distribution of nectins in the middle turn of cochleae of wild-type and nectin KO mice at
P1. Arrowheads point to examples of up-regulated nectin-3 signals at the boundaries between supporting
cells. Arrows point to nectin-1 signals abnormally concentrated at the boundaries between hair cells.
Fig. 4. Mosaic patterning of cells formed by hetero-
philic interaction of nectin-1 and -3. (A to D) Bound-
aries between two colonies of HEK293 cells, which
were doubly transfected with a nectin and EGFP (green)
or mCherry (red) in the combinations indicated. (A)
Control cells expressing EGFP and mCherry; (B) cells
expressing nectin-1/EGFP and nectin-1/mCherry; (C)
cells expressing nectin-3/EGFP and nectin-3/mCherry;
(D) cells expressing nectin-1/EGFP and nectin-3/mCherry.
(E) Time-lapse microscopy of a coculture of MDCK cells
expressing nectin-1/EGFP and those expressing nectin-3/
mCherry. Arrowhead, cell expressing nectin-1; asterisks,
pair of cells expressing nectin-3. Time (t) is in minutes.
(F) Hypothetical profiles of nectin-nectin interactions at
various cell-cell boundaries. Bidirectional arrows
indicate heterophilic or homophilic interactions be-
tween nectins. The thickness of the arrows represents
the relative strength of nectin-nectin interactions. The
broken arrow indicates less frequent cases. When het-
erophilic interaction between nectin-1 and -3 occur,
the majority of these molecules are recruited to
heterophilic binding sites; therefore, their potential
homophilic interactions are not depicted in the wild-
type cartoon. This may also be the case for interactions
between nectin-2 and -3, but this has not been tested.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1146
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

-3, into which the fluorescent proteins enhanced
green fluorescent protein (EGFP) or mCherry
were introduced to distinguish between the two
cell lines. These cells were sparsely cultured to
allow the formation of independent colonies.
When their colony edges came into contact with
one another, their boundaries were examined.
Cells expressing identical nectin types did not
intermingle at the border, whereas those express-
ing nectin-1 and -3 mutually invaded the counter
colony, resulting in the formation of a mosaic
pattern (Fig. 4, A to E, and fig. S6). We also per-
formed time-lapse video microscopy using a
coculture of MDCK cells expressing nectin-1 or
-3 (N1- and N3-MDCK cells). In the supporting
movie (movie S1 and Fig. 4E), one N1-MDCK
cell (arrowhead) initially adhered to one of a pair
of N3-MDCK cells (asterisks); subsequently, the
former cell invaded the space between the two
N3-MDCKcells. As a result, N1- and N3-MDCK
cells were rearranged into a mosaic pattern. Sim-
ilar behavior of cells was repeatedly observed in
multiple experiments.
Thus, we propose that the heterophilic in-
teractions between nectin-1 and -3 are critical
for establishing the checkerboard-like pattern
of hair cells and supporting cells. The molec-
ular interaction between nectin-1 and -3 is the
strongest of all possible combinations of the
three nectins, which is likely to be responsible
for the checkerboard-like assembly of these
cells (Fig. 4F), as predicted by the mathemat-
ical model (8). The loss of nectin-3 removed such
biased cell-cell adhesion, leading to cell rear-
rangement, including attachments between hair
cells (Fig. 2D), as explained by the differential
adhesiveness hypothesis (18). Nectin-1 KO mice
displayed milder phenotypes. In these mice, the
relatively strong interaction between nectin-3
and -2 probably retained the adhesion between
hair cells and supporting cells; on the other hand,
the adhesion between supporting cells should
have been enhanced as a result of the redistri-
bution of nectin-3 to these sites. These combi-
natory situations probably suppressed adhesion
between hair cells (Fig. 4F). In nectin-2 KO mice,
the heterophilic interactions between nectin-1
and -3 persisted; this explains the absence of a
phenotype in these mice. In the absence of
nectins, the cell junctions were not disrupted.
This is most likely due to the coexpression of
classic cadherins in the auditory epithelia. Hair
cells and supporting cells are thought to be seg-
regated through the process of lateral inhibi-
tion mediated by Notch-Delta signaling (4, 19),
and such processes themselves might contribute
to the spatial separation of these cells (20–22).
However, genetic inactivation of Notch signal-
ing does not impair the checkerboard-like pat-
tern, although it does result in an increase in
the number of hair cells (4). This suggests that
lateral inhibition is insufficient to create the
checkerboard-like cellular pattern, stressing the
importance of nectins in this patterning process.
It is of note that heterophilic interactions be-
tween Hibris and Roughest, other members
of the immunoglobulin superfamily, also con-
tribute to the cell arrangement in the Drosophila
eye (23, 24), suggesting that similar mecha-
nisms are conserved for cellular patterning across
species.
References and Notes
1. H. I. Yamanaka, H. Honda, Int. J. Dev. Biol. 34, 377
(1990).
2. R. L. Gulley, T. S. Reese, J. Neurocytol. 5, 479
(1976).
3. E. McKenzie, A. Krupin, M. W. Kelley, Dev. Dyn. 229,
802 (2004).
4. P. J. Lanford et al., Nat. Genet. 21, 289 (1999).
5. P. Chen, J. E. Johnson, H. Y. Zoghbi, N. Segil,
Development 129, 2495 (2002).
6. M. Montcouquiol et al., Nature 423, 173 (2003).
7. M. Montcouquiol, M. W. Kelley, J. Neurosci. 23, 9469
(2003).
8. H. Honda, H. Yamanaka, G. Eguchi, J. Embryol.
Exp. Morphol. 98, 1 (1986).
9. M. Takeichi, Science 251, 1451 (1991).
10. B. M. Gumbiner, Nat. Rev. Mol. Cell Biol. 6, 622
(2005).
11. Y. Takai, H. Nakanishi, J. Cell Sci. 116, 17 (2003).
12. Y. Takai, W. Ikeda, H. Ogita, Y. Rikitake, Annu. Rev. Cell
Dev. Biol. 24, 309 (2008).
13. Y. Takai, J. Miyoshi, W. Ikeda, H. Ogita, Nat. Rev. Mol.
Cell Biol. 9, 603 (2008).
14. K. Satoh-Horikawa et al., J. Biol. Chem. 275, 10291
(2000).
15. H. Togashi et al., J. Cell Biol. 174, 141 (2006).
16. N. Okabe, K. Ozaki-Kuroda, H. Nakanishi, K. Shimizu,
Y. Takai, Dev. Dyn. 230, 174 (2004).
17. F. D. Nunes et al., J. Cell Sci. 119, 4819 (2006).
18. M. S. Steinberg, Science 141, 401 (1963).
19. M. W. Kelley, D. R. Talreja, J. T. Corwin, J. Neurosci. 15,
3013 (1995).
20. J. Kimble, P. Simpson, Annu. Rev. Cell Dev. Biol. 13,
333 (1997).
21. R. Kopan, R. Cagan, Trends Genet. 13, 465 (1997).
22. J. Lewis, Curr. Opin. Neurobiol. 6, 3 (1996).
23. S. Bao, R. Cagan, Dev. Cell 8, 925 (2005).
24. S. Bao, K. F. Fischbach, V. Corbin, R. L. Cagan, Dev. Biol.
344, 948 (2010).
Acknowledgments: This work was supported by the
Global Centers of Excellence Program “Global Center
for Education and Research in Integrative Membrane
Biology” and the Targeted Proteins Research Program
from the Ministry of Education, Culture, Sports,
Science and Technology in Japan; by Grants-in-Aid
from the Japan Society for the Promotion of Science;
and by the Core Research for Evolutional Science and
Technology from the Japanese Science and Technology
Agency.
Supporting Online Material
www.sciencemag.org/cgi/content/full/science.1208467/DC1
Materials and Methods
Figs. S1 to S6
References (25–27)
Movie S1
16 May 2011; accepted 8 July 2011
Published online 28 July 2011;
10.1126/science.1208467
Impacts of Fishing
Low–Trophic Level Species on
Marine Ecosystems
Anthony D. M. Smith,
1
* Christopher J. Brown,
2,3
Catherine M. Bulman,
1
Elizabeth A. Fulton,
1
Penny Johnson,
1
Isaac C. Kaplan,
4
Hector Lozano-Montes,
5
Steven Mackinson,
6
Martin Marzloff,
1,7
Lynne J. Shannon,
8
Yunne-Jai Shin,
8,9
Jorge Tam
10
Low–trophic level species account for more than 30% of global fisheries production and
contribute substantially to global food security. We used a range of ecosystem models to
explore the effects of fishing low–trophic level species on marine ecosystems, including marine
mammals and seabirds, and on other commercially important species. In five well-studied
ecosystems, we found that fishing these species at conventional maximum sustainable yield
(MSY) levels can have large impacts on other parts of the ecosystem, particularly when they
constitute a high proportion of the biomass in the ecosystem or are highly connected in the
food web. Halving exploitation rates would result in much lower impacts on marine ecosystems
while still achieving 80% of MSY.
C
oncerns about the trophic impact of har-
vesting marine species were recognized
more than three decades ago (1). Despite
recent successes in reducing exploitation rates in
some marine ecosystems (2), concerns remain
over the effects of fishing on the structure and
function of marine ecosystems (3, 4).
Low–trophic level (LTL) species in marine
ecosystems comprise species that are generally
plankton feeders for the larger part of their life
cycle. They are often present in high abundance
and tend to form dense schools or aggregations.
They include small pelagic “forage” fish such as
anchovy, sardine, herring, mackerel, and capelin
but also invertebrate species such as krill. Hu-
mans harvest across the trophic levels in marine
food webs, and landings of LTL species have
been increasing generally in proportion with
global catches (5). Forage fish account for over
30% of global fish landings, most of which is
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1147
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

now used for fishmeal production as feed for
livestock industries and aquaculture rather than
being consumed directly (6). However, LTL spe-
cies also contribute directly to food security in
many developing countries, and between 10
and 20% of global landings are consumed di-
rectly by humans (7). One species alone, Peru-
vian anchovy, contributes up to 50% of global
landings used for fishmeal production. Driven by
global markets for fertilizer, animal feed, and
increases in the production of seafood from
aquaculture, demand for fishmeal continues to
increase (8).
LTL species play an important role in marine
food webs because they are the principal means
of transferring production fromplankton to larger
predatory fish and to marine mammals and sea-
birds. Several studies have raised concerns about
the impacts on seabirds of local depletion of
forage fish [anchovy in Perú (9), sand eels in the
North Sea (10), and anchovy and sardines in
South Africa (11)]. Similar concerns have been
raised about the prospects of a large increase in
catch of krill in the Southern Ocean and its po-
tential impact on recovery of depleted marine
mammals such as whales (12). Of particular con-
cern are “wasp waist” systems, where a large part
of the plankton production is funnelled through a
small number of LTL species to higher trophic
levels (13, 14).
Although studies in individual ecosystems
have raised concerns about the ecological effects
of fishing LTL species, there has been no sys-
tematic attempt to examine and summarize what
these broader effects might be or under what cir-
cumstances various effects might be expected to
arise. In this study, we used ecosystem models in
five well-studied regions to examine systemic ef-
fects of fishing LTL species. The regions include
three eastern boundary current ecosystems—
the northern Humboldt, the southern Benguela,
and the California current—and two systems less
dominated by upwelling, including the North Sea
and the southeast Australian shelf and conti-
nental slope (Fig. 1). To avoid conclusions being
dominated by structural assumptions in partic-
ular types of model, we used three different eco-
system models to explore the responses: Ecopath
with EcoSim(EwE) (15, 16), OSMOSE (17, 18),
and Atlantis (19, 20). For each ecosystem and
model, we selected up to five LTL species or
groups and subjected them one by one to a range
of fishing pressures, resulting in depletion levels
relative to unfished biomass from zero (no
fishing) to 100% (extirpated). The LTL species
selected included some that are currently fished
(such as anchovy) and others that are not cur-
rently exploited in those ecosystems (such as
krill and mesopelagic fishes). We did not in-
clude harvested shellfish such as scallops and
prawns, notwithstanding their commercial impor-
tance (21), because most of the models did not
resolve these species well. Impacts on other eco-
logical groups in the ecosystem were measured
relative to biomass levels of those groups pro-
duced by simulations in which the focal LTL
species was unfished, and all other groups were
fished at current levels. Details of the ecosys-
tems, models, groups, and experiments are pro-
vided in (22).
We found widespread impacts of harvesting
LTL species across the ecosystems and LTL spe-
cies selected (Fig. 2). The percent of ecological
groups exhibiting effects greater than 40% in-
creased with the level of depletion of the LTL
species, but the extent of impact also varied
across LTL species. Impacts on other ecological
groups were both positive and negative (fig. S1),
ranging up to very severe impacts for some
groups (>60% change in biomass) even at rel-
atively low levels of depletion (25% below un-
fished levels—that is, biomass reduced to 75%of
unfished levels) of the LTL species. Negative im-
pacts (reductions in abundance) tended to pre-
dominate for marine mammals and seabirds,
although the majority of impacts on such groups
were small. Some commercial species could also
be negatively affected, although again impacts
on most commercial species were small. Results
Fig. 1. Global map showing location of study ecosystems. From left to right
are the California current, northern Humboldt, North Sea, southern Benguela,
and southeast Australia. Graph shows trend in landings of forage species from
1950 to 2009. [Source: Sea Around Us Project, www.seaaroundus.org/global/
1/3.aspx. Images of forage fish are copyright Casson Trenor, 2010, at www.
sustainablesushi.net]
1
Commonwealth Scientific and Industrial Research Organiza-
tion, Wealth fromOceans Flagship, Hobart, TAS 7001, Australia.
2
School of Biological Sciences, University of Queensland, St
Lucia, QLD 4072, Australia.
3
Commonwealth Scientific and
Industrial Research Organisation, Climate Adaptations Flag-
ship, Brisbane, QLD 4163, Australia.
4
National Oceanographic
and Atmospheric Administration, Northwest Fisheries Science
Center, Seattle, WA 98112, USA.
5
Commonwealth Scientific
and Industrial Research Organization, Wealth from Oceans
Flagship, Floreat, WA 6014, Australia.
6
Centre for Environ-
ment, Fisheries and Aquaculture Science, Lowestoft NR33
0HT, UK.
7
Institute for Marine and Antarctic Studies, Uni-
versity of Tasmania, Sandy Bay, TAS 7005, Australia.
8
Uni-
versity of Cape Town, Marine Research (MA-RE) Institute
and Department of Zoology, Rondebosch 7701, South Africa.
9
Institut de Recherche pour le Développement, UMR EME
212, Ecosystèmes Marins Exploités, 34203 Sète, France.
10
In-
stituto del Mar del Perú, Esquina Gamarra y General Valle s/n,
Callao, Perú.
*To whom correspondence should be addressed. E-mail:
tony.d.smith@csiro.edu
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1148
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

were generally robust to the three types of model
used in the analysis (fig. S2).
The variation in impact of harvesting different
LTL species has potentially important manage-
ment implications; large impacts may require a
change in harvest levels, whereas LTL species
with small impacts could be harvested at con-
ventional single-species levels. In each ecosys-
tem, harvesting several of the LTL species was
found to have high impacts, although the species
with high impacts were not always consistent
across ecosystems (Fig. 2). For example, in the
northern Humboldt ecosystem, harvesting ancho-
vy had high impacts, and harvesting sardine had
low impacts, but in the southern Benguela eco-
system, harvesting sardines had the larger impact,
whereas the impacts of fishing both species were
low in the southeast Australian and California
current ecosystems. Impacts of harvesting meso-
pelagic fishes (a group not currently targeted in
any of these ecosystems or generally in global
fisheries) were consistently high across ecosys-
tems, and impacts of harvesting krill (large zoo-
plankton), also not currently exploited in these
ecosystems, also tended to be medium to high.
Fishing sand eels had the highest impact in the
North Sea.
To explain this range of impacts across LTL
species, we looked for more generic properties of
these groups (other than taxonomy) that might
explain and predict the variation. Three potential
predictors were the relative abundance of the
group in the ecosystem (for example, Peruvian
anchovy accounts for up to 35% of the consumer
biomass in the northern Humboldt ecosystem),
the trophic level of the group, and the connec-
tivity of the group in the food web. Trophic level
was not a good predictor of impact, but the other
two factors appear to be important. Abundant
groups have consistently large impacts, whereas
smaller groups can have either small or large im-
pacts (Fig. 3A). There appears to be a threshold
effect for connectance (the proportion of total
trophic connections in the food web for each LTL
species), with species that have a connectance
value greater than ~0.04 having larger impacts
(Fig. 3B). However, factors other than total con-
nectance are likely to be important, including the
presence of groups with trophic niches similar to
those of the exploited species that can dampen
the ecosystem effects of depleting the targeted
species.
There are important tradeoffs to examine in
considering the wider implications of these results
for exploitation of LTL species. In particular, im-
pacts on other parts of the ecosystem will be
smaller at lower exploitation rates, but yields also
will be lower (Fig. 4). There is a tension here be-
tween achieving broader goals of protecting and
maintaining biodiversity (including ecosystem
structure and function) and global food security.
LTL species support the latter both through direct
human consumption and through providing feed
for livestock and aquaculture production. Consid-
erable reductions in impact can be achieved by
moving fromexploitation at MSYlevels (achieved
at close to 60%depletion levels) to a target of 75%
of unexploited biomass (25% depletion) for an
LTL species, as shown in Fig. 4. The cost of
such a change would be slightly less than 20%
of long-term yield. This target could be achieved
at significantly lower exploitation rates (most-
ly less than half MSY rates) (fig. S3), which
would imply much lower fishing effort and may
be closer to long-term economic optimum levels.
There could also be some benefit of a reduction
in harvest rate of LTL species to yields for other
Fig. 2. Effects of level
of depletion of LTL spe-
cies on the proportion
of other trophic groups
whose biomass varied
by more than 40% rela-
tive to their level where
the LTL species was not
fished. Results are shown
for a variety of LTL spe-
cies fished in each mod-
eled ecosystem.
Fig. 3. Relationships between attributes of depleted LTL species and their ecosystem impact. Impacts
are scored as the rank of the largest effect: rank 1, no change greater than 20% in any other ecological
group; rank 2, no change greater than 60% in any other ecological group; and rank 3, change greater
than 60% in at least one other ecological group. Each point corresponds to one ecosystem, model, and
LTL species. All LTL species are depleted by 60%. (A) Impact of relative biomass of LTL species (biomass
as a percent of total consumer biomass in the ecosystem) on rank of largest effect. (B) Impact of
connectance (proportion of all ecosystem trophic links involving the LTL species) on rank of largest
effect.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1149
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

commercially targeted species (fig. S1). Although
we did not explicitly examine multi-species har-
vest strategies, exploitation rates well below MSY
levels are consistent with previous findings that
lower exploitation rates should be adopted for
most species (2).
These results are based on model predictions.
Each of the models has been validated against
time-series data from well-studied systems, and
additional empirical validation for impacts on
seabirds and marine mammals is provided in
(22). Clearly, the details of which groups respond
to depletion of LTL species is sensitive to both
model parameterization and to choice of model
structure (22). For this reason, we do not consider
that these models should be used to determine
tactical management decisions. However, the over-
all findings reported here are robust to details of
model choice.
The conclusion that lower exploitation rates
are needed for forage species also finds support
from a wider set of model types (23). Spatial
structure in marine ecosystems is an important
factor in species interactions, and local prey de-
pletion may be particularly important for land-
based predators such as penguins and seals (24).
Two of the models used in this study (OSMOSE
and Atlantis) incorporate spatial structure, but not
always at the resolution needed to address such
issues. All of the models incorporate environ-
mental forcing and variability, which is also an
important feature driving the dynamics of many
LTL species (25, 26). The finding that con-
nectance influences which species are likely to
have larger impacts is potentially important,
but although the measure of connectance is easi-
ly derived in models, it may be more difficult
to determine empirically (and the empirical va-
lidity of the indicator would need verification).
Previous studies have shown that the ways in
which species are connected in the food web
can influence system properties (27, 28). Pre-
vious studies have also emphasized that ad-
ditional protection may be needed for forage
species (29).
The exploitation patterns examined in this
study have involved constant fishing mortality
rates. Initial explorations of other forms of
exploitation, including use of biomass thresholds
or “set asides” (biomass levels below which no
exploitation will occur), suggest that lower eco-
logical impacts could be achieved for similar
long-term average yields, but at the cost of high-
er year-to-year variation in catches. Use of such
set asides is already a feature of some LTL fish-
eries, including a 5-million-ton-minimumspawn-
ing stock biomass level for Peruvian anchovy
(30) and 150,000 tons for California sardine
(31). Closed areas are also used in some fish-
eries so as to reduce impacts on predators, such
as closures for sand eels in some parts of the
North Sea to improve the breeding success of
sea birds (32).
Although harvest strategies for LTL species
vary widely, many stocks are currently fished at
levels below the biomass that achieves MSY
(22). The results of this study combined with set
asides and targeted spatial closures should help
inform harvest strategies that achieve ecologi-
cal objectives while ensuring ongoing substantial
yields from LTL groups in support of the on-
going challenge of feeding the global human
population (33).
References and Notes
1. R. M. May, J. R. Beddington, C. W. Clark, S. J. Holt,
R. M. Laws, Science 205, 267 (1979).
2. B. Worm et al., Science 325, 578 (2009).
3. D. G. Ainley, L. K. Blight, Fish Fish. 10, 13
(2009).
4. M. Llope et al., Glob. Change Biol., published online
14 October 2010 (10.1111/j.1365-2486.2010.02331.x).
5. T. A. Branch et al., Nature 468, 431 (2010).
6. J. Alder, B. Campbell, V. Karpouzi, K. Kaschner,
D. Pauly, Annu. Rev. Environ. Resour. 33, 153
(2008).
7. A. G. J. Tacon, M. Metian, Ambio 38, 294 (2009).
8. G. Merino, M. Barange, C. Mullon, J. Mar. Syst. 81,
196 (2010).
9. J. Jahncke, D. M. Checkley, G. L. Hunt, Fish. Oceanogr.
13, 208 (2004).
10. M. Frederiksen et al., Mar. Ecol. Prog. Ser. 300, 201
(2005).
11. R. J. M. Crawford, L. G. Underhill, L. Upfold, B. M. Dyer,
ICES J. Mar. Sci. 64, 570 (2007).
12. A. J. Constable et al., ICES J. Mar. Sci. 57, 778
(2000).
13. P. Cury et al., ICES J. Mar. Sci. 57, 603 (2000).
14. L. J. Shannon, P. M. Cury, A. Jarre, ICES J. Mar. Sci. 57,
720 (2000).
15. C. Walters, D. Pauly, V. Christensen, J. F. Kitchell,
Ecosystems (N. Y.) 3, 70 (2000).
16. V. Christensen, C. J. Walters, Ecol. Modell. 172, 109
(2004).
17. Y.-J. Shin, P. Cury, Can. J. Fish. Aquat. Sci. 61, 414
(2004).
18. M. Travers et al., Ecol. Modell. 220, 3089
(2009).
19. E. A. Fulton et al., Ecol. Modell. 173, 371
(2004).
20. E. A. Fulton, A. D. M. Smith, A. E. Punt, ICES J. Mar. Sci.
62, 540 (2005).
21. S. C. Anderson, J. M. Flemming, R. Watson, H. K. Lotze,
PLoS ONE 6, e14735 (2011).
22. Materials and methods are available as supporting
material on Science Online.
23. M. C. Tyrrell, J. S. Link, H. Moustahfid, Fish. Res. 108,
1 (2011).
24. J. A. Santora, C. S. Reiss, A. M. Cossio, R. R. Veit,
Fish. Oceanogr. 18, 20 (2009).
25. R. T. Barber, F. P. Chavez, Science 222, 1203
(1983).
26. F. P. Chavez, J. Ryan, S. E. Lluch-Cota, M. Niquen C,
Science 299, 217 (2003).
27. T. Gross, L. Rudolf, S. A. Levin, U. Dieckmann, Science
325, 747 (2009).
28. J. M. Montoya, G. Woodward, M. C. Emmerson, R. V. Solé,
Ecology 90, 2426 (2009).
29. C. J. Walters, V. Christensen, S. J. Martell, J. F. Kitchell,
ICES J. Mar. Sci. 62, 558 (2005).
30. P. Freon, M. Bouchon, C. Mullon, C. García, M. Ñiquen,
Prog. Oceanogr. 79, 401 (2008).
31. K. T. Hill et al., Assessment of the Pacific Sardine
Resource in 2008, for U.S. Management in 2009 (2009);
available at www.pcouncil.org/wp-content/uploads/
2009_CPS_SAFE_APP1_Sardine.pdf.
32. S. P. R. Greenstreet et al., ICES J. Mar. Sci. 63, 1530
(2006).
33. H. C. J. Godfray et al., Science 327, 812
(2010).
Acknowledgments: This work was partly funded by the
Marine Stewardship Council (MSC) to investigate
“Developing best practice management for low
trophic level fisheries: evaluation of harvest strategies.”
The authors also acknowledge funding from the
Commonwealth Scientific and Industrial Research
Organization (Australia), National Oceanographic
and Atmospheric Administration (USA), Institut de
Recherche pour le Développement (France), and
Instituto del Mar del Perú (Perú). I.C.K. (not funded
by MSC) and E.A.F. acknowledge support from the
Gordon and Betty Moore Foundation and the David
and Lucile Packard Foundation. S.M. acknowledges
support from European Union (EU) FP7 grant FACTS
(Forage Fish Interactions), grant agreement 244966,
and Department of Environment Food and Rural
Affairs contract M1202. L.S. acknowledges the
support of the South African Research Chair Initiative,
funded through the South African Department of
Science and Technology and administered by the
South African National Research Foundation
(NRF). Y.J.S. acknowledges support from EU FP7
project MEECE (Marine Ecosystem Evolution
in a Changing Environment), grant agreement
212085. J. Field kindly provided the EcoSim model
for the northern California current and advised on
its use. C. de Moor provided information on
management targets for sardines and anchovy
in the Southern Benguela.
Supporting Online Material
www.sciencemag.org/cgi/content/full/science.1209395/DC1
Materials and Methods
SOM Text
Figs. S1 to S3
Table S1
References
6 June 2011; accepted 12 July 2011
Published online 21 July 2011;
10.1126/science.1209395
Fig. 4. Tradeoff between
yield and ecological im-
pact as level of LTL deple-
tion varies. Yield (blue) is
shown as a proportion of
MSY. Ecological impact
(gray) is measured as the
proportion of other eco-
logical groups whose bio-
mass varied by more than
40%. Shaded zones show
T1.96 times SE. Results
arefor all ecosystems, mod-
els, and LTL species.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1150
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Expanding the Genetic Code of
Escherichia coli with Phosphoserine
Hee-Sung Park,
1
*† Michael J. Hohn,
1
* Takuya Umehara,
1
Li-Tao Guo,
1
Edith M. Osborne,
2

Jack Benner,
2
Christopher J. Noren,
2
§ Jesse Rinehart,
3,4
§ Dieter Söll
1,5
§
O-Phosphoserine (Sep), the most abundant phosphoamino acid in the eukaryotic
phosphoproteome, is not encoded in the genetic code, but synthesized posttranslationally.
Here, we present an engineered system for specific cotranslational Sep incorporation (directed
by UAG) into any desired position in a protein by an Escherichia coli strain that harbors a
Sep-accepting transfer RNA (tRNA
Sep
), its cognate Sep–tRNA synthetase (SepRS), and an
engineered EF-Tu (EF-Sep). Expanding the genetic code rested on reengineering EF-Tu to relax
its quality-control function and permit Sep-tRNA
Sep
binding. To test our system, we synthesized
the activated form of human mitogen-activated ERK activating kinase 1 (MEK1) with either one
or two Sep residues cotranslationally inserted in their canonical positions (Sep
218
, Sep
222
).
This system has general utility in protein engineering, molecular biology, and disease research.
O
-Phosphoserine (Sep) was identified 80
years ago as a constituent of phospho-
proteins from egg yolk (1). Since then,
the extent and importance of the eukaryotic phos-
phoproteome has been realized and has provided
insight into large interconnected networks of
kinases and phosphatases (2). Protein kinases
represent one of the largest eukaryotic gene fam-
ilies, composing nearly 2% of all human genes
(3). Sep is the most abundant phosphoamino
acid; based on an analysis of >2000 HeLa cell
phosphoproteins, the relative abundance of Sep is
7.3 and 48 times higher than that of phospho-
threonine and phosphotyrosine, respectively (2).
A major research limitation is the inability to
biosynthesize these phosphoproteins for detailed
studies of their enzyme or substrate properties.
The discovery of Sep–tRNA synthetase
(SepRS), a unique aminoacyl–tRNA synthetase
devoted to Sep-tRNA
Cys
formation in methano-
genic archaea, provided an opportunity to de-
velop our Sep-insertion strategy. The natural role
of SepRS is the formation of Sep-tRNA
Cys
,
which is then converted to Cys-tRNA
Cys
by the
enzyme SepCysS in the presence of a sulfur
donor (4) (Fig. 1A). Given the high specificity of
SepRS for Sep and for tRNA
Cys
and our
knowledge of the identity elements in this tRNA
(5), and based on our understanding of the
structure of this enzyme and its catalytic site (6, 7),
we devised a system to incorporate Sep into
proteins directed by the UAG(amber) codon. For
this, we chose Methanocaldococcus jannaschii
(Mj) tRNA
Cys
and the mesophilic Methanococ-
cus maripaludis (Mmp) SepRS as the orthogonal
pair [reviewed in (8)] for the synthesis of phos-
phoserylated amber suppressor tRNA.
We first designed tRNA
Sep
(Fig. 1B), an am-
ber suppressor tRNA derived from Mj tRNA
Cys
by two mutations in the anticodon, and an ad-
ditional C20U change that improves amino-
acylation by SepRS (5). In vitro aminoacylation
by Mmp SepRS showed (Fig. 1C) that the an-
ticodon change lowered (to about 40%) the
ability of tRNA
Sep
to be aminoacylated when
compared to tRNA
Cys
. In agreement with earlier
data (5), total Escherichia coli tRNAcould not be
acylated with Sep (Fig. 1C). On the basis of these
in vitro data, Mj tRNA
Sep
and Mmp SepRS appear
to be an orthogonal pair.
Efficient and selective addition of Sep to the
E. coli genetic repertoire requires exclusive in-
teraction of SepRS with tRNA
Sep
for Sep-tRNA
Sep
formation without interfering in the host transla-
tion system, as well as a sufficient intracellular
concentration of Sep. Because E. coli has a Sep-
compatible transporter (9), Sep (2 mM) was
added to the LB growth medium, and the endog-
enous serB gene encoding phosphoserine phos-
phatase was deleted in the E. coli test strain
without affecting growth. To assess whether the
Mj tRNA
Sep
–Mmp SepRS pair is functional and
orthogonal in E. coli in vivo, we performed a
suppression assay that used a gene encoding
chloramphenicol acetyltransferase (CAT) with a
UAG stop codon at the permissive position 112
(wild-type amino acid: Asp) to produce the CAT
enzyme. Cell survival was measured in the pres-
ence of Sep and varying amounts of chloram-
phenicol (Cm) where the different half-maximal
inhibitory concentration (IC
50
) values and the
1
Department of Molecular Biophysics and Biochemistry, Yale
University, NewHaven, CT 06520, USA.
2
NewEngland BioLabs,
Ipswich, MA 01938, USA.
3
Department of Cellular and Mo-
lecular Physiology, Yale University, New Haven, CT 06520,
USA.
4
Systems Biology Institute, Yale University, West Haven,
CT 06516, USA.
5
Department of Chemistry, Yale University,
New Haven, CT 06520, USA.
*These authors contributed equally to this work.
†Present address: Department of Chemistry, Korea Advanced
Institute of Science and Technology, Daejeon 305-701, Korea.
‡Present address: Department of Chemistry, Angelo State
University, San Angelo, TX 76909, USA.
§To whom correspondence should be addressed. E-mail:
dieter.soll@yale.edu (D.S.); noren@neb.com (C.J.N.); jesse.
rinehart@yale.edu ( J.R.)
A
B
10 0 20 30 40 50 60
0
2
4
6
8
12
10
Time (min)
S
e
p

a
c
c
e
p
t
a
n
c
e

(
%
)

i
n

t
o
t
a
l

t
R
N
A
C
E. coli total tRNA
+M. jannaschii tRNA
Cys
+M. jannaschii tRNA
Sep
+ ATP AMP + PP
i
SepRS
Lys Sep
Lys
Cys
tRNA
Cys
?
SepCysS
S-donor + PLP
Sep
GCA GCA GCA
tRNA
Sep
Fig. 1. Design of tRNA
Sep
and its aminoacylation by M. maripaludis SepRS. (A) Pathway of Cys-tRNA
Cys
formation in M. maripaludis. ATP, adenosine 5′-triphosphate; AMP, adenosine 5′-monophosphate; PPi,
inorganic pyrophosphate; PLP, pyridoxal phosphate. (B) Cloverleaf structure of tRNA
Sep
. Arrows indicate
the three nucleotide changes compared to M. jannaschii tRNA
Cys
. (C) Acylation with Sep of M. jannaschii
tRNA
Cys
and tRNA
Sep
catalyzed by M. maripaludis SepRS. Total E. coli tRNA (

), or tRNA from E. coli
strains expressing the M. jannaschii tRNA
Cys
(

) or the tRNA
Sep
(

) gene was acylated by M. maripaludis
SepRS with [
14
C]Sep (0.1 mM) in the presence of ATP (10 mM).
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1151
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

tRNA
Sep
-dependent CATsynthesis correlate with
suppression efficiency (Fig. 2). When only tRNA
Sep
was expressed (Fig. 2, column B), Cm resistance
increased about 3.3-fold over background (Fig. 2,
column A). Thus, tRNA
Sep
can be aminoacyl-
ated to a certain degree by an unknown E. coli
aminoacyl–tRNA synthetase (we later found
that Gln is being incorporated at the amber stop
codon). In contrast, simultaneous expression of
tRNA
Sep
and SepRS did not provide Cm re-
sistance (Fig. 2, column C). This may indicate
that SepRS can outcompete any endogenous
aminoacyl–tRNAsynthetase andformSep-tRNA
Sep
;
however, this aminoacyl-tRNA is neither deliv-
ered to the ribosome nor accommodated on it.
Providing additional EF-Tu did not improve the
result (Fig. 2, column D). Coexpression of tRNA
Sep
,
SepRS, and SepCysS should result in forma-
tion of Sep-tRNA
Sep
and subsequent SepCysS-
mediated conversion to Cys-tRNA
Sep
(4). Indeed,
a 2.3-fold increase in Cmresistance was observed
(Fig. 2, column E). This further supports the no-
tion that although Sep-tRNA
Sep
is synthesized, it
cannot be used properly by the E. coli protein bio-
synthesis machinery. By contrast, coexpression of
tRNA
Sep
and Mmp CysRS generated a 12.3-fold
increase in Cm resistance (Fig 2, column F), dem-
onstrating that Cys-tRNA
Sep
can be readily used
for amber codon suppression in the CAT gene.
Given that EF-Tu is a component of quality
control in protein synthesis (10), it is plausible
that Sep-tRNA
Sep
may be rejected by EF-Tu.
Chemically synthesized Sep-tRNA
Gln
was a
poor substrate for in vitro protein synthesis (11).
tRNAs carrying negatively charged amino acids
are bound poorly by EF-Tu (12), and molecular
dynamics simulations suggested that Sep-tRNA
Cys
may not be bound by EF-Tu (13). We tested this
assumption in EF-Tu–mediated Sep-tRNA hy-
drolysis protection experiments (14). Although
EF-Tu protected [
35
S]Cys-tRNA
Cys
from deacyl-
ation at pH 8.2 (fig. S1), [
14
C]Sep-tRNA
Cys
was
significantly deacylated irrespective of the pres-
ence of EF-Tu (Fig. 3B and fig. S1). Thus, insuf-
ficient binding of Sep-tRNA
Sep
to EF-Tu may
explain the lack of Sep insertion into protein.
This observation required the generation of
EF-Tuvariants able toproductivelybindSep-tRNA.
We were encouraged by reports that EF-Tu var-
iants allow binding of tRNAs charged with cer-
tain unusual amino acids (15, 16). Guided by the
structure of the E. coli EF-Tu:Phe-tRNA
Phe
com-
plex (17), we selected six residues (His
67
, Asp
216
,
Glu
217
, Phe
219
, Thr
229
, and Asn
274
) in the amino
acid binding pocket of EF-Tu (Fig. 3A) for com-
plete randomization in order to generate EF-Tu
variants that bind Sep-tRNA. Variants that permitted
SepRS and tRNA
Sep
-dependent Sep incorpora-
tion were selected in vivo (see SOM). One clone,
designated EF-Sep (amino acid variants shown in
Fig. 3A), led to a 10-fold increase in Cm resist-
ance (Fig. 2, column H), whereas the combina-
tion of SepRS and EF-Sep without tRNA
Sep
was
not active in the CAT suppression assay (Fig. 2,
column G). Thus, it appeared that EF-Sep did
bind Sep-tRNA
Sep
, a conclusion that was con-
firmed in the hydrolysis protection assay (Fig.
3B). This assay also shows that EF-Sep still
retained some ability to bind Cys-tRNA(fig. S1).
To prove that the observed suppression is due
to Sep incorporation, we expressed myoglobin
with an amber codon in the Asp
127
position (fig.
S2A). The expected full-length protein was syn-
thesized only when EF-Sep, SepRS, and tRNA
Sep
were coexpressed (fig. S2A). Mass spectrometry–
time-of-flight (MS-TOF) and MS/MS analysis
showed that Sep is present at the position spec-
ified by UAG in both the intact and trypsin-
digested proteins (Fig. S2, B and C).
Final validation of our strategy was the syn-
thesis of a Sep-containing human protein MEK1
(mitogen-activated ERKactivating kinase 1). This
key eukaryotic enzyme of the mitogen-activated
Fig. 2. In vivo synthesis
of chloramphenicol acetyl-
transferase (measured by
IC
50
value) by tRNA
Sep
-
dependent read-through of
an amber mutation in the
CAT gene. tRNA
Sep
was co-
expressed in the E. coli
Top10DserB strain with the
proteins indicatedinthefig-
ure (SepRS, M. maripaludis
CysRS, SepCysS, EF-Sep,
EF-Tu). Selection was car-
ried out on LB agar plates
containing 2 mMSep and
various concentrations of
chloramphenicol. Error bars
indicate SEM.
I
C
5
0

(
C
h
l
o
r
a
m
p
h
e
n
i
c
o
l

µ
g

m
l
-
1
)
tRNA
Sep
SepRS
CysRS (Mmp)
SepCysS
EF-Sep
EF-Tu






+





+
+




+
+

+


+
+



+
+
+


+


+


+

+

+



A B C D H G F E
0
10
20
30
40
50
A
Time (min)
0 20 40 60
0
20
40
60
80
100
S
e
p
-
t
R
N
A
C
y
s

r
e
m
a
i
n
i
n
g

(
%
)
EF-Sep
EF-Tu
BSA
B
N274 W
H67 R E216 N
D217 G
F219 Y
T229 S
Phe-tRNA
Phe
EF-Tu
tRNA
Phe
Fig. 3. Design of EF-Sep. (A) Model of the amino acid binding pocket of
E. coli EF-Tu bound to Phe-tRNA (based on Protein Data Bank structure
1OB2). To accommodate Sep-tRNA, an E. coli tufB library was constructed
that would allow the six highlighted amino acid residues to change to any
of the 20 canonical amino acids. The six mutations in our final EF-Sep are
indicated by an arrow. (B) EF-Sep protects Sep-tRNA
Cys
from deacylation.
M. jannaschii [
14
C]Sep-tRNA
Cys
was incubated at pH 8.2 and at room
temperature in the presence or absence of wild-type EF-Tu or EF-Sep.
Deacylation of Sep-tRNA
Cys
was measured by acid precipitability. Error
bars indicate SEM. Abbreviations for the amino acid residues are as
follows: D, Asp; E, Glu; F, Phe; G, Gly; H, His; N, Asn; R, Arg; S, Ser; T, Thr;
W, Trp; and Y, Tyr.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1152
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

signaling cascade is crucial for cell proliferation,
development, differentiation, cell cycle progres-
sion, and oncogenesis (18). Activation of MEK1
requires posttranslational phosphorylation of Ser
218
and Ser
222
by MEKactivating kinases (e.g., Raf-1,
MEKK, or MOS). Substitution of both Ser resi-
dues with Glu yields a constitutively active enzyme,
albeit with lower activity (19). We generated a
clone encoding a MEK1 fusion protein [with the
maltose binding protein (MBP) at the Nterminus
and a His
6
tag at the C terminus] in which Ser
222
was changed to Glu, and the Ser
218
codon was
replaced by UAG to encode Sep. After expres-
sion in the presence of SepRS, tRNA
Sep
, and
EF-Sep, 25 mg of full-length MBP-MEK1(Sep
218
,
Glu
222
) was isolated from 1 liter of culture. The
presence of Sep in MBP-MEK1(Sep
218
,Glu
222
)
protein was demonstrated by its ability to phos-
phorylate ERK2, which then phosphorylates mye-
lin basic protein. MBP-MEK1(Sep
218
,Glu
222
)
had a 2500-fold higher specific activity than
MBP-MEK1(Ser
218
,Ser
222
) and a 70-fold higher
specific activity than the constitutively active
MBP-MEK1(Glu
218
,Glu
222
) (Fig. 4A). MS/MS
analysis confirmed the incorporation of Sep at
position 218 (Fig. 4B). To determine if our E. coli
expression system would allow the simultaneous
insertion of two Sep residues into the protein, we
changed the Ser codons in positions 218 and 222
to UAG. As expected, the expression efficiency
of MBP-MEK1(Sep
218
,Sep
222
) was markedly re-
duced compared to that of wild-type MBP-MEK1
(only about 1 mg of full-length protein was obtained
from1 liter of culture). The presence of Sep at both
active-site positions of MEK1 was tested by
Western blot analysis with a monoclonal antibody
specific to phosphorylation at these two residues
(Fig 4C). Only MBP-MEK1(Sep
218
,Sep
222
), and
to a weaker extent MBP-MEK1(Sep
218
,Ser
222
), were
detected in this experiment, whereas neither MBP-
MEK1(Ser
218
,Ser
222
), MBP-MEK(Sep
218
,Glu
222
),
or MBP-MEK(Glu
218
,Glu
222
) was recognized by
the antibody. This demonstrates that the addition
of SepRS, tRNA
Sep
, and EF-Sep endows E. coli
with the ability to read UAG as a phosphoserine
codon.
Our work underscores the key role of EF-Tu
in quality control of protein synthesis by ensuring
facile delivery of the correct cognate aminoacyl-
tRNA to the ribosome (10, 20). Generating an
orthogonal aminoacyl–tRNA synthetase:tRNA
pair was insufficient to genetically encode Sep.
Expansion of the genetic code to include Sep
depended critically on reengineering of EF-Tu to
bind Sep-tRNA
Sep
. This situation is precisely
paralleled in the naturally evolved genetic encod-
ing system for selenocysteine, which requires a
specialized elongation factor [SelBin prokaryotes,
and EFSec in eukaryotes; reviewed in (21)]. In-
spired careful manipulation of components of the
protein-synthesizing systemwill allowfurther ex-
pansion of the genetic code without sacrificing
organismal fitness.
Orthogonal aminoacyl–tRNAsynthetase:tRNA
pairs (in our case, SepRS:tRNA
Sep
) are critical
elements for genetic code expansion, whether
produced by evolutionary processes and found in
nature [e.g., pyrrolysyl–tRNAsynthetase:tRNA
Pyl
(22)] or designed in the laboratory by synthetic
biologists [reviewed in (8)]. The tRNA of choice
is typically an amber suppressor tRNA; however,
a general limitation in product yield is associated
with recoding the new amino acid by a stop co-
don (e.g., UAG), because peptide chain elonga-
tion by the designed aminoacylated suppressor
tRNA competes with chain termination by the
release factor (e.g., RF1). Yet this impediment
may soon be removed by the advent of E. coli
expression strains with genome-wide UAG co-
don reassignments and an RF1 deletion (23, 24).
The ability to generate physiologically rele-
vant active kinases and stoichiometrically phos-
phorylated protein domains could reveal new
types of kinase inhibitors for drug development,
allow systematic dissection of phosphorylation-
dependent protein-protein interactions, and expose
new structure-function relationships (18, 25, 26).
References and Notes
1. F. Lipmann, Biochem. Z. 262, 3 (1933).
2. J. V. Olsen et al., Cell 127, 635 (2006).
3. G. Manning, D. B. Whyte, R. Martinez, T. Hunter,
S. Sudarsanam, Science 298, 1912 (2002).
4. A. Sauerwald et al., Science 307, 1969 (2005).
5. M. J. Hohn, H. S. Park, P. O’Donoghue, M. Schnitzbauer,
D. Söll, Proc. Natl. Acad. Sci. U.S.A. 103, 18095 (2006).
6. S. Kamtekar et al., Proc. Natl. Acad. Sci. U.S.A. 104,
2620 (2007).
7. R. Fukunaga, S. Yokoyama, Nat. Struct. Mol. Biol. 14,
272 (2007).
8. C. C. Liu, P. G. Schultz, Annu. Rev. Biochem. 79, 413 (2010).
9. B. L. Wanner, W. W. Metcalf, FEMS Microbiol. Lett. 79,
133 (1992).
10. F. J. LaRiviere, A. D. Wolfson, O. C. Uhlenbeck, Science
294, 165 (2001).
11. D. M. Rothman et al., J. Am. Chem. Soc. 127, 846 (2005).
12. T. Dale, L. E. Sanderson, O. C. Uhlenbeck, Biochemistry
43, 6159 (2004).
13. J. Eargle, A. A. Black, A. Sethi, L. G. Trabuco,
Z. Luthey-Schulten, J. Mol. Biol. 377, 1382 (2008).
14. J. Ling et al., Proc. Natl. Acad. Sci. U.S.A. 104, 15299
(2007).
15. Y. Doi, T. Ohtsuki, Y. Shimizu, T. Ueda, M. Sisido, J. Am.
Chem. Soc. 129, 14458 (2007).
16. T. Ohtsuki, H. Yamamoto, Y. Doi, M. Sisido, J. Biochem.
148, 239 (2010).
0 1 2 3 4 5
0
5
10
15
MEK1 (µg/assay)
P
h
o
s
p
h
a
t
e

i
n
c
o
r
p
o
r
a
t
e
d

i
n
t
o

M
y
B
P

(
p
m
o
l
/
m
i
n
)
MEK1
Sep218,Glu222
MEK1
Ser218,Ser222
MEK1
Glu218,Glu222
A C
80
58
80
58
80
58
MBP-MEK1-His
6
G
S
T
-
M
E
K
1
kDa
Sep 222:
218: Sep
Ser
Ser
Ser
Sep
Glu Glu
Glu Sep
R
e
l
a
t
i
v
e

i
n
t
e
n
s
i
t
y

(
%
)
m/z
B
a
n
t
i
-
M
B
P
a
n
t
i
-
P
-
M
E
K
Fig. 4. Properties of E. coli–producedSep-containing
human MEK1. (A) Kinase activity assay. MEK1 was
produced in E. coli as a fusion protein with an
N-terminal maltose-binding protein (MBP) tag and
a C-terminal His
6
tag. Residues Ser
218
and Ser
222
,
which are targets to phosphorylation by MEK1
activators, were mutated to either Glu
218
/Glu
222
or Sep
218
/Glu
222
to produce active MEK1 variants.
Various amounts of MBP-MEK1-His
6
with active-site residues Ser
218
/Ser
222
(

), Glu
218
/Glu
222
(

), or
Sep
218
/Glu
222
(

) were used to phosphorylate inactive ERK2 in vitro. ERK2 activity was then measured in a
radiometric assay using g-[
32
P]ATP and myelin basic protein (MyBP) as substrates. Error bars indicate SEM.
(B) MS/MS spectrum confirming the presence of Sep
218
in MBP-MEK1(Sep
218
/Ser
222
). m/z, mass-to-charge
ratio. (C) MBP-MEK1-His
6
variants with genetically encoded active-site residues Sep
218
/Sep
222
(lane 1),
Ser
218
/Ser
222
(lane 2), Sep
218
/Ser
222
(lane 3), Sep
218
/Glu
222
(lane 4), and Glu
218
/Glu
222
(lane 5) were
produced in E. coli and partially purified by Ni
2+
affinity chromatography. Proteins were separated by
SDS–polyacrylamide gel electrophoresis and either stained with Coomassie (top) or transferred to a nylon
membrane and detected with monoclonal antibodies specific for the phosphorylated active site of human
MEK (center) or the MBP tag (bottom). Purchased activated glutathione S-transferase (GST)–MEK was used
as a control (lane 6). Dark and light arrowheads indicate the positions of MBP-MEK1-His
6
and GST-MEK1,
respectively. The strong bands (~70 kDsize) are probable truncation products caused by termination at UAG.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1153
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

17. P. Nissen et al., Science 270, 1464 (1995).
18. J. S. Sebolt-Leopold, R. Herrera, Nat. Rev. Cancer 4,
937 (2004).
19. D. R. Alessi et al., EMBO J. 13, 1610 (1994).
20. J. M. Schrader, S. J. Chapman, O. C. Uhlenbeck,
Proc. Natl. Acad. Sci. U.S.A. 108, 5215 (2011).
21. S. Yoshizawa, A. Böck, Biochim. Biophys. Acta 1790,
1404 (2009).
22. K. Nozawa et al., Nature 457, 1163 (2009).
23. T. Mukai et al., Nucleic Acids Res. 38, 8188
(2010).
24. F. J. Isaacs et al., Science 333, 348 (2011).
25. J. D. Scott, T. Pawson, Science 326, 1220 (2009).
26. M. B. Yaffe et al., Cell 91, 961 (1997).
Acknowledgments: We thank P. Dennis, P. O’Donoghue,
and J. Ling for enthusiastic discussions. M.J.H. was a
Feodor Lynen Postdoctoral Fellow of the Alexander von
Humboldt Foundation (Bonn, Germany). H.-S.P. held
a postdoctoral fellowship of the Korean Science
Foundation. This work was supported by grants from
NSF (MCB-0645283 and MCB-0950474) (to D.S.),
National Institute of General Medical Sciences (GM
22854) (to D.S.), National Research Foundation of Korea
(ABC-20100029737) (to H.-S.P.), and National Institute
of Diabetes and Digestive and Kidney Diseases
(K01DK089006) (to J.R.). Yale University holds the U.S.
Patent 7,723,069 B2: “Site Specific Incorporation of
Phosphoserine into Polypeptides Using Phosphoseryl-tRNA
Synthetase” by D. Soll and J. Rinehart. Yale University has
applied for a patent that covers the engineered EF-Tu
described in this manuscript. Reagents are available under a
Yale Material Transfer Agreement.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6046/1151/DC1
Materials and Methods
Figs. S1 and S2
Table S1
References (27–37)
19 April 2011; accepted 7 July 2011
10.1126/science.1207203
Exome Sequencing of Head and Neck
Squamous Cell Carcinoma Reveals
Inactivating Mutations in NOTCH1
Nishant Agrawal,
1,2
*† Mitchell J. Frederick,
3
* Curtis R. Pickering,
3
* Chetan Bettegowda,
2,4
*
Kyle Chang,
5
Ryan J. Li,
1
Carole Fakhry,
1
Tong-Xin Xie,
3
Jiexin Zhang,
6
Jing Wang,
6
Nianxiang Zhang,
6
Adel K. El-Naggar,
7
Samar A. Jasser,
3
John N. Weinstein,
6
Lisa Treviño,
5
Jennifer A. Drummond,
5
Donna M. Muzny,
5
Yuanqing Wu,
5
Laura D. Wood,
8
Ralph H. Hruban,
8
William H. Westra,
8
Wayne M. Koch,
1
Joseph A. Califano,
1,9
Richard A. Gibbs,
5,9
David Sidransky,
1
Bert Vogelstein,
2
Victor E. Velculescu,
2
† Nickolas Papadopoulos,
2
David A. Wheeler,
5
Kenneth W. Kinzler,
2
† Jeffrey N. Myers
3

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer worldwide.
To explore the genetic origins of this cancer, we used whole-exome sequencing and gene copy
number analyses to study 32 primary tumors. Tumors from patients with a history of tobacco use
had more mutations than did tumors from patients who did not use tobacco, and tumors that were
negative for human papillomavirus (HPV) had more mutations than did HPV-positive tumors.
Six of the genes that were mutated in multiple tumors were assessed in up to 88 additional
HNSCCs. In addition to previously described mutations in TP53, CDKN2A, PIK3CA, and HRAS, we
identified mutations in FBXW7 and NOTCH1. Nearly 40% of the 28 mutations identified in NOTCH1
were predicted to truncate the gene product, suggesting that NOTCH1 may function as a tumor
suppressor gene rather than an oncogene in this tumor type.
M
ore than half a million new cases of
head and neck squamous cell carcino-
ma (HNSCC) will occur in 2011, in-
cluding 50,000 cases in the United States, making
it the sixth most common cancer in the world
(1–3). HNSCC and its treatment can result in
cosmetic deformity and functional impairment of
vital functions, including breathing, swallowing,
speech, phonation, taste, hearing, and smell.
These cancers are frequently lethal, with a five-
year survival of only ~50%(4). HNSCCs, like all
solid tumors, are thought to be initiated and to
progress through a series of genetic alterations.
Indeed, several cellular signaling pathways are
dysregulated in this tumor type through genetic
and epigenetic alterations, such as those involv-
ing TP53 and CDK2NA (4). HNSCCs also ex-
hibit many chromosomal abnormalities, including
amplifications of region 11q13, which contains
the cyclin D1 gene, and region 7p11, which en-
codes epidermal growth factor receptor (EGFR)
(5). Tobacco use and excessive alcohol consump-
tion are major risk factors for HNSCC in the
United States (6). More recently, human papilloma-
virus (HPV) has emerged as an additional risk
factor for the development of cancers of the oro-
pharynx (7). Patients with HPV-associated cancers
have an improved overall and disease-specific
survival, suggesting that these tumors have dis-
tinct biological features (8).
To gain a comprehensive view of the genetic
alterations underlying HNSCC, we sequenced
~18,000 protein-encoding genes in tumors from
32 patients. Thirty of the 32 patients had not been
treated with chemotherapy or radiation before
their tumor biopsy, so the spectrum of changes
we observed largely reflects those of tumors in
their naturally occurring state. Tumor samples
were carefully selected or microdissected in order
to achieve a neoplastic cellularity of >60%.
DNA was purified from these tumors as well as
matched nonneoplastic tissue and used to gen-
erate libraries suitable for massively parallel se-
quencing. After capture of the coding sequences
with a SureSelect (Agilent, Santa Clara, California)
or CCDS (NimbleGen, Madison, Wisconsin) En-
richment System, the DNAwas sequenced by using
an Illumina (San Diego, California) GAIIx/HiSEq
(17 tumors) or SOLiD(Carlsbad, California) V3/V4
(15 tumors) instruments. The average coverage of
each base in the targeted regions was 77-fold and
44-fold for the Illumina and SOLiD instruments,
and 92.6% and 90% of targeted bases were rep-
resented by at least 10 reads in these platforms,
respectively (table S1).
Using stringent criteria for analysis of these
data (9), we identified 911 candidate somatic mu-
tations in 725 genes among the 32 tumors. To
ensure that our algorithms for identifying muta-
tions were reliable, we evaluated the candidate
mutations by Sanger sequencing or by 454 se-
quencing and confirmed 609 of them(67%) (table
S2). One hundred and fifty-two (17%) mutations
did not confirm, and 150 (16%) mutations could
not be tested because of an unusually high GC
content, difficulty in the design of unique primers,
or other unknown factors preventing specific
amplification and sequencing of the locus. The
range of confirmed mutations per tumor was 2 to
78, with a mean and SD of 19 T 16.5 mutations
per tumor (table S1).
There were obvious differences in the genetic
landscapes of HPV-associated and HPV-negative
HNSCCs. First, far fewer genes were mutated per
tumor in the HPV-associated tumors as compared
with those tumors not epidemiologically related
1
Department of Otolaryngology–Head and Neck Surgery,
Johns Hopkins University School of Medicine, 600 North Wolfe
Street, Baltimore, MD 21287, USA.
2
Ludwig Center for Cancer
Genetics and Howard Hughes Medical Institutions, Johns
Hopkins Kimmel Cancer Center, Baltimore, MD 21231, USA.
3
Department of Head and Neck Surgery, University of Texas
M.D. Anderson Cancer Center, 1515 Holcombe, Houston, TX
77030, USA.
4
Department of Neurosurgery, Johns Hopkins
University School of Medicine, 600 North Wolfe Street, Bal-
timore, MD 21287, USA.
5
Human Genome Sequencing Center,
Baylor College of Medicine, One Baylor Plaza, Houston, TX
77030, USA.
6
Department of Bioinformatics and Computa-
tional Biology, University of Texas M.D. Anderson Cancer Cen-
ter, 1515 Holcombe, Houston, TX 77030, USA.
7
Department of
Pathology, University of Texas M.D. Anderson Cancer Center,
1515 Holcombe, Houston, TX 77030, USA.
8
Department of
Pathology, Johns Hopkins University School of Medicine, 600
North Wolfe Street, Baltimore, MD21287, USA.
9
Milton J. Dance
Head and Neck Center, Greater Baltimore Medical Center,
Baltimore, MD 21204, USA.
10
Department of Molecular and
Human Genetics, Baylor College of Medicine, One Baylor Plaza,
Houston, TX 77030, USA.
*These authors contributed equally to this work.
†To whom correspondence should be addressed. E-mail:
nagrawal@jhmi.edu (N.A.); velculescu@jhmi.edu (V.E.V.);
kinzlke@jhmi.edu (K.W.K.); jmyers@mdanderson.org (J.N.M.)
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1154
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

to HPV (4.8 T 3 versus 20.6 T 16.7, P < 0.05,
Welch two sample t test) (table S3A). The dif-
ference in the number of mutations between
HPV-associated and HPV-negative tumors was
independent of smoking status. Second, TP53 mu-
tations were not identified in any of the HPV-
associated tumors but were found in 78% of the
HPV-negative tumors. These data are consistent
with previous results on HNSCCas well as HPV-
associated cervical cancers (10–12).
As expected, more mutations were identified
in tumors from patients with a history of tobacco
use as compared with those from patients who
did not use tobacco (21.6 T 17.8 versus 9.5 T 6.5,
P < 0.05, Welch two sample t test) (table S3B).
Surprisingly, and in contrast to data from lung
cancer, the mutational spectrumwas not enriched
for G:C>T:A transversions in those tumors as-
sociated with smoking (table S4). Nearly all of
the HNSCC tumors analyzed had a mutation
spectrum similar to that of non–smoking-related
lung cancers and other non–smoking-related
tumors. These data suggest that the effects of
tobacco on the mutational spectra vary among dif-
ferent tumor types.
We selected genes for further analysis if they
or closely related genes were altered in at least 2
of the 32 tumors sequenced. The genes included
were PIK3AP1, RIMBP2, SI, NRXN2, NRXN3,
EPHA7, RASA1, RXFP3, PIK3CA, HRAS, TP53,
CDKN2A, NOTCH1, and FBXW7 (table S2).
We then analyzed the sequences of these genes
in additional HNSCCs and their corresponding
normal tissues (9). In total, somatic mutations in
TP53, NOTCH1, CDKN2A, PIK3CA, FBXW7,
and HRAS were identified in 47, 15, 9, 6, 5, and
4% of patients, respectively (table S5). The re-
maining genes were not observed to be mutated
in more than one of the additional samples ana-
lyzed. Of the 63 TP53 mutations in the 120 sam-
ples analyzed (32 in the discovery set and 88 in
the prevalence set), 25 were predicted to be in-
activating mutations [10 nonsense, 10 insertions
or deletions (“indels”), and five splice-site mu-
tations], and 38 were missense. Two of these
were homozygous—that is, there was no remain-
ing normal allele—and seven tumors demon-
strated two mutations in TP53, likely representing
inactivation of both alleles. In the other samples,
the presence of contaminating DNA from non-
neoplastic cells may have made it difficult to
reliably distinguish heterozygous from homozy-
gous changes. There were 11 CDKN2A muta-
tions observed among the 120 tumors analyzed,
of which nine were definitely inactivating (three
nonsense, four indels, and two splice site), and
two were missense. The frequency and types of
mutations we observed in TP53 and CDKN2A, as
well as in HRAS and PIK3CA, were consistent
with previous studies of HNSCC (4).
A total of 28 NOTCH1 mutations were iden-
tified. Seven of 21 patients with NOTCH1 muta-
tions had two independent mutations, presumably
on different alleles. Eleven of the NOTCH1 mu-
tations were predicted to truncate the protein pro-
duct (seven nonsense and four indels), whereas
17 were missense (Tables 1 and 2). Next to TP53,
NOTCH1 was the most frequently mutated gene
found in the combined discovery and prevalence
sets, with alterations present in 15%of patients. To
date, NOTCH1 has not been reported to be mu-
tated at a significant frequency in other solid tu-
mor types (13).
FBXW7 mutations have not been previously
observed in HNSCC, although they are frequent
in other tumor types. Of the six FBXW7 muta-
tions we identified, two were indels, and the other
four were missense; none were homozygous.
Although both alleles of most tumor suppressor
genes are mutated in tumors in which they play a
role, FBXW7 is an exception. FBXW7 is a mem-
ber of the F-box protein family and constitutes a
component of the ubiquitin protein ligase com-
plex. It acts as a tumor suppressor in several tu-
mors, and one of its major targets is NOTCH1,
which it targets for degradation. The FBXW7
mutations we observed were in a hotspot known
to block the degradation of active NOTCH1 (14).
It is attractive to hypothesize that FBXW7 muta-
tions are modulating the Notch pathway, although
FBXW7 also targets other cancer-related proteins
for degradation, including cyclin E and c-myc.
Further studies will be necessary to elucidate the
function of these FBXW7 mutations in HNSCC.
To complement the sequencing data, we per-
formed copy number analysis with Affymetrix
(Santa Clara, California) SNP6.0 microarrays on
42 tumor and normal sample pairs, including 25
of the samples used for massively parallel se-
quencing. We found that the most frequently mu-
tated genes often were affected by copy number
changes. For example, loss of heterozygosity(LOH)
was observed in the tumor samples with muta-
tions in TP53, and CDKN2A was frequently de-
leted (table S6). In addition, LOHat the NOTCH1
locus was detected in two of the three tumors
with NOTCH1 mutations that were analyzed for
Table 2. Missense mutations in NOTCH1. Samples in bold were used for the initial (discovery set) screen.
Underlined samples were from the Human Genome Sequencing Center, and the remaining samples were
from Johns Hopkins University. OC, oral cavity; OP, oropharynx; HP, hypopharynx; N, no; Y, yes; WT, wild
type; unk, unknown; na, not applicable; +, positive.
Sample Site Tobacco TP53 status HPV status Amino acid (protein)
HN14PT OC N Mutant na p.P391S
478 OC N WT na p.G484V
HN 102 PT OC N WT na p.E450K
HN 105 PT OC Y WT na p.G812W
HN 115 PT OC Y Mutant na p.G1340A
HN 117 PT OC N WT na p.G310R
HN 117 PT OC N WT na p.C456R
HN 142 PT OP Y WT + p.R353H
HN 142 PT OP Y WT + p.M2011R
HN 148 PT OP Y Mutant unk p.G1638V
HN 148 PT OP Y Mutant unk p.P2272S
HN 208 PT OC Y Mutant na p.R365C
HN 208 PT OC Y Mutant na p.R1280C
HN 227 PT HP Y Mutant na p.R365C
HN 251 PT OP N WT + p.F1292L
HN 255 PT OP Y WT unk p.V2039L
HN 255 PT OP Y WT unk p.V2039E
Table 1. Truncating mutations in NOTCH1. Samples in bold were used for the initial (discovery set) screen.
Underlined samples were from the Human Genome Sequencing Center, and the remaining samples were
fromJohns Hopkins University. OC, oral cavity; OP, oropharynx; L, larynx; N, no; Y, yes; WT, wild type; unk,
unknown; na, not applicable; +, positive.
Sample Site Tobacco TP53 status HPV status Amino acid (protein) Mutation type
HN12PT OC Y WT na p.W1843X Nonsense
385 OC N Mutant na p.R1984X Nonsense
600 OC Y WT unk Q290X Nonsense
HN 105 PT OC Y WT na fs Indel
HN 107 PT OC Y Mutant na p.S402X Nonsense
HN 115 PT OC Y Mutant na fs Indel
HN 130 PT L Y WT na p.E949X Nonsense
HN 139 PT OP Y Mutant unk p.C554X Nonsense
HN 183 PT OC unk Mutant na p.Q1958X Nonsense
HN 194 PT OC N Mutant na fs Indel
HN 245 PT OP Y WT + fs Indel
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1155
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

copy number. Given that seven other tumors in
our cohort had two NOTCH1 mutations, in-
activation of both alleles probably occurred in at
least 9 of the 21 patients with NOTCH1 muta-
tions. These data support the idea that NOTCH1
acts as a tumor suppressor in HNSCC. Addition-
ally, recurrent gains and losses were observed in
several specific regions of the genome (table S7
and figs. S1 to S6). Recurrent focal changes were
identified by filtering for segments with more
than three copies or losses of one or more copies
in at least three samples. The focal changes in-
cluded deletions of 9p21.3 (containing CDKN2A)
and amplifications of small segments on 11q, 3q,
and 7p. These regions have previously been im-
plicated in HNSCCand contain known oncogenes
(CCND1, PIK3CA, and EGFR, respectively).
Our data raise questions about the role of
NOTCH1 in tumorigenesis. An involvement of
NOTCH1 in human cancers was first demon-
strated through the discovery of translocations in
Tcell leukemias (15). Subsequently, more subtle
mutations of NOTCH1 were identified in a varie-
ty of hematopoietic tumors. Most of the muta-
tions in hematopoietic tumors are clustered in
two hot spots in the heterodimerization (HD) and
C-terminal polypeptide-enriched proline, gluta-
mate, serine, and threonine (PEST) domain (Fig.
1A) (14). Exogenous expression of these mu-
tants, as well as of the translocated NOTCH1,
lead to neoplastic transformation in vitro and in
vivo. In contrast, while this manuscript was un-
der review, Klinakis et al. showed that reduced
activity of the Notch signaling pathway is asso-
ciated with the development of chronic myelo-
monocytic leukemia (CMML), suggesting a tumor
suppressor role for Notch (16). Additionally, a
small number of mutations, some truncating, have
been previously observed in solid tumors (Fig. 1B)
(17). This pattern is consistent with a suppressor
gene rather than an oncogene function, but the
mutations were present in only a small fraction of
any individual tumor type and were difficult to
distinguish from passenger mutations. In con-
trast, the relatively large number of mutations we
observed in HNSCC strongly implicates them
as drivers (P < 10
−8
) (8). Moreover, the spectrum
of NOTCH1 mutations we observed in HNSCC
was fundamentally different fromthose identified
in hematopoietic tumors because a high fraction
were in the N-terminal epidermal growth factor
(EGF)–like ligand binding domain, and the ma-
jority of the mutations were predicted to alter the
protein N-terminal to the transmembrane region
(Fig. 1C). The location and nature of these alter-
ations, together with the observation that two
NOTCH1 mutations were found in each of seven
patients, provides strong genetic evidence that
NOTCH1 often acts as a tumor suppressor gene
in HNSCC. This interpretation is consistent with
functional studies of the role of NOTCH1 in squa-
mous epithelial cells because NOTCH1
−/−
mice
develop epithelial tumors (18). The function of
NOTCH1 in cancer—as an oncogene in some leu-
kemias and tumor suppressor gene in CMML,
HNSCC, and perhaps other cancers—may reca-
pitulate its dual role in normal biology, in which
activation can lead to stem cell maintenance in
some tissues but terminal differentiation in others
(19). These results also emphasize the importance
of assessing the functional attributes of cancer-
associated mutations in a cell-type–specific fash-
ion, either in tissue culture or in model organisms.
The results of this study provide evidence that
HNSCCs, although morphologically similar, are
composed of distinct diseases at the molecular
level. The different genetic landscapes associated
with HPV and tobacco exposure are consistent
with clinical and epidemiologic data that has
suggested the importance of these environmental
factors in prognosis and response to therapeutic
interventions. Another important observation is
that only 18% of the 28 tumors not associated
with HPV from the discovery set had activating
mutations in a bona fide oncogene. In contrast,
89% of these tumors harbored inactivating mu-
tations in at least one bona fide tumor suppressor
gene predominately in TP53. This distinction is
critical because the new generation of molecu-
larly targeted therapies is directed toward acti-
vated oncogenes, but such drugs cannot directly
target mutated tumor suppressor genes because
they are already inactivated. Tumor suppressor
gene predominance is not limited to HNSCC be-
cause exomic analysis of other tumor types, such
as renal cell carcinomas and pancreatic endo-
crine neoplasms, have revealed similar patterns
(20, 21). Our finding that HNSCCs have few
directly targetable mutations has implications
for controlling this disease in the future. In par-
ticular, it suggests that prevention, careful surveil-
lance of patients at risk, and early detection are
the optimal approaches for reducing morbidity
and mortality from this disease.
References and Notes
1. R. I. Haddad, D. M. Shin, N. Engl. J. Med. 359, 1143
(2008).
2. A. Jemal, R. Siegel, J. Xu, E. Ward, CA Cancer J. Clin. 60,
277 (2010).
3. D. M. Parkin, F. Bray, J. Ferlay, P. Pisani, CA Cancer
J. Clin. 55, 74 (2005).
4. C. R. Leemans, B. J. Braakhuis, R. H. Brakenhoff,
Nat. Rev. Cancer 11, 9 (2011).
5. J. R. Berenson, J. Yang, R. A. Mickel, Oncogene 4, 1111
(1989).
6. J. A. Brennan et al., N. Engl. J. Med. 332, 712 (1995).
7. G. D’Souza et al., N. Engl. J. Med. 356, 1944 (2007).
8. K. K. Ang et al., N. Engl. J. Med. 363, 24 (2010).
9. Materials and methods are available as supporting
material on Science Online
10. J. P. Klussmann et al., Clin. Cancer Res. 15, 1779
(2009).
11. H. C. Hafkamp et al., Int. J. Cancer 107, 394 (2003).
12. V. Balz et al., Cancer Res. 63, 1188 (2003).
13. www.sanger.ac.uk/perl/genetics/CGP/cosmic
14. C. D. Baldus et al., Haematologica 94, 1383 (2009).
15. L. W. Ellisen et al., Cell 66, 649 (1991).
16. A. Klinakis et al., Nature 473, 230 (2011).
17. B. Westhoff et al., Proc. Natl. Acad. Sci. U.S.A. 106,
22293 (2009).
18. M. Nicolas et al., Nat. Genet. 33, 416 (2003).
19. M. Roy, W. S. Pear, J. C. Aster, Curr. Opin. Genet. Dev.
17, 52 (2007).
20. I. Varela et al., Nature 469, 539 (2011).
21. Y. Jiao et al., Science 331, 1199 (2011).
22. J. C. Aster et al., J. Biol. Chem. 272, 11336 (1997).
23. R. Kopan, M. X. Ilagan, Cell 137, 216 (2009).
24. L. Y. Cohen et al., Cell Death Differ. 12, 243 (2005).
347 Non-synonymous mutations:
347 Missense or in-frame indels
A
N
C
262 Non-synonymous mutations:
46 Missense or in-frame indels
216 Truncating mutations
EGF-like repeats
LNR
RAM
NLS
Ankyrin repeats NLS PEST
B
N
C
C
C
N
TMD
Fig. 1. Schematic depiction of mutations in NOTCH1. (A) Previously observed NOTCH1 mutations in
hematopoietic malignancies. EGF, epidermal growth factor; LNR, Lin12-Notch repeats; TMD, trans-
membrane domain; RAM, recombination signal-binding protein 1 for J-k (RBPjk) association module;
NLS, nuclear localization signal; PEST, proline, glutamic acid, serine/threonine-rich motifs. Red bars
represent previously described mutation hotspots (amino acids 1575 to 1630 and 2250 to 2550). (B)
Previously observed NOTCH1 mutations in solid tumors. Colored arrow (missense mutation) and “X”
(truncating mutation) depict mutations found in different tumor types: pink, breast cancer; black, glioma;
blue, lung cancer; green, pancreatic adenocarcinoma; red, esophageal squamous cell carcinoma; purple,
tongue squamous cell carcinoma. (C) Mutations in NOTCH1 in HNSCC observed in this study. Black arrow
indicates missense mutation and red “X” indicates truncating mutation (13, 22–24).
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1156
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Acknowledgments: We thank our patients for their courage
and generosity. We thank N. Silliman, J. Ptak, M. Whalen,
L. Dobbyn, J. Schaeffer, X. Li, O. Folawiyo, and Z. Khan for
expert technical assistance. This work was supported by the
National Institutes of Health (NIH)/National Institute of
Dental and Craniofacial Research grants RC2DE020957,
and RC2DE020958; NIH grants CA121113, CA43460,
CA57345, and CN43302; NIH Specialized Program of
Research Excellence grants P50DE019032 and
5P50CA09700708; Cancer Prevention Research Institute of
Texas grant RP100233; Cancer Center Support grant
CA16672; AACR Stand Up To Cancer–Dream Team
Translational Cancer Research grant; and the Virginia and
D. K. Ludwig Fund for Cancer Research. Under agreements
between the Johns Hopkins University, Genzyme, Exact
Sciences, Inostics, Qiagen, Invitrogen, and Personal Genome
Diagnostics, N.P., B.V., K.W.K., and V.E.V are entitled to a
share of the royalties received by the university on sales
of products related to genes and technologies described
in this manuscript. N.P., B.V., K.W.K., and V.E.V are
co-founders of Inostics and Personal Genome Diagnostics,
are members of their Scientific Advisory Boards, and own
Inostics and Personal Genome Diagnostics stock, which
is subject to certain restrictions under Johns Hopkins
University policy. J.C. is the Director of Research of the
Milton J. Dance Head and Neck Endowment. The terms of
these arrangements are managed by the Johns Hopkins
University in accordance with its conflict-of-interest
policies. C.B. is a recipient of T32 CA009574
NIH/National Cancer Institute National Research Service
Award. C.R.P. is a TRIUMPH Fellow and supported by the
GSK Translational Research Fellowship. This paper is
based on a web database application provided by
Research Information Technology Systems (RITS)
www.rits.onc.jhmi.edu/.
Supporting Online Material
www.sciencemag.org/cgi/content/full/science.1206923/DC1
Materials and Methods
Figs. S1 to S6
Tables S1 to S8
References
13 April 2011; accepted July 5 2011
Published online 28 July 2011;
10.1126/science.1206923
The Mutational Landscape of Head
and Neck Squamous Cell Carcinoma
Nicolas Stransky,
1
* Ann Marie Egloff,
2
* Aaron D. Tward,
1,3,4
* Aleksandar D. Kostic,
1,5
Kristian Cibulskis,
1
Andrey Sivachenko,
1
Gregory V. Kryukov,
1,5
Michael S. Lawrence,
1
Carrie Sougnez,
1
Aaron McKenna,
1
Erica Shefler,
1
Alex H. Ramos,
1
Petar Stojanov,
1
Scott L. Carter,
1
Douglas Voet,
1
Maria L. Cortés,
1
Daniel Auclair,
1
Michael F. Berger,
1
Gordon Saksena,
1
Candace Guiducci,
1
Robert C. Onofrio,
1
Melissa Parkin,
1
Marjorie Romkes,
6
Joel L. Weissfeld,
7
Raja R. Seethala,
8
Lin Wang,
8
Claudia Rangel-Escareño,
9
Juan Carlos Fernandez-Lopez,
9
Alfredo Hidalgo-Miranda,
9
Jorge Melendez-Zajgla,
9
Wendy Winckler,
1
Kristin Ardlie,
1
Stacey B. Gabriel,
1
Matthew Meyerson,
1,5,10,11
Eric S. Lander,
1,5,12
Gad Getz,
1
Todd R. Golub,
1,5,11,13,14
† Levi A. Garraway,
1,5,10,11
†‡ Jennifer R. Grandis
2,15
†‡
Head and neck squamous cell carcinoma (HNSCC) is a common, morbid, and frequently lethal
malignancy. To uncover its mutational spectrum, we analyzed whole-exome sequencing data
from 74 tumor-normal pairs. The majority exhibited a mutational profile consistent with tobacco
exposure; human papillomavirus was detectable by sequencing DNA from infected tumors. In addition
to identifying previously known HNSCC genes (TP53, CDKN2A, PTEN, PIK3CA, and HRAS), our analysis
revealed many genes not previously implicated in this malignancy. At least 30% of cases harbored
mutations in genes that regulate squamous differentiation (for example, NOTCH1, IRF6, and TP63),
implicating its dysregulation as a major driver of HNSCC carcinogenesis. More generally, the results
indicate the ability of large-scale sequencing to reveal fundamental tumorigenic mechanisms.
H
ead and neck squamous cell carcinoma
(HNSCC) is the sixth most common non-
skin cancer in the world, with an inci-
dence of ~600,000 cases per year and a mortality
rate of ~50% (1). The major risk factors for
HNSCC are tobacco use, alcohol consumption,
and infection with human papillomavirus (HPV)
(2). Despite advances in our knowledge of its
epidemiology and pathogenesis, the survival rates
for many types of HNSCC have improved little
over the past 40 years (3). As such, a deeper
understanding of HNSCCpathogenesis is needed
to promote the development of improved thera-
peutic approaches.
We performed solution-phase hybrid capture
and whole-exome sequencing on paired DNA
samples (tumors and matched whole blood) from
92 HNSCC patients. Most anatomic sites were
represented (oral cavity, oropharynx, hypopharynx,
larynx, and sinonasal cavity) (Fig. 1C and table
S1). Of the patients profiled in this study, 89 and
79% reported a history of tobacco and alcohol
use, respectively (table S1). Initially, 14% of all tu-
mors and 53% of oropharyngeal tumors were
found to be positive for HPV based on HPV-16
polymerase chainreaction(PCR)/insituhybridization
(Fig. 1 and table S1). Tumor copy-number analy-
sis with the use of single-nucleotide polymorphism
(SNP) arrays (fig. S1) replicated previous findings
of frequent CCND1 amplifications; CDKN2Adele-
tions; and rarer MYC, EGFR, ERBB2, or CCNE1
amplifications (4), indicating that the collection
is genetically representative of HNSCC.
We achieved 150-fold mean sequence cov-
erage of targeted exonic regions, with 87% of
loci covered at >20-fold (figs. S2 and S3 and
table S2). We excluded from further analysis 18
tumors in which initial analysis revealed extensive
stromal admixture (figs. S3 and S4 and supple-
mental methods), leaving 74 samples for analysis.
We also performed whole-genome sequencing
(31-fold mean coverage) (table S3) on an oro-
pharyngeal tumor and a hypopharyngeal tumor.
On average, we identified 130 coding muta-
tions per tumor, 25%of which were synonymous
(Fig. 1A). We queried 321 of these mutations by
mass spectrometric genotyping and validated 288
(89.7%). However, the validation rate increased
to 95.7%for mutations whose allelic fraction was
>20% of total DNA, suggesting that the sensi-
tivity of mass spectrometric genotyping may be re-
duced in the setting of increased stromal admixture.
The overall HNSCC mutation rate was com-
parable to other smoking-related malignancies
such as small-cell lung cancer and lung adeno-
carcinoma (5, 6). The mutation rate of HPV-
positive tumors was approximately half of that
found in HPV-negative HNSCC (mean = 2.28
mutations per megabase compared with 4.83 mu-
tations per megabase; P = 0.004, rank sum test),
consistent with epidemiologic studies suggestive
of biological differences between HPV-positive
and -negative disease. The two tumors that un-
derwent whole-genome sequencing harbored
19 (HN_62469) and 111 (HN_62699) high-
confidence somatic rearrangements, respectively
(fig. S5 and tables S4 and S5).
Although base mutation rates varied widely
(0.59 to 24 mutations per megabase; Fig. 1A), the
average rate of guanosine-to-thymidine (G →T)
transversions at non-CpG sites (12 T 6%, stan-
dard deviation) was characteristic of tobacco ex-
posure (Fig. 1B). Among patients who reported
a smoking history, tumors with the highest
fraction of G →T transversions showed a tend-
ency toward increased overall mutation rates
(P = 0.02, Spearman rank correlation) (Fig. 1,
Band C). Thus, the G→Ttransversion frequency
1
The Broad Institute of MIT and Harvard, Cambridge, MA 02142,
USA.
2
Department of Otolaryngology, University of Pittsburgh
and University of Pittsburgh Cancer Institute, Pittsburgh, PA
15213, USA.
3
Department of Otology and Laryngology, Harvard
Medical School, Boston, MA 02114, USA.
4
Department of
Otolaryngology, Massachusetts Eye and Ear Infirmary, Boston,
MA 02114, USA.
5
Harvard Medical School, Boston, MA 02115,
USA.
6
Department of Medicine, University of Pittsburgh and
University of Pittsburgh Cancer Institute, Pittsburgh, PA 15261,
USA.
7
Department of Epidemiology, University of Pittsburgh
and University of Pittsburgh Cancer Institute, Pittsburgh, PA
15232, USA.
8
Department of Pathology, University of Pittsburgh
and University of Pittsburgh Cancer Institute, Pittsburgh, PA
15213, USA.
9
Instituto Nacional de Medicina Genómica, Mexico
City, 01900, Mexico.
10
Department of Medical Oncology, Dana-
Farber Cancer Institute, Boston, MA 02115, USA.
11
Center for
Cancer Genome Discovery, Dana-Farber Cancer Institute, Boston,
MA 02115, USA.
12
Massachusetts Institute of Technology,
Cambridge, MA 02142, USA.
13
Department of Pediatric On-
cology, Dana-Farber Cancer Institute, Boston, MA 02115, USA.
14
Howard Hughes Medical Institute, Chevy Chase, MD 20815,
USA.
15
Department of Pharmacology and Chemical Biology,
University of Pittsburgh and University of Pittsburgh Cancer
Institute, Pittsburgh, PA 15213, USA.
*These authors contributed equally to this work.
†These authors contributed equally to this work.
‡To whom correspondence should be addressed. E-mail:
levi_garraway@dfci.harvard.edu (L.A.G.); grandisjr@
upmc.edu ( J.R.G.)
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1157
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

may represent a robust readout of “functional”
tobacco exposure. We observed differences in
mutation rates and G→Ttransversion frequencies
by tumor site even when restricting the analysis to
HPV-negative tumors. In particular, HPV-negative
laryngeal cancers exhibited higher mutation rates
and G → T transversion frequencies compared
with HPV-negative cancers found in the oral cav-
ity, oropharynx, hypoharynx, or sinonasal cavity
(P = 0.008 and P < 0.0001, respectively, rank
sum tests) (Fig. 1, A to C, and fig. S8).
Notwithstanding the overall apparent corre-
lation between G→Ttransversions and mutation
rates, several “outlier” tumors showed elevated
mutation rates, despite a low fraction of G →T
transversions. Some of these tumors contained
mutations in one or more DNA repair genes. No-
tably, both HNSCC tumors with the highest mu-
tation rates occurred in nonsmokers (Fig. 1). These
results raise the possibility that some HNSCC
tumors may contain genetic alterations that pro-
mote elevated mutation rates apart from the ef-
fects of tobacco (supporting online material).
To explore the biological basis of HNSCC in
an unbiased manner, we used the MutSig algorithm
(7) to identify genes harboring more mutations
than expected by chance, given the total number
of mutations detected. This analysis revealed 39
genes with high statistical significance of mutations
(false discovery rate q < 0.1) (figs. S6 and S7 and
tables S6 and S7). Compared with recent cancer
Fig. 1. Mutation rates and base-substitution frequencies in head and neck can-
cers. (A) Rate of synonymous and nonsynonymous mutations, expressed in number
of mutations per megabase (Mb) of covered target sequence. Nonsynonymous
mutation rates range from0.43 to 17.1mutations per megabase (mean =3.3). (B)
Breakdown of individual base-substitution rates used for statistical significance of
mutation recurrence, for the same samples as in (A). The samples were ordered by
the rate of G→T transversions, which are indicative of smoking-induced mutations.
(C) Key clinical parameters for the samples described in (A) and (B) (table S10). The
first row indicates HPV detection by sequencing; the second row indicates HPV
detection by real-time PCR (RT-PCR). WGS, whole-genome shotgun sequencing.
EGF like repeats LNR
HD-N
HD-C
NOTCH1
TM
RAM ANK TAD PEST
Cleavage sites
Ligand binding
Nonsense
Missense
Splice site
Indel
Deletion
NOTCH2
Nonsense
Missense
NOTCH3
Nonsense
Missense
Deletion
Regions of NOTCH1 activating mutations:
Regions of alterations in HNSCC:
T-ALL
CLL
Fig. 2. NOTCHgene mutations identified in head and neck cancer. (Top) Schematic diagramof the domain
structure of NOTCH1 (domain structures of NOTCH2 and NOTCH3 are similar). (Middle and Bottom)
NOTCH genetic alterations in T-ALL, CLL, and HNSCC. All nonsense mutations occur upstream of the
transcriptional activation domain (TAD), which is required for transactivation of target genes. Each
arrowhead represents a single point mutation in an individual tumor of the class indicated to the left. EGF,
epidermal growth factor; LNR, Lin-12 NOTCH repeats; HD-N, N-terminal heterodimerization domain; HD-C,
C-terminal heterodimerization domain; TM, transmembrane; RAM, RBP-Jκ–associated module; ANK, ankyrin
repeats; PEST, Pro-Glu-Ser-Thr degradation motif.
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1158
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

genome projects such as those dealing with
ovarian cancer and multiple myeloma (7, 8), our
analysis of HNSCC revealed a larger number of
significantlymutatedgenes. However, the majority
of mutated genes did not reach statistical signif-
icance (table S6), suggestingthat manymaycontain
passenger events. Thus, we hypothesized that the
MutSig algorithm identified an enriched set of
genes that probably underwent positive selection
during tumorigenesis. Toward this end, numerous
significantly mutated genes had previously been im-
plicated in HNSCC—including TP53, CDKN2A,
HRAS, PTEN, andPIK3CA(q<0.1) (4)—providing
support for the validity of the approach. TP53,
the most commonly mutated gene in HNSCC, was
also disrupted by a 100-kb deletion detected by
whole-genome sequencing and validated with a
focal copy-number change detected by SNP array
(fig. S9). However, most significantly mutated genes
had not previously been implicated in HNSCC.
To explore their biological importance, we
first considered mutated HNSCC genes that also
undergo frequent genetic alterations in other can-
cers. NOTCH1 was particularly noteworthy: Point
mutations affecting this gene occurred in 11% of
the HNSCC tumors (Figs. 2 and 3 and tables S6
and S7), and focal deletions were seen in two
additional tumors (Fig. 2). Previous evidence
from animal models had implicated Notch dys-
regulation in cutaneous squamous cell carcinoma
(9), but somatic NOTCH1 mutations had not
previously been identified in squamous malig-
nancies. In addition, we found nonsynonymous
point mutations in NOTCH2 or NOTCH3 in 11%
of the samples (Figs. 2 and 3 and table S6) and a
focal deletion of NOTCH3 in one additional case
(Fig. 2). Whereas NOTCH1 contains activating
mutations in Tcell acute lymphoblastic leukemia
and chronic lymphocytic leukemia (10, 11) and
NOTCH2 contains activating mutations in dif-
fuse large Bcell lymphoma (12), the mutations in
HNSCC appeared to be loss-of-function muta-
tions, consistent with those recently described for
myeloid leukemia (13).
Several NOTCH1 nonsense mutations in
HNSCC are predicted to generate truncated pro-
teins that lack the C-terminal ankyrin repeat
domain, a region critical for transactivation of
target genes (Fig. 2) (14). Five additional mu-
tations (four missense and one in-frame deletion)
cluster in highly conserved residues situated with-
in or nearby the extracellular ligand binding
domain (Fig. 2). Two others are splice-site muta-
tions that may generate truncated proteins or
delete critical functional residues (for instance,
ligand binding or activation by proteolytic cleav-
age) (Fig. 2). Together, these findings suggested
that NOTCHdysregulation—and, more generally,
mechanisms governed by NOTCH signaling—
contribute to the genesis or progression of HNSCC.
A
q = 0.1
% mutated samples
HPV (Real-time PCR)
1 10
-2
10
-4
10
-6
q−value
**
1%
3%
5%
3%
4%
5%
5%
5%
7%
11%
5%
8%
5%
20%
7%
14%
12%
8%
12%
62%
50 40 30 20 10 0
# mutations
RB1
DICER1
NOTCH2
RIPK4
NOTCH3
SYNE2
EZH2
IRF6
TP63
MLL2
MED1
PIK3CA
HRAS
SYNE1
PTEN
NOTCH1
FAT1
CASP8
CDKN2A
TP53
6
2
4
2
1
6
2
5
0
6
6
2
7
3
9
6
3
0
8
0
6
2
8
1
4
6
2
8
9
7
6
3
0
2
1
6
2
7
4
1
6
2
6
9
9
6
2
4
6
9
6
2
7
5
5
0
0
0
7
6
6
3
0
8
1
6
3
0
5
8
6
2
6
0
1
6
2
8
6
3
6
2
9
9
5
6
2
4
3
2
6
2
2
9
8
6
2
9
2
6
6
3
0
9
5
6
2
3
1
8
0
0
3
6
1
6
2
8
5
4
6
2
6
8
6
6
2
4
1
5
6
2
5
0
5
0
0
1
9
0
6
2
2
3
7
6
2
6
4
6
6
2
6
2
4
6
2
7
5
6
6
2
4
8
1
0
1
0
0
0
6
2
4
1
7
0

0
4
6
6
2
6
7
2
6
2
8
6
0
6
2
5
3
2
6
2
8
3
2
6
2
9
2
1
0
0
7
6
1
6
2
6
5
2
0
0
3
7
8
6
2
3
3
8
6
2
3
7
6
6
2
9
9
6
0
0
3
1
3
6
3
0
4
8
0
0
1
2
2
6
3
0
3
9
0
0
4
4
3
6
2
8
6
1
6
2
8
5
7
_
2
6
2
3
3
2
6
2
8
0
7
6
2
4
9
3
6
2
6
7
1
6
2
5
1
5
6
2
5
3
9
6
3
0
0
7
6
3
1
1
4
6
2
9
8
4
0

0
6
4
6
2
6
0
2
6
2
4
2
6
6
3
1
1
5
6
2
8
2
5
6
2
8
1
0
6
2
7
4
0
0
0
3
3
8
0
0
4
6
6
6
2
3
7
4
6
3
0
2
7
Syn.
Missense
Splice Site
Nonsense
Frame shift
In frame Indel
CDKN2A
Mut/Del 25%
RB1
Mut/Del 3%
Proliferation
HPV16 E7
Infected 15%
HPV16 E6
Infected 15%
TP53
Mut/Del 63%
HRAS
Mut 4%
Cell Death
PIK3CA
Mut 8%
Terminal
Differentiation
CCND1
Amp 22%
CASP8
Mut 8%
PTEN
Mut/Del 8%
B
?
NOTCH1,2,3
Mut/Del 22%
TP63
Mut/Amp 8%
IRF6
Mut 5%
DICER1
Mut 3%
RIPK4
Mut 3%
SYNE1,2
Mut 24%
?
M
IR
2
0
3
Samples mutated in the core
differentiation genes
Fig. 3. Genetic disruption of a squamous differ-
entiation program in head and neck cancers. (A)
Heat-map representation of individual mutations
present in a series of 74 tumors, represented in
columns. (Top) HPV status by tumor. (Middle) Matrix
of mutations in individual genes by type of mutation
and tumor. Asterisks indicate tumors characterized
by whole-exome and whole-genome sequencing.
(Left) Number of mutations in each gene. Percent-
ages represent the fraction of tumors harboring at
least one mutation in the specified gene. (Right) Se-
lected recurrently mutated genes ranked by q value.
Genes that define the core differentiation cluster are
shown in red. (B) Proposed partial wiring diagramof
the molecular circuitry of HNSCC. Darker blue, loss
of function; red, gain of function; yellow, viral genes;
light blue, genes putatively implicated in HNSCC.
Numbers listed beneath each protein represent the
fraction of tumors harboring mutations, amplifica-
tions, or deletions in the corresponding genes.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1159
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

To further interpret the mutations identified
in HNSCC, we looked for functionally related
gene sets harboring an excess of mutations. For
this purpose, we considered an expanded list of
76 genes (q < 0.25) (table S7) and looked for
enrichment in functional gene sets. The highest-
scoring gene set contained genes related to epi-
dermal development (table S8). The significantly
mutated genes (q < 0.25) in this set included
NOTCH1, IRF6, and TP63. These genes are all
clearly related to squamous differentiation. The
most abundant TP63 protein product in squamous
epithelia, known as DNp63, promotes renewal
of basal keratinocytes by a mechanism that re-
quires down-regulation of NOTCH1and CDKN2A
(15–17). IRF6, in turn, has been implicated in the
proteasomal degradation of DNp63 (18). Further-
more, terminal differentiation in squamous epi-
thelia is induced in response to genotoxic stress
by a mechanism involving p53-dependent trans-
activation of NOTCH1, an activity antagonized
by DNp63 (19). Because HNSCC involves trans-
formation of the squamous epithelial lineage,
which is histologically similar to the epidermis,
these findings led us to hypothesize that muta-
tions in such genes disrupt a stratified squamous
development/differentiation program in precursor
cells of this malignancy.
Further inspection of recurrent mutations
identified 11 additional genes carrying disruptive
mutations that function in the squamous differ-
entiation program. The evidence includes mouse
knockouts with defects in squamous epithelial
differentiation (Notch1, Notch2, Irf6, Tp63, Ripk4,
Cdh1, Ezh2, and Dicer1) (Fig. 3A) (20–25), hu-
man germline mutations causing orofacial cleft-
ing syndromes (IRF6, TP63, CDH1, and MLL2)
(26), and knockdown or deregulated expression
leading to a differentiation block and increased
proliferation in cultured human keratinocytes
(TP63, NOTCH1, IRF6, MED1) (15, 27). Thus,
many mutated genes in HNSCC may govern
squamous differentiation. These mutations may
promote an immature and more proliferative
basal-like phenotype, consistent with known stages
of progression and markers of differentiation in
HNSCC (Fig. 3B).
We also found recurrent mutations in less well-
characterized genes. For example, we observed
mutations in SYNE1 and SYNE2 in 20 and 8% of
HNSCC samples, respectively (fig. S7 and tables
S6 and S7). These genes have been implicated in
the regulation of nuclear polarity (28), a process
that operates upstream of NOTCH1 in squamous
epithelia (Fig. 3B) (29). RIMS2 and PCLO mu-
tations were seen in 11 and 12% of cases, re-
spectively; the corresponding proteins mediate
calcium sensing (30), another crucial process for
terminal squamous differentiation (20).
Beyond the genes directly involved in squa-
mous differentiation, we found mutations involv-
ing two apoptosis-related genes: CASP8 (8%)
and DDX3X (4%) (fig. S7 and table S7). Thus,
suppression of apoptosis may also contribute to
HNSCC pathogenesis, perhaps in concert with
disrupted squamous maturation (Fig. 3B). The
histone methyltransferases PRDM9 (11%) and
EZH2 (6%) are also recurrently mutated.
Viral infection by HPV figures prominently
into the etiology of a subset of HNSCC and is
most frequently detected by in situ hybridization
or p16 immunohistochemistry. We reasoned that
HNSCC genome sequencing might also offer a
robust HPVdetection method. We therefore used
the PathSeq algorithm (31) and a viral sequence
database to identify HNSCC sequencing reads
that aligned to HPVgenomes. We observed HPV-
16 sequence reads in 14 tumors (19%) (range: 1 to
40,000 reads), 11 of which were also positive by
HPV-16 PCR (P < 0.0001) (table S9). The three
tumors that were HPV-negative by PCRhad very
lowHPV-16 sequence read counts (fig. S10); this
may reflect reduced HPVdosage or technical con-
tamination. We observed an inverse correlation
between HPV status (determined by sequencing)
and TP53 mutation, as shown previously (P =
0.001, Fisher’s exact test) (32). These data
underscore the potential utility of massively
parallel sequencing to detect both human and
nonhuman etiologic agents in tumor specimens.
Given that NOTCH pathway inhibitors have
entered clinical trials, the discovery of loss-of-
function NOTCH1 mutations in HNSCC may
have important therapeutic implications. A re-
cent clinical trial of a g-secretase inhibitor (which
inhibits NOTCH) was halted in part because of
an increased frequency of skin cancers in the
treatment arm (33). This clinical observation is
consistent with those from mouse models, in
which cutaneous knockout of Notch1 promotes
skin tumor formation (24). Our results suggest
that patients taking g-secretase inhibitors may
require monitoring for the development of both
cutaneous and head/neck squamous malignancies.
Despite the anatomical distinctions that dom-
inate current clinical management of HNSCC,
our results point to several unifying features at
the molecular level. For example, TP53 inacti-
vation, either through somatic mutation or HPV
infection, appears nearly universal in this malig-
nancy. The present study suggests that disruption
of the squamous differentiation program may
represent an additional overarching feature that
occurs by numerous genetic mechanisms across
tumors from multiple anatomic sites. Thus,
HNSCC pathogenesis may involve a maturation
arrest or a lineage dependency similar to that seen
in other cancer types (34). However, HNSCC
appears to be unusual in that the mutational
etiology is diverse, in contrast to leukemia and
prostate cancer in which developmental patholo-
gies appear to be caused by lesions in only a few
target genes. Rational therapeutic avenues tar-
geting this block in squamous differentiation
may require synthetic lethal approaches to iden-
tify specific cellular dependencies arising from
NOTCH inactivation, TP63 alteration, or other
events that deregulate the program. Finally, our
results demonstrate that whole-exome sequencing
of large numbers of tumor/normal pairs should
enable fundamental new insights into tumor bi-
ology that are relevant to many human cancers.
References and Notes
1. J. Ferlay et al., Int. J. Cancer 127, 2893 (2010).
2. A. Argiris, M. V. Karamouzis, D. Raben, R. L. Ferris,
Lancet 371, 1695 (2008).
3. S. Gupta, W. Kong, Y. Peng, Q. Miao, W. J. Mackillop,
Int. J. Cancer 125, 2159 (2009).
4. C. R. Leemans, B. J. Braakhuis, R. H. Brakenhoff,
Nat. Rev. Cancer 11, 9 (2011).
5. W. Lee et al., Nature 465, 473 (2010).
6. E. D. Pleasance et al., Nature 463, 184 (2010).
7. M. A. Chapman et al., Nature 471, 467 (2011).
8. The Cancer Genome Atlas Research Network, Nature 474,
609 (2011).
9. U. Koch, F. Radtke, Curr. Top. Dev. Biol. 92, 411 (2010).
10. X. S. Puente et al., Nature 475, 101 (2011).
11. J. C. Sok et al., Clin. Cancer Res. 12, 5064 (2006).
12. S. Y. Lee et al., Cancer Sci. 100, 920 (2009).
13. A. Klinakis et al., Nature 473, 230 (2011).
14. R. A. Kovall, S. C. Blacklow, Curr. Top. Dev. Biol. 92, 31 (2010).
15. B. C. Nguyen et al., Genes Dev. 20, 1028 (2006).
16. R. Okuyama et al., Oncogene 26, 4478 (2007).
17. X. Su et al., EMBO J. 28, 1904 (2009).
18. F. Moretti et al., J. Clin. Invest. 120, 1570 (2010).
19. T. Yugawa et al., Cancer Res. 70, 4034 (2010).
20. C. Blanpain, E. Fuchs, Nat. Rev. Mol. Cell Biol. 10, 207 (2009).
21. M. Chidgey et al., J. Cell Biol. 155, 821 (2001).
22. A. Dumortier et al., PLoS ONE 5, e9258 (2010).
23. E. Ezhkova et al., Genes Dev. 25, 485 (2011).
24. M. Nicolas et al., Nat. Genet. 33, 416 (2003).
25. C. L. Tinkle, T. Lechler, H. A. Pasolli, E. Fuchs, Proc. Natl.
Acad. Sci. U.S.A. 101, 552 (2004).
26. Online Mendelian Inheritance in Man, www.ncbi.nlm.nih.gov.
27. C. L. Tu, W. Chang, Z. Xie, D. D. Bikle, J. Biol. Chem. 283,
3519 (2008).
28. Y. Lüke et al., J. Cell Sci. 121, 1887 (2008).
29. S. E. Williams, S. Beronja, H. A. Pasolli, E. Fuchs, Nature
470, 353 (2011).
30. K. Fujimoto et al., J. Biol. Chem. 277, 50497 (2002).
31. A. D. Kostic et al., Nat. Biotechnol. 29, 393 (2011).
32. W. H. Westra et al., Clin. Cancer Res. 14, 366 (2008).
33. A. Extance, Nat. Rev. Drug Discov. 9, 749 (2010).
34. L. A. Garraway, W. R. Sellers, Nat. Rev. Cancer 6, 593 (2006).
Acknowledgments: We thank the Broad Institute Sequencing
Platform for the generation of sequencing data, the staff of
the Biological Samples Platform and the Genetic Analysis
Platform at the Broad Institute, and J. Aster for helpful
discussions. This work was conducted as part of the Slim
Initiative for Genomic Medicine, a joint U.S.-Mexico project
funded by the Carlos Slim Health Institute, and is part of a
global effort in collaboration with the International Cancer
Genome Consortium. The work was also supported by grants
from the National Human Genome Research Institute (S.B.G.,
E.S.L.), the National Cancer Institute (M.M., S.B.G., L.A.G., J.R.G.,
A.M.E.), the Starr Cancer Consortium (L.A.G.), the Novartis
Institutes for BioMedical Research (N.S.), the Howard
Hughes Medical Institute (T.R.G.), the American Cancer
Society ( J.R.G.), and the NIH Director’s New Innovator Award
(L.A.G.). L.A.G. and M.M. are paid consultants for and
equity holders in Foundation Medicine, a genomics-based
oncology diagnostics company, and are also paid consultants
for Novartis. M.F.B. is a paid consultant for Foundation
Medicine. E.S.L. is an equity holder in Foundation Medicine.
The compressed binary Sequence Alignment/Map files
(BAM) and SNP array data are available in dbGaP under
accession no. phs000370.v1.p1.
Supporting Online Material
www.sciencemag.org/cgi/content/full/science.1208130/DC1
Materials and Methods
SOM Text
Figs. S1 to S10
Tables S1 to S11
References (35–66)
10 May 2011; accepted 5 July 2011
Published online 28 July 2011;
10.1126/science.1208130
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1160
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

MED23 Mutation Links Intellectual
Disability to Dysregulation of
Immediate Early Gene Expression
Satoru Hashimoto,
1
* Sarah Boissel,
2
* Mohammed Zarhrate,
2
Marlène Rio,
2
Arnold Munnich,
2
Jean-Marc Egly,
1
† Laurence Colleaux
2

MED23 is a subunit of the Mediator complex, a key regulator of protein-coding gene expression.
Here, we report a missense mutation (p. R617Q) in MED23 that cosegregates with nonsyndromic
autosomal recessive intellectual disability. This mutation specifically impaired the response of
JUN and FOS immediate early genes (IEGs) to serum mitogens by altering the interaction
between enhancer-bound transcription factors (TCF4 and ELK1, respectively) and Mediator.
Transcriptional dysregulation of these genes was also observed in cells derived from patients
presenting with other neurological disorders linked to mutations in other Mediator subunits or
proteins interacting with MED. These findings highlight the crucial role of Mediator in brain
development and functioning and suggest that altered IEG expression might be a common
molecular hallmark of cognitive deficit.
M
ammalian Mediator (MED) is an evo-
lutionary conserved multiprotein com-
plexcomposedof more than 20 subunits
that form four distinct modules (fig. S1, upper
scheme) (1–3). This large complex is a key reg-
ulator of gene expression that functions as an
adaptor and conveys essential information from
transcription factors (TFs) bound at upstream
responsive elements to the basal RNA polymer-
ase II (Pol II) transcription machineries (4, 5).
Mediator also plays a key role in sensing devel-
opmental and environmental signals to deliver
adequate outputs to the transcription machinery
(6), and specific subunits are likely to be dedi-
cated to fine-tuning of distinct expression pro-
grams. Reports linking Mediator subunits to human
diseases may thus allow a deeper understanding
of how cell-specific expression programs are
regulated.
We ascertained a large Algerian consanguin-
eous multiplex family in which five affected in-
dividuals presented with nonsyndromic intellectual
disability (ID) (Fig. 1A). Genetic analyses identified
a disease-causing mutation (c.1850G>A, p.R617Q)
within the MED23 gene (NM_015979.2) that en-
codes one of the tail module’s Mediator subunits
(Fig. 1B) [see supporting online material (SOM)
text]. This variation cosegregated with the dis-
ease and was not present in dbSNP build 132, nor
was it detected in any of 608 control chromosomes,
including 242 chromosomes from individuals of
Algerian origin. Moreover, R617 was absolute-
ly conserved across all MED23 orthologs, from
Xenopus to Drosophila and human. This mutation
does not affect MED23 expression, protein sta-
bility or architecture, or composition of the whole
Mediator complex (see SOM text and fig. S1).
MED23 was originally identified as a genet-
ic suppressor of a hyperactive ras phenotype in
Caenorhabditis elegans (7) and mediates the im-
mediate early gene (IEG) response to serum mi-
togens (8). Gene expression analysis by microarray
experiments after serum addition to serum-starved
control- (wildtype) andpatient-derived(M23/R617Q)
cultured skin fibroblasts revealed a difference in
the serum response of a small number of genes
(SOM and fig. S2). We thus further evaluated the
effect of this mutation on JUN, FOS, and EGR1
expression by quantitative reverse transcription–
polymerase chain reaction (RT-PCR). Expression
of EGR1 was unchanged in response to serum
addition in both wild-type and M23/R617Q cells
(Fig. 1C). However, in response to serum, JUN
expression was down-regulated and FOS expres-
sion was up-regulated in M23/R617Q cell com-
pared to controls (Fig. 1, Dand E). M23/R617Q
cells engineered to express wild-type MED23 pro-
tein responded to serum treatment as did wild-
type cells [EGR1 expression was unchanged; JUN
and FOS RNAsyntheses were restored to control
levels (Fig. 1)]. Finally, the R617Q mutation had
no effect on transcriptional activation induced by
other stimuli such as ultraviolet (UV) irradiation
or nuclear receptor activation, suggesting that it
only causes a selective impairment of a subset of
transcription activation processes (see SOM text
and fig. S3).
To test whether dysregulation of IEG ex-
pression in M23/R617Q cells was the result of
defective recruitment of transcriptional factors, we
next performed chromatin immunoprecipitation
(ChIP) experiments with antibodies directed against
various proteins involved in the formation of the
transcriptional preinitiation complex (PIC). ChIP
showed decreased recruitment of MED proteins
to the JUN promoter region in M23/R617Q cells
compared with controls (Fig. 2, Ato C). Reduced
amounts of bound TFIIH (CDK7) and p-Tef-b
complex (CDK9) were also observed at the JUN
promoter in M23/R617Q cells (Fig. 2, E and F).
These two complexes are responsible for the Ser
2
and Ser
5
phosphorylation of the Pol II C-terminal
domain (CTD) and play a key role in promoting
its escape into the productive elongation phase
(9). Although recruitment of Pol II to the pro-
moter appears equally effective in wild-type or
1
Institut de Génétique et de Biologie Moléculaire et Cellulaire,
CNRS/INSERM/Universitéde Strasbourg, BP 163, 67404 Illkirch
Cedex, C. U. Strasbourg, France.
2
INSERMU781andDépartement
de Génétique, Fondation IMAGINE, UniversitéParis Descartes,
Hôpital Necker-Enfants Malades, Paris 75015, France.
*These authors contributed equally to this work.
†To whom correspondence should be addressed. E-mail:
laurence.colleaux@inserm.fr (L. C.); egly@igbmc.fr ( J. M. E.)
A B
C D E
affected individual
control individual
f
o
l
d

m
R
N
A
Time (min)
EGR1
30
0
60
120
180
0 30 60 90
Time (min)
JUN
0
2
4
6
30
0
3
6
9
0 30 60 90
Time (min)
FOS
30
0
7
14
21
0 30 60 90
WT *
* *
*
*
**
M23/R617Q
M23/R617Q
(rescue)
Fig. 1. Identification of the MED23 mutation in an intellectually disabled family and characterization
of MED23/R617Q mutated protein in Mediator complex. (A) Pedigree of the family. Filled symbols
indicate affected individuals and slashes, deceased individuals (B) Electrophoregrams showing the
c.1850G>A variation in the MED23 sequence in an affected individual and a healthy control. (C to E)
Time-course analysis of EGR1 (C), JUN (D), and FOS (E) expression after serum addition to serum-
starved wild-type (WT), M23/R617Q, and M23/R617Q cells transfected with MED23 cDNA. The
values of three independent experiments are shown relative to serum-starved cells at time 0. *P <
0.05 and **P < 0.01.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1161
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

M23/R617Q cells (Fig. 2D), Pol II at the JUN
promoter in M23/R617Q cells showed defective
phosphorylation (fig. S4, A and B). The aber-
rant phosphorylation of Pol II diminished its abil-
ity to initiate transcription and to further escape
into the productive elongation phase (fig. S4C).
Pol II at the FOS promoter was indistinguishable
between wild-type and M23/R617Q fibroblasts
(Fig. 2, J to O).
Mediator couples chromatin remodeling and
PIC formation (10). Acetylation of histone H3
lysine 9 (H3K9Ac) or di- and trimethylation of
histone H3 lysine 4 (H3K4M) are common hall-
marks of transcription activation, whereas meth-
ylation of histone H3 lysine 9 (H3K9M) has been
shown to correlate with transcriptional repres-
sion (11). ChIP using antibodies raised against
H3K9Ac, H3K4M, or H3K9M showed that cells
carrying the R617Q mutation displayed reduced
H3K9 acetylation and increased H3K9 methyla-
tion (but no change in H3K4 methylation) com-
pared with controls (Fig. 2, G to I), suggesting
impaired chromatin remodeling in the vicinity
of the JUN promoter in M23/R617Q cells. No
such differences were noted at the FOS promoter
(Fig. 2, P to R).
We next investigated the effect of the R617Q
mutation on the binding of several transcription
factors on JUN and FOS promoter regions. The
JUN promoter includes several AP1, SP1, and
MEF2/RSRF proximal responsive elements, as
well as a distal TCF regulatory element located 3
kb upstream of the transcription start site (Fig. 3,
upper scheme) (12, 13). Phosphorylated JUN
(JUN-P), MEF2, and SP1 proteins are normally
present on JUN promoter upon serum induction
of wild-type and M23/R617Q cells (Fig. 3, A to
C). Serum treatment of wild-type cells also caused
TCF4 recruitment at its responsive distal TCF
element (Fig. 3D), and ChIP revealed incidental
binding of TCF4 to the proximal promoter (Fig.
3E) owing to the formation of a chromatin loop
(SOMtext and fig. S5) (13, 14). By contrast, TCF4
recruitment at both distal and proximal promot-
ers was impaired in M23/R617Qcells (Fig. 3, D
and E). These results, combined with the obser-
vation that TCF4 coprecipitated with the wild-
type MED23 subunit (Fig. 3I), led us to conclude
that Mediator is required to link enhancer-bound
TCF4 with the basal transcription machinery, a link-
age that fails in the presence of the MED23/R617Q
mutation.
FOS transcription is also controlled by AP1
sites and is mediated by cooperative binding of
serum response factor (SRF) and Ets ternary tran-
scription factors (TCFs) [ELK1, SAP1 (ELK4),
and NET (SAP2/ERP/ELK3)] to serum response
elements (SREs) (15). The MED23 mutation does
not affect binding of JUN-P to the AP1 sites in the
J K L
M N O
MED1
MED17
CDK8
Pol II CDK7 CDK9
0
0.4
0.8
1.2
- +
P H3K9Ac R
H3K4M
Q H3K9M
0
3
6
9
- +
FOS
0
4
8
12
- +
0
0.3
0.6
0.9
- +
0
0.8
1.6
2.4
- +
0
0.4
0.8
1.2
- +
0
0.4
0.8
1.2
- +
A
%

i
n
p
u
t
MED1 B MED17
C
CDK8
D
%

i
n
p
u
t
Pol II E F
CDK7 CDK9
G
H3K9Ac
I
H3K4M H
H3K9M
%

i
n
p
u
t
JUN
0
2
4
6
- +
0
0.1
0.2
0.3
- +
0
3
6
9
- +
WT
M23/R617Q
WT
M23/R617Q
0
0.3
0.6
0.9
- +
0
0.5
1
1.5
- +
0
0.4
0.8
1.2
- +
0
0.4
0.8
1.2
- +
0
0.3
0.6
0.9
- +
0
5
10
15
- +
0
3
6
9
- +
0
0.5
1
1.5
- +
*
*
*
*
*
*
*
Fig. 2. Response of IEGs to serum in WT and M23/R617Q skin fibroblasts. ChIP analysis using primer
sets specific for JUN (A to I) and FOS promoters ( J to R) was performed on the chromatin fraction from
WT and M23/R617Q cells that were untreated (−) or treated with serum for 10 min (+). Values from
real-time PCR were normalized to the percentage of input chromatin. Error bars represent the SD of at
least two independent experiments. *P < 0.05 and **P < 0.01.
Fig. 3. Binding of transcription ac-
tivators to IEG promoter regions.
ChIP experiments using primer sets
specific for JUN proximal promoter
(A to C and E) or distal promoters
(D) and FOS promoter (F to H) were
performed on the chromatin frac-
tion from WT and M23/R617Q cells
that were untreated (−) or treated
with serum for 10 min (+). Values
from real-time PCR were normalized
to the percentage of input chroma-
tin. Error bars represent the SD of at
least two independent experiments.
* P <0.05 and ** P < 0.01. (I) Coim-
munoprecipitationexperiments show-
ing interaction between MED23 and
TCF4 or ELK1 in HeLa cells.
JUN
-3,000
TCF MEF2/
RSRF
Sp1
+1 -200
AP-1 AP-1
A
%

i
n
p
u
t
JUN-P
B
MEF2
D
SP1 C
0
0.6
1.2
1.8
- +
0
1.1
2.2
3.3
- +
0
0.5
1
1.5
- +
E
TCF4-distal
%

i
n
p
u
t
distal proximal
0
0.6
1.2
1.8
- +
FOS
SRF
SRE
+1 -400
Ets AP-1 AP-1
%

i
n
p
u
t
F G
ELK1
H
ELK3
0
0.6
1.2
1.8
- +
0
0.6
1.2
1.8
- +
0
0.6
1.2
1.8
- +
I
I
n
p
u
t
A
b

C
t
r
l
A
b

T
C
F
4
MED23
A
b

E
L
K
1
JUN-P
TCF4-proximal
0
0.6
1.2
1.8
0 10
WT
M23/R617Q
*
*
**
26 AUGUST 2011 VOL 333 SCIENCE www.sciencemag.org 1162
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

FOS promoter (Fig. 3F). By contrast, we observed
reduced binding of ELK1, which interacts with
MED23 (Fig. 3I) (16), to the FOS promoter in
M23/R617Q cells (Fig. 3G), and increased bind-
ing of its related paralog ELK3 (Fig. 3, Gand H).
Notably, these data are consistent with previous
analyses in mouse demonstrating that ELK3 was
the predominant ternary complex in mouse fibro-
blasts and that loss of MED23 nearly eliminated
activation by ELK1 (17). Taken together, our re-
sults suggest that the dysregulation of FOS ex-
pression observed in M23/R617Qcells upon serum
induction is due to altered constitution of what
remains nonetheless a functional PIC.
IEG expression affects brain development
and plasticity (18). We questioned whether ab-
normal serum-induced JUN and FOS expression
could be also observed in neurological disorders
caused by mutations in genes encoding other
Mediator subunits or proteins interacting with
MED. Mutations of MED12 are responsible for
two allelic forms of X-linked IDsyndromes: Opitz-
Kaveggia syndrome [Mendelian Inheritance in
Man (MIM) 305450] and Fryns-Lujan syndrome
(MIM309520) (19, 20). Mutations in the XPD
subunit of TFIIH, which interacts with Mediator
(Fig. 4A) (21–23), result in the autosomal reces-
sive disorder diseases xeroderma pigmentosum
(XP-D, MIM278730) and trichothiodystrophy
(TTD, MIM601675) (24). Approximately 30% of
XP patients have progressive neurological de-
generation, and all TTD cases are mentally re-
tarded. Similar EGR1 expression (Fig. 4, Band E)
but impaired regulation of JUN and FOS expres-
sion upon serum induction was observed in
lymphoblastoid cell lines derived from an Opitz-
Kaveggia patient (M12-L; Fig. 4, C and D) and
in skin fibroblasts derived from a XP-D case
(Fig. 4, F and G). M12-L and XP-D cells also
showed impaired transcriptional activation induced
by other stimuli such as UVirradiation or nuclear
receptor ligands (fig. S6) (25, 26). Fibroblasts de-
rived from a TTD patient exhibited a partial up-
regulation of EGR1 and a down-regulation of JUN
expression (Fig. 4, E and F). As a control, no dif-
ference was observed in fibroblasts derived from
a XP patient who presented with multiple skin
cancers but no neurological abnormalities and
who carried a mutation in the xeroderma pigmen-
tosumgroup Cprotein (XP-C, MIM279720) (Fig.
4, H to J).
We have shown that a single MED23 muta-
tion may have distinct effects on transactivation
complex formation and opposite consequences
for gene expression owing to impaired interaction
between mutated MED23 protein and cognate spe-
cific DNA binding protein (fig. S7). This study
underlines, therefore, the key role of the MED23
subunit in organizing protein interactions within
the transactivation complex.
The nonsyndromic ID identified here corre-
lates with transcriptional defects in the IE genes.
However, patients with syndromic conditions
showed disruptions in a larger spectrum of pro-
cesses, including nuclear receptor–mediated re-
sponse and UV-induced gene expression.
IEGs may couple short-termneuronal activity
with changes in gene transcription of neurons.
Moreover, JUN and FOS are implicated in learn-
ing and consolidation of a long-term memory
trace (27, 28). We thus propose that the intellec-
tual disability observed in patients with Mediator
and TFIIHmutations could be, at least in part, the
result of impaired fine-tuning of IEG expression
during development. We propose that altered IEG
expression might provide a molecular signature for
cognitive deficits.
References and Notes
1. R. C. Conaway, S. Sato, C. Tomomori-Sato, T. Yao,
J. W. Conaway, Trends Biochem. Sci. 30, 250 (2005).
2. R. D. Kornberg, Trends Biochem. Sci. 30, 235
(2005).
3. D. J. Taatjes, Trends Biochem. Sci. 35, 315 (2010).
4. G. T. Cantin, J. L. Stevens, A. J. Berk, Proc. Natl. Acad.
Sci. U.S.A. 100, 12003 (2003).
5. G. Cai, T. Imasaki, Y. Takagi, F. J. Asturias, Structure 17,
559 (2009).
6. S. Malik, R. G. Roeder, Nat. Rev. Genet. 11, 761
(2010).
7. N. Singh, M. Han, Genes Dev. 9, 2251 (1995).
8. J. L. Stevens et al., Science 296, 755 (2002).
9. H. P. Phatnani, A. L. Greenleaf, Genes Dev. 20, 2922
(2006).
10. J. C. Black, J. E. Choi, S. R. Lombardo, M. Carey, Mol. Cell
23, 809 (2006).
11. T. Kouzarides, Cell 128, 693 (2007).
12. C. A. Hazzalin, L. C. Mahadevan, Nat. Rev. Mol. Cell Biol.
3, 30 (2002).
13. A. S. Nateri, B. Spencer-Dene, A. Behrens, Nature 437,
281 (2005).
14. M. H. Kagey et al., Nature 467, 430 (2010).
15. G. Buchwalter, C. Gross, B. Wasylyk, Gene 324, 1
(2004).
16. G. Wang et al., Mol. Cell 17, 683 (2005).
17. M. A. Balamotis et al., Sci. Signal. 2, ra20
(2009).
18. T. Herdegen, J. D. Leah, Brain Res. Brain Res. Rev. 28,
370 (1998).
19. H. Risheg et al., Nat. Genet. 39, 451 (2007).
20. C. E. Schwartz et al., J. Med. Genet. 44, 472
(2007).
21. C. Esnault et al., Mol. Cell 31, 337 (2008).
22. S. Akoulitchev, S. Chuikov, D. Reinberg, Nature 407,
102 (2000).
23. J. Soutourina, S. Wydau, Y. Ambroise, C. Boschiero,
M. Werner, Science 331, 1451 (2011).
24. J. E. Cleaver, E. T. Lam, I. Revet, Nat. Rev. Genet. 10,
756 (2009).
25. N. Le May et al., Mol. Cell 38, 54 (2010).
26. E. Compe et al., Nat. Neurosci. 10, 1414 (2007).
27. W. Tischmeyer, R. Grimm, Cell. Mol. Life Sci. 55, 564
(1999).
28. S. Sanyal, D. J. Sandstrom, C. A. Hoeffer, M. Ramaswami,
Nature 416, 870 (2002).
Acknowledgments: We are grateful to the patients and
their families for their participation in the study. We
thank J. P. Jais for LOD-score calculation. Work in the
J.-M. Egly’s laboratory was supported by a European
Research Council Advanced Grant 2009, the French
National Research Agency (no. ANR-08-GENOPAT-042),
and the Association pour la Recherche sur le Cancer
(SL220100601335). L. Colleaux’s laboratory was
supported in part by the Centre National de la Recherche
Scientifique and the French National Research Agency
(no. ANR-08-MNP-010). S.H. was supported by an
INSERM young investigator grant. S.B. was supported
by the Région Ile-de France, the Fondation pour la
Recherche Médicale, and the Fondation Jerome Lejeune.
The authors declare that they have no conflicts
of interest.
Supporting Online Materials
www.sciencemag.org/cgi/content/full/333/6046/1161/DC1
Materials and methods
SOM Text
Figs. S1 to S7
Table S1
References (29–36)
6 April 2011; accepted 1 July 2011
10.1126/science.1206638
f
o
l
d

m
R
N
A
f
o
l
d

m
R
N
A
JUN FOS EGR1
WT-L
M12-L
f
o
l
d

m
R
N
A
0
2
4
6
- +
0
2
4
6
- +
0
60
120
180
- +
0
60
120
180
- +
0
0.7
1.4
2.1
- +
0
6
12
18
- +
0
5
10
15
- +
I
n
p
u
t
A
b

C
t
r
l
A
b

c
d
k
7
MED12
MED23
p62
WT
XP-D
TTD
WT
XP-C
* *
*
*
**
*
230
130
M.W.
60
0
4
8
12
- +
0
4
8
12
- +
A
B C D
E F G
H I J
Fig. 4. Defective IEG expres-
sion in other neurological dis-
orders. (A) Mediator subunits
MED12 and MED23 were co-
immunoprecipitated with TFIIH
subunits CDK7 and p62 using
500-mg HeLa extracts. Induc-
tion of EGR1 (B, E, H), JUN (C,
F, I), and FOS (D, G, J) expres-
sion after serum treatment of
lymphoblastoid cell lines derived
fromahealthy subject (WT-L) or an
Opitz-Kaveggia patient (M12-L),
and of skin fibroblasts derived
froma healthy subject (WT), a pa-
tient with xeroderma pigmento-
sum group D (XP-D), a patient with trichothiodystrophy (TTD), and a patient with xeroderma pigmentosum
group C (XP-C). Values are shown relative to nontreated cells. Error bars represent the SDof three independent
experiments.
www.sciencemag.org SCIENCE VOL 333 26 AUGUST 2011 1163
REPORTS

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m

www.sciencemag.org/products 1167
Protein and PePtide Microarray
the Proarray Ultra is a uniquely powerful protein and peptide array
platform, which is more sensitive than optimized ecL eLiSa. the
new platform allows analysis of more than 30,000 peptides or pro-
teins at once, and typically requires only 1µl of analyte sample. With
its sensitivity and fexibility, Proarray Ultra overcomes the traditional
trade off between optimized 96-well plate-based assays, which are
reliable and have good performance characteristics but lack the re-
quired throughput, and high throughput array technologies, which
are typically only semiquantitative. analysis with Proarray Ultra is
provided by Proimmune to customers as a turnkey service and will
enable a broad new range of research programs that are in many
cases simply not feasible with existing approaches. applications in-
clude antigen and monoclonal antibody characterization, analysis of
serum antibody specifcity, correlation of antibody responses with
disease onset, progression and outcome, protein-protein interaction
site mapping, iterative peptide and protein design, and general bio-
assay design.
Proimmune
For info: 888-505-7765 www.proimmune.com
cytoSkeLeton and aSSociated
Protein enrichMent
two new Proteoextract kits enable researchers to more rapidly and
effectively enrich, isolate, and analyze the cytoskeleton and associ-
ated proteins. the Proteoextract cytoskeleton enrichment and iso-
lation kit provides the necessary extraction buffers to quickly and
selectively remove soluble cytoplasmic and nuclear protein fractions
for downstream biochemical analysis including Western blot and
mass spectrometry. the assay is performed in as little as 20 minutes
and foregoes the need for harsh buffers and centrifugation, thus
retaining more of the native focal adhesion- and actin-associated
proteins, as compared to traditional protein fractionation methods.
the Proteoextract native cytoskeleton enrichment and Staining kit
includes the necessary reagents to effciently enrich, retain, and
visualize the cytoskeleton and associated proteins in their native,
adhered conformations with the same gentle technique. the kit
also includes labeled phalloidin and antibodies against vinculin and
glyceraldehyde-3-phosphate dehydrogenase (GaPdh) as markers
for visualization and colocalization of the actin cytoskeleton, focal
adhesions, and cytosol for immunofuorescence analysis.
eMd Millipore
For info: 800-645-5476 www.millipore.com
MicroWave PePtide SyntheSizer
the Biotage initiator+ SP Wave is a versatile synthesis solution as
it can be used as a stand-alone peptide synthesizer or easily con-
verted to a high specifcation microwave assisted organic synthe-
sis (MaoS) system. it is the ideal development tool for chemists
synthesizing peptides, peptidomimetics, and small molecules using
single or multistep procedures. the initiator+ SP Wave is perfectly
suited for the small scale synthesis of peptides and peptidomimet-
ics including peptide nucleic acid using microwave irradiation, from
a scale as low as 5 µmol. the semiautomated operation can reduce
the consumption and waste of expensive reagents that occur in oth-
er systems due to priming or use of large dead volumes. the system
includes a large 10-inch touchscreen, intuitive software, pre- or self-
defned methods, inert gas capability, vortex, and magnetic stirring.
Biotage
For info: 800-446-4752 www.biotage.com
3d ceLL cULtUre ScaffoLd
alvetex is a unique cell culture scaffold that enables genuine three-
dimensional (3d) cell culture to be grown routinely and cost-effec-
tively. alvetex provides a stable and inert scaffold for cells to grow,
mimicking in vivo growth and formation of tissues in the body. alve-
tex has a highly porous architecture that is specifcally tailored for 3d
cell culture and designed to be compatible with existing cell culture
techniques and post cell culture analysis. extensively tested and vali-
dated with a large number of cell types, alvetex enables the forma-
tion of complex 3d cellular structures, increases viability, decreases
cell death, and enhances cell function. Using this unique material,
cells can be visualized by confocal or live-cell imaging or treated like
tissue. With alvetex, it’s easy to process cells for histological stain-
ing, immunocytochemistry, and electron microscopy. viable cells can
also be isolated by alvetex for passaging or fow cytometry.
aMS Biotechnology
For info: +44-1235-828200 www.amsbio.com
electronically submit your new product description or product literature information! 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 availability of products and materials. Endorsement by Science or AAAS of any products or
materials mentioned is not implied. Additional information may be obtained from the manufacturer or supplier.
hiGh-PerforMance SeParation-eSi ModULe
the ceSi 8000 high-Performance Separation-eSi Module with optiMS technology con-
sists of the frst commercial ceSi sprayer combined with new capillary electrophoresis
instrumentation specifcally designed for mass spectrometry (MS). this gives researchers
a robust, high-sensitivity interface of capillary electrophoresis (ce) with MS, while provid-
ing a broad range of analyte coverage complementary to liquid chromatography-MS and
gas chromatography-MS. optiMS technology integrates an intrinsically low fow ce separa-
tion with electrospray ionization (eSi) in a single, dynamic process that takes place within
a capillary. the new process is referred to as ceSi, refecting the combination. ceSi-MS is
capable of generating separations of peptides with a high peak capacity (>320), allowing
the analysis of very complex samples at the proteome level. the ultralow fow rates used
by ce effectively reduce ion suppression and make it possible to detect posttranslational
modifcations that typically hamper effcient ionization. high-separation effciency and sen-
sitivity permit the detection of peptides present at picoMolar concentrations.
Beckman coulter
For info: 714-961-6436 www.beckmancoulter.com
LI FE SC I EN C E TEC HN O LO G I ES
N EW PRO D UC TS: PRO TEO M IC S
A A A S/ Sc ienc e Bu sin e ss O f f ic e Fe a t u re

o
n

A
u
g
u
s
t

2
5
,

2
0
1
1
w
w
w
.
s
c
i
e
n
c
e
m
a
g
.
o
r
g
D
o
w
n
l
o
a
d
e
d

f
r
o
m


Your responsive CROpartner,
deliveringcustomizedsolutions and
adaptability to changing needs.
Customized, Responsive, and On-Time.

MPI Research is the CROthat defines responsiveness, moving your drug development program
forward with customized solutions for all your preclinical research and early clinical support
needs. From discovery services to safety evaluation, including analytical and bioanalytical
support, you can count on MPI Research for quick quotes, frequent updates, rapid turnaround,
and scientific rigor. At every stage, and on every level, we adapt to your most exacting needs.
Explore the breadth of capabilities that make us your responsive CRO at www.MPIResearch.com.
CRM197 from
Corynebacterium
diphtheriae
Reagent
Proteins’
Recombinant
CRM197
Intact Mass
Analysis*
58,407 58,407
LysC peptide
mapping
Identical peptide fingerprint
Polysaccharide
conjugation,
epitope density by
RP-HPLC
10 to 12 11.2
Immunogenicity
of polysaccharide
conjugates
Comparable IgG and OPA (opso-
nization activities) titers
for all serotypes tested
GE Healthcare
Life Sciences
Confidence
comes with
the right
interactions.
Join leading scientists from academia and
industry at Developments in Protein Interaction
Analysis (DiPIA), November 12-15, Boston,
Massachusetts, USA.
This scientic conference oers opportunities
to meet and discuss the latest developments in
label-free interaction and stability analysis using
Biacore™ Surface Plasmon Resonance (SPR) and
MicroCal™ microcalorimetry systems.
Register at www.gelifesciences.com/dipia
© 2011 General Electric Company – All rights reserved.
GE Healthcare Bio-Sciences AB, Björkgatan 30,
751 84 Uppsala, Sweden
K11074. First published May 2011.
Moving science forward
©
2
0
1
1
T
h
e
r
m
o
F
i
s
h
e
r
S
c
i
e
n
t
i
f
i
c
I
n
c
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
C
o
p
y
r
i
g
h
t
s
i
n
a
n
d
t
o
t
h
e
s
c
i
e
n
t
i
s
t
s
i
m
a
g
e
a
r
e
o
w
n
e
d
b
y
a
t
h
i
r
d
p
a
r
t
y
a
n
d
l
i
c
e
n
s
e
d
f
o
r
l
i
m
i
t
e
d
u
s
e
o
n
l
y
t
o
T
h
e
r
m
o
F
i
s
h
e
r
S
c
i
e
n
t
i
f
i
c
b
y
G
e
t
t
y
I
m
a
g
e
s
.
Limited by an outdated PCR system?
You can run 15 minute PCR protocols and use 50% less
consumables and energy by getting away from the old standard.
Choose Thermo Scientific Piko Thermal Cyclers, Phusion reagents,
and ultra thin wall (UTW) plastic consumables for the most
rapid and accurate PCR results. These technologically advanced
components can be used individually to improve your current
workflow, or as a complete solution for advanced PCR.
• Ultra fast PCR cycling protocols optimized to <15 minutes
• 96-well PCR runs using 50% less plastics, reagents and energy
• PCR kits that allow amplification directly from plant, blood,
and tissue samples
Start your PCR experiment today
www.thermoscientific.com/advancedpcr
Everything for PCR
Thermo Scientific PCR portfolio has
everything you need for successful PCR
including industry leading reagents,
high-quality instruments and trusted
plastic consumables.
*Certain terms and conditions apply. To obtain a copy of the full terms and conditions of the guarantee, visit www.appliedbiosystems.com/taqmanguarantee.
For Research Use Only. Not intended for any animal or human therapeutic or diagnostic use. ©2011 Life Technologies Corporation. All rights reserved. The
trademarks mentioned herein are the property of Life Technologies Corporation or their respective owners. TaqMan®is a registered trademark of Roche
Molecular Systems, Inc., used under permission and license. CO10003 0711
There’s no time for troubleshooting or retesting samples. That’s why the TaqMan®family of
products offers 20 times more pre-designed and pre-optimized assays than any other brand.
With multiple sizes and formats developed to help you detect your desired target virtually
100%of the time, you can get the right answers right away. And as always, all TaqMan®
assays carry our TaqMan®Assays QPCR Guarantee.*
www.appliedbiosystems.com/taqman
i get data faster with 20x more assay choices
i amTaqMan
®
Speed through samples with the largest collection of pre-optimized assays
Make no
mistake.
www.NEBPhusion.com
Phusion DNA Polymerase was developed by Finnzymes Oy, now a part of Thermo Fisher Scientific. This product
is manufactured by New England Biolabs, Inc. under agreement with, and under the performance specifications of
Thermo Fisher Scientific. Phusion
®
is a registered trademark and property of Thermo Fisher Scientific.
Phusion
®
DNA Polymerase
from New England Biolabs.
Choosing the right polymerase for PCR can be a challenge.
For high fidelity amplification, turn to Phusion
®
DNA Polymerase,
now manufactured by NEB. Phusion offers robust amplification,
faster cycling times and fewer mistakes in PCR.
Phusion High-Fidelity DNA Polymerase from NEB –
the enzyme that puts fidelity first from the company that puts science first.
Taq Pfu Phusion
50x
6x
1x
F
i
d
e
l
i
t
y
50
40
30
20
10
Fidelity assays were performed using a lacI-based
method modified from Frey & Suppman, 1995.
Experience the extreme fidelity
of Phusion.
Building upon your voice and our vision,
Roche scientists have developed new
reagents that position reliability and
comprehensive efficiency across
common and hard-to-transfect cell lines.
Created to meet the needs of researchers who require
high efficiency, ease of use and reproducibility,
X-tremeGENE reagents provide more viable cells with
more physiologically relevant results. Our newest
transfection reagents deliver results-driven performance,
allowing you to focus your efforts on discovery, and
helping you to shape the future. With Roche, you can
have complete confidence in the technology behind you.
Transfect with confidence.
For life science research only.
Not for use in diagnostic procedures.
X-TREMEGENE is a trademark of Roche.
© 2011 Roche Diagnostics. All rights reserved.
Roche Diagnostics Corporation
Roche Applied Science
Indianapolis, Indiana
Introducing X-tremeGENE DNA Transfection Reagents
Request your free sample today at
x-tremegene.roche.com
The Brain Prize recognizes and rewards outstanding contributions to European neuroscience, frombasic to clinical
Grete Lundbeck European Brain Research Foundation
Call for Nominations for
THE PRIZE OF €1 MILLIONWILL BE AWARDEDINCOPENHAGEN9 MAY2012
Nominations by 15 September 2011
Nominations will be reviewed by the Selection Committee:
YVES AGID FRANCE
HUDAAKIL USA
COLINBLAKEMORE UNITEDKINGDOM CHAIRMAN
FRED. H. GAGE USA
TOMAS HÖKFELT SWEDEN VICE-CHAIRMAN
FLORIANHOLSBOER GERMANY
RANGAR. KRISHNAN SINGAPORE
JES OLESEN DENMARK
FOR THE NOMINATIONFORMANDDETAILS OF THE NOMINATIONPROCEDURE PLEASE VISIT:
WWW.THEBRAINPRIZE.ORG
Prize Winners 2011
Péter Somogyi, Oxford, UK, Tamás Freund, Budapest, Hungary and György Buzsáki, Newark, NJ, USA
THEBRAINPR1ZE
GeneArt
®
gene synthesis, build your
next breakthrough
©2011 Life Technologies Corporation. All rights reserved. The trademarks mentioned herein are the property of Life Technologies Corporation or their
respective owners. For research use only. Not intended for any animal or human therapeutic or diagnostic use. CO18747
Go to www.invitrogen.com/geneart
Experience the newgeneration of gene synthesis and assembly tools
GeneArt
®
solutions give you predictability, adaptability, and confidence. Whether
increasing protein expression rates as much as 100X or assembling genetic constructs
with unprecedented precision, GeneArt
®
gene synthesis and assembly tools enable you
to break through your most complex challenges.
NanoDrop

2000 and 2000C
UV-Vis Spectrophotometers
Measure samples as small
as 0.5 µL up to 2 µL
NanoDrop 3300
Fluorospectrometer
Measure fluorescent samples
as small as 1 µL
NanoDrop 8000
UV-Vis Spectrophotometer
Measure up to 8 microvolume
samples at a time
In science there are always essential steps in your workflow. QA/QC of DNA,
RNA and protein samples is critical for confidence in qPCR, sequencing,
microarray analysis or bioproduction, but there’s a better alternative to the
time and complexity of conventional methods. Using just a minimal sample
and almost no sample prep, the Thermo Scientific NanoDrop and its built-
in methods make QA/QC so incredibly easy, and so much faster, you won’t
notice this step on the way to your ultimate discovery.
Quantify, verify,
simplify
• learn about our free trial program • www.thermoscientific.com/nanodrop
©
2
0
1
1
T
h
e
r
m
o
F
i
s
h
e
r
S
c
i
e
n
t
i

c
I
n
c
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
C
o
p
y
r
i
g
h
t
t
o
t
h
e
p
r
o
t
e
i
n
i
m
a
g
e
a
r
e
o
w
n
e
d
b
y
a
t
h
i
r
d
p
a
r
t
y
a
n
d
l
i
c
e
n
s
e
d
f
o
r
l
i
m
i
t
e
d
u
s
e
o
n
l
y
t
o
T
h
e
r
m
o
F
i
s
h
e
r
S
c
i
e
n
t
i

c
b
y
i
S
t
o
c
k
.
www.sciencemag.org/products 1164
“There are beautiful techniques
like X-ray crystallography that
do a wonderful job of telling us
what the folded state is.”
Armed with powerful algorithms, massive networks of com-
puters, and a battery of new laboratory techniques, biochem-
ists are gradually figuring out how proteins get their shapes.
By Alan Dove
Newton’s equations of motion for understanding how all of
the atoms are moving,” says Pande. Simulating each nano-
second of protein movement this way in a very fast comput-
er, though, would take a day of real time; simulating a com-
plete folding process for one molecule could take millennia.
To get around that, the investigators broke the problem into
thousands of smaller pieces, each simulating a tiny portion
of the folding process. They then created a simple and free
program, called Folding@Home, that anyone could download
and run on a computer or Playstation 3 console. Millions did.
The program obtains a fragment of the team’s current simu-
lation, runs it during the computer’s idle time, then returns
the result to the researchers and gets a new fragment. “In-
stead of something taking a million days on one computer,
it might take 100 days on 10,000 computers, or 10 days on
100,000 computers,” says Pande.
Besides splitting up the problem, the Folding@Home
project also manages it intelligently. Once a batch of results
comes in, the algorithm uses them to adjust the simulation,
focusing on the most likely kinetic pathways for the protein
to take next. Along the way, the experiment is also adjusting
biochemists’ understanding of protein behavior. In particular,
Pande says researchers have come to understand that
“proteins aren’t necessarily just rigid rocks. The dynamics of
the protein itself [are] interesting and important.”
C
R
E
D
I
T
:

R
E
P
R
I
N
T
E
D

W
I
T
H

P
E
R
M
I
S
S
I
O
N

F
R
O
M

V
.

A
.

V
O
E
L
Z
,

G
.

R
.

B
O
W
M
A
N
,

K
.

B
E
A
U
C
H
A
M
P
,

V
.

S
.

P
A
N
D
E
.


M
O
L
E
C
U
L
A
R

S
I
M
U
L
A
T
I
O
N

O
F

A
B

I
N
I
T
I
O

P
R
O
T
E
I
N

F
O
L
D
I
N
G

F
O
R

A

M
I
L
L
I
S
E
C
O
N
D

F
O
L
D
E
R

N
T
L
9
(
1
-
3
9
)
,

J
.

A
M
.

C
H
E
M
.

S
O
C
.
,

1
3
2

1
5
2
6
-
1
5
2
8

(
2
0
1
0
)
.

C
O
P
Y
R
I
G
H
T

2
0
1
0

A
M
E
R
I
C
A
N

C
H
E
M
I
C
A
L

S
O
C
I
E
T
Y
.
UPCOMING FEATURES
qPCR—October 7
Genomics: Gene Expression—November 4
Outsourcing Lab Services—November 25
PROTEIN CHEMISTRY:
GETTING ABOVE THE FOLD
F
ifty years ago, biochemist Christian Anfinsen dena-
tured some pure globular proteins, and observed
that they refolded into their proper three-dimensional
shapes without assistance, demonstrating that the
amino acid sequence alone provides all of the information
necessary to direct the folding of at least certain proteins.
All researchers had to do was figure out the rules for that
process, and they would be able to predict any structure
from the protein’s sequence.
That turned out to be a tough problem. The brute-force
approach, simply testing all possible conformations of the
protein and seeing which one has the lowest energy state,
is a nonstarter. Even a relatively small protein can have an
astronomically large number of conformations; testing each
one, even in a very fast computer, would take longer than
the age of the universe. Meanwhile, traditional biochemical
methods, such as X-ray crystallography and nuclear mag-
netic resonance (NMR), yield definitive results but can take
years to solve a single structure.
Recently, however, the combination of increased comput-
er power, more sophisticated algorithms, and advances in
high throughput structural biology have pushed the bound-
aries of structure prediction steadily forward. Along the
way, scientists have racked up some impressive successes
and have identified a host of potential applications for their
new technologies.
KINETICS IN THE CLOUD
Among computational biologists, protein folders fall into two
camps: Those who want to solve complete structures and
those who want to understand folding kinetics. Vijay Pande,
associate professor of chemistry at Stanford University in
Stanford, California, is in the latter group. “There are beauti-
ful techniques like X-ray crystallography that do a wonderful
job of telling us what the folded state is,” he says. “What is
often challenging is that it’s very difficult to understand what
the intermediate states along that pathway are.”
When Pande and his colleagues started studying folding
kinetics, the field was mainly of academic interest. That
changed as researchers discovered the medical importance
of intermediate folding states. “Nowadays there are many
diseases involving protein misfolding, where it’s believed that
the intermediate states themselves are the toxic elements,”
says Pande. As an example, he cites Alzheimer’s disease,
with its hallmark fibrils of misfolded amyloid proteins.
In order to study folding kinetics, Pande’s team had to
confront a major computational challenge. “A traditional
brute-force approach to go after this is to just run a molec-
ular dynamic simulation, which is basically just integrating
LI FE SC I EN C E TEC HN O LO G I ES
PRO TEO M IC S
A A A S/ Sc ienc e Bu sin e ss O f f ic e Fe a t u re
www.sciencemag.org/products 1165
GAMERS, FOR THE WIN
Pande is not the only researcher using distributed computing
to fold proteins. At the University of Washington in Seattle,
Washington, Professor of Biochemistry David Baker and his
colleagues run the Rosetta@Home project, which also bor-
rows time on millions of personal computers to perform fold-
ing calculations. While Pande is studying protein-folding dynam-
ics, though, Baker’s team focuses on solving complete protein
structures—the final state of the process rather than how
it got there.
Merely having a lot of distributed computing power isn’t
enough, however. As with the folding dynamics problem,
testing all possible folding solutions for a protein by brute
force would simply take too long. Instead, Baker’s algorithm
assigns each participating computer to explore a small set of
the possible solutions. “They sort of start in a random part of
this protein conformational space, and then they look around
and they send us back the lowest energy structure that
they’ve found,” says Baker. Using those results as a guide,
the algorithm chooses new points to test, and proceeds until
it reaches the lowest-energy conformation possible, which
should be the correct structure.
The team has achieved good results with Rosetta for rel-
atively small proteins, but Baker is also trying other tech-
niques to solve larger structures and complete them faster.
For example, the researchers have now developed a new
program called FoldIt, which enlists human volunteers to
assist the computers. The volunteers interact with a video
game-like interface, and “win” by finding the lowest-energy
conformations. Hundreds of players are typically working on
the game at any given time, providing useful guidance for
more complex folding problems.
Besides human intelligence-derived data, the investigators
have also tried including laboratory-based results rather than
making the algorithm start from scratch. “What we’ve been
doing over the last two years is basically taking our protein
folding algorithm and just adding in a small amount of data
on the protein structure. It turns out that small amounts of
data can totally change the ball game so that we can now ac-
tually compute structures accurately,” says Baker. The tech-
nique is particularly useful for proteins that only yield low-
resolution NMR or X-ray data; the computer can take those
rough results and determine a high-resolution structure.
Although both Rosetta@Home and Folding@Home rely
extensively on volunteers and borrowed computing power,
Pande and Baker insist that they are not competing for us-
ers. Indeed, the researchers calculate that there are enough
computers in the world to allow both projects to expand ten-
fold, while both the power and number of available machines
continue to increase.
FLEX TO FIT
Other researchers are also building on the established fold-
ing algorithms to extend them into other areas of protein
chemistry. Ora Schueler-Furman, professor of microbiology
at the Hadassah Medical School of Hebrew University in
Jerusalem, Israel, is one of them.
Schueler-Furman and her colleagues focused specifically
on the problem of proteins binding to each other, where one
or both partners flex into a final structure only on binding.
“You can think of it as a fishing rod with a bait at the end of
it; the bait would be a motif that binds to another protein,
and in order to get there you’d have this flexible fishing rod,
the unstructured region in the protein that allows you to bind
to a partner and then to obtain a completely different func-
tion,” says Schueler-Furman.
Phosphatases, for example, generally lack specificity
for particular phosphates until they bind their regulatory
subunits. These subunits use unstructured regions that
only fold into their final shapes by binding to their asso-
ciated phosphatases. Understanding that final shape could
allow researchers to develop specific phosphatase agonists
or antagonists.
To do that, Schueler-Furman and her colleagues developed
FlexPepDock, an algorithm that runs inside Baker’s Rosetta
system. FlexPepDock starts with a rigid model of the target
protein’s binding site, then explores all possible conforma-
tions for the ligand’s flexible binding domain. Because it’s
sampling only a small region of one protein and looking for a
specific endpoint, the program can run this brute-force simu-
lation in a reasonable time.
There’s a catch, of course. “The major restriction that we
encounter is that we need a good structure of the receptor,”
says Schueler-Furman. For receptors whose structures have
been solved, though, the FlexPepDock algorithm is prov-
ing very useful; researchers around the world have already
solved more than 1,000 binding site structures using the
team’s computing cluster.
FOLD THE PROTEIN, HAL
Not everyone in the algorithmic folding field is using clusters
of computers. At IBM, researchers have been building on
the company’s long history of high-performance mainframe
computing to create a system that can solve folding prob-
lems on a single machine. The project, dubbed Blue Gene,
began in 1999 with a goal of scaling up supercomputer pow-
er by a thousandfold.
While such an advance could have numerous applications,
the Blue Gene team identified protein folding as one of their
main goals. “Almost immediately we realized that Blue Gene
was not going to be a single-purpose machine; its best use
would be to provide some general capabilities that would
allow not just one simulation application but as many as
possible,” says Ajay Royyuru, Ph.D., senior manager for
IBM’s Computational Biology Center in Yorktown Heights,
New York. continued
»
LI FE SC I EN C E TEC HN O LO G I ES
PRO TEO M IC S
A A A S/ Sc ienc e Bu sin e ss O f f ic e Fe a t u re
FEATURED PARTICIPANTS
Brandeis University
www.brandeis.edu
Hebrew University of
Jerusalem Faculty of
Medicine
medicine.huji.ac.il/
Default.aspx
IBM’s Computational
Biology Center
www.research.ibm.com/
bioinformatics
Stanford University
www.stanford.edu
U.S. National Institutes
of Health
www.nih.gov
University of
Washington
www.washington.edu
www.sciencemag.org/products 1166
In the project’s first decade, the team has built two suc-
cessive generations of machines, dubbed Blue Gene L and
Blue Gene P, and more than accomplished the thousandfold
speed increase they sought initially. Now, Royyuru and his
colleagues are working on the next model, Blue Gene Q,
which will exceed the P model’s 3 petaflop top speed.
“That change in computing capability has allowed us to in-
vestigate biomolecular phenomena that were not accessible
to simulation before, that then became a tool for hypoth-
esis generation,” says Royyuru. As an example, he points to
simulations of rhodopsin folding over the course of a few mi-
croseconds, as the protein goes from dark-adapted to light-
adapted conformations.
The simulation revealed surprisingly large changes that ex-
perimenters were then able to check in the lab. “That was
an example where the hypothesis generated through simula-
tion led to validation through experiments,” says Royyuru,
adding that “we’re stepping into a new realm of science,
where in the absence of simulation you wouldn’t have a par-
ticular hypothesis.”
Royyuru says that Blue Gene’s supercomputing approach is
complementary to distributed computing strategies. Indeed,
IBM’s World Community Grid project also pursues complex
simulations, but not the same ones the researchers run on
Blue Gene. “The science problems that you can tackle are
different with different styles of computing,” he explains.
Because distributed computing systems rely on breaking a
problem into many small parts, they are ideal for searching
large numbers of possible protein conformations. The com-
puters do not need to communicate anything except their
final answers. In contrast, protein dynamic simulations, such
as the rhodopsin experiment, require constant interchanges
of data through the system and mandate a highly connected
architecture such as Blue Gene.
BIG SCIENCE AND ITS DISSENTERS
Regardless of the specific architecture they use, algorithmic
folding researchers generally agree that their approach will
never replace traditional biochemical methods. In order to DOI: 10.1126/science.opms.p1100057
Alan Dove is a science writer and editor based in Massachusetts.
improve those methods, researchers at the U.S. National
Institutes of Health (NIH) established the Protein Structure
Initiative (PSI) in 2000. Since then, the Initiative has solved
several technical problems, but has also drawn some out-
spoken critics.
Initially, the $764 million, 10-year project set out to stream-
line protein structure determination methods and solve as
many structures as possible. By 2008, PSI researchers had
solved more than 3,400 structures using NMR and X-ray
crystallography. In 2010, the Initiative began a new phase,
called PSI-3, extending the effort by another five years and
$290 million. PSI-3 focuses heavily on collaborative projects,
membrane protein structures, and studying the biological
context of protein folding.
“There are a lot of very difficult problems that are now
trying to be addressed by these methods, in particular the
structure of integral membrane proteins, which has been a
very difficult scientific endeavor. The hope is that the meth-
ods that we’ve developed can make a difference in improving
the efficiency and speed with which membrane proteins are
determined, but we won’t know that until towards the end
of the program,” says Ward Smith, Ph.D., director of the PSI.
Not everyone is happy with the Initiative’s open-ended
approach, though. “It was stupid from the beginning, and
they keep changing their goal depending on how urgently
they think they need to get funding,” says Gregory Petsko,
Ph.D., professor of biochemistry at Brandeis University in
Waltham, Massachusetts.
Petsko also objects to the high cost of the PSI. To illustrate
that, he points to the $290 million cost of the project’s cur-
rent phase. That’s equivalent to over four hundred investiga-
tor-initiated grants, which Petsko says would have produced
many more useful results than the PSI.
Smith defends the Initiative’s track record. “This really sort
of boils down to a philosophical difference, whether you be-
lieve that discovery science programs have a place in the
academic research environment. Greg’s argument is that the
single investigator with an [individual] R0-1 grant is able to
accomplish more than a large center, but I think you’ll find
that the technologies that have developed in the last 10 years
of PSI-1 and PSI-2 have benefited all of the research com-
munity,” says Smith.
Smith cites a plethora of technological developments listed
on the PSI website as well as commercial products that have
capitalized on the Initiative’s findings. For example, compa-
nies such as Emerald Biosystems, Hamilton Robotics, and
others now offer automated and semiautomated worksta-
tions for high throughput crystallization, based partly on the
PSI’s work.
While they may not agree on which method is currently
working best, researchers in the protein folding field are gen-
erally optimistic about their progress. “Some people have
been looking for the day when the protein folding problem
is solved. It might never be quite that stark,” says Pande.
He adds that “it is this sort of gradual shift, of maybe in the
next five or 10 years getting to the point where we can ad-
dress really complicated problems that no one would have
dreamed of 10 years ago.”
C
R
E
D
I
T
:

F
R
O
M

R
A
V
E
H

B
,

L
O
N
D
O
N

N
,

Z
I
M
M
E
R
M
A
N

L
,

S
C
H
U
E
L
E
R
-
F
U
R
M
A
N

O

(
2
0
1
1
)

R
O
S
E
T
T
A

F
L
E
X
P
E
P
D
O
C
K

A
B
-
I
N
I
T
I
O
:

S
I
M
U
L
T
A
N
E
O
U
S

F
O
L
D
I
N
G
,

D
O
C
K
I
N
G

A
N
D

R
E
F
I
N
E
M
E
N
T

O
F

P
E
P
T
I
D
E
S

O
N
T
O

T
H
E
I
R

R
E
C
E
P
T
O
R
S
.

P
L
O
S

O
N
E

6
(
4
)
:

E
1
8
9
3
4
.

D
O
I
:
1
0
.
1
3
7
1
/
J
O
U
R
N
A
L
.
P
O
N
E
.
0
0
1
8
9
3
4
.

©

2
0
1
1

R
A
V
E
H

E
T

A
L
.
LI FE SC I EN C E TEC HN O LO G I ES
PRO TEO M IC S
A A A S/ Sc ienc e Bu sin e ss O f f ic e Fe a t u re
“We’re stepping into a new realm of science,
where in the absence of simulation you
wouldn’t have a particular hypothesis.”
www.sciencemag.org/products 1167
PROTEIN AND PEPTIDE MICROARRAY
The ProArray Ultra is a uniquely powerful protein and peptide array
platform, which is more sensitive than optimized ECL ELISA. The
new platform allows analysis of more than 30,000 peptides or pro-
teins at once, and typically requires only 1µl of analyte sample. With
its sensitivity and flexibility, ProArray Ultra overcomes the traditional
trade off between optimized 96-well plate-based assays, which are
reliable and have good performance characteristics but lack the re-
quired throughput, and high throughput array technologies, which
are typically only semiquantitative. Analysis with ProArray Ultra is
provided by ProImmune to customers as a turnkey service and will
enable a broad new range of research programs that are in many
cases simply not feasible with existing approaches. Applications in-
clude antigen and monoclonal antibody characterization, analysis of
serum antibody specificity, correlation of antibody responses with
disease onset, progression and outcome, protein-protein interaction
site mapping, iterative peptide and protein design, and general bio-
assay design.
ProImmune
For info: 888-505-7765 www.proimmune.com
CYTOSKELETON AND ASSOCIATED
PROTEIN ENRICHMENT
Two new ProteoExtract Kits enable researchers to more rapidly and
effectively enrich, isolate, and analyze the cytoskeleton and associ-
ated proteins. The ProteoExtract Cytoskeleton Enrichment and Iso-
lation kit provides the necessary extraction buffers to quickly and
selectively remove soluble cytoplasmic and nuclear protein fractions
for downstream biochemical analysis including Western blot and
mass spectrometry. The assay is performed in as little as 20 minutes
and foregoes the need for harsh buffers and centrifugation, thus
retaining more of the native focal adhesion- and actin-associated
proteins, as compared to traditional protein fractionation methods.
The ProteoExtract Native Cytoskeleton Enrichment and Staining Kit
includes the necessary reagents to efficiently enrich, retain, and
visualize the cytoskeleton and associated proteins in their native,
adhered conformations with the same gentle technique. The kit
also includes labeled phalloidin and antibodies against vinculin and
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as markers
for visualization and colocalization of the actin cytoskeleton, focal
adhesions, and cytosol for immunofluorescence analysis.
EMD Millipore
For info: 800-645-5476 www.millipore.com
MICROWAVE PEPTIDE SYNTHESIZER
The Biotage Initiator+ SP Wave is a versatile synthesis solution as
it can be used as a stand-alone peptide synthesizer or easily con-
verted to a high specification microwave assisted organic synthe-
sis (MAOS) system. It is the ideal development tool for chemists
synthesizing peptides, peptidomimetics, and small molecules using
single or multistep procedures. The Initiator+ SP Wave is perfectly
suited for the small scale synthesis of peptides and peptidomimet-
ics including peptide nucleic acid using microwave irradiation, from
a scale as low as 5 µmol. The semiautomated operation can reduce
the consumption and waste of expensive reagents that occur in oth-
er systems due to priming or use of large dead volumes. The system
includes a large 10-inch touchscreen, intuitive software, pre- or self-
defined methods, inert gas capability, vortex, and magnetic stirring.
Biotage
For info: 800-446-4752 www.biotage.com
3D CELL CULTURE SCAFFOLD
Alvetex is a unique cell culture scaffold that enables genuine three-
dimensional (3D) cell culture to be grown routinely and cost-effec-
tively. Alvetex provides a stable and inert scaffold for cells to grow,
mimicking in vivo growth and formation of tissues in the body. Alve-
tex has a highly porous architecture that is specifically tailored for 3D
cell culture and designed to be compatible with existing cell culture
techniques and post cell culture analysis. Extensively tested and vali-
dated with a large number of cell types, Alvetex enables the forma-
tion of complex 3D cellular structures, increases viability, decreases
cell death, and enhances cell function. Using this unique material,
cells can be visualized by confocal or live-cell imaging or treated like
tissue. With Alvetex, it’s easy to process cells for histological stain-
ing, immunocytochemistry, and electron microscopy. Viable cells can
also be isolated by Alvetex for passaging or flow cytometry.
AMS Biotechnology
For info: +44-1235-828200 www.amsbio.com
Electronically submit your new product description or product literature information! 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 availability of products and materials. Endorsement by Science or AAAS of any products or
materials mentioned is not implied. Additional information may be obtained from the manufacturer or supplier.
HIGH-PERFORMANCE SEPARATION-ESI MODULE
The CESI 8000 High-Performance Separation-ESI Module with OptiMS Technology con-
sists of the first commercial CESI sprayer combined with new capillary electrophoresis
instrumentation specifically designed for mass spectrometry (MS). This gives researchers
a robust, high-sensitivity interface of capillary electrophoresis (CE) with MS, while provid-
ing a broad range of analyte coverage complementary to liquid chromatography-MS and
gas chromatography-MS. OptiMS Technology integrates an intrinsically low flow CE separa-
tion with electrospray ionization (ESI) in a single, dynamic process that takes place within
a capillary. The new process is referred to as CESI, reflecting the combination. CESI-MS is
capable of generating separations of peptides with a high peak capacity (>320), allowing
the analysis of very complex samples at the proteome level. The ultralow flow rates used
by CE effectively reduce ion suppression and make it possible to detect posttranslational
modifications that typically hamper efficient ionization. High-separation efficiency and sen-
sitivity permit the detection of peptides present at picoMolar concentrations.
Beckman Coulter
For info: 714-961-6436 www.beckmancoulter.com
LI FE SC I EN C E TEC HN O LO G I ES
N EW PRO D UC TS: PRO TEO M IC S
A A A S/ Sc ienc e Bu sin e ss O f f ic e Fe a t u re
September 26–27, 2011 | Sheraton Boston Hotel, Boston, MA
Register Now!
Register now for Screening Stem Cells 2011: From Reprogramming to Regenerative Medicine, September 26–27, Boston, MA.
Hosted by the Society for Laboratory Automation and Screening (SLAS) in collaboration with ve world-renowned stem cell
organizations, Screening Stem Cells 2011 features a scientic program comprising 18 podium speakers and thought-provoking
poster sessions. There will also be an exhibition showcasing leading edge technologies. Screening Stem Cells 2011 offers an
in-depth exploration of this emerging eld of drug discovery research.
Program Sessions Include:
In Collaboration With:
For a complete list of the
distinguished speaker
line-up and to register,
visit SLAS.org.
•Reprogramming
Session Chair: Chad Cowan, Ph.D.,
Harvard Medical School
•Business Models for Success
Session Chair: Brock Reeve, Ph.D., HSCI
•Differentiation
Session Chair: Deepak Srivastava, M.D.,
Gladstone/UCSF
•Cell-Based Therapeutics
Session Chair: Hans Keirstead, Ph.D., UC Irvine
Lambda XL
Extended Life Light Source
FEATURES
º 10,000 |ou| e·pec¦ed ||¦e
º F|ç||, s¦ab|e
º No ||ç|·.o|¦açe pu|se
º No a||çureu¦ uecessa|,
º bu||¦·|u d||.e| ¦o| op¦|oua| ¦|¦e| w|ee| aud s|u¦¦e|
º /dap¦ab|e ¦o ros¦ r|c|oscopes
lFCNE. 415.883.0128 | f/\. 415.883.0572 | EV/lL. lNfC@'u!!ER.CCV | www.'u!!ER.CCV
!|e Larbda \L |s a b|oad
spec¦|ur, ||ç||, s¦ab|e ||ç|¦
sou|ce (±1% pea|·¦o·pea|
¦uc¦ua¦|ous) w|¦| au
e·pec¦ed |arp ||¦e o¦
10,000 |ou|s. !|e ||ç|¦
|u¦eus|¦, cau be adjus¦ed
¦o d|¦¦e|eu¦ |e.e|s o¦
a¦¦euua¦|ou aud ¦|e
||qu|d ||ç|¦ çu|de
couuec¦|ou assu|es
ou¦pu¦ uu|¦o|r|¦,
|u ¦|e ¦e|d o¦ .|ew.
AAAS is here –
preparing minority students for careers in science.
Part of AAAS’s mission is to strengthen and diversify
the scientific work force. To help achieve this goal, AAAS
partners with NSF to present the Historically Black Colleges
and Universities Undergraduate Program, a conference
where students from HBCUs get experience presenting
their research, networking with peers, meeting with repre-
sentatives from graduate schools, and learning about career
opportunities. As a AAAS member, your dues support these
efforts. If you’re not yet a member, join us. Together we can
make a difference.
To learn more, visit aaas.org/plusyou/hbcuup
Orders (toll-free) 1-877-616-2355
|
Technical support (toll-free) 1-877-678-8324 support@cellsignal.com
|
Inquiries info@cellsignal.com
|
Environmental Commitment eco.cellsignal.com
www.cellsignal.com
for quality products you can trust...
©
2
0
1
1
C
e
l
l
S
i
g
n
a
l
i
n
g
T
e
c
h
n
o
l
o
g
y
,
I
n
c
.
C
e
l
l
S
i
g
n
a
l
i
n
g
T
e
c
h
n
o
l
o
g
y
®
a
n
d
X
P
®
a
r
e
t
r
a
d
e
m
a
r
k
s
o
f
C
e
l
l
S
i
g
n
a
l
i
n
g
T
e
c
h
n
o
l
o
g
y
,
I
n
c
.
Select antibodies from these
pathways and protein classes
Chromatin Regulation /
Epigenetics
MAP Kinase Signaling
Apoptosis /Autophagy
PI3K/Akt Signaling
Translational Control
Calcium, cAMP &
Lipid Signaling
Cell Cycle /Checkpoint
DNA Damage
Jak /Stat Pathway
NF-kB Signaling
TGF-β /Smad Signaling
Lymphocyte Signaling
Neuroscience
Angiogenesis
Vesicle Trafficking
Cytoskeletal Signaling
Adhesion
Glucose Metabolism
Wnt /Hedgehog/Notch
Protein Folding
Protein Stability
Nuclear Receptors
Transcription Factors
Adaptor /Scaffold
Phosphatases
Stem Cell Markers
Build Your Own
Antibody Sampler Kit
Unparalleled product quality, validation,
and technical support.
Custom Antibody Sampler Kits are now available from
Cell Signaling Technology. Start with a minimum purchase
of five sample size vials of your choice, and get further
discounts with 10 and 15 vials purchased.
:: Each vial contains enough primary antibody
to perform four western blots per target
:: Select 5 or more antibodies specific to your research
needs from over 700 sample size products
:: Antibody selection includes many of our
highest quality antibodies, including
our premium XP
®
monoclonal antibodies
For a complete list of available sample size products, visit
www.cellsignal.com/custom-sampler-kit This offer is valid within the U.S. only. Outside of the U.S.,
please contact your local representative for related offers.
Antibodies and Related Reagents for Signal Transduction Research
Stony Brook University’s Department of Pharmacolog-
ical Sciences is seeking a Postdoctoral Associate
(PDA). The position is available for those holding
Ph.D./D.Sc. or equivalent in the biological sciences
and experience in cell signaling analyses (required)
with up to two years’ prior postdoctoral training
(preferred). The PDA will be part of a team interrogat-
ing the A-Kinase-Anchoring Protein (AKAP) family.
AKAPs are multivalent molecular scaffolds that dock
protein kinases, phosphoprotein phosphatases,
GPCRs, and adaptor molecules. The work will
focus on AKAP5 and AKAP12, which are intimately
involved in cell signaling events. The PDA will make
use of autofluorescently tagged AKAPs to examine in
live cells, dynamic changes in response to activation
and estimates of MW (by fcs). The PDA will make
use of fPLC, radioligand binding and proteomic
means to identify/characterize novel signaling
molecules that dock to AKAPs. The PDA will employ
novel biosensor AKAPs that can report on the docking
of signaling molecules and phosphorylation of
the AKAP in response to activation. Knockdown
(siRNA) and rescue with mutant AKAPs will
be employed.
For a full position description, application procedures,
or to apply online, visit www.stonybrook.edu/jobs
(Job Ref. #: WC-R-6888-11-07-S) or send cover letter
and C.V. to: Dr. Craig C. Malbon, Leading Professor
and Ethicist, Department of Pharmacological
Sciences, Basic Sciences Tower, T-7, Room 162,
Stony Brook University, Stony Brook, NY
11794-8651
Stony Brook University/SUNY
is an equal opportunity,
affirmative action employer.
Postdoctoral Associate
Trudeau Institute, Inc. has openings for Ph.D., M.D. or D.V.MPostdoctoral Fellows in the following
laboratories.
FACULTY RESEARCH INTERESTS
Andrea Cooper, Ph.D. – Immunopathogenesis of mycobacterial disease
Laura Haynes, Ph.D. – The effect of age on the function of CD4 T cells
Mithilesh Jha, Ph.D. – Identification and characterization of calciumchannels and their complexes
in cells of the Immune system; Roles of adrenergic receptor signaling in the immune system
Elizabeth Leadbetter, PhD. – Cooperative iNKTand Bcell responses to lipid antigens; Lipid-based
vaccine strategies for infectious diseases
Markus Mohrs, Ph.D. – Cellular and molecular mechanisms governing cytokine responses to
infection
Stephen Smiley, Ph.D. – T cell responses to bacteria; Coagulation during immunity/
immunopathology.
Alexei Tumanov, M.D., Ph.D. – Mucosal immunology, liver diseases
Trudeau Institute, Inc. is located in northern New York State’s Adirondack Mountains and offers
competitive salaries, a full complement of benefits and on-site childcare and housing. Further details
may be found at www.trudeauinstitute.org.
Please send a CV, cover letter, a statement of research interests and three references (with e-mail
addresses included) to Amy Richardson, HR Manager, Trudeau Institute, Inc., 154 Algonquin
Avenue, Saranac Lake, NY 12983 or arichardson@trudeauinstitute.org.
If interested in interviewing with a particular lab, please specify in the cover letter.
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
p
o
s
t
d
o
c
t
o
r
a
l

o
p
p
o
r
t
u
n
i
t
i
e
s
Tracy Holmes
Worldwide Associate Director
Science Careers
Phone: +44 (0) 1223 326525
UNITED STATES & CANADA
E-mail: advertise@sciencecareers.org
Fax: 202-289-6742
Tina Burks
Midwest/West Coast/
South Central/Canada
Phone: 202-326-6577
Elizabeth Early
East Coast & Industry
Phone: 202-326-6578
Marci Gallun
Sales Administrator
Phone: 202-326-6582
Online Job Posting Questions
Phone: 202-312-6375
EUROPE & REST OF WORLD
E-mail: ads@science-int.co.uk
Fax: +44 (0) 1223 326532
Alex Palmer
Phone: +44 (0) 1223 326527
Susanne Kharraz
Phone: +44 (0) 1223 326529
Lisa Patterson
Phone: +44 (0) 1223 326528
JAPAN
ASCA Corporation
Phone: +81-3-6802-4616
Fax: +81-3-6802-4615
E-mail: mhara@aaas.org
CHINA &TAIWAN
Ruolei Wu
Phone: +86-1367-1015-294
E-mail: rwu@aaas.org
For full advertising details, go to
ScienceCareers.org and click For Employers,
or call one of our representatives.
All ads submitted for publication must comply
with applicable U.S. and non-U.S. 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 encour-
ages our readers to alert us to any ads that
they feel may be discriminatory or offensive.
S
c
ie
n
c
e
C
a
r
e
e
r
s
A
d
v
e
r
t
is
in
g
Roche, Switzerland
“Make your mark.
improve lives.”
Roche post-doctoral Fellowship Program
Who we are
At Roche, 80,000 people across 150 countries are pushing back the frontiers of healthcare. Working together,
we’ve become one of the world’s leading research-focused healthcare groups. Our success is built on inno-
vation, curiosity and diversity, and on seeing each other’s differences as an advantage. To innovate healthcare,
Roche has ambitious plans to keep learning and growing – and is seeking people who have the same goals for
themselves.
The headquarters in Basel is one of Roche’s largest sites, over 8,000 people from approximately 80 countries
work at Roche Basel. Favored by its geographic location in the heart of Europe, the Basel area is one of the
most dynamic economic regions in Switzerland – a great place to live and work.
The Roche post-doctoral Fellowship Program (RPF)
It is our ambition to build and maintain «Scientific Leadership» for the creation of innovative healthcare products
and solutions, now and in the future.
The RPF Program will foster creative science, engage talented scientists as postdocs, reward Roche‘s best scien-
tists as mentors and strengthen and expand scientific relationship with academia. We mean to
• Foster creative science as a basis for innovation. Seek and develop new scientific knowledge;
exploit emerging science, technologies and methods.
• Collaborate with Academia Research Groups, partner with them for a defined period of time on a
RPF R&D project
• Develop scientific talents with a Roche Postdoc Fellowship for 2 to 4 years. Postdocs are mentored both
by an excellent scientist at Roche and an excellent scientist in the Academic Research Group.
• Aim to publish scientific articles in top rated, peer reviewed journals, if necessary preceeded by
patent priority applications.
Current Positions
• Post-doctoral Scientist Modelling & Simulation (Job ID 378345)
• Post-doctoral Scientist Modelling & Simulation within Clinical Research (Job ID 378466)
• Post-doctoral Scientist Mass-Spectrometry (Job ID 378487)
• Post-doctoral Scientist Mass-Spectrometry within Oncology Research (Job ID 378491)
• Post-doctoral Scientist within cardiovascular/inflammatory biology (Job ID 378488)
• Post-doctoral Scientist Position in Neuroimaging Research (Job ID 372374)
Contact HR: W. Kinzy, Phone: + 41 61 687 94 03
The next step is yours. To apply online for this position visit www.careers.roche.ch
Roche is an Equal Opportunity Employer
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
p
o
s
t
d
o
c
t
o
r
a
l

o
p
p
o
r
t
u
n
i
t
i
e
s
www.sciencecareers.org 1170
continued on page 1172
»
C
R
E
D
I
T
:

(
P
H
O
T
O

I
L
L
U
S
T
R
A
T
I
O
N
)

S
C
I
E
N
C
E
;

(
I
M
A
G
E
S
)

©

I
S
T
O
C
K
P
H
O
T
O
.
C
O
M
UPCOMING FEATURES
Focus on Japan—September 2
Faculty: Financial Planning for Scientists—September 9
Top Employers Survey—October 7
S
ince Science Ca-
reers started con-
ducti ng annual
sur veys seven
years ago, alter-
nating between polling post-
docs and postdoc advisors,
the attributes that survey
respondents select as being
most important to a success-
ful postdoc have not varied
much.
This year’s record number
of survey participants, com-
prising 798 postdoc supervi-
sors, ranked the importance
of twelve attributes to having
a successful postdoc experi-
ence. The attributes supervi-
sors ranked most highly, and
the ones selected by the post-
docs polled in 2010, are listed
in the chart on the right.
Only 37 percent of supervisors thought helping postdocs’
spouses and partners find a job was important or very impor-
tant to a successful postdoc experience; in comparison, 86 per-
cent of postdocs felt that way (see “The Postdoc Experience:
Taking A Long Term View,” Science, 2010, doi:10.1126/science.
opms.r1000093). The “spouses and partners” category has been
rising in ranks among postdocs in their surveys. For example, in
the 2006 survey only 16 percent of postdocs polled noted help for
spouses or partners to find jobs as a key factor. This increase may
reflect a change in the value that postdocs, but not their supervi-
sors, place on their personal lives relative to their work.
That was the case for Simona Casarosa, currently an assis-
tant professor at the University of Trento, Italy. After complet-
ing her Ph.D. in Italy, she accepted a position as a postdoc in
Strasburg, France. Her choice was dictated by the fact that
her husband, also a scientist, had finished his Ph.D. months
earlier and had already landed a post in France. Casarosa says
the choice of a postdoc lab depends on what is important
to someone. “You have to ask yourself what you want from
your life. If you want to have a striking career and be famous
then you should choose on that basis,” she says. “But if your
personal life is important to you, then you need to take that
into consideration.”
Traditionally postdocs have sought positions in the scientific
powerhouses of the US, UK, and Germany. But postdocs who
are looking for something different and a chance to see faraway
places might want to consider nontraditional choices like Australia
or Singapore.
POSTDOCS: STRIVING FOR SUCCESS
IN A TOUGH ECONOMY
SURVEY
METHODOLOGY
This year’s survey was launched on
March 15, 2011, with e-mail invi-
tations sent out to about 40,000
current and former postdoc ad-
visors worldwide. Of the 798
completed surveys that were
collected, 71 percent came
from Europe (39 percent) and
North America (32 percent).
The remaining respondents
were located in Asia/Austra-
lia/Pacific Rim (20 percent)
or other areas of the world (9
percent). Most were males (72
percent) 40 years of age and
older (76 percent) and worked in
academic institutions (70 percent)
and government organizations (13
percent). The primary area represented
was the life sciences (57 percent).
One thing that’s on everyone’s mind these days—scientists included—is the dire economic situation. But despite tighter
budgets, the attributes contributing to a successful postdoc experience, such as communication and mentoring, have not
changed, according to the postdoc supervisors who completed this year’s annual survey for Science Careers. Yet supervisors
say it’s all the more important for postdocs to carefully plan their career moves ahead of time and make sure that they get all
the training they need during their postdoc years. By Laura Bonetta
Supervisors Rate Factors That Contribute to a Successful Postdoc
Experience, Plus Rankings Compared to 2010 Postdoc Survey
Supervisor
Rankings
2011
1
2
3
4
5
6
7
8
9
11
12
10
6 (tie)
3 (tie)
n/a
9
1 (tie)
3 (tie)
1 (tie)
n/a
6 (tie)
10
8
3 (tie)
Attribute
Very
Important
Important
39%
39%
39%
46%
47%
51%
56%
50%
28% 9%
17%
52%
50%
43% 46%
39% 48%
48%
42%
44%
37%
30%
42%
25%
Postdoc
Rankings
2010
FO C US O N C A REERS
PO STD O C SURV EY
AAAS/ Sc ienc e Business Offic e Feat ure
Announcing the Chicago Fellows
Postdoctoral Program.
Our faculty is dedicated to one powerful goal: to train the
next generation of researchers and help them launch their
careers. Apply for a coveted fellowship in the biological
and biomedical sciences.
Fellows receive:
• Dedicated research funds
• Mentorship at laboratory, departmental and University levels
• Highly competitive compensation
Eligible fields of study include Cancer Biology, Evolutionary
Biology, Neurosciences, Systems Biology, Genetics,
Molecular Biology, Biochemistry and Translational Studies.
Application deadlines are Mar. 1, Jul. 1, and Nov. 1.
To apply, go to http://chicagofellows.bsd.uchicago.edu.
THE FUTURE
OF BIOMEDICAL
RESEARCH
STARTS HERE.
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
p
o
s
t
d
o
c
t
o
r
a
l

o
p
p
o
r
t
u
n
i
t
i
e
s
University of Pittsburgh
Schools of the Health Sciences
Premier Postdoctoral Opportunities in
Biomedical Research
Join a world-class center for pioneering biomedical research.
Postdocs are vital members of our institutional culture,
valued and respected for their signicant contributions to the
university’s research enterprise. A customized high quality
and enriching postdoctoral training experience is assured
within our multicultural and multidisciplinary academic
health sciences community.
Training opportunities are available in the Schools of
Medicine, Dental Medicine, Health and Rehabilitation
Sciences, Nursing, Pharmacy, and the Graduate School of
Public Health.
e university is part of the fabric of Pittsburgh, a vibrant
and aordable city that oers a multitude of local amenities
including major league sports, a dynamic cultural district,
and old-world neighborhood charm. Rated “Most Livable
City” in the US by Forbes Magazine and e Economist, as
well as twice holding the top position on the Places Rated
Almanac!
Pittsburgh is a great place to live and work. e University
of Pittsburgh ranked among the top 10 universities on the
“Best Places to Work Postdocs 2011” survey by e Scientist
magazine.
Maximize Your Postdoctoral Experience!
• A community of 675 postdocs in the health sciences
• Among the top 10 academic institutions in NIHFunding
• State-of-the-art research facilities and equipment
• Competitive compensation and benets
• Strong collaborative relationships
• Training grant opportunities
• Career development programs and services
• Annual Postdoc Career and Research Symposia
• Active postdoctoral association and social events
Visit our website at
www.postdocjobs.hs.pitt.edu
to view open postdoctoral positions.
For additional information about career development services
and programs available to biomedical scientists and clinical
researchers, visit the Oce of Academic Career Development
at www.oacd.health.pitt.edu.
www.sciencecareers.org 1172
After hearing rumors in Australia that there would be Aus$400
million cut from the National Health and Medical Research Coun-
cil budget, researchers mounted a vigorous protest campaign that
was successful in maintaining current spending for the major re-
search agencies. “Right now success rates for grant applications
are about 23 percent, which is not bad in an international context,”
says Joan Heath, a zebrafish researcher at the Parkville Branch of
the Ludwig Institute for Cancer Research in Melbourne, Australia.
“We have just escaped a difficult situation.” (In comparison, the
success rate for NIH grants in 2010 was 20.6 percent; see: http://
report.nih.gov/award/success/Success_ByIC.cfm).
Most postdocs in Australia are appointed on principal investiga-
tor (PI) grants, typically at the time that the grant starts until its end
in three years, explains Heath. All of her postdocs have salaries
of at least Aus$70,000 (US$74,500) a year, which is higher than
typical postdoc salaries in the US. “I don’t know why our doors are
not being beaten down.” Yet Australia is not a popular destination
among American postdocs.
Another good place for postdocs may be Singapore. “It’s very
nice for postdocs who want to explore another part of the world,”
says Philipp Kaldis, a PI at the Institute of Molecular and Cell
Biology (IMCB). “Public transportation is very well organized, the
weather is always nice, it’s very close to many exotic vacation
spots, research is well supported by the government, and salaries
for postdocs are more than what they are in the United States.” On
the downside, apartments are very expensive and going home or
to a conference can be a long trip.
THE VALUE OF MENTORS
Year after year, most supervisor respondents in the Science Careers
survey select mentoring as one of the key attributes for a success-
ful postdoc experience. In this year’s survey, two thirds (65 per-
cent) of postdoc supervisors indicated that they spent 20 percent
or less of their professional time on supervisory responsibilities
(see table above). The remaining one third (34 percent) spent more
than 20 percent of their time supervising postdocs. Most (80 per-
cent) supervisors felt that the time they spent advising postdocs
was adequate.
“Mentoring is really important to me,” says Gail Bishop, pro-
fessor of microbiology and internal medicine at The University of
Iowa. “I belong to the subgroup of PIs who put a lot into mentor-
ing.” Bishop, who indicated in the survey that she spends over
20 percent of her time mentoring trainees, acknowledges that
mentoring is one of those things that a PI typically learns on the
job. “We are all selected for our productivity, creative ability at the
bench, and our ability to communicate scientifically,” she explains.
“Then instantly you become a manager and often have no experi-
ence doing it.” A key to learning how to manage people, she says,
is to start gradually with only a few people in the lab. “One pitfall
for many beginning faculty is building up a lab too fast,” she adds.
The number of postdocs a PI feels they can effectively manage
changes depending upon the level of their experience, according to
survey respondents. When they were asked about the maximum
number of postdocs they could manage, very few (3 percent) felt
they should be responsible for only one postdoc, whereas most
(63 percent) felt that they could supervise two or three postdocs.
Seventeen percent of the survey respondents felt that they could
supervise four or more postdocs. The remaining 17 percent were
not sure about the maximum number of postdocs they should su-
pervise. In general, as the age of a supervisor increased, so did the
maximum number of postdocs that the supervisor thought he/she
should be responsible for.
Interestingly, the 2011 survey revealed that female supervisors,
on average, spend more time mentoring postdocs than male
supervisors. Mentoring may come more easily to female faculty,
says Heath, who has been the postgraduate student coordinator
at her institute since 1995. She doesn’t think many of her male
colleagues would spend as much time as she does mentoring
other people’s graduate students. “We have 20 to 25 students at
our institute, and I take a personal interest in each one of them.
I sit down with them and mentor their progress and careers,”
she explains.
That is not, however, to say that men don’t mentor. “I like men-
toring,” says Michael Stumpf, a systems biologist at Imperial Col-
lege London in the UK, who also indicated in the survey he spends
over 20 percent of his time mentoring postdocs. Stumpf says he
tries to get all his postdocs to supervise Ph.D. or Master’s stu-
dents “so that they see if they like it,” he says. “It will be an impor-
tant component of what they do as PIs.” He also makes sure they
present their research findings at scientific conferences. Because
Stumpf holds a joint grant in collaboration with a group in Japan, he
also sends all his graduate students and postdocs to spend time in
his collaborator’s lab in Japan. “It is important to be exposed to a
different way of doing science,” says Stumpf.
ENSURING SUCCESS
Each year the Science Careers survey asks participants to rank
those factors, such as learning new skills or publishing papers,
which contribute to a successful scientific career. The top-rated
factors have stayed the same over the years, regardless of eco-
nomic conditions.
This year’s survey respondents felt that conducting high quality
research was most important for a successful postdoc (79 per-
cent), followed by learning to work independently (65 percent),
and publishing work (60 percent). Significantly fewer supervisors
felt that learning how to write grants and obtain funding (36 per-
cent), developing new research skills (35 percent), gaining a deep
knowledge of a specific area of research (32 percent), and learning
to manage or supervise others (23%) were among the top attri-
butes important for a successful postdoc experience.
FO C US O N C A REERS
PO STD O C SURV EY
AAAS/ Sc ienc e Business Offic e Feat ure
continued
»
% OF TIME
SPENT
% OF
TOTAL
% OF
MALES
% OF
FEMALES
20% or less 65% 69% 57%
21 % or
more
34% 31% 43%
Portion of Professional Time Spent
Supervising Postdocs
“We have 20 to
25 students at our
institute, and I take a
personal interest in
each one of them. I sit
down with them and
mentor their progress
and careers.”
—Joan Heath
Simona Casarosa Joan Heath
NW210508R
The EGL Foundation invites you to apply to the
Gruss Lipper Post-Doctoral
Fellowship Program
Eligibility
• Israeli citizenship
• Candidates must have completed PhD
and/or MD/PhD degrees in the Biomedical
Sciences at an accredited Israeli
University/Medical School or be in their
final year of study
• Candidates must have been awarded a
postdoctoral position in the U.S. host
research institution.
Details regarding the fellowship are available
at www.eglcf.org
Our Department of Immunology is strongly involved in international
research projects on immunity, pathogenicity, and vaccine design in
tuberculosis. We are looking for
Scientists (Postdoc level)
- Salary (German Federal Salary Scale) up to TVöD E14, dependent on qualication -
in the eld of tuberculosis, particularly:
1. Pathology and immunology (human system)
2. Gene expression proling
3. Deep sequencing
4. Data mining / evaluation
Applicants must hold a Ph.D. or M.D. and have a proven record of perfor-
mance in at least one of these areas. Initial appointment will be two years.
Our research is integrated in successful international consortia of leading
institutions from Europe, the U.S. and Africa. The positions are particularly
attractive for those interested in infection biology in an integrative, multidis-
ciplinary international environment.
Submit your application including C.V., publication record, and names and
addresses of at least two referees no later than 6 weeks after publication of
this advert exclusively by email to pd-imm@mpiib-berlin.mpg.de.
Max Planck Institute for Infection Biology
Prof. Dr. Dr. h.c. Stefan H.E. Kaufmann
Charitéplatz 1
10117 Berlin (Germany)
http://www.mpiib-berlin.mpg.de/research/immunology.htm
The Max Planck Society (MPS) is committed to employing more handicapped individu-
als and encourages them to apply. MPS also seeks to increase the number of women
in areas where they are underrepresented and explicitly encourages women to apply.
www.medimmune.com/careers
Announcing:
The MedImmune Postdoctoral Program
Postdoctoral Fellowships in Research and Development
Advancing Science and Medicine to Help People Live Better Lives
Copyright © 2011 MedImmune, LLC. All rights reserved.
MedImmune is an Equal Opportunity Employer.
“This is My
MedImmune.”
“This is an
opportunity
to
develop
my
future
in biomedical
research.

We are now accepting applications for our Postdoctoral Program where you will
have an opportunity to develop your professional potential and gain valuable
experience from some of the top minds in the biopharmaceutical industry.
We are seeking PhD talent for a 3-year Postdoctoral Fellowship. Positions are
available in Gaithersburg, Maryland; Santa Clara and Mountain View, California;
and Cambridge, UK.
Explore our Fellowship opportunities in areas of biotechnology discovery and
product development including:
• Infectious Disease • Oncology
• Respiratory & Inflammation • Biopharmaceutical Development
• Translational Science • Protein Engineering
World-class mentoring • Competitive salary and benefits, 401(k)/pension plan,
and bonus potential!
Learn much more at: www.medimmune.com/careers
Click on the Postdoctoral box on the home page, or on the link under
“Our Opportunities”.
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
p
o
s
t
d
o
c
t
o
r
a
l

o
p
p
o
r
t
u
n
i
t
i
e
s
www.sciencecareers.org 1174
Learning to work independently was more important to postdoc
supervisors (65 percent) than to postdocs (44 percent according
to the 2010 postdoc study) for a successful postdoc experience.
In addition to these “traditional academic accomplishments,”
says Hank Seifert, a professor at the Northwestern University
Feinberg School of Medicine in Chicago, postdocs have to learn
to “communicate effectively in writing, when giving talks, and in
interpersonal discussions. An understanding of the big picture-
perspective of their chosen profession is critical to making good
career decisions.” These additional skills make a postdoc more
likely to land a faculty job when there is tough competition.
Although what it takes to be successful in science has not
changed much over the years, in these tough economic times,
it is more important than ever for graduate students to start
looking for a postdoc position at least a year before finishing
their graduate work. “I am always amazed at the number of our
graduate students who apply to my lab and people who look
for a postdoc with three to six months lead time,” says Kevin
Gardner, professor of biochemistry at the University of Texas
Southwestern Medical Center. Sometimes graduate students
delay their search for postdoc positions because they are waiting
to have a paper in press, but this strategy does not always work
to their advantage because “if the lab is already full and finances
are limited, they will not get the job, even if they have good
publications,” warns Gardner. In times when funding is good,
says Gardner, there is more flexibility for PIs to provide positions
on a short notice.
Finding the right lab is a two-way street. The survey showed that
the most common attribute PIs look for when selecting a postdoc
to join their lab is strong research experience (77 percent). Other
sought-after factors include: an interest in working in new fields (56
percent), a graduate adviser with a good reputation (43 percent),
and a good research institution (43 percent).
In addition to these professional considerations, it is important
that the postdoc and PI get along on a personal level, says IMCB’s
Kaldis. For this reason, he asks postdoc candidates a lot of
questions over the course of a two- to three-hour-long interview
to get insights into their personalities. “A lot of it is gut feeling. It’s
a very individual process, but in general I look for people who are
open to collaboration, and who want to talk to each other and to
me,” he says.
AFTER THE POSTDOC
According to 81 percent of the survey respondents, the average DOI: 10.1126/science.opms.r1100106
Laura Bonetta is a scientist turned freelance writer based in the
Washington, D.C., area.
length of a postdoc in their labs was around three years or less;
however, 16 percent placed it at four years or more. In the 2009
survey, 76 percent of postdoc supervisors said the average length
of their postdocs was one to three years, and 19 percent said four-
plus years. That difference might reflect shorter terms imposed by
tighter funding.
Whether someone is successful in finding a job following a post-
doc depends on many factors, including “being realistic and pro-
active, understanding what is needed and what is expected, and
what hiring committees are looking for in a resume,” says Bishop.
“Regardless of the job, postdocs need to take the initiative early
on and set themselves up to have what employers want. The big-
gest mistake I see is when people do not think about job options
ahead of time or do not think about it realistically.”
In addition to being proactive, it is important to be flexible. “My
sense is that a lot of postdocs come with unrealistic expectations,”
says Martin Dove, professor of earth sciences at the University
of Cambridge, United Kingdom. “You have to have some idea of
where your career might go. You have to come in having a plan A,
but also having a plan B so that you don’t hang on to something
that is not viable.” He adds that it is also important to set a time-
line. “For example, say [to yourself]: ‘If I don’t get an academic
post in five to six years, I will switch to plan B.’”
To help postdocs reach their goals, many mentors have career
guidance meetings every 2 to 12 months, sometimes using ca-
reer development plans or simply outlining the goals that postdocs
want to reach. “I view a career development plan as a tool, or
hammer, to do something constructive, or to hit yourself on the
head,” says Marder. Marder meets with his postdocs annually to
list the goals that they will need to achieve during the year; then at
the next meeting they review those goals to see if they were met
and set out new ones.
Despite all the best planning and good intentions, jobs may be
hard to come by, so postdocs have to be prepared to be persistent
and not get discouraged. “My advice is that if you really enjoy re-
search in an academic environment and are good at it, then perse-
vere, there will always be quality research opportunities available
for well-trained researchers,” says Seifert.
Imperial College London
www3.imperial.ac.uk
Institute of Molecular and
Cell Biology
www.imcb.a-star.edu.sg
Ludwig Institute for Cancer
Research
www.licr.org
Northwestern University
Feinberg School of Medicine
www.feinberg.northwestern.
edu
FEATURED PARTICIPANTS
The University of Iowa
www.uiowa.edu
University of Cambridge
www.cam.ac.uk
University of Texas
Southwestern Medical
Center
www.swmed.edu
University of Trento
www.unitn.it/en
FO C US O N C A REERS
PO STD O C SURV EY
AAAS/ Sc ienc e Business Offic e Feat ure
“It is important to be exposed to a different way
of doing science.” —Michael Stumpf
A postdoctoral fellowship posion is available at the Center for iPS Cell
Research and Applicaon (CiRA) at Kyoto University, to lead a project on
germcell dierenaon fromhuman pluripotent stemcells together with
Prof. Yamanaka and Dr. Takahashi. The aim of the project is generaon of
an ecient and highly reproducible protocol for germcell dierenaon.
This project will be collaborave with Prof. Minori Saitou at Graduate
School of Medicine, Kyoto University.
* About Prof. Saitou’s research, please refer to the arcle in the journal
Cell below.
hp://www.sciencedirect.com/science/arcle/pii/
S0092867411007719
Job Descripons - Establishment of in vitro dierenaon system from
human pluripotent stem cells
(1. Establishing ecient dierenaon condions using germcell –specic
reporter cell lines and high-throughput screening with small molecule
libraries, 2. Characterizing germ cells derived from human pluripotent
stem cells with single cell based analyses, 3. Opmizing dierenaon
protocols for specic germ cell stages)
Applicants must submit a CV, a list of research achievements, and reason
for applying and future vision of your research in CiRA by September 30,
2011 (Applying by e-mail is not acceptable.)
Please refer to our website for details:
hp://www.cira.kyoto-u.ac.jp/e/employment.html
Contact: Hiromi Kurokawa, Research Promoon Group, Director’s
Oce, Center for iPS Cell Research and Applicaon, Kyoto University
E-mail: collaboraon@cira.kyoto-u.ac.jp
Postdoctral Fellowship for Germ Cell Research Available
in Yamanaka Group (Kyoto university)
Center for iPS Cell Research and Applicaon (CiRA), Kyoto University
Research Associateship Programs
Graduate, Postdoctoral and
Senior Research Awards
offered for research at
US Federal Laboratories
and affiliated institutions
Qualified applicants will be reviewed without regard to race,
religion, color, age, sex or national origin.
Questions should be directed to the :
National Research Council
TEL: (202) 334-2760
E-MAIL: rap@nas.edu
Detailed program information, including instructions on
how to apply online, is available on the NRC Web site at :
www.nationalacademies.org/rap
• Awards for independent research at over 100 participating
laboratory locations
• 12-month awards renewable for up to 3 years
• Annual stipend $42,000 to $75,000 - higher for senior researchers;
graduate entry level stipend begins at $30,000
• Relocation, professional travel, health insurance
• Annual application deadlines Feb. 1, May 1, Aug. 1, Nov. 1
• Open to US and non-US citizens
Opportunities for graduate, postdoctoral and senior research
in all areas of science and engineering
Applicants must contact Advisor(s) at the lab(s)
prior to application deadline.
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
p
o
s
t
d
o
c
t
o
r
a
l

o
p
p
o
r
t
u
n
i
t
i
e
s
Postdoctoral
Fellowships Available
The Lombardi Comprehensive
Cancer Center at Georgetown Uni-
versity, a multidisciplinary NCI-
designated cancer research center, is currently
recruiting postdoctoral fellows into positions
funded by an NCI training grant. The goal is to
develop strong basic and translational scientists
with an interest in cancer research. Successful
applicants will choose a mentor from an interdis-
ciplinary group of investigators who are committed
to cancer research. Research programs include:
• The role of growth factor signal pathways
• The development of hormone and drug
resistance
• The genetic and molecular mechanisms of
malignant progression
• Invasion metastasis angiogenesis
• Stem cells in cancer
• Development of novel immunological and
anticancer therapies
• The etiology of cancer, biomarkers, and
molecular epidemiology
• Bioinformatics and cancer
Go to http://lombardi.georgetown.edu/
education/TBIO/postdoc.html for further
information.
Salary is competitive and commensurate with
qualifications and experience. Applicants should
send curriculumvitae, a short statement of research
interests and career goals, and the names and
addresses of three references to Karen Shepherd
at bivinsk@georgetown.edu.
Minorities and women are strongly encouraged
to apply. U.S. citizenship or permanent residency
is required.
Advance Your Career and Your Education?
Corporate Postdoctoral Residency Program
A unique 24-month corporate experience and networking opportunity
for outstanding Postdoctoral scientists and engineers seeking success in
Bioscience Management.
• Begin with a paid summer internship in the laboratory of a senior
investigator at Life Technologies.
• Continue with fall enrollment at Keck Graduate Institute, earning the
innovative Postdoctoral Professional Masters (PPM) degree; a 9-month
accredited program giving PhDs the business and management skills
needed to pursue high-level careers in the life sciences industry.
• Upon PPM completion, gain crucial corporate experience in Life
Technologies’ 1-year paid residency program.
The Corporate Postdoctoral Residency Program was designed by the
Chief Scientific Officer and the Global Head of R&D of Life Technolo-
gies along with KGI’s academic leadership.
Please apply by December 15, 2011 to be considered for this pro-
gram.
For further information about the program and our PPM visitation day,
see www.kgi.edu/ppm_residency or email admissions@kgi.edu.
The Faculty of Natural Sciences, invites applications for a tenured
W2-Professorship in Inorganic Chemistry
(succession of Prof. Rau)
in the Department of Chemistry and Pharmacy.
Successful candidates should represent the field of Inorganic Chemistry in research and teaching. The pre-
ferred research speciality area is modern inorganic chemistry, of particular interest are investigations
towards molecular, semi-conducting or nanostructured systems for sustainable chemistry. The applicant
can expect a strongly interdisciplinary research environment with the Center of Excellence “Engineering of
Advanced Materials” and the Graduate School “Molecular Science”. Active collaborations and participa-
tions within these activities and future initiatives are appreciated.
Prerequisites to the job are a university degree, good teaching skills, a doctorate and additional scientific
attainments. The latter are evidenced by a post-doctoral habilitation or similar academic qualifications,
which may have been attained in specialised fields other than at a university, or a junior professorship.
With respect to further preconditions for the job, legal provisions apply.
The University of Erlangen-Nuremberg is an Equal Opportunities Employer committed to increasing the
number of female academic staff and would expressly welcome applications from women. We also offer
encouragement for Dual-Carreer-Couples.
Preference will be given to disabled applicants with commensurate qualifications.
The position is to be filled as soon as possible.
Applications should include a CV, a list of publications, public lectures and degree courses tought, together
with certified copies of degree certificates as well as a brief statement of research interests, but no publica-
tions, should be sent not later than October 15, 2011 (date of stamp) to: The Dean, Naturwissenschaftliche
Fakultät der Universität Erlangen-Nürnberg, Universitätsstr. 40, D-91054 Erlangen, Germany.
www.fau.de
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
POSITIONS OPEN POSTDOC OPPORTUNITIES
POSTDOCTORAL OPPORTUNITIES
Faculty Position in Developmental Neuroscience
Institute for Neuroscience and
Department of Pharmacology and Physiology
School of Medicine and Health Sciences
The George Washington University
The newly formed Institute for Neuroscience and the Department of Pharma-
cology and Physiology are accepting applications for a tenure-eligible faculty
member at the rank of Assistant or Associate Professor with expertise in the
genetic and cellular analysis of neurodevelopmental disorders in animal
models. This position will be one of five added to the existing community of
GWU neuroscientists as part of an initiative to expand research in develop-
mental disorders including autism. We seek an investigator whose research
focuses on analysis in animal models of genes known to cause human neu-
rodevelopmental disorders. This individual will participate in medical and
graduate education in the Department of Pharmacology and Physiology as
well as the Institute for Biomedical Sciences.
Basic Qualifications: Applicants must have a terminal degree (Ph.D. or M.D.)
in an appropriate discipline and substantial accomplishments in biomedical
research as demonstrated by a significant number of first and/or senior author
publications in outstanding peer-reviewed journals as well as initial success
in obtaining external research support. Preferred Qualifications: Preference
will be given to candidates with a growing research program focused on
genetic, molecular and cell biological analysis of mouse models of human
neurodevelopmental disorders. Integration of mouse genetic analysis with
ongoing human genetic analysis is encouraged. The successful candidate will
participate in Institute for Neuroscience research activities including develop-
ment of multi-investigator projects for extramural funding. Salary and start
up funds will be commensurate with experience.
To be considered, please send a complete curriculum vitae plus names
and contact information for 3 references electronically to Anthony-Samuel
LaMantia, Ph.D., Professor of Pharmacology and Physiology, Director, GWU
Institute for Neuroscience at recruit2011@gwumc.edu. If possible, please
send this information in the PDF format. Review of Applications will begin
on October 1, and will continue until the position is filled. Only complete
applications will be considered.
GW is an Equal Opportunity/Affirmative Action Employer.
EARTH SCIENCES DIVISION
The Earth Sciences Division at Berkeley Lab is looking for individuals to
fulll strategic scientist and manager positions as well as postdoctoral
positions in research programs that crosscut many areas of the
geosciences disciplines. These include climate, carbon sequestration,
ecology, geothermal energy, hydrogeology, and more. Individuals
whose experience matches qualications in computational modeling
(specically of coupled processes, reactive transport, and multi‐phase
ow), geotechnical engineering, and laboratory and eld‐work
applications should apply.
See our open positions:
esd.lbl.gov/resources/workplace/hr/new_employees/job_opportunities.html
Berkeley Lab has over 150 openings in both science and operations. Visit
our career web site for a complete listing (jobs.lbl.gov). Berkeley Lab is an
armative action/equal opportunity employer committed to the
development of a diverse workforce.
Like us on Facebook
facebook.com/pages/Earth‐Sciences
‐Division/185214424867798
Follow us on Twitter
twitter.com/esdwebmaster
Watch us on Vimeo
http://www.vimeo.com/user4438779
/videos
Learn more on our Blog
earthsciences.typepad.com
Explore the ESD website
esd.lbl.gov
See all jobs at LBNL
jobs.lbl.gov
ASSISTANT or ASSOCIATE PROFESSOR
The newly created Center of Excellence in Cancer Research at the Paul L.
Foster School of Medicine at El Paso is seeking highly qualified applicants
for tenure-track faculty positions at the Assistant or Associate Professor
level in the Department of Biomedical Sciences. Successful candidates are
expected to develop an independently funded research programin cancer
research or a related field. The positions report to the Director of Basic
Science Research of the Center of Excellence in Cancer Research.
Minimum qualifications: M.D., Ph.D., or M.D. /Ph.D. degree in a field
related to cancer, three years of postdoctoral experience, and a strong
publication record.
Preferred qualifications: Candidates with experience in cancer research,
experience using the latest technologies and funded grant support are
particularly encouraged to apply. (Only candidates with funded grant
support will be considered for Associate Professor Rank.)
A competitive salary, startup package, and comprehensive benefits are
available. Interested candidates must apply online at website: http:
//jobs.texastech.edu, requisition #80343 - Assistant Professor and
#81927 Associate Professor.
For further information, potential applicants may inquire confiden-
tially to: Rajkumar Lakshmanaswamy Ph.D., e-mail: rajkumar.
lakshmanaswamy@ttuhsc.edu, Director of Basic Science Research
of the Center of Excellence in Cancer Research. The position is open
until filled. Application review will begin immediately.
Texas Tech University Health Sciences Center is an Equal Opportunity/
Affirmative Action Employer. NW195484
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
Marijanna Kumerich
Chair in Leukaemia &
Lymphoma Research
Senior research opportunities do not
come more remarkable than this.
As a result of a substantial donation, our Faculty of
Medical and Health Science is set to embark on a
comprehensive research programme into Leukaemia &
Lymphoma biology and treatment. Central to this
programme’s success is this senior academic position
– a remarkable career opportunity for you and a unique
chair for this area of research in New Zealand.
For an internationally acclaimed expert in the study of
leukaemia and lymphoma, this is what can only be
described as a ‘blank canvas’ opportunity - the chance
to build, lead and develop a team focussed on
fundamental and translational leukaemia and
lymphoma research. Lump sum funding has been
guaranteed for five years, after which we envisage the
programme will be self-sustaining. As such, full devotion
can be directed toward leading a programme of
internationally recognised research, and the discovery
and translation of new therapies. Furthermore, the
specific area of focus is not defined, so you will have the
mandate to direct and develop the programme in line
with your experience and interests.
As networking and collaboration will be imperative to
achieving programme milestones, expect strong links
with Auckland Hospital, resourcing support across the
University, and unrestricted scope to leverage expertise
from the wider research community.
Considering the calibre and seniority of this position, it
is essential you have demonstrated international
eminence in your field of study, and have been widely
published in the most influential journals. While you
may come to us from either a clinical or non-clinical
background, you must hold a higher research degree
and have a record of international invitations to
conferences and professional meetings. The ability to
influence policy development, and service design and
delivery with respect to haematological malignancies, is
equally important.
For a confidential discussion about this role please
contact Professor Peter Browett ph: +64 9 923 6281,
email: p.browett@auckland.ac.nz
For further information go to
www.auckland.ac.nz/opportunities
The University has an equity policy and welcomes
applications from all qualified persons.
Vanderbilt University School of Medicine announces a national search for the
Chair of the Department of Biochemistry.
The Department of Biochemistry at Vanderbilt is among the top 5 depart-
ments in the U.S., as measured by NIH funding. Its 19 primary faculty are
supported by over $35M in sponsored research and many of its faculty also
serve as Directors of key institutional Centers. The Department has particular
scientific strengths in chemical and structural biology, drug discovery, genome
maintenance, and cancer biology.
The next Chair of Biochemistry will have exceptional scientific accomplishments
and reputation, and will demonstrate the visionary leadership skills commensurate
with the quality of the Department. The Chair will be prepared to devote this
phase of his/her career to the advancement of the faculty, the Department, and
the University. He/she will have been active in national activities, having a pro-
fessional network that will be an important asset in recruitment and reputation
building. The Chair will be expected to recruit as many as 12 primary faculty in
the next five years, dramatically shaping the Department’s scientific focus, culture,
and demographics. Successful candidates will possess a Ph.D. with a sustained
record of excellence in scholarly activity in the field of biochemistry.
Korn/Ferry International is assisting Vanderbilt with this important search. Please
forward, as soon as possible, nominations of appropriate candidates to: Warren E.
Ross, M.D., c/o Shannon Yeatman (shannon.yeatman@kornferry.com),
Korn/Ferry International, 1835 Market Street, Suite 2000, Philadelphia,
PA19103.
Vanderbilt University is an equal opportunity, affirmative action employer.
CHAIR
Department of Biochemistry
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
Yale University
School of Medicine
FACULTY POSITION AT THE JUNIOR OR SENIOR
LEVEL
DEPARTMENT OF CELLULAR AND MOLECULAR
PHYSIOLOGY
The Department of Cellular and Molecular Physiology is
conducting a search for new faculty members at the junior or
senior level.
The search seeks candidates whose research connects the prop-
erties of molecules to the properties of physiological systems,
with particular emphasis on integrative or translational research,
or research that makes use of genetic techniques. Investigators
interested in the cardiovascular, respiratory, musculoskeletal,
gastrointestinal, renal, endocrine or reproductive systems are
especially encouraged to apply
Excellent opportunities are available for collaborative research,
as well as for graduate and medical student teaching. Candidates
must hold a Ph.D., M.D., or equivalent degree. Applicants should
include a curriculum vitae, a statement of research interests and
goals, and three letters of reference. Applications should be
emailed to leisa.strohmaier@yale.edu in PDF format.
Application Deadline: October 20, 2011
Yale University is an
Affirmative Action/Equal Opportunity Employer.
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
Mayo Foundation is
an afrmative action
and equal opportunity
employer and
educator. Post-offer/
pre-employment drug
screening is required.
Faculty Positions in Cartilage Biology/Regeneration
Mayo Clinic seeks outstanding candidates for research faculty positions in the eld of cartilage biology/regeneration, a
strategic priority of the institution. This is part of a coordinated, multi-department recruitment in fundamental and diseased-
focused areas pertinent to stem cell biology (e.g., nuclear reprogramming, developmental biology, tissue bioengineering) as
they apply to the broader elds of arthritis, musculoskeletal (cartilage biology), neurologic (neuroregeneration) and
cardiovascular (heart repair) diseases. We also encourage candidates utilizing interdisciplinary approaches relevant to human
disease with expertise in regenerative diagnostics or therapeutics. Candidates should hold the M.D., Ph.D., M.D./Ph.D. or
equivalent degree and have a demonstrated record of relevant research accomplishments as evidenced by their training,
publications and peer-review funding. Appointment and rank in an academic department will be determined based on the
applicant’s qualications.
Mayo Clinic is an internationally renowned, integrated, multidisciplinary academic medical center with comprehensive programs
in medical education and research, with campuses in Rochester, Minnesota; Jacksonville, Florida; and Scottsdale/Phoenix,
Arizona. Mayo Clinic supports ve schools spanning undergraduate, graduate, medical, postgraduate medical and continuing
medical education and has a vibrant research enterprise with programs in clinical, basic and population sciences. In 2010, the
institution received more than $340 million in extramural research awards.
Mayo Clinic’s location in Rochester, MN, combines the ease of small-city living with easy access to additional cultural and
entertainment opportunities in nearby Minneapolis/St. Paul. Mayo Clinic offers a competitive salary, excellent benets, and
has been recognized by Fortune magazine as one of the “100 Best Companies to Work For.” To learn more about Mayo Clinic
and Rochester, MN, please visit http://www.mayoclinic.org/scientist-jobs/
Interested candidates are invited to submit a cover letter, curriculum vitae and brief description of research plans to:
Email: RegenerativeMedicinePositions@Mayo.edu
Heal the sick, advance the science, share the knowledge.
Opportunity for Postdoctoral Fellowship 2012/2013
The Azrieli Fellows Program welcomes the best and the brightest
postdoctoral scholars who wish to undertake postdoctoral research in
Israel. The time candidates spend conducting research in Israel will
provide them with valuable tools to realize their professional goals
through collaborative association with Israel’s world-class academic
institutions and researchers.
The fellowships are awarded on the basis of academic excellence.
Candidates are assessed on their potential to make cutting-edge
contributions to their respective fields. Aspects of personal merit and
leadership abilities are also taken into consideration without regard to
race, religion, gender, ethnicity or age.
Eligibility:
The Azrieli Fellows Program is currently open to scholars in any field
of study who are:
• at the postdoctoral level.
• Canadian citizens or who have completed their doctorate at a
Canadian university.
• able to obtain an academic sponsor who is a faculty member at one of
the seven participating Israeli academic institutions.
Participating Israeli Institutions:
Bar-Ilan University | Ben-Gurion University | Technion-Israel Institute
of Technology | Tel Aviv University | The Hebrew University of
Jerusalem | University of Haifa | Weizmann Institute of Science
For more information and to apply, please visit our website at:
www.azrielifoundation.org
TENURE-TRACK FACULTY POSITIONS
The Department of Biochemistry and Molecular Biology is planning a major
expansion and is seeking applicants for several TENURE-TRACKFACULTY
POSITIONS at the associate or assistant professor level. We offer a highly
collaborative scientific environment with many multidisciplinary centers and
state-of-the-art facilities including transgenic, in vivo and in vitro imaging,
genomic, and proteomic (http://tulane.edu/som/) cores. Candidates with
research interests that complement and strengthen existing research activities
(http://tulane.edu/som/departments/biochemistry/), as well as bring novel
approaches and expertise in cancer signaling mechanisms, structural biology,
microRNA, epigenetics, cancer stemcells, medicinal chemistry, chromosome
dynamics, or genetic/disease model systems are encouraged to apply. Success-
ful applicants will be expected to develop strong and independent extramurally
funded research programs, and participate in graduate and medical student
teaching. Highly competitive salaries, start-up funds and laboratory space
will be provided. Applicants must have a Ph.D., M.D. or equivalent degree,
post-doctoral experience, and clear evidence of research productivity and
independence. The search committee will begin consideration of applications
immediately and continue until positions are filled.
Applicants should submit curriculum vitae, a brief summary of past accom-
plishments and future research plans, and the names and email addresses of
five references by email to: biochem@tulane.edu or by mail to:
The Faculty Search Committee
c/o Gilbert Estrada
Department of Biochemistry and Molecular Biology SL43
Tulane University Health Sciences Center
1430 Tulane Ave.
New Orleans, LA70112
Tulane University is an Affirmative Action/Equal Opportunity Employer
and encourages applications from minorities, women, and other
qualified persons.
POSTDOCTORAL OPPORTUNITIES
AAAS is here –
bringing educational infrastructure
to the developing world.
AAAS is helping the Rwandan government
rebuild its educational infrastructure as
a way to help drive economic growth and
development. By providing materials
such as the Project 2061 Atlas of Science
Literacy, lesson plans from Science
NetLinks, and access to Science digital
libraries, AAAS is helping the people of
Rwanda work toward a future built around
science and technology. As a AAAS
member your dues support these efforts.
If you’re not yet a AAAS member, join us.
Together we can make a difference.
To learn more, visit
aaas.org/plusyou/rwanda
FACULTY POSITION
DEPARTMENT OF PHARMACOLOGY
The Department of Pharmacology invites applications fromoutstanding
scientists for a tenure track position at the Assistant Professor rank.
A competitive laboratory start up package will be provided to the
successful candidate to support the development of an independent,
funded research programin drug discovery, chemical biology, medicinal
chemistry, nanomedicine, combinatorial chemistry, pharmacogenomics
or other areas broadly relevant to pharmacology. Candidates should
have a Ph.D. and/or M.D. degree and postdoctoral experience as well
as a strong record of research accomplishments. Baylor College of
Medicine is located in the Texas Medical Center and offers a highly
interactive environment and a strong infrastructure for research. Review
of applications will occur as they are received, however, completed
applications should be received by November 1, 2011.
Applicants should submit a statement of research interests and curricu-
lum vitae as a single PDF to pharmacology@bcm.edu. Three letters
of reference should be sent separately to pharmacology@bcm.edu,
attention: Timothy Palzkill, Ph.D., Chair, Department of Pharma-
cology, Baylor College of Medicine, One Baylor Plaza, Houston,
TX 77030.
Baylor College of Medicine is an Equal Opportunity/Affirmative
Action And Equal Access Employer.
FACULTY POSITIONS
The Department of Cancer Biology at the Perelman
School of Medicine at the University of Pennsylvania
seeks candidates for an Assistant, Associate,
and/or Full Professor position in the tenure track.
Rank will be commensurate with experience.
The successful applicant will have experience
in the eld of cancer biology, including but not
limited to cancer genetics and genomics, tumor
microenvironment, epigenetics, angiogenesis,
cancer cell metabolism and oncogenic signaling.
Responsibilities include the development of an
independent research program. Teaching duties
comprise mentoring students and course lecturing.
Applicants must have an M.D. and/or Ph.D degree
and have demonstrated excellent qualications in
Education and Research.
The University of Pennsylvania is an equal
opportunity, afrmative action employer. Women
and minority candidates are strongly encouraged
to apply.
Lewis A. Chodosh, MD, PhD
J. Samuel Staub Professor
Chair, Department of Cancer Biology
Perelman School of Medicine,
421 Curie Boulevard,
Philadelphia, PA 19104-6160
Contact:
Apply for this position online at:
http://www.med.upenn.edu/apps/faculty_ad/
index.php/g304/d2690
FACULTY POSITION AT VANDERBILT UNIVERSITY
IN
DEVELOPMENTAL GENETICS
The Department of Biological Sciences at Vanderbilt Univer-
sity seeks candidates to fill a tenure-track, assistant professor
faculty position in developmental genetics. Candidates will
be considered in all areas of developmental biology using any
genetic model. We desire candidates whose research broadly
overlaps and complements existing areas of interest within
the department (http://sitemason.vanderbilt.edu/biosci). For
information about developmental biology at Vanderbilt, see
http://www.mc.vanderbilt.edu/devbio/. The central criteria
for the position are excellence in research and the ability to
teach undergraduate and graduate students with a high level of
effectiveness.
Applicants should send a PDF containing a letter of application,
curriculum vitae, and a statement of current and future research
interests to angela.thomason@vanderbilt.edu. Applicants
also should arrange for three letters of recommendation to be
sent to the same email address. Review of applicants will begin
November 1, 2011, and will continue until the position has
been filled.
Vanderbilt University is an Affirmative Action/Equal Oppor-
tunity Employer. Women and under-represented minority
candidates are especially encouraged to apply.
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
Faculty of Medicine
The University of Zurich invites applications for a position as
Tenure-Track Assistant
Professor in Biochemistry/
Structural Biology
The University of Zurich is committed to further strengthening
its research efforts in protein science, particularly structural
biology. It provides a stimulating and attractive environment
for interdisciplinary research at the highest international
level. Creative researchers with a proven track record who
use X-ray crystallography in the context of innovative
approaches and important biological problems are encouraged
to apply.
The Department of Biochemistry is located on the Irchel
campus, the location of the science and preclinical departments
of the University. The Department also maintains strong ties
in research and teaching with the science departments of the
ETH Zurich and the Paul-Scherrer-Institute, which operates
the Swiss Synchrotron Light Source for structural studies of
biomolecules.
The position is accompanied by an attractive startup package,
including funding for coworkers, consumables and equipment.
Tenure evaluation is after 6 years. The new professor will be
expected to contribute to teaching. Applications including a
curriculum vitae, a list of publications, a detailed description
of future research plans, and a list of potential referees
should be received no later than November 15, 2011 by:
The Dean, Professor Dr. Dr. K. W. Grätz, Pestalozzistrasse 3/5,
CH-8091 Zurich, Switzerland. (Guidelines for submission of
applications: www.med.uzh.ch/FormulareundRichtlinien/
Bewerbung/application_form.doc)
For additional information, see www.bioc.uzh.ch or contact
Prof. Andreas Plückthun (plueckthun@bioc.uzh.ch)
University of Zurich, Winterthurerstrasse 190,
CH-8057 Zurich.
TWO ASSISTANT PROFESSOR POSITIONS
UNIVERSITY OF NEBRASKA-LINCOLN (UNL)
SCHOOL OF BIOLOGICAL SCIENCES
The School of Biological Sciences continues to expand its faculty and
invites nominations and applications for two tenure-track, Assistant
Professor positions.
Algal/Microbial Molecular Biology. This academic-year position will
be supported in part by a National Science Foundation Experimental
Program to Stimulate Competitive Research (EPSCoR) Grant. We
seek a broadly trained molecular/microbial biologist who will conduct
vigorous basic research using an algal systemto address significant cel-
lular, metabolic or physiological processes and employ modern ‘omics’
technologies for uncovering gene function and system-level pathway
analyses. The successful candidate will be part of an interdisciplinary
team addressing the underlying basic biology of eukaryotic algae and
how those findings can be exploited for the production of biofuels or
other high-value products. The successful candidate will be housed
in the George W. Beadle Center, which includes state-of-the-art core
facilities in proteomics, metabolomics, genomics, crystallography,
bioimaging, flowcytometry, bioinformatics, biophysical spectroscopy,
and trace element analysis. APhD(or equivalent) in molecular biology,
biochemistry, microbiology, cell biology or related field and two years
postdoctoral experience (or equivalent) is required.
Ecological Modeling. For this academic-year position we seek an ecol-
ogist with expertise in mathematical/simulation modeling. Research
can be in either a basic or applied context in natural or human-domi-
nated systems, addressing questions in conservation biology, disease
dynamics, ecological risk assessment, or other areas of population,
community or ecosystemecology. The interdisciplinary position would
build on the existing interactions between the School of Biological Sci-
ences, Dept of Mathematics, School of Natural Resources, and Dept of
Computer Science and Engineering through the applicant’s interests in
mathematical biology, ecosystemmanagement, environmental policy,
human dominated ecosystems or distributed sensor networks. The suc-
cessful candidate is expected to develop strong collaborations and in
so doing contribute significantly to UNL’s Life Sciences Initiative. A
PhD (or equivalent) and experience in mathematical modeling and/or
computer programming is required.
Both positions are part of the strategic plans of UNL, the College of
Arts and Sciences and the School of Biological Sciences directed to
strengthen the life sciences. It is expected that the successful candidates
will establish nationally recognized and extramurally funded research
programs and contribute to the undergraduate and graduate teaching
missions of the School of Biological Sciences.
Both positions come with highly competitive start-up packages. Lin-
coln Nebraska boasts an outstanding quality of life that includes fine
culinary and artistic treasures, a budding live music scene and numer-
ous parks, golf courses and bike trails. In 2008, WebMDreported that
Lincoln was the healthiest city in the United States.
To learn more about the University of Nebraska and the School of
Biological Sciences, visit http://biosci.unl.edu. Applicants should
go to http://employment.unl.edu, search for requisition number
110506 and 110596 respectively, complete the Faculty Academic/
Administrative Information form, attach a letter of application, Cur-
riculumVitae, a statement of research plans, and statement of teaching
interests. Applicants must arrange for three confidential letters of refer-
ence to be sent directly to: Search Committee, School of Biological
Sciences, University of Nebraska, 348 Manter Hall, Lincoln, NE
68588-0118, USA or by e-mail to biologysearch@unl.edu. Review
of applications will begin on September 15, 2011 and continue until
the positions are filled or the search is closed.
The University of Nebraska has an active National Science
Foundation ADVANCE gender equity program, and is committed to
a pluralistic campus community through affirmative action, equal
opportunity, work-life balance, and dual careers.
Who is # this year?
Science publishes its 10th annual Top Employers Survey on October 7, 2011. Science has a
long history of providing a forum for scientists to express their opinions about the biotech
and pharma industry.
Advertise in this issue to reach both ACTIVE and PASSIVE job seekers. Here’s how:
1. Scientists in the biotech/pharma community eagerly anticipate the results of this survey
every year. By advertising, your reach goes beyond active job seekers to those involved
in the field, targeting elusive passive job seekers.
2. Your association with this issue tells prospective recruits that you are among the best.
Reach the scientists that your competitors are reaching and promote your advantages.
3. This issue will be distributed to attendees at the American Society of Human Genetics
meeting in Montreal, so you reach beyond Science’s 700,000 weekly readers.
Start building your pipeline today with Science.
ScienceCareers.org
Produced by the Science/AAAS CustomPublishing Office
O
ver
9
8
,0
0
0 S
cience
print
r
ead
ers
w
ork
in
ind
us
t
r

r
each
t
hes
e
pas
s
ive
job
s
eekers
!
**
* * Publisher’s Own Data, August 2011
2
0
1
1 A
n
n
u
a
l
T
o
p
E
m
p
lo
y
e
r
s
S
u
r
v
e
y
S
pecial C
areer
F
eat
ure: O
ct
o
b
er
7
, 2
0
1
1
R
eserve your
ad
by
September
2
0
to
guarantee space
*
*Ads accepted until October 3 if space is still available
To Book Your Ad, Contact:
In the United States, Canada, and Latin America
E-mail: advertise@sciencecareers.org
In Japan:
E-mail: mhara@aaas.org
In China and Taiwan:
E-mail: rwu@aaas.org
In Europe and Rest of World
E-mail: ads@science-int.co.uk
o
n
l
i
n
e

@
s
c
i
e
n
c
e
c
a
r
e
e
r
s
.
o
r
g
Nontraditional Careers:
Opportunities Away
Fromthe Bench
Want to learn more about exciting and
rewarding careers outside of academic/
industrial research? Viewa roundtable
discussion that looks at the various career
options open to scientists across different
sectors and strategies you can use to
pursue a nonresearch career.
Webinar
Produced by the
Science/AAAS Business Office.
NowAvailable
On Demand
www.sciencecareers.org/webinar
AQUATIC MICROBIAL ECOLOGIST
Baylor University is pleased to accept applications for a new
tenure-trackAssistant Professor strategic hire within the Depart-
ment of Biology and associated with the Center for Reservoir
andAquatic Systems Research (CRASR) to beginAugust 2012.
Candidates in aquatic microbial ecology or aquatic microbiol-
ogy who use molecular techniques to study the ecology of
microorganisms and their relationship to the environment
and/or human health are encouraged to apply. A strong inter-
disciplinary research focus is essential. For full consideration
applications must be received by November 15, 2011. Addi-
tional information is available at http://www.baylor.edu/hr/
index.php?id=79678 or from Ryan_S_King@baylor.edu.
Chartered in 1845 by the Republic of Texas, Baylor Univer-
sity is the oldest university in Texas and the world’s largest
Baptist University. Baylor’s mission is to educate men and
women for worldwide leadership and service by integrating
academic excellence and Christian commitment within a caring
community. Baylor is actively recruiting new faculty with a
strong commitment to the classroom and an equally strong
commitment to discovering new knowledge as Baylor aspires
to become a top tier research university while reaffirming and
strengthening its distinctive Christian mission as described in
Baylor 2012.
Baylor is a Baptist university affiliated with the Baptist
General Convention of Texas. As an Affirmative Action/
Equal Employment Opportunity Employer, Baylor
encourages minorities, women, veterans, and persons with
disabilities to apply.
Assistant Professor of
Evolutionary or Ecological Genomics
The Department of Evolution, Ecology, and Organismal Biology (EEOB)
invites applications for a TENURE-TRACK POSITION at the Assistant
Professor Rank in Evolutionary or Ecological Genomics. We seek
outstanding individuals who address fundamental questions in evolution
and/or ecology either through the generation of large-scale sequence
or gene expression data or by using either computational/statistical
approaches for the