Semiaquatic adaptations in a giant predatory dinosaur

Abstract

We describe adaptations for a semiaquatic lifestyle in the dinosaur Spinosaurus aegyptiacus. These adaptations include retraction of the fleshy nostrils to a position near the mid-region of the skull and an elongate
neck and trunk that shift the center of body mass anterior to the knee joint. Unlike terrestrial theropods, the pelvic girdle
is downsized, the hindlimbs are short, and all of the limb bones are solid without an open medullary cavity, for buoyancy
control in water. The short, robust femur with hypertrophied flexor attachment and the low, flat-bottomed pedal claws are
consistent with aquatic foot-propelled locomotion. Surface striations and bone microstructure suggest that the dorsal “sail”
may have been enveloped in skin that functioned primarily for display on land and in water.

Full text

DOI: 10.1126/science.1258750
, 1613 (2014);345 Science et al.Nizar Ibrahim
Semiaquatic adaptations in a giant predatory dinosaur
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others
here.following the guidelines can be obtained byPermission to republish or repurpose articles or portions of articles
): September 25, 2014 www.sciencemag.org (this information is current as of
The following resources related to this article are available online at
http://www.sciencemag.org/content/345/6204/1613.full.html
version of this article at: including high-resolution figures, can be found in the onlineUpdated information and services,
http://www.sciencemag.org/content/suppl/2014/09/10/science.1258750.DC1.html
can be found at: Supporting Online Material
http://www.sciencemag.org/content/345/6204/1613.full.html#related
found at: can berelated to this article A list of selected additional articles on the Science Web sites
http://www.sciencemag.org/content/345/6204/1613.full.html#ref-list-1
, 15 of which can be accessed free:cites 40 articlesThis article
http://www.sciencemag.org/cgi/collection/paleo
Paleontology subject collections:This article appears in the following
registered trademark of AAAS. is aScience2014 by the American Association for the Advancement of Science; all rights reserved. The title CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on September 25, 2014www.sciencemag.orgDownloaded from on September 25, 2014www.sciencemag.orgDownloaded from on September 25, 2014www.sciencemag.orgDownloaded from on September 25, 2014www.sciencemag.orgDownloaded from on September 25, 2014www.sciencemag.orgDownloaded from

7. M. White, N. Ashton, Curr. Anthropol. 44, 598–609 (2003).
8. F. Fontana et al., J. Anthropol. Archaeol. 32, 478–498
(2013).
9. A. Picin, M. Peresani, C. Falguères, G. Gruppioni, J.-J. Bahain,
PLOS ONE 8, e76182 (2013).
10. R. Barkai, A. Gopher, S. E. Lauritzen, A. Frumkin, Nature 423,
977–979 (2003).
11. R. Shimelmitz, R. Barkai, A. Gopher, J. Hum. Evol. 61, 458–479
(2011).
12. S. L. Kuhn, Curr. Anthropol. 54, S255–S268 (2013).
13. V. B. Doronichev, PaleoAnthropol. 2008, 107 (2008).
14. E. Boëda, in The Definition and Interpretation of Levallois
Technology, H. L. Dibble, O. Bar-Yosef, Eds. (Prehistory Press,
Madison, WI, 1995), pp. 41–69.
15. E. Boëda, Le Concept Levallois: Variabilité des Méthodes (CNRS
Éditions, Paris, 1994).
16. P. J. Brantingham, S. L. Kuhn, J. Archaeol. Sci. 28, 747–761
(2001).
17. S. J. Lycett, M. I. Eren, J. Archaeol. Sci. 40, 2384–2392
(2013).
18. S. J. Lycett, M. I. Eren, World Archaeol. 45, 519–538 (2013).
19. M. I. Eren, S. J. Lycett, PLOS ONE 7, e29273 (2012).
20. R. Foley, M. M. Lahr, Camb. Archaeol. J. 7, 3 (1997).
21. S. J. Armitage et al., Science 331, 453–456 (2011).
22. H. Valladas et al., J. Hum. Evol. 65, 585–593 (2013).
23. M. Meyer et al., Nature 505, 403–406 (2014).
24. K. Prüfer et al., Nature 505,43–49 (2014).
25. Materials and methods are available as supporting material on
Science Online.
26. R. Badalian et al., Radiat. Meas. 34, 373–378 (2001).
27. E. V. Arutyunyan, V. A. Lebedev, I. V. Chernyshev,
A. K. Sagatelyan, Dokl. Earth Sci. 416, 1042–1046 (2007).
28. V. A. Lebedev, I. V. Chernyshev, K. N. Shatagin, S. N. Bubnov,
A. I. Yakushev, J. Volcanol. Seismol. 7, 204–229 (2013).
29. S. Joannin et al., Earth Planet. Sci. Lett. 291,149–158
(2010).
30. V. Ollivier et al., Quat. Int. 223-224, 312–326 (2010).
31. E. Frahm et al., J. Archaeol. Sci. 41, 333–348 (2014).
32. A. J. Jelinek, Science 216, 1369–1375 (1982).
33. S. McBrearty, J. Archaeol. Res. 69, 7 (2013).
34. J. J. Shea, Curr. Anthropol. 52,1–35 (2011).
35. T. Hopkinson, A. Nowell, M. White, PaleoAnthropology 2013,61
(2013).
36. S. Lycett, J. Anthropol. Archaeol. 26, 541–575 (2007).
37. R. Stöckli, E. Vermote, N. Saleous, R. Simmon, D. Herring,
The Blue Marble Next Generation - A true color earth dataset
including seasonal dynamics from MODIS (NASA Earth
Observatory, 2005); http://earthobservatory.nasa.gov/
Features/BlueMarble/bmng.pdf.
38. L. Augustin et al., Nature 429, 623–628 (2004).
ACKNO WLEDGM ENTS
Data discussed in this paper can be found in the supplementary
materials. All artifacts are stored at the Institute of Archeology
and Ethnography, Yerevan, Armenia. We thank the following
organizations for their financial support: the University of
Connecticut [2008–2014: Norian Armenian Programs Committee,
College of Liberal Arts and Sciences (CLAS), Office of Global
Affairs, Study Abroad; and CLAS Book Committee]; the UK
Natural Environment Research Council (grant IP-1186-0510), the
L. S. B. Leakey Foundation (2010 and 2011), the Irish Research
Council (2008 and 2009), and the University of Winchester.
We also thank P. Avetisyan and B. Yeritsyan, Institute of Archeology
and Ethnography, Republic of Armenia, for their collaboration.
SUPPLEMENTARY MATERIALS
www.sciencemag.org/content/345/6204/1609/suppl/DC1
Materials and Methods
Supplementary Text
Figs. S1 to S16
Tables S1 to S7
References (39–192)
Databases S1 and S2
27 May 2014; accepted 19 August 2014
10.1126/science.1256484
PALEONTOLOGY
Semiaquatic adaptations in a giant
predatory dinosaur
Nizar Ibrahim,
1
*Paul C. Sereno,
1
Cristiano Dal Sasso,
2
Simone Maganuco,
2
Matteo Fabbri,
3
David M. Martill,
4
Samir Zouhri,
5
Nathan Myhrvold,
6
Dawid A. Iurino
7
We describe adaptations for a semiaquatic lifestyle in the dinosaur Spinosaurus
aegyptiacus. These adaptations include retraction of the fleshy nostrils to a position
near the mid-region of the skull and an elongate neck and trunk that shift the center of
body mass anterior to the knee joint. Unlike terrestrial theropods, the pelvic girdle is
downsized, the hindlimbs are short, and all of the limb bones are solid without an open
medullary cavity, for buoyancy control in water. The short, robust femur with hypertrophied
flexor attachment and the low, flat-bottomed pedal claws are consistent with aquatic
foot-propelled locomotion. Surface striations and bone microstructure suggest that
the dorsal “sail”may have been enveloped in skin that functioned primarily for display
on land and in water.
Bones of the predatory dinosaur Spinosaurus
aegyptiacus first came to light over a cen-
tury ago from Upper Cretaceous rocks in
Egypt (1–3) but were destroyed in World
War II (4). More recently, isolated teeth
and bones (5) and the anterior half of an adult
skull (6) have been discovered in the Kem Kem
beds of eastern Morocco (Fig. 1A) and equiv-
alent horizons in Algeria, but are insufficiently
complete to estimate the size, proportions, and
functional adaptations of this species. Here
we report the discovery of a partial skeleton of
S. aegyptiacus fromthemiddleoftheKemKem
sequence(Fig.1B),whichisprobablyCenomanian
in age (~97 million years ago) (7).
The subadult skeleton, here designated the neo-
type of S. aegyptiacus (8), preserves portions of the
skull, axial column, pelvic girdle, and limbs. It was
discovered in fluvial sandstone that has yielded re-
mains of the sauropod Rebbachisaurus (9)andthree
other medium-to-large theropods (an abelisaurid,
Deltadromeus,andCarcharodontosaurus)(7,10).
We regard two additional Kem Kem theropods,
Sigilmassasaurus brevicollis and S. maroccanus
(11,12), to be referable to S. aegyptiacus (8).
The neotype skeleton and isolated bones refer-
able to S. aegyptiacus were scanned with com-
puted tomography, size-adjusted, and combined
with a digital recreation of the original Egyptian
fossils (Fig. 2A, red). Missing bones were extrap-
olated between known bones or estimated from
those of other spinosaurids (6,13,14). The digi-
tal model of the adult skeleton of Spinosaurus
(Fig. 2A), when printed and mounted, measures
over 15 m in length, longer than Tyrannosaurus
specimens (~12.5 m) (15).
A concentrated array of neurovascular foramina
open on the anterior end of the snout and ap-
pear similar to foramina in crocodilians that
house pressure receptors that detect water move-
ment (8,16)(Fig.2Bandfig.S6).Theenlarged,
procumbent, interlocking anterior teeth are well
adapted for snaring fish (5,6)(Fig.2Bandfig.S4).
The fossa for the fleshy nostril is small and, unlike
any other nonavian dinosaur, is retracted to a
posterior position to inhibit the intake of water
(Fig. 2C and figs. S4 and S6) (8).
Most cervical and dorsal centra are elongate
compared to the sacral centra, resulting in a pro-
portionately long neck and trunk (Figs. 2A and 3
and tables S1 and S2). The anteriormost dorsal
centra, however, are proportionately short, ex-
ceptionally broad, and concavoconvex (Fig. 2D).
These characteristic vertebrae, the affinity of which
has been controversial (7,11,12), are referred
here to S. aegyptiacus, based on their association
with spinosaurid skeletons in Niger (8)andEgypt
(2).Thehorizontalcervicodorsalhingecreated
by these broad centra would facilitate dorsoven-
tral excursion of the neck and skull in the pur-
suit of prey underwater.
Thedistaltwo-thirdsofthetailcomprisesver-
tebrae with relatively short centra, diminutive
zygapophyses, and anteroposteriorly compressed
neural spines (Fig. 2G). The affinity of these
caudal elements has been uncertain (17), but
comparisons with associated remains from Egypt
(2) and more proximal caudals in the neotype
(Fig. 2A) allow referral to Spinosaurus. Short
centra and reduced neural arch articulations
enhance lateral bending during tail propulsion
in bony fish (18).
The forelimb has hypertrophied deltopectoral
and olecranon processes for powerful flexion and
extension (Fig. 2A). Elongate manual phalanges
(Fig. 2H) and less recurved, manual unguals that
SCIENCE sciencemag.org 26 SEPTEMBER 2014 •VOL 345 ISSUE 6204 1613
1
Department of Organismal Biology and Anatomy, University
of Chicago, Chicago, IL 60637, USA.
2
Museo di Storia
Naturale di Milano, Corso Venezia 55, 20121 Milan, Italy.
3
School of Earth Sciences, University of Bristol, Queen’s
Road, Bristol, BS8 1RJ, UK.
4
School of Earth and
Environmental Sciences, University of Portsmouth, Burnaby
Road, Portsmouth, PO1 3QL, UK.
5
Laboratoire de
Géosciences, Faculté des Sciences Aïn Chock, Université
Hassan II, Casablanca, Morocco.
6
Intellectual Ventures, 3150
139th Avenue Southeast, Bellevue, WA 98005, USA.
7
Dipartimento di Scienze della Terra, Sapienza Università di
Roma, Piazzale Aldo Moro 5, 00185 Rome, Italy.
*Corresponding author. E-mail: nibrahim@uchicago.edu
RESEARCH |REPORTS

areprobablyreferabletoSpinosaurus (11)and
were possibly used in gaffing and slicing aquatic
prey suggest that the manus is proportionately
longer than in earlier spinosaurids (13,14).
Thepelvicgirdleandhindlimbareconsiderably
reduced in Spinosaurus (Fig. 2A). The surface
area of the iliac blade is approximately one-half
that in most other theropods (table S1), and the
supraacetabular crest that supports the hindlimb
is low (Fig. 2F). Hindlimb length is just over 25%
of body length (table S1). In a plot of forelimb,
hindlimb, and body length (Fig. 3), Spinosaurus
and other large theropods maintain fairly similar
forelimb lengths. Relative hindlimb length, however,
is noticeably less in the spinosaurid Suchomimus
(25%) and especially in Spinosaurus (19%) than
in other large tetanuran theropods.
Unlike other mid- or large-sized dinosaurs, the
femur in Spinosaurus is substantially shorter
than the tibia (Fig. 2, I and J, and table S1). In
smaller-bodied bipedal dinosaurs, short femoral
proportions indicate increased stride length and
enhanced speed. In Spinosaurus this is clearly
not the case, given the short hindlimb. The femur
in Spinosaurus has an unusually robust attach-
ment for the caudofemoral musculature, which
is anchored along nearly one-third of the femoral
shaft (Fig. 2I), suggesting powerful posterior
flexionofthehindlimb.Thearticulationatthe
knee joint for vertical limb support, in contrast,
is reduced. The distal condyles of the femur are
narrow, and the cnemial crest of the tibia is only
moderatelyexpanded(Fig.2,IandJ).Together
these features recall the shortened condition of
the femur in early cetaceans (19,20) and in extant
semiaquatic mammals that use their hindlimbs
in foot-propelled paddling (21).
Pedal digit I is unusually robust and long in
Spinosaurus:UnlikeAllosaurus or Tyrannosaurus,
1614 26 SEPTEMBER 2014 •VOL 345 ISSUE 6204 sciencemag.org SCIENCE
Fig. 1. Geographic location and stratigraphic
position of the neotype skeleton of S. aegyptiacus.
(A) Locality (X), situated 18 km northeast of Erfoud
in southeastern Morocco. (B) Stratigraphic position
at the base of the upper unit of the Kem Kem beds,
with correlative positions of associated remains
of contemporary dinosaurs. Abbreviations: c, clay;
CT, Cenomanian-Turonian limestone; p, pebbles; P,
Paleozoic; sd, sandstone; st, siltstone.
Fig. 2. Semiaquatic skeletal adaptations in S. aegyptiacus.(A) Skeletal reconstruction in swimming
pose showing known bones (red) based on size-adjusted, computed tomographic scans of the neotype
(FSAC-KK 11888), referred specimens, and drawings of original bones (1). (B) Rostral neurovascular
foramina in lateral view (MSNM V4047 and a digital restoration of the holotypic lower jaw). (C)Narial
fossa in lateral view (MSNM V4047). (D) Anterior dorsal vertebra (~D1) in lateral, anterior, and posterior
views (UCRC PV601). (E) Dorsal neural spine (D8) in left lateral view (FSAC-KK 11888). (F) Left ilium in
lateral view (FSAC-KK 11888). (G) Mid-caudal vertebra (~CA30, reversed) in anterior and left lateral
views (UCRC PV5). (H) Right manual II-1 phalanx in proximal, lateral, and dorsal views (FSAC-KK 11888).
(I) Left femur in lateral view (FSAC-KK 11888). (J) Right tibia (reversed) in lateral view (FSAC-KK 11888).
(K) Right pedal digit III ungual in dorsal, lateral, and proximal views (FSAC-KK 11888). Abbreviations: af,
articular facet; ag, attachment groove; at, anterior trochanter; C2, 10, cervical vertebra 2, 10; CA1, caudal
vertebra 1; cc, cnemial crest; ce, centrum; clp, collateral ligament pit; D13, dorsal vertebra 13; ded, dorsal
extensor depression; dip, dorsal intercondylar process; fl, flange; ft, fourth trochanter; ftu, flexor tubercle;
lco, lateral condyle; nf, narial fossa; ns, neural spine; nvfo, neurovascular foramina; poz, postzygapophysis;
prz, prezygapophysis; S1, 5, sacral vertebra 1, 5; sac, supraacetabular crest; tp, transverse process.
Institutional abbreviations: FSAC, Faculté des Sciences Aïn Chock, Casablanca; MSNM, Museo di
Storia Naturale di Milano; UCRC, University of Chicago Research Collection, Chicago. Scale bars, 10 cm
in(B)to(D),(G),(H),and(K);and20cmin(E),(F),(I),and(J).
RESEARCH |REPORTS

the first phalanx of digit I in Spinosaurus is
the longest nonungual phalanx in the pes (fig.
S1) and would have been in contact with the
substrate in a stationary pose. The pedal un-
guals are proportionally large, long, low, and
flat-bottomed (Fig. 2K and figs. S1 and S2),
features that differ markedly from the deeper
recurved unguals in other large theropods.
The unguals in Spinosaurus are reminiscent of
the flattened pedal unguals of shorebirds that
do not perch (22). In addition, the toes of some
shorebirds have fleshy lobes and interdigital
webbing that enhance foot-propelled propulsion.
The lengthened digit I and flattened pedal un-
guals in Spinosaurus suggestthatthefootmay
have been adapted to traversing soft substrates
or webbed for paddling.
Increasesinbonemassanddensityarecom-
mon skeletal modifications in terrestrial verte-
brates transitioning to a semiaquatic existence (23).
In Spinosaurus, this was achieved by enlarging
midline display structures, eliminating open med-
ullary cavities in the long bones, and increasing
bone density. In subadult Spinosaurus,thedorsal
neural spines are composed primarily of dense
bone with only a narrow central zone of cancel-
lous bone (Fig. 4D), and long bones have solid
shafts (Fig. 4, A and C) with no development of
the open medullary cavity that is present in other
theropods, including early spinosaurids (Fig. 4B).
Bone density within the long bones, in addition,
is 30 to 40% greater in Spinosaurus than in other
theropods (8).
We estimated a center-of-body mass for a
flesh rendering of Spinosaurus created over
the digital skeleton (8). Center-of-mass estimates
for several theropods have been expressed as
a percentage of femoral length measured an-
teriorly from the hip joint (24). The center of
massinabipedmustbelocatedoverthemid-
dle one-third of the pes to generate a plausible
mid-stance pose (25). In our flesh rendering of
Spinosaurus, the center of body mass is po-
sitioned in front of both the hip and knee joints
at a distance greater than femur length (fig. S3),
suggesting that forelimb support was required
during terrestrial locomotion. Spinosaurus ap-
pears to have been poorly adapted to bipedal
terrestrial locomotion. The forward position of
the center of mass within the ribcage may have
enhanced balance during foot-propelled locomo-
tion in water.
These adaptations suggest that Spinosaurus
was primarily a piscivore, subsisting on sharks,
sawfish, coelacanths, lungfish, and actinopterygian s
that were common in the Kem Kem river system
(5,7,11). A long narrow skull and powerful fore-
limbs are also present in earlier spinosaurids, which
like Spinosaurus (26) have been interpreted as
predominantly piscivorous (13,14,27,28).
The locomotor adaptations outlined above,
however, mark a profound departure in form and
function from early spinosaurids. Prominent
among these are the reduced pelvic girdle; short
hindlimb; short femur; and long, low, flat-bottomed
pedal unguals, all of which can be verified in
the second partial skeleton described by Stromer
as “Spinosaurus B”(2,8). We note here that Spino-
saurus must have been an obligate quadruped
on land, the first discovered among theropod
dinosaurs, given the usual horizontal sacroiliac
joint and the anterior location of the estimated
center of body mass (8). Baryonyx was interpreted
as a facultative quadruped, based on its long skull
and neck and robust humerus (27), but this was
not confirmed by the discovery of more complete
hindlimb remains of the related Suchomimus (13).
In Spinosaurus we infer foot-powered paddling
from the relatively short femur with hypertro-
phied flexor attachment and strong pedal digit
I, as occurs in semiaquatic mammals such as early
cetaceans (19–21). Low, flat-bottomed pedal unguals
are coincident with digital lobes or webbing in
SCIENCE sciencemag.org 26 SEPTEMBER 2014 •VOL 345 ISSUE 6204 1615
Fig. 3. Ternary morphospace plot comparing
forelimb, hindlimb, and body length. Fore li mb
(humerus + radius + metacarpal II), hindlimb (femur +
tibia + metatarsal III), and body length (from snout
tip to posterior extremity of pelvic girdle) are plotted
as percentages of the sum of forelimb, hindlimb,
and body lengths in S. aegyptiacus and other large
tetanuran theropods (data from Table 1). Blue zone
shows the range of forelimb length, from 7% (Tyran-
nosaurus) to 12% (Allosaurus). Hindlimb length
(red zone) ranges from 34% (Allosaurus)to19%
(Spinosaurus). Abbreviations: Ac, Acrocanthosaurus;
Al, Allosaurus;Sp,Spinosaurus;Su,Suchomimus;Ty,
Tyrannosaurus.
hind limb
forelimb
body
Ty
Ac
Al
10
60
20
Sp
Su
70
80
40
30
Fig. 4. Bone microstructure and dorsal spine form. (A) Mid-shaft thin section of the right femur of
S. aegyptiacus (FSAC-KK 11888). (B) Mid-shaft thin section of the right femur of Suchomimus tenerensis
(MNN GAD608). (C) Cross-sectional view of right manual II-1 phalanx of S. aegyptiacus (FSAC-KK 11888).
(D) Thin section of a dorsal neural spine (distal section) in S. aegyptiacus (FSAC-KK 11888). (E)Dorsal
vertebrae with tall neural spines and spinal tendons in a cleared and stained specimen of Trioceros
(Chamaeleo)cristatus (FMNH 19886). Abbreviations: cb, cancellous bone; ec, erosional cavities; Hb,
Haversian bone; mc, medullary cavity; ns, neural spine; pb, primary bone; sc, scapula; st, striae; te,
tendon of multisegment spinal muscle. Institutional abbreviations: FMNH, Field Museum of Natural
History. Scale bars, 2 cm in (A) and (C), 3 cm in (B), 5 mm in (D), and 1 cm in (E).
RESEARCH |REPORTS

shore birds (22), and interdigital webbing has
been reported in theropod dinosaurs (29).
Reduction of the pelvic girdle and hindlimb and
the concomitant enhancement of axial-powered
locomotion are common among semiaquatic
vertebrates. The flexibility of the tail and the
form of the neural spines in Spinosaurus suggest
tail-assisted swimming. Like extinct and extant
semiaquatic reptiles, Spinosaurus used lateral
undulation of the tail, in contrast to the vertical
axial undulation adopted repeatedly by semi-
aquatic mammals (20,21).
The dorsal “sail”in Spinosaurus, the tallest
axial structure documented among dinosaurs,
has been argued to be a thermoregulatory sur-
face, a muscle- or fat-lined hump (30), or a dis-
play structure. Stromer (1) drew an analogy to
the skin-covered neural spines of the crested
chameleon, Trioceros cristatus (Fig. 4E). As in
T. cristatus,thesailofSpinosaurus is centered
over the trunk (Fig. 2A). The shape and position-
ing of the spine are also similar, and the base of
the neural spine is expanded anteroposteriorly,
with edges marked by ligament scars (Fig. 2E).
In Trioceros, a tendon of multisegmental axial
musculature attaches to the expanded base of
the neural spine (Fig. 4E). The upper portion of
the spine has sharp anterior and posterior edges,
is marked by fine vertical striae (Figs. 2E and 4D),
and is spaced away from adjacent spines, un-
lik e the broader, contiguous, paddle-shaped dorsal
spines of other spinosaurids (13). The striated
surface, sharp edges, and dense, poorly vascular-
ized internal bone of the spines suggest that they
were wrapped snugly in skin and functioned as
a display structure that would have remained
visible while swimming.
REFERENCES AND NOTES
1. E. Stromer, Ahb. Königl. Bayer. Akad. Wissen. Math-Phys. Kl.28,
1–32 (1915).
2. E. Stromer, Abh. Königl. Bayer. Akad. Wissen. Math-Naturwissen.
Abt.22,1–79 (1934).
3. J. B. Smith, M. C. Lamanna, H. Mayr, K. J. Lacovara,
J. Paleontol. 80, 400–406 (2006).
4. W. Nothdurft, J. Smith, The Lost Dinosaurs of Egypt (Random
House, New York, 2002).
5. P. Taquet, D. Russell,C. R. Acad. Sci. Paris 299,347–353 (1998).
6. C. Dal Sasso, S. Maganuco, E. Buffetaut, M. A. Mendez, J. Vert.
Paleontol. 25, 888–896 (2005).
7. P. C. Sereno et al., Science 272, 986–991 (1996).
8. See the supplementary materials on Science Online.
9. R. Lavocat, in Comptes Rendus de la 19ème Congrès
Géologique International, Alger, 1952, session XII-3, 15 (1954),
pp. 65–68.
10. L. Mahler, J. Vert. Paleont. 25, 236–239 (2005).
11. D. A. Russell, Bull. Mus. Hist. Nat. Paris 18, 349–402 (1996).
12. B. McFeeters, M. J. Ryan, S. Hinic-Frlog, C. Schröder-Adams,
H. Sues, Can. J. Earth Sci. 50, 636–649 (2013).
13. P. C. Sereno et al., Science 282, 1298–1302 (1998).
14. A. J. Charig, A. C. Milner, Bull. Nat. Hist. Mus 53,11–70 (1997).
15. C. Brochu, J. Vert. Paleontol. Mem. 7,22 (suppl. 4), 1–138
(2002).
16. D. B. Leitch, K. C. Catania, J. Exp. Biol. 215, 4217–4230 (2012).
17. F. E. Novas, F. Dalla Vecchia, D. F. Pais, Rev. Mus. Argent.
Cien. Nat. 7, 167–175 (2005).
18. F. E. Fish, G. V. Lauder, Annu. Rev. Fluid Mech. 38, 193–224
(2006).
19. S. I. Madar, Adv. Vert. Paleobiol. 1, 353–378 (1998).
20. P. D. Gingerich, Paleobiology 29, 429–454 (2003).
21. F. E. Fish, IEEE J. Oceanic Eng. 29, 605–621 (2004).
22. A. Manegold, Acta Ornithol. 41,79–82 (2006).
23. E. Amson, C. de Muizon, M. Laurin, C. Argot, V. de Buffrénil,
Proc. Biol. Sci. 281, 20140192 (2014).
24. K. T. Bates, R. B. J. Benson, P. L. Falkingham, Paleobiology 38,
486–507 (2012).
25. S. M. Gatesy, M. Bäker, J. R. Hutchinson, Paleobiology 29,
535–544 (2009).
26. E. Stromer, Abh. Königl. Bayer. Akad. Wissen. Math.-Naturwissen.
Abt. 33,1–102 (1936).
27. A. J. Charig, A. C. Milner, Nature 324, 359–361 (1986).
28. E. J. Rayfield, A. C. Milner, V. B. Xuan, P. G. Young,
J. Vert. Paleontol. 27, 892–901 (2007).
29. M. L. Casanovas Cladellas et al., España Zub. Monogr. 5,
135–163 (1993).
30. J. B. Bailey, J. Paleontol. 71, 1124–1146 (1997).
ACKNOW LEDGMENT S
We thank C. Abraczinskas for final drafts of all text figures;
M. Auditore for discussions and drawings; T. Keillor, L. Conroy,
and E. Fitzgerald for image processing and modeling; R. Masek,
T. Keillor, E. Fitzgerald, and F. Bacchia for fossil preparation;
C. Straus, N. Gruszauskas, D. Klein, and the University of Chicago
Medical Center for computed tomographic scanning; M. Zilioli,
F. Marchesini, M. Pacini, E. Lamm, and P. Vignola for preparation of
histological samples; A. Di Marzio (Siemens Milano) and P. Biondetti
(Fondazione Ospedale Maggiore Istituto di Ricovero e Cura a
Carattere Scientifico, Milan) for computed tomography scanning and
rendering of MSNM V4047; and the Island Fund of the New York
Community Trust and National Geographic Society (grant
SP-13-12) for support of this research. N.I. was also supported by
NSF grant DBI-1062542. We also thank the embassy of the
Kingdom of Morocco in Washington, DC, for their continued
interest in this project. Skeletal measurements and geologic data
are included in the supplementary materials. The neotype is going to
be deposited at the Faculté des Sciences Aïn Chock (University of
Casablanca), Casablanca, Morocco.
SUPPLEMENTARY MATERIALS
www.sciencemag.org/content/345/6204/1613/suppl/DC1
Supplementary Text
Figs. S1 to S8
Tables S1 to S5
References (31–48)
15 July 2014; accepted 3 September 2014
10.1126/science.1258750
NEUROSCIENCE
A critical time window for dopamine
actions on the structural plasticity
of dendritic spines
Sho Yagishita,
1,2
Akiko Hayashi-Takagi,
1,2,3
Graham C.R. Ellis-Davies,
4
Hidetoshi Urakubo,
5
Shin Ishii,
5
Haruo Kasai
1,2
*
Animal behaviors are reinforced by subsequent rewards following within a narrow time
window. Such reward signals are primarily coded by dopamine, which modulates the
synaptic connections of medium spiny neurons in the striatum. The mechanisms of the
narrow timing detection, however, remain unknown. Here, we optically stimulated
dopaminergic and glutamatergic inputs separately and found that dopamine promoted
spine enlargement only during a narrow time window (0.3 to 2 seconds) after the
glutamatergic inputs. The temporal contingency was detected by rapid regulation of
adenosine 3′,5′-cyclic monophosphate in thin distal dendrites, in which protein-kinase
A was activated only within the time window because of a high phosphodiesterase
activity. Thus, we describe a molecular basis of reinforcement plasticity at the level of
single dendritic spines.
Animal behaviors are reinforced only when
rewarded shortly after a motor or sensory
event (1,2). The neocortex, hippocampus,
and amygdala process the sensorimotor
signals and send glutamatergic synaptic out-
put to the striatum (3), where connections can
be modified by Hebbian learning mechanisms,
such as spike-timing-dependent plasticity (STDP)
(4). Animals learn to associate the sensorimotor
signals with subsequent rewards through rein-
forcement of the neuronal circuits involving do-
pamine (5–7). Despite its importance, this narrow
timing detection has never been demonstrated at
the cellular level and might be ascribed to neural
network properties (6,8).
Dendritic spine morphology is correlated with
spine function (9), and dendritic spines enlarge
during long-term potentiation in the cortices
(10–12). We examined the effects of dopamine
on the structural plasticity in striatal medium
spiny neurons (MSNs). Results show that do-
pamine affected spine structural plasticity in a
narrow time window consistent with behav-
ioral conditioning (5). Functional imaging revealed
the molecular interrelationships between the re-
inforcement and Hebbian plasticity.
We investigated dopamine actions on glutama-
tergic synapses on MSNs using optogenetics and
1616 26 SEPTEMBER 2014 •VOL 345 ISSUE 6204 sciencemag.org SCIENCE
1
Laboratory of Structural Physiology, Center for Disease
Biology and Integrative Medicine, Faculty of Medicine, The
University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan.
2
Core Research for Evolutional Science and Technology,
Japan Science and Technology Agency, Japan Science and
Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama
332-0012, Japan.
3
Precursory Research for Embryonic
Science and Technology, Japan Science and Technology
Agency, Japan Science and Technology Agency, 4-1-8
Honcho, Kawaguchi, Saitama 332-0012, Japan.
4
Department
of Neuroscience, Mount Sinai School of Medicine, New York,
NY 10029, USA.
5
Integrated Systems Biology Laboratory,
Department of Systems Science, Graduate School of
Informatics, Kyoto University, Sakyo-ku, Kyoto 606-8501,
Japan.
*Corresponding author. E-mail: hkasai@m.u-tokyo.ac.jp
RESEARCH |REPORTS