CNS Evolution: New Insight from the Mud
Whether the highly centralised nervous systems of chordates and protostomes
arose from a common ancestral precursor or independently has been a long-
standing debate. Now, analysis of neural gene expression in an evolutionarily
important chordate outgroup — the sand-dwelling, hemichordate acorn
worms — reveals the presence of a central and peripheral nervous system,
suggesting a common origin of central nervous systems.
E`lia Benito-Gutie ´rrez
and Detlev Arendt*
The origin of the chordate central
nervous system (CNS) has remained
a controversial topic in evolutionary
biology. One major unsolved question
is how it is related to the CNS of other
animal groups (Figure 1). Did the
chordate CNS originate independently
or from a shared ancestral CNS?
Although concentrations of neurons
are found in all major animal phyla,
there are some basal groups in which
the nervous system is only scarcely
centralised or not centralised at all. In
addition, these neuron concentrations
are found at different sites in different
animal groups: in chordates, neurons
are concentrated at the dorsal side of
the body, while in non-chordate
invertebrates a strand of centralised
neurons is found on the ventral side.
This is why many authors believe that
the chordate CNS evolved separately
in the chordate lineage. One of these
authors, Romer , proposed that the
chordate ancestors were animals with
no or only a rudimentary CNS
resembling that of today’s
pterobranchs and ascidians. Another
author, Garstang , proposed that the
chordate neural tube evolved by the
dorsal fusion of ciliary bands, like those
found in echinoderm or enteropneust
larvae. However, there is an alternative
view. Inspired by Geoffroy St. Hilaire’s
‘unite ´ de plan’ , and despite the
opposite locations, comparative
anatomist Anton Dohrn  proposed
that the chordate CNS was
homologous to that of protostome
annelids and arthropods. He assumed
that vertebrates evolved from
worm-shaped, annelid-like ancestors
that turned upside-down during their
evolution such that the dorsal and
ventral sides became inverted (DV-axis
inversion). In contrast to Romer’s
scenario, Dohrn considered ascidians
to be secondarily modified forms with
a simplified larval CNS. Dohrn’s
proposed homology between the
dorsal and ventral CNS of vertebrates
with time, but a paper published by
Current Biology  might well yield new
perspectives on this old idea.
The revival of Dohrn’s ideas could
have been already anticipated based
on gene expression data from different
neuronal cell-type specification, which
revealed striking similarities between
the vertebrate, insect and annelid
nervous systems. The notion of the
DV-axis inversion found great support
from this pool of data, particularly after
the comparison of DV patterning genes
between flies and frogs, most notably
BMP, which is expressed ventrally in
frogs and dorsally in flies [6–9]. This
observation is corroborated by
inverted left-right patterning, as
revealed by left-sided Pitx and Nodal
Figure 1. Revised evolutionary tree of deuterostome groups.
This simplified phylogenetic tree of deuterostome groups discussed in the text is based
on recent molecular phylogenies [12,13]. One species of each branch has been selected for
illustration, except for hemichordates, where the tree has been expanded to illustrate that
pterobranchs have evolved within enteropneusts. Note that ascidians are the closest living
relatives of the vertebrates.
Current Biology Vol 19 No 15
expression in chordates as opposed to
sea urchin and starfish larvae where
these genes are expressed on the right
side of the body [10,11]. Also, refined
deuterostome phylogeny now
indicates that ascidians are indeed
secondarily simplified  and that
the sessile pterobranchs stem from
free-living acorn worms , safely
ruling out that pterobranch- or
ascidian-like organisms were among
the chordate ancestors. These findings
eliminated Romer’s influential theory
from the list of plausible explanations
for the origin of chordates.
However, these findings were not
enough to stop Dohrn’s views being
disputed, as there is still a major
objection regarding the common origin
for the chordate, arthropod and annelid
CNS. If they were indeed homologous,
strands of centralised neurons that are
similar tothe chordate CNS should also
exist in the more basal deuterostome
groups, such as echinoderms or
hemichordates (Figure 1). Clearly, the
echinoderm CNS is very divergent
and bears no resemblance to that of
the chordates, but how about that of
the hemichordates? In juveniles of the
enteropneust worm Saccoglossus
kowalevski — a hemichordate
species commonly known as ‘acorn
worms’ — the scattered epithelial
expression of neural markers and
nervous system patterning genes
seemed to suggest that these
worm-shaped animals develop only a
diffuse nervous system, with neurons
distributed over the entire epidermis
instead of being concentrated in a
nerve cord . This prompted the
idea of ancient ‘skin brains’ , which
proposes a non-centralised ancestral
nervous system, where scattered
neurons within the ectoderm would
have evolved into a proper internal
CNS independently in the more
advanced protostome and
The new work by Brunet and
collaborators  is now adding a fresh
brushstroke to the picture by showing
that adult acorn worms — unlike
juveniles — actually possess a
fully-formed CNS, which features what
could be interpreted as a transition
dorsal chordate CNS (Figure 2) . By
analysing the expression patterns
of genes in the CNS, the authors
demonstrate that in adult Ptychodera
flava and S. kowalevski worms a ventral
and a dorsal strand of centralised
neurons are present that merge
anterior-dorsally at the level of the
worm’s collar. Both species display
a tripartite body that consists of an
they use for burrowing in the sand,
followed by a short thick collar and
a very long trunk. Further anteriorly, the
collar cord extends into a neural plate-
like concentration of neurons in the
proboscis stem, which harbors the
thickest layers of neurons and
underlying axons, and which extends
ventrally to fully encircle the proboscis
stem just behind the proboscis
(Figure 2). In other body regions,
are interpreted as a peripheral nervous
system (PNS). In line with a clear
separation into a CNS and a PNS, CNS
markers such as hb9 and Drg,
expressed by somatic motor and
sensory neurons of the dorsal root
ganglia, or VACht, expressed by
cholinergic neurons, are only found
within the cords. Otherwise, serotonin
expressing cells are only found outside
the cords. This centralisation of the
nervous system is apparent from the
earliest stages of metamorphosis
between larva and adult. The authors
conclude that the previously described
‘diffuse’ nervous system present at
earlier developmental stages in
Figure 2. Comparative anatomy of the CNS.
Schematic comparison of centralised nerve cords in an (A) annelid, (B) enteropneust  and
(C) chordate (note that this schematic is dorsoventrally inverted). Superficial and internalised
portions of the CNS are depicted in bright and dark yellow, respectively. Left panels: Lateral
views. Right panels: Animals cut open along the midline (black dashed line; dorsal in the
annelid and enteropneust, ventral in the chordate) and flattened. Red dashed line indicates
the midline on the opposite body side, internalised to form the floorplate of the neural tube
in the chordate. A grey dotted line demarcates possibly homologous CNS strands, as initially
put forward by Nu ¨bler-Jung and Arendt . In the chordate, the asterisks indicate the position
of the ancient (now dorsal) mouth at the bottom of the brain and the arrow depicts where
a modified gill slit will form the new mouth. Note that the ventral neurogenic strand along
the ventral midline will disintegrate (yellow dashed line). cc: collar cord; ct: circumesophageal
tract; dc: dorsal cord; np: neural plate; vc: ventral cord; vs: ventral strand.
Saccoglossusisa transitoryfeature that Download full-text
may correspond to the larval nervous
system of other enteropneusts.
This work opens up new avenues of
comparative CNS research. Clearly,
these data, together with the inverted
BMP patterning in acorn worms ,
are consistent with the view that the
neural plate of the proboscis stem, the
collar cord, the circumesophageal tract
and ventral cord together correspond
to the chordate CNS as a whole and to
the CNS of other invertebrates where
inversion has not occurred, as
proposed earlier . Yet, a more
detailed comparative picture still
of neuron types in enteropneusts and
of their differential distribution is rather
scarce and will require a much closer
inspection of a larger number of
neuronal markers. Also, a link with the
detailed orthologous gene expression
data in vertebrates, similar to that
described for Saccoglossus , will
have to be established. Only then will it
be possible to firmly homologise any
portion of the enteropneust CNS with
that of chordates or even annelids
or arthropods. As a start, the
concentration of GABAergic neurons in
the proboscis stem, apparently located
territory , may correspond to
 and in the annelid forebrain (R.
Tomer and D.A., unpublished results).
If indeed the CNS represents ancient
bilaterian heritage and vertebrates
inverted their DV axis, one prominent
problem still remains, as discussed by
Brunet and colleagues  (Figure 2):
CNS are located exactly where the
chordates would have evolved their
(new) mouth — on their new ventral
(formerly dorsal) body side now facing
the substrate. How can we reconcile
this? Dohrn  had suggested that the
new chordate mouth evolved from the
ventral relocation of gill slits (Figure 2),
as is suggested by the amphioxus
mouth, which is thought to represent
a ventrally shifted gill slit  — hence
the name Branchiostoma, meaning ‘gill
slit mouth’. Interestingly, a strand of
neurogenic tissue has recently been
discovered along the amphioxus
ventral midline giving rise to scattered
neuronal precursors that further
migrate dorsally  before the mouth
takes its place. Future molecular
comparisons of the neuronal cell types
involved will reveal whether this
transitory neurogenic ventral strand
in amphioxus might be related to the
dorsal strand of neurons in acorn
worms or rather represents an
independent acquisition that either
could be an apomorphy or could be
related to a second wave of
centralisation: namely the dorsal
reunion of a primitive neuronal
population with placode-neural crest
derived from mud- and sand-living
acorn worms, comparative research on
chordate nervous system evolution
appears more exciting than ever.
1. Romer, A.S. (1972). The vertebrate as a dual
animal - somatic and visceral. Evol. Biol. 6,
2. Garstang, W. (1894). Preliminary note on a new
theory of the ancestry of the Chordata. Zool.
Anz. 17, 122–125.
3. Geoffroy St.-Hilaire, E. (1822). Conside ´rations
ge ´ne ´rales sur la verte `bre. Me ´m. Mus. Hist. Nat.
4. Dohrn, A. (1875). Der Ursprung der Wirbelthiere
und das Princip des Functionswechsels
(Leipzig: Verlag von Wilhelm Engelmann).
5. Nomaksteinsky, M., Ro ¨ttinger, E., Dufour, H.D.,
Chettouh, Z., Lowe, C.J., Martindale, M.Q., and
Brunet, J.-F. (2009). Centralization of the
deuterostome nervous system predates
chordates. Curr. Biol. 19, 1264–1269.
6. Arendt, D., and Nu ¨bler-Jung, K. (1994).
Inversion of dorsoventral axis? Nature 371, 26.
7. Arendt, D., and Nu ¨bler-Jung, K. (1999).
Comparison of early nerve cord development
in insects and vertebrates. Development
8. Denes, A.S., Jekely, G., Steinmetz, P.R.,
Raible, F., Snyman, H., Prud’homme, B.,
Ferrier, D.E., Balavoine, G., and Arendt, D.
(2007). Molecular architecture of annelid nerve
centralization in bilateria. Cell 129, 277–288.
9. Holley, S.A., Jackson, P.D., Sasai, Y., Lu, B.,
De Robertis, E.M., Hoffmann, F.M., and
Ferguson, E.L. (1995). A conserved system for
dorsal-ventral patterning in insects and
vertebrates involving sog and chordin. Nature
10. Duboc, V., Rottinger, E., Lapraz, F.,
Besnardeau, L., and Lepage, T. (2005).
Left-right asymmetry in the sea urchin embryo
is regulated by nodal signaling on the right side.
Dev. Cell 9, 147–158.
11. Hibino, T., Nishino, A., and Amemiya, S. (2006).
Phylogenetic correspondence of the body axes
in bilaterians is revealed by the right-sided
expression of Pitx genes in echinoderm larvae.
Dev. Growth Differ. 48, 587–595.
12. Delsuc, F., Brinkmann, H., Chourrout, D., and
Philippe, H. (2006). Tunicates and not
cephalochordates are the closest living
relatives of vertebrates. Nature 439, 965–968.
13. Cannon, J.T., Rychel, A.L., Eccleston, H.,
Halanych, K.M., and Swalla, B.J. (2009).
Molecular phylogeny of hemichordata, with
updated status of deep-sea enteropneusts.
Mol. Phylogenet. Evol. 52, 17–24.
Stange-Thomann, N., Gruber, C.E., Gerhart, J.,
and Kirschner, M. (2003). Anteroposterior
patterning in Hemichordates and the origins of
the chordate nervous system. Cell 113, 853–865.
15. Holland, N.D. (2003). Early central nervous
system evolution: an era of skin brains? Nat.
Rev. Neurosci. 4, 1–11.
16. Nu ¨bler-Jung, K., and Arendt, D. (1996).
Enteropneusts and chordate evolution.
Curr. Biol. 6, 352–353.
17. Lowe, C.J., Terasaki, M., Wu, M.,
Freeman, R.M., Jr., Runft, L., Kwan, K.,
Haigo, S., Aronowicz, J., Lander, E., Gruber, C.,
et al. (2006). Dorsoventral patterning in
hemichordates: insights into early chordate
evolution. PLoS Biol. 4, e291.
18. Sugino, K., Hempel, C.M., Miller, M.N.,
Hattox, A.M., Shapiro, P., Wu, C., Huang, Z.J.,
and Nelson, S.B. (2006). Molecular taxonomy of
major neuronal classes in the adult mouse
forebrain. Nat. Neurosci. 9, 99–107.
19. Conklin, E.G. (1932). The embryology of
amphioxus. J. Morph. 54, 69–118.
20. Benito-Gutierrez, E., Nake, C., Llovera, M.,
Comella, J.X., and Garcia-Fernandez, J. (2005).
The single AmphiTrk receptor highlights
increased complexity of neurotrophin signalling
in vertebrates and suggests an early role in
developing sensory neuroepidermal cells.
Development 132, 2191–2202.
Developmental Biology Unit, European
Molecular Biology Laboratory, D-69117
Cancer: CINful Centrosomes
The regulation of centrosome number is lost in many tumors and the presence
of extra centrosomes correlates with chromosomal instability. Recent work
now reveals how extra centrosomes cause chromosome mis-segregation
in tumor cells.
Samuel F. Bakhoum
and Duane A. Compton*
Centrosomes are pivotal organizers
of the microtubule cytoskeleton and
their duplication and inheritance is
strictly controlled during the cell cycle
in a manner that parallels genome
duplication . This control is lost
in many cancer cells, making the
presence of extra centrosomes
a discernible feature of many tumors
. This defect has long been
associated with aneuploidy in cancer
and it is postulated that additional
centrosomes induce chromosome
mis-segregation, which then
contributes to tumorigenesis [3–6].
Current Biology Vol 19 No 15