Ancient Evolutionary Origin of the
Neural Crest Gene Regulatory Network
Tatjana Sauka-Spengler,1Daniel Meulemans,1Matthew Jones,1and Marianne Bronner-Fraser1,*
1Division of Biology 139-74, California Institute of Technology, Pasadena, CA 91125, USA
The vertebrate neural crest migrates from its
origin, the neural plate border, to form diverse
derivatives. We previously hypothesized that a
neural crest gene regulatory network (NC-GRN)
guides neural crest formation. Here, we investi-
gate when during evolution this hypothetical
network emerged by analyzing neural crest for-
mation in lamprey, a basal extant vertebrate.
We identify 50NC-GRN homologs and usemor-
pholinos to demonstrate a critical role for eight
transcriptional regulators. The results reveal
conservation in deployment of upstream fac-
tors, suggesting that proximal portions of the
network arose early in vertebrate evolution
and have been conserved for >500 million
years. We found biphasic expression of neural
crest specifiers and differences in deployment
of some specifiers and effectors expected to
confer species-specific properties. By testing
the collective expression and function of neural
crest genes in a single, basal vertebrate, we
reveal the ground state of the NC-GRN and re-
solve ambiguities between model organisms.
The neural crest is a migratory, multipotent cell population
that gives rise to diverse derivatives, including melano-
cytes, sensory and autonomic neurons, and mineralized
matrices like bone and dentine (Knecht and Bronner-
Fraser, 2002). These features represent many of the defin-
ing characters of vertebrates. Neural crest cells are spec-
ified at the edges of the forming central nervous system,
from which they delaminate, migrate extensively, and dif-
ferentiate. Appearance of the neural crest has been linked
to the evolution of predation in vertebrates, as its deriva-
tives form much of the jaw and peripheral nervous system
(Gans and Northcutt, 1983). Consistent with this notion,
nonvertebrate chordates are sessile filter-feeders which
lack definitive neural crest. Recent work in urochordates
suggests that the neural crest did not evolve de novo,
but arose from a population of neural tube cells possess-
ing a subset of their molecular and migratory properties
(Jeffery et al., 2004). Understanding the evolution of defin-
itive neural crest cells is critical to understanding verte-
brate origins; however, the gene regulatory changes that
resulted in this vertebrate novelty remain a mystery.
Based upon data from Xenopus, zebrafish, chick, and
mouse, we have hypothesized that a gene regulatory net-
work underlies formation of the neural crest in jawed ver-
tebrates (gnathostomes), taking into account their unique
site of origin at the neural plate border, their extensive
migratory ability, and their multipotency (Meulemans and
Bronner-Fraser, 2004; Sauka-Spengler and Bronner-
Fraser, 2006). This putative neural crest gene regulatory
network (NC-GRN) describes a set of interacting signals,
that confer properties like multipotency and migratory ca-
pacity to nascent neural crest cells. The proposed NC-
GRN begins with inductive signals (e.g., Wnt, BMP, and
FGF) that establish the neural plate border, upregulating
transcription factors such as Msx1/2, Pax3/7, and Zic.
These neural plate border specifiers in turn upregulate
neural crest specifier genes, such as Slug/Snail, FoxD3,
and members of the SoxE family. At lower levels of the
NC-GRN, the neural crest specifiers turn on specific
downstream effector genes that render the neural crest
migratory and multipotent. This proposed NC-GRN has
proven useful in organizing experimental approaches in
a variety of species.
Critical unresolved issues concerning the evolution of
and protochordates suggest that the NC-GRN evolved
sometime during the approximately 200 million years
between the divergence of protochordates and the ap-
pearance of ray-finned bony fish. However, it is unknown
when during this period the NC-GRN evolved, or if this
of the evolutionary history of the NC-GRN via comparative
studies offers direct insights into the genetic bases of ver-
tebrate origins and can help solve some of these unre-
Lamprey and hagfish are jawless vertebrates (agna-
thans) and thus constitute the most basal vertebrate
group, the cyclostomes, bearing many characteristics of
the ancestral vertebrate. Although it is clear that lamprey
has a bona fide neural crest, the pathways of migration
in the hindbrain region are quite different than those in
other vertebrates (Horigome et al., 1999; McCauley and
Bronner-Fraser, 2003), and lampreys lack several impor-
tant neural crest derivatives, including sympathetic chain
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 405
ganglia (Butler and Hodos, 1996; Johnels, 1956). This rai-
ses the intriguing question of whether or not the molecular
mechanisms of neural crest specification are similar or
different between jawed and jawless vertebrates.
thetical NC-GRN in one of the most basal extant verte-
brates, the lamprey. We have isolated homologs of each
class of patterning molecules in the putative NC-GRN:
signaling molecules, neural plate border specifiers, neural
crest specifiers, and neural crest effector genes. Impor-
of the transcription factors expressed at the neural plate
border or presumptive neural crest in the lamprey using
morpholino-mediated protein knockdown. By examining
the resultant phenotypic changes using a wide range of
neural crest marker genes, we present, to our knowledge,
the first collective analysis of the function of a set of tran-
scription factors within a putative NC-GRN in a single ver-
tebrate. Comparison with published data from higher ver-
tebrates suggests that the upstream core of the network
was fixed at the base of the vertebrate lineage more than
the regulatory network, appear to be more divergent and
less evolutionarily constrained in their deployment.
To examine the deployment and role of lamprey genes in
neural crest development, we constructed an arrayed di-
rectional cDNA library prepared from lamprey embryos
2–24 days of age. We performed extensive low-stringency
of genes involved in neural crest formation in gnathos-
tomes. A list of the 50 genes identified here is provided in
Table S2 (see the Supplemental Data available with this
article online). We analyzed their expression patterns by
in situ hybridization at stages encompassing neural crest
inductive (E4), premigratory (E5–E6), and early (E6.5) mi-
gratory phases, as well as neural crest differentiation (later
larvae). Then, the function of a subset of transcription fac-
tors was tested using morpholino-mediated knockdown.
Inlamprey, theneural tube formsbysecondaryneurula-
tion, whereby a solid rod-like structure transforms into
of gastrulation (E4), the ectoderm flattens in the dorsum of
neural groove that is replaced by the neural ridge. At E5,
the neural rod elevates, gradually detaching from the
epithelium, concomitant with appearance of neural crest
precursors. The head morphologically extends, becoming
visible at E6 during cavitation, when the neural crest
primordia appear as bulges on the dorsal neural tube.
Lamprey Homologs of Patterning Signals Involved
in Neural Crest Formation
In gnathostomes, neural crest induction is mediated by
a varietyof signals, includingWnt,BMP,FGFs, and Notch/
Delta. All model vertebrates examined express compo-
nents of these signaling pathways, though functional stud-
ies imply some differences in their relative importance be-
tween species (review in Meulemans and Bronner-Fraser
). To examine if gene deployment is conserved in
logs (Wnt4–9) (Figure 1). At the onset of neurulation at E4,
Wnt6 is expressed in the neural plate border and through-
out the presumptive neural plate (Figure 1E); subsequently
tube(E5),where neuralcrestprecursors first appearahalf-
day later (Figures 1G and 1I). Expression is similar to that
of gnathostome Wnt1 (Hollyday et al., 1995; Saint-Jeannet
et al., 1997). Paralleling avian Wnt6 (Garcia-Castro et al.,
2002), lamprey Wnt8 is expressed in the nonneural ecto-
derm, adjacent to the neural crest progenitors in the edges
of the neural rod (Figures 1H and 1J). As in Xenopus and
zebrafish (LaBonne and Bronner-Fraser, 1998; Liem
et al., 1995; Nguyen et al., 1998) BMP2/4A is expressed
neurula (E4), as the neural groove and folds form, but is
more prominent in the neural plate border. In contrast,
pressed in a punctate pattern in the nonneural ectoderm,
whereas Notch1 is ubiquitously expressed throughout the
embryo, with slightly higher levels in the nonneural ecto-
derm and neural plate border (Figures 1C and 1D).
Lamprey Homologs of Neural Plate Border
Neural plate border specifiers are transcription factors
(Msx1/2, Pax3/7, Dlx3/5, and Zic) expressed in patterns
that encompass the border region plus either the nonneu-
ral or neural ectoderm. Their overlapping patterns at the
neural plate border are thought to uniquely define this re-
is expressed in the early embryo and homologous to
LjMsxA, previously shown to be in the upper and lower
lip in the late lamprey larvae (Shigetani et al., 2002). In
the early neurulae (E4), MsxA is expressed at the neural
plate border and in the ectoderm (Figures 2A, 2E, and
2F), where it persists through midneurulae (E4.5), but it is
in a rostrocaudal wave starting at E5. A Zic homolog is ex-
pressed in the early neurulae (E4), throughout the neural
plate, and overlaps with MsxA at the neural plate border
(Figures 2B and 2F). The ZicA- and MsxA-positive border
domain also expresses Pax3/7 and abuts Dlx-positive
nonneural ectoderm (Figures 2C, 2D, 2F, and 2G). Both
MsxA and DlxA/DlxB genes become downregulated in
and dorsal neural ridge regions throughout neurulation.
Later, all are coexpressed in the dorsal neural tube, but
not in migrating neural crest cells (Figures 4A–4C). These
expression patterns are highly reminiscent of those previ-
tion to their expression in the neural plate border and neu-
ral tube, ZicA and Pax3/7 are expressed in somites
(Figures 4B and 4C, Figure 6).
Lamprey Neural Crest Gene Regulatory Network
406 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Homologs of Neural Crest Specifier
In the hypothetical NC-GRN, neural plate border genes
are thought to upregulate neural crest specifiers like
c-Myc, AP2, Id, Snail, FoxD3, and SoxEs, collectively ex-
pressed bybona fide neural crest progenitors and thought
to confer migratory ability and specify cell fate (see Dis-
cussion). Neural crest specifier genes differ from border
specifiers in that they tend to be retained in the migrating
neural crest population and are thought to function after
formation of the nascent neural plate border (reviewed in
Meulemans and Bronner-Fraser ). Interestingly, we
noted that lamprey neural crest specifier genes exhibit
two modes of expression: there are those expressed early
in the neural plate border (E4) and those first expressed in
E5.5–E6 neurulae, in bona fide neural crest progenitors.
Figure 1. Expression of Signaling Molecules in Early Lamprey Neurula
Dorsalview of E4neurula; anterioris to thetop. (A) BMP2/4A is expressed athigherlevels inthe border territory abutting theopen neural plate (np).(B)
BMP2/4B is in the neural plate and the border. (C) DeltaA is in the nonneural ectoderm surrounding the neural plate. (D) Notch1 is expressed through-
out the embryo, but expression is highest in nonneural ectoderm and neural plate border. (E) At E4, Wnt6 is expressed in the border and neural plate.
(F) At E4, Wnt8 is expressed at low levels in nonneural ectoderm. (G) By E5, Wnt6 is confined to the dorsolateral edges of the neural tube (nt; black
arrowheads), withfaint expression in theoverlying ectoderm (ec),as seenintransversesection (K).(H) By E5Wnt8 isin thenonneural ectoderm (white
arrowhead) adjacent to neural crest progenitors in the dorsolateral neural tube, observed the transverse section (N). (I) Cross-section through an
E4 embryo showing Bmp2/4A expression. The early gastrula has large cells with significant amounts of yolk; the mRNA tends to be throughout
the cytoplasm, but it also concentrated in and around the nuclei. This leads to the patchy staining in whole-mount. (J) Same section as (I) overlaid with
DAPI. (L) A cross-section of the embryo in (E) shows strong punctate staining that largely overlaps with nuclear DAPI label, as seen in the overlay (M).
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 407
Lamprey Neural Crest Gene Regulatory Network
Early Neural Crest Specifiers
Several neural crest specifiers are observed in the neural
plate border, initiating expression shortly after, and then
overlapping with, the border specifiers, such as MsxA,
ZicA, and Pax3/7 (Figure 3). For example, AP2 is observed
in the nonneural ectoderm plus neural plate border at E4,
similar to other vertebrates (Figures 3A, 3F, 3H, and 3J).
Later, AP2 is downregulated in the ectoderm but remains
in the dorsal neural tube and migrating neural crest
(Figure 4D) (Meulemans and Bronner-Fraser, 2002). Initial
expression of Id at E4 is similar to that of AP2 in the neural
Later, it becomes restricted to the dorsal neural tube and
migrating neural crest cells (Figure 4E; Meulemans et al.,
2003). Early onset of the expression of n-Myc and Snail
is confined to the neural plate, encompassing the neural
plate border (Figures 3C, 3D, and 3I). n-Myc is maintained
in the dorsal neural tube and in neural crest proper
(Figure 4F), whereas Snail is upregulated and expressed
broadly in many tissues including the dorsal neural tube
and mesoderm (data not shown); similar to its hagfish ho-
molog, it is not specific to nascent neural crest (Ota et al.,
2007). In contrast to neural crest specifiers, SoxB1 marks
the neural plate but is not observed within the neural plate
border region (Figure 2G and Figure 3E).
Late Neural Crest Specifiers
As the neural rod cavitates to form a neural tube at E6,
lamprey homologs of the neural crest specifiers FoxD3
(FoxD-A) and Sox9/10 (SoxE1, SoxE2, and SoxE3) are up-
or migrating neural crest cells in the head (SoxE2) and
trunk (SoxE2 and SoxE3) (Figures 4G–4I). Subsequently,
they are differentially maintained in different neural crest
derivatives (Figure 8 and data not shown). In late neurulae,
these factors are coexpressed with n-Myc, AP2, and Id in
streams of cranial and early migrating trunk neural crest
cells. Their deployment gradually becomes restricted to
particular neural crest subpopulations; e.g., FoxD-A
strongly labels cranial ganglia in the head and dorsal
root ganglia in the trunk at E10–E12, but is downregulated
in chondrogenic cranial neural crest (data not shown). In
contrast, lamprey n-Myc persists in pharyngeal neural
crest where it is coexpressed with SoxE genes and homo-
logs of Col2a1 (Figures 5C and 5F) (McCauley and Bron-
ner-Fraser, 2006; Zhang et al., 2006).
Two genes that function as neural crest specifiers in
jawed vertebrates, TwistA and Ets1a, are expressed only
in late-migratory and postmigratory cranial neural crest
(Figures 4K–4M). This contrasts with the early onset of
Figure 2. Expression of Neural Plate Border Specifiers in E4 Neurula
Dorsal (A–F) and lateral (G) views of the open neural plate, with anterior toward the top. (A) MsxA is expressed at the neural plate border and in the
nonneural ectoderm. (B) ZicA is present throughout the neural plate and border, where it overlaps with MsxA (E). (C) Pax3/7 marks distinct regions of
dorsal ectoderm and overlaps with MsxA and ZicA at the neural plate edges (F). DlxB (D) and DlxA (G) in the ventrolateral ectoderm abuts, but is
excluded from, the neural plate border (white arrowheads) and SoxB1-expressing neural plate proper.
408 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Neural Crest Gene Regulatory Network
cells in other vertebrates (Remy and Baltzinger, 2000;
Tahtakran and Selleck, 2003) and the early expression of
Twist-1 in delaminating neural crest in Xenopus and ze-
brafish (Hopwood et al., 1989; Linker et al., 2000). The
conspicuous absence of TwistA and Ets1a expression
from premigratory and early migrating neural crest may
be due to their late deployment in lamprey compared
with other vertebrates. Alternatively, there may be other
unidentified homologs that play an early role in neural
crest specification in lamprey. To address the latter possi-
bility, we performed additional library screens, as well as
a detailed survey of the draft Petromyzon genome
sequence, by scanning the Trace Archive database
BlastSearch). This permitted identification of two addi-
tional Ets1 and three additional Twist homologs. Ets1a
and Ets1b are first detected in the neural crest lineage in
the postmigratory cells within the forming branchial
as E4.5 in hematopoietic and endothelial precursors that
populate the peripheral yolk surface and subsequently
coalesce to form the vascular system (Figure 4O and
Figure S1). Similarly, none of the Twist homologs are
expressed in the premigratory or early migrating neural
Figure 3. Expression of Early Neural Crest Specifiers
Initial expression of early crest specifiers overlaps with that of neural plate border specifiers within the border territory. (A) AP2 is observed in the non-
neural ectoderm plus neural plate border at E4–E4.5 (white arrowheads), where it overlaps with border specifiers such as ZicA (H), but not with the
neural marker SoxB1 (F). (B) Id is found at lower levels in the nonneural ectoderm, but is strongly upregulated at the border, where it overlaps with
MsxA (G). Early onset of SnailA at E4 (C) and n-Myc at E4.5 (D) is confined to the open neural plate and border, which is a much broader range
than that of SoxB1 expression (E). (I) n-Myc expression at the border abuts the DlxB-expressing nonneural ectoderm expression of, and overlaps
with, the AP2 transcripts (J). (K) Cross-section through an embryo showing SnailA transcripts distributed throughout the cytoplasm of large yolky
cells, but more concentrated aroundnuclei, leading to thepatchy appearance ofsignal observed inwhole-mount. (L)The same section as (K) overlaid
with DAPI. (M) A thick cross-section (?16 mm) through embryo processed by double in situ hybridization with MsxA and Id probes; white arrowheads
indicate transcript overlap in the neural plate border. (N) The same section as (M) overlaid with DAPI.
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 409
Lamprey Neural Crest Gene Regulatory Network
Figure 4. A. Expression of Neural Plate Border and Neural Crest Specifiers in Lamprey E6–E6.5 Neurula and Twist and Ets Homo-
logues in Late Migrating Neural Crest Cells
Lateral view of the embryo, anterior to the right. (A–C) The border specifiers MsxA, ZicA, and Pax3/7 are restricted to the dorsal neural tube after the
neural rod condenses. (D–F) Early neural crest specifiers AP2, Id, and n-Myc are in the dorsal neural tube and migrating neural crest. Late neural crest
specifiers are first upregulated in the premigratory (SoxE1, [G]; FoxD-A, [I]), migrating neural crest (SoxE2, [H]), or both. (J) Schematic cross-sections
summarizing neural plate border and neural crest specifier expression in E4 and E6.5 neurulae. DlxC, a paralog of DlxB, is expressed in epidermal
ectoderm after neural tube closure. Late onset of expression of the transcription factors TwistA (K and L) and Ets1a (M) and Ets1b (N and O), two
Ets-1 homologs, in the postmigratory neural crest settling within branchial arches is shown. (L) TwistA transcripts are confined to the neural-crest-
(nc) derived portion of branchial arches (dotted line in the frontal section through the embryo in [K]). While no Ets1 homolog is expressed in premi-
gratory or early migrating crest (Figure S1), both Ets1a and Ets1b share expression in postmigratory branchial arch crest (M, N, and O) and forming
410 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Neural Crest Gene Regulatory Network
crest; instead, both TwistA and TwistB are expressed in
the crest lining the branchial arches (Figures 4K and 4L
and Figure S1). TwistC and TwistD are only found in mes-
enchymal populations that will give rise to cartilage of
the upper and lower lip, with TwistC appearing slightly
earlier, within the anterior ridge of the mandibular arch.
cranial placodes (black arrow, [M]) and epibranchial placodes (white asterisk, [M] and [N]). Only Ets1a transcripts are found in forming dorsal root
ganglia (black arrowheads, [M]), whereas Ets1b is expressed in endothelial and hematopoietic precursors that populate the peripheral yolk surface
(white arrowheads, [O]) and the interior of forming nephrotomes (white arrow, [O]).
Figure 5. Expression of Neural Crest Effector Genes
A type II Cadherin homolog (Cad IIA) is expressed in premigratory and early migrating neural crest (A), as well as in condensing cranial ganglia (D),
where it overlaps with the lamprey Brn3b (G) and Neurogenin (NgnA, [J]), found in both cranial and epibranchial ganglia, in the frontal section through
the cephalic region as indicated (M). Signaling receptors Robo (B) and Neuropilin2 (Npn2, [E]) are expressed in migrating and postmigratory branchial
arch neural crest (nc; arrow on the frontal section through the branchialbasket) (K). Conversely, Npn2 ligand Sema3 is expressed inmesodermal core
ofthearches (H)andtheectodermalpits(ec),asseeninfrontalsection(N).(I)Type ICadherinhomolog(Cad IA)isexcluded frombranchialarchneural
crest (nc), but is expressed in the dorsal aspect of the branchial mesoderm (m) (dotted line and arrow in the frontal section [O]). Col2a1 expression
starts at E10–E11 (C), and by E12 Col2a1 is strongly expressed by the entire cephalic ectoderm, with the exception of the branchial arches (F). Within
the branchial arches, Col2a1 is expressed in both the outer and newly forming inner posterior portion of the arch neural crest (nc) (arrowheads in fron-
tal section through the region [L]). The overlying branchial arch ectoderm (ec) lacks Col2a1 transcripts. epi, epibranchial ganglia; fg, facial ganglion;
nt, neural tube; ov, otic vesicle; tg, trigeminal ganglion; vg, vagal ganglion.
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 411
Lamprey Neural Crest Gene Regulatory Network
Neural crest specifiers turn on specific downstream effec-
tor genes that render the neural crest migratory and multi-
potent. We cloned homologs of 18 genes expressed by
neural crest or its derivatives (or both) in gnathostomes:
transcription factors like Brn-3b and Neurogenin, several
tors such as Neuropilin2 (Npn2) and Robo, and several
Hairy genes and collagen molecules, including Col2a1.
We found a type II Cadherin, expressed by premigratory
and early migrating cranial neural crest, much like gna-
thostome Cad-6, Cad-7, and Cad-11; it is also present in
condensing cranial and dorsal root ganglia (Figures 5A
and 5D). Conversely, an N-cadherin-like molecule is ex-
pressed in the neural tube, but is absent from migrating
neural crest cells entering the branchial arches (Figures
5Iand 5O).Both Npn2 andRobo areexpressed onmigrat-
ing and differentiating neural crest cells (Figures 5B, 5E,
and 5K), similar to the expression seen in higher verte-
brates (see Discussion), while the Npn2 ligand Sema3 is
found in the mesodermal core of the brancial arches and
in the ectodermal pits where branchial openings will form
(Figures 5H and 5N). Col2a1 homologs are expressed in
neural-crest-derived cartilage within the arches (Figures
5F and 5L), whereas transcripts of NeurogeninA and the
lamprey Pou4f2 homolog Brn-3b are found in forming cra-
nial ganglia (Figures 5G, 5J, and 5M). Expression of multi-
ple Hairy genes is observed in the forming cranial and
trunk sensory ganglia (Figure 8B and data not shown).
Functional Interactions within the NC-GRN
We tested functional relationships between components
of this putative gene regulatory network using antisense
morpholino oligonucleotides (MOs) to selectively block
translation of three neural plate border genes (MsxA,
Pax3/7, and ZicA) and five neural crest specifier genes
MOs were injected into a single blastomere at the two- or
four-cell stage. Those embryos with MO restricted to the
left or right half, such that the uninjected side served as
an internal control relative to the MO-treated side, were
selected. We then examined the expression of a wide
range of transcription factors in morphant embryos; the
monly used as a neural crest marker in other species, this
was not used routinely in the present study because, in
lamprey, Snail expression is selective to prospective
neural crest early on, but subsequently becomes some-
what ubiquitous, limiting its usefulness. Therefore, we
analyzed effects of morpholino knockdown on expression
of SoxE1 and FoxD-A, as well as SoxE2, AP2, n-Myc, Id,
Msx, Pax3/7, ZicA, and SoxB1. Control morpholinos
showed no effect on lamprey development and neural
crest marker gene expression (Figure 6).
Neural Plate Border Specifier Pax3/7
Pax3/7 morpholinos caused a marked reduction in ex-
pression of FoxD-A (78% of embryos; n = 80) and SoxE1
(71% of embryos; n = 94) on the injected side in mid to
late neurulae (E6–E6.5). The expression domain of ZicA
and SoxB1 on the injected side was expanded, whereas
n-Myc staining in postmigratory branchial arch crest cells
matches within the Pax3/7 target sequence caused no
change in neural crest gene expression (data not shown),
demonstrating that the observed effects are not due to
nonspecific disruption of neural plate border and neural
crest formation (Figure 6).
Neural Plate Border Specifier MsxA
MsxA MO caused a striking decrease in expression of the
neural crest specifier gene FoxD-A and a slightly milder
phenotype in the case of SoxE1. FoxD-A expression was
absent on the injected side at early stages (E5.5–E6)
through E6.5 (84% of embryos; n = 86). The effect of
MsxA depletion on SoxE1 expression was more subtle,
with a clear gap in SoxE1 expression at E5.5–E6 (70% of
embryos; n = 60), but much less evident by E6.5 (61% of
embryos; n = 66; Figure 6). This may indicate that MsxA
is necessary for the early onset of SoxE1 expression, but
not essential for its maintenance in the delaminating neu-
ral crest. Similar to effects of Msx1 depletion in Xenopus,
the neural tube on the injected side often appeared to
bulge with a concomitant expansion of the Pax3/7 domain
in the cephalic neural tube (Figure 6).
Neural Plate Border Specifier ZicA
ZicA MO caused reduction of both FoxD-A and SoxE1 ex-
pression (79% and 76% of embryos, n = 85 and n = 100,
respectively; Figure 6), as well as that of other early crest
specifiers AP2, n-Myc, and Id (data not shown). In addi-
tion, it often caused slight overgrowth of the neural tube
and expansion of Pax3/7 and SoxB1 expression. The de-
fect in these markers was long-lasting, with left/right dif-
ferences observed from E5.5 and persisting through
E6.5, suggesting that ZicA is necessary for both early
onset of FoxD-A and SoxE1 and their maintenance in the
Neural Crest Specifier FoxD-A
FoxD-A MO embryos showed either complete depletion
or marked reduction in SoxE1 expression at 6 days
(81% of embryos; n = 68; Figure 6), as well as a significant
decrease in expression of migrating neural crest marker,
on expression of border specifiers Pax3/7 and ZicA. Al-
though the FoxD-A morpholino caused no expansion of
the neural plate border at early specification stages, it
did downregulate later expression of ZicA and Pax3/7 in
the dorsal neural tube (Table S1).
Neural Crest Specifier AP2
AP2 MO altered expression of late (FoxD-A, SoxE1, and
SoxE2) and early (n-Myc and Id) neural crest specifiers,
but it had no effect on Pax3/7. The effects on SoxE1 and
FoxD-A expression were most obvious at E5.5, consistent
with a role in initial onset of, but not maintenance of,
SoxE1/FoxD-A expression (73% and 70% of embryos;
vious depletion of SoxE2, a migrating neural crest marker,
which reflects a decrease in the numbers of migrating
412 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Neural Crest Gene Regulatory Network
crest cells on the injected side, though migrating neural
crest cells from the contralateral side begin to compen-
sate for the deficiency by E6.5.
Neural Crest Specifier Id
Id MO caused reduction of FoxD-A and SoxE1 in the pre-
migratory neural crest cells atE5.5–E6.5 (75% and 71%of
embryos; n=50 andn=45,respectively;Figure 6), butdid
not seem to alter the expression of n-Myc (data not
shown). In addition, there appeared to be a dramatic de-
crease in the progenitor pool. The cephalic neural tube
was somewhat expanded, with a concomitant expansion
of Pax3/7, ZicA, and the neural marker SoxB1. At later
stages, ZicA and SoxB1 were downregulated in the dorsal
neural tube (data not shown).
Neural Crest Specifier n-Myc
n-Myc MO caused a reduction in FoxD-A and SoxE1 ex-
pression (75% and 80% of embryos; n =57 and n = 61, re-
ZicA expression. The effect was most prominent early, at
E5.5–E6. At later stages, there was downregulation of
Pax3/7 in the dorsal aspect of the neural tube (data not
Figure 6. Gene Regulatory Interactions between Neural Plate Border and Neural Crest Specifiers as Revealed by Morpholino-
Mediated Gene Knockdown
crest markers SoxE1 and FoxD-A on the injected sideat E6–E6.5, as shown byinsitu hybridization. Pax3/7 MO2 anteriorintegration causes complete
abolishment of FoxD-A expression. Injection of control morpholino (CoMO) at two-cell stage causes no change in neural crest marker gene expres-
sion. In MsxA-depleted embryos both FoxD-A and SoxE1 expression are inhibited on the injected side. In embryos injected with FoxD-A MO, SoxE1
expression was reduced or completely absent, whereas SoxE1 MO-treated embryos had a defect in FoxD-A expression. Embryos injected with
MsxA, ZicA, and Id MOs showed slight overgrowth of the neural tube on the injected side and concomitant expansion of Pax3/7-expressing neural
cells. No such expansion was observed with AP2, n-Myc, and FoxD-A MOs.
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 413
Lamprey Neural Crest Gene Regulatory Network
Neural Crest Specifier SoxE1
SoxE1 MO caused a decrease in FoxD-A expression in
74% of embryos (n = 78), indicating that early expression
of SoxE1 is essential for expression or maintenance
(or both) of another neural crest specifier (Figure 6), but
it did not alter the expression of Pax3/7. Later effects
of SoxE1 MO have been described elsewhere (McCauley
and Bronner-Fraser, 2006).
In summary, the results show (1) that blocking neural
plate border specifiers causes a reduction in neural crest
specifier expression, and often expands the neural terri-
tory, affecting expression of border specifiers such as
ral crestspecifiers AP2, Id,n-Myc,SoxE1,and FoxD-Aare
required for continued expression of neural crest speci-
fiers (markers of the premigratory neural crest such as
SoxE1 and FoxD-A), but not for early expression of neural
plate border specifiers. In general, neural crest specifers
do not affect border specifiers, with the exception of Id,
which causes some expansion of Pax3/7.
Rescue with Cognate RNAs of Morpholino-
To confirm that the effects of the antisense MOs are spe-
cific, we carried out rescue experiments using coexpres-
sion of cognate mRNAs from other species. The results
show that morpholino-dependent loss of neural crest
specifier gene expression is rescued by coinjection of
the corresponding orthologous mRNA from other verte-
brates that lacks the 50-morpholino target sequence. In
all cases, the percentage of morpholino-treated embryos
with abnormalities in marker gene expression was drasti-
cally reduced after rescue with cognate mRNA (see Fig-
ure 7). In mRNA-rescued morphants, the number of
hypomorphicembryos increasedand extremedifferences
in left/right gene expression were rare (<30% of rescued
embryos). This not only demonstrates specificity of the
morpholino, but also a remarkable conservation of protein
function across large evolutionary distances.
Effects of Morpholino-Mediated Knockdown
on Neural Crest Derivatives
Finally, we asked whether knockdown of neural plate
border and neural crest specifiers caused long-term
deficiencies in neural crest derivatives. To this end, we
allowed MO-injected embryos to develop to early larval
stages and examined markers for nascent cranial and
dorsal root ganglia and pharyngeal cartilages, as well as
the presence of pigment cells.
Consistent with interference in neural plate border
specification, embryos treated with morpholinos against
derivatives (Figures 8A, 8B, and 8C). There was loss of
pigment cells and cranial ganglia and defects in formation
of neural-crest-derived elements of branchial arches.
These effects were most pronounced with Pax3/7 and
In embryos injected with MOs against neural crest
specifiers, a less severe global reduction of neural crest
derivatives was observed. This is likely to reflect partial re-
generation of the neural crest progenitor pool in these em-
bryos by neighboring precursors. Close examination of
neural-crest-specifier-depleted embryos revealed more
severe reductions in particular neural crest derivatives
Figure 7. Coexpression of Cognate Het-
erospecific mRNA Rescues Morpholino-
Mediated Neural Crest Phenotypes
The results of rescue experiments show that
morpholino-dependent loss of neural crest
specifier gene expression can be rescued by
coinjection of the corresponding orthologous
notypes are divided into three categories: nor-
mal, extremely abnormal, and mildly abnormal
marker expression and results, expressed in
percentage of total number of embryos treated
(MO-only treated embryos in black, MO +
mRNA-treated in red). Inall cases,thepercent-
abnormalities in marker gene expression was
drastically reduced after rescue with cognate
mRNA. Cognate heterospecific mRNA used
was from Xenopus Zic1, Msx1, AP2, and
Sox9 and chick Pax3, and scoring markers
were either lamprey SoxE1 or FoxD-A.
414 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Neural Crest Gene Regulatory Network
depending onthefactorblocked. WhileinAP2-and FoxD-
A-depleted larvae, clear reductions in pigment cells were
observed, the pharyngeal arches appeared less effected
(Figure 8A). Embryos treated with Id and FoxD-A MOs
showed more significant reductions in cranial ganglia
than either AP2- or n-Myc-depleted embryos (Figure 8B).
arch cartilages was seen, but melanocyte numbers were
near normal (Figure 8C).
A uniquely vertebrate innovation, the neural crest is
defined by its origin at the neural plate border, migratory
capability, multipotentiality, and combinatorial gene ex-
pression. As a basal jawless vertebrate, lamprey pos-
and form many, but not all, neural-crest-derived struc-
tures found in jawed vertebrates (McCauley and Bron-
ner-Fraser, 2003, 2006; Meulemans and Bronner-Fraser,
2002; Meulemans et al., 2003; Myojin et al., 2001; Neidert
et al., 2001; Newth, 1950). However, there is little or no in-
formation about early steps in neural crest specification in
the lamprey. Analysis of a hypothetical NC-GRN in this
basal vertebrate promises to inform on the general archi-
tecture and evolutionary history of an archetypical verte-
brate gene regulatory network. As both a critical test of
this putative network and a representation of its ground
state, we have performed functional tests involving multi-
ple interactions within a single, basal vertebrate.
Conservation ofthe UpstreamCore of the NC-GRN
We identified 50 genes involved in neural crest formation
in lamprey. Our findings are consistent with several
features of a putative NC-GRN proposed to function in
jawed vertebrates (Meulemans and Bronner-Fraser,
2004), particularly with respect to its proximal elements.
Expression of signaling molecules and neural plate border
specifiers is highly conserved, as are the functions of bor-
der specifiers tested in our study. We find BMP, Wnt, and
Delta expression in similar patterns to those noted in frog
and zebrafish (Garcia-Castro et al., 2002; Glavic et al.,
2004; LaBonne and Bronner-Fraser, 1998; Nguyen et al.,
1998; Saint-Jeannet et al., 1997), suggesting that signal-
to play analogous functions in neural crest specification to
those in other vertebrates; e.g., Wnt8 is expressed in the
nonneural ectoderm abutting the neural rod, much like
chick Wnt6 (Garcia-Castro et al., 2002). Similarly, lamprey
to the neural plate border, implying that their combinato-
rial presence in the border is highly conserved across all
vertebrate neurulae (Aruga et al., 2002; Bang et al.,
1999; Basch et al., 2006; Luo et al., 2001; Monsoro-Burq
et al., 2005; Sato et al., 2005).
gene regulatory network exhibit both conserved and di-
vergent features. Our results suggest that neural crest
specifiers are activated in two phases, with one set of
transcription factors activated at the neural plate border
of the early neurula and the other during a second later
Figure 8. Depletion of Neural Plate Border and Neural Crest Specifiers Results in a Reduction or Loss of Neural Crest Derivatives
(A) Unilateral or dorsal targeting of gene-specific morpholinos results in embryos with dramatically reduced numbers of melanocytes compared with
embryos injected with control morpholino (CoMO). Stage-matched control embryos injected with CoMO are shown to the left of perturbed embryos.
Lateralview ofPax3/7-, FoxD-A-, or AP2-depleted larvae examined atE15.5,E11.5, and E13, respectively. Dorsal view of E13.5 ZicA-depletedlarvae
shows unilateral or complete ablation of pigment cells.
(B) Unilateral depletion of NC-GRN elements results in reduction of cranial ganglia (arrowheads). In situ hybridization with DeltaA (MsxA-, Id-, and
FoxD-A-depleted embryos) or Hairy1/2B probe (ZicA-depleted embryos) shows decreased or absent marker expression on the injected side (right
side each panel) in facial ganglion (fg), vagal ganglion (vg), and sometimes trigeminal ganglion (tg). Left panels show control side of each embryo. A
pronounced effect obtained with Pax3/7 MO2 shows absence of SoxE2 expression in the dorsal root ganglia on the injected side (black arrows).
(C) EmbryoswithunilateralMsxA-, ZicA-,and n-Myc-MO integration show defects in pharyngeal archcartilages,asdemonstrated bythedecreaseor
loss of SoxE1 or SoxE2 expression in these structures on the injected side (black arrows).
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 415
Lamprey Neural Crest Gene Regulatory Network
phase wherein the neural crest in the dorsal neural tube is
forming. This differs from the previous formulation of the
of deployment of neural crest specifier genes into early
(neural plate border) and late (bona fide neural crest pre-
cursor) categories. We note that the expression patterns
and functions of late neural crest specifiers, like FoxD3
and SoxE family members, in lamprey resemble those ob-
2003; Lee et al., 2004; Montero-Balaguer et al., 2006;
Sasai et al., 2001; Spokony et al., 2002; Taylor and Lab-
onne, 2005; Yan et al., 2005), whereas c-Myc, Id, AP2,
and Snail are first deployed in the early neurula at the neu-
ral plate border rather than in nascent neural crest cells
(Aybar et al., 2003; Bellmeyer et al., 2003; Knight et al.,
2003; LaBonne and Bronner-Fraser, 2000; Light et al.,
2005; O’Brien et al., 2004). These early-activated neural
specifiers, suggesting they may be their direct targets.
Furthermore, these genes are involved in cell cycle control
and therefore may play a role in maintaining multipotency
of neural crest progenitors by acting as a cell cycle control
switch between proliferation, cell death, and cell fate de-
cisions (Kee and Bronner-Fraser, 2005; Light et al., 2005).
The slow development of lamprey offers the advantage
of allowing exquisite temporal resolution not possible in
rapidly developing organisms like Xenopus and zebrafish.
In jawed vertebrates, c-Myc and its direct target Id3 are
expressed at the neural/nonneural ectoderm border prior
to Snail1 and Sox8 (Bellmeyer et al., 2003; Kee and Bron-
2005), whereas Snail2, Sox9, and FoxD3 areexpressed by
premigratory neural crest (O’Donnell et al., 2006). How-
ever, the rapid development of Xenopus makes the exact
timing of these expression patterns much more difficult to
resolve. In amniotes like chick, Id family members are
expressed at the neural plate border (Kee and Bronner-
Fraser, 2001a, 2001b, 2001c), together with proto-onco-
genes c-Myc and n-Myc (not shown) and bHLH transcrip-
tion factor AP2a (Shen et al., 1997). In contrast, Sox9,
FoxD3, and Snail2 are first expressed in the neural folds,
while Sox10 is first expressed in delaminating neural crest
(Aybar et al., 2003; Honore et al., 2003; O’Donnell et al.,
2006; Sakai et al., 2006). Thus, subdivision of lamprey
neural crest specifiers into early- and late-acting cate-
gories may reflect either a lack of conservation or a previ-
ously unrecognized characteristic of the vertebrate neural
crest network in general.
A difference in gene expression between lamprey and
other species is that Snail is expressed earlier at the lam-
prey neural plate border, in contrast to its expression in
premigratory neural crest in frogs, fish, and birds. Further-
more, the Snail homolog we identified does not display
a neural-crest-specific pattern at premigratory stages,
but rather appears to be ubiquitous, similar to hagfish
SnailA (Ota et al., 2007), and thus may represent an inter-
esting regulatory difference between cyclostomes and
gnathostomes. Similarly, the transcription factor Ets1 is
expressed in premigratory, migrating, and postmigratory
neural crest in Xenopus and chick and proposed to func-
tion in neural crest cell specification (Maroulakou et al.,
1994; Tahtakran and Selleck, 2003). In contrast, no lam-
prey Ets1 homologs areexpressed in the neural crest cells
during specification stages; rather, the first expression of
both Ets1a and Ets1b isin populations of earlydifferentiat-
ing neural crest within the branchial arches. In addition,
Ets1b is expressed in hematopoietic and endothelial pre-
in hematopoiesis, vasculogenesis, and angiogenesis
(Maroulakou and Bowe, 2000); this suggests that the lam-
prey gene functions hematopoetically while lacking early
neural crest specifier function. Along the same lines, Twist
is expressed in the premigratory crest in Xenopus (Hop-
wood et al., 1989), whereas lamprey Twist homologs
appear to be expressed only in postmigratory crest cells
lining branchial arches and persist in mesenchyme form-
ing buccal cartilage. Extensive searches have yielded
four Twist and two Ets1 homologs plus one Ets1-related
factor. Given the current genome coverage (?95%), we
suggesting that the lamprey genes lack early neural crest
specifier function. Intriguingly, these genes may have
been co-opted to an earlier function in gnathostomes,
lost early specification function in lampreys, or both. In
contrast to this apparent lack of conservation, signaling
receptors and adhesion and matrix molecules like Neuro-
pilin2, Robo, and Col2a1 have similar expression patterns
in gnathostomes and lamprey (Gammill et al., 2006; Jia
et al., 2005; Zhang et al., 2006). An N-cadherin-like adhe-
sion molecule, Cadherin IA, is expressed in neural tube
and periocular region, but is absent from branchial arch
neural crest population (Xu et al., 2001). A type II Cadherin
homolog (Cad IIA), similar to Cad-6/7/10/11, shares simi-
larities with all of its gnathostome counterparts and is
found in premigratory (in the case of Cad-6) and early mi-
grating (in the case of Cad-7 and Cad-11) neural crest, as
well as in differentiating neurogenic derivatives (in the
case of Cad-7 and Cad-10) (Borchers et al., 2001; Liu
et al., 2006; Nakagawa and Takeichi, 1998).
The cumulative results suggest that lamprey possesses
a NC-GRN which is a modified version of that hypothe-
sized to function in gnathostomes.
Interactions within the NC-GRN
Gain- and loss-of-function experiments performed in
various jawed vertebrates give clues about the genetic
interactions leading to neural crest specification; e.g.,
morpholino knockdown of neural plate border specifiers
Msx1, Msx2, Pax3, and Zic1 in Xenopus, as well as
Pax7 in chick, causes alterations in expression of neural
crest specifiers Slug, FoxD3, Sox9, and Sox10 (Basch
et al., 2006; Khadka et al., 2006; Monsoro-Burq et al.,
2005; Sato et al., 2005). Concomitantly, inactivation of
of Pax3 and Zic1 and the neural marker Sox2 (Monsoro-
Burq et al., 2005; Sato et al., 2005). Inactivation of early
(c-Myc, Ids, or AP2) and late (Sox8, Sox9, and Sox10)
416 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Neural Crest Gene Regulatory Network
gnathostome neural crest specifiers affects expression of
all neural crest specifiers (Bellmeyer et al., 2003; Honore
et al., 2003; Kee and Bronner-Fraser, 2005; Luo et al.,
2003; Yan et al., 2005). However, functional evidence for
aspects of this putative network often conflicts between
different jawed vertebrates; e.g, depletions of AP2,
FoxD3, or Sox10 in Xenopus disrupts neural crest induc-
tion (Honore et al., 2003; Luo et al., 2003; Sasai et al.,
2001), while in zebrafish, their knockdown impinges on
differentiation but has no apparent effect on induction
(Dutton et al., 2001; Knight et al., 2003; Montero-Balaguer
et al., 2006). These differences may be due to the tetra-
ploidy of zebrafish, compensation by redundant paralogs,
or both (Yan et al., 2005). The emerging data suggest that
the neural crest specifiers extensively cross-regulate to
maintain their expression, though hierarchical relation-
ships remain difficult to ascribe.
To better understand the network and obtain a more
comprehensive picture of the relationships between its el-
ements, we analyzed the effects of knockdown of three
neural plate border and five neural crest specifier genes
on more neural crest markers than has previously been
done in any other vertebrate (SoxE1, SoxE2, FoxD-A,
AP2, n-Myc, and Id, summarized in Table S1). Although
Snail is typically used as a crest marker in Xenopus, its
quasi-ubiquitous presence during premigratory stages in
lamprey obviated its usefulness in this study.
In comparing our results with those previously de-
scribed in gnathostomes, we find that inactivation of bor-
der specifiers MsxA, Pax3/7, and ZicA results in depletion
of neural crest specifier expression, consistent with ob-
servations in Xenopus. However, lamprey neural plate
border specifiers do not appear to mutually coactivate.
We did observe an expansion of the dorsal neural tube
and, correspondingly, of Pax3/7 expression therein, sug-
gesting that inactivation of border specifiers may result
in a fate conversion from neural crest to neural tube. Be-
cause many of the neural plate border specifiers are later
expressed in the dorsal neural tube, they appear to have
later and separate functions in the developing nervous
system. Our data show that inactivation of FoxD-A,
n-Myc, or Id decreases expression of Pax3/7, ZicA, and
SoxB1 in the roof plate, in agreement with findings in Xen-
opus that FoxD3 induces Zic1 and neural markers (Sasai
et al., 2001), whereas Sox9 is required for later expression
of Pax3 and Msx1 (Spokony et al., 2002).
Interestingly, our rescue experiments using Xenopus
Zic1, Msx1, AP2, and Sox9, as well as chick Pax3
mRNA, suggest that these heterospecific proteins can
functionally compensate for the loss of their lamprey or-
thologs. These experiments imply that the protein struc-
ture of these transcription factors has been sufficiently
conserved during vertebrate evolution to be interchange-
able in the context of neural-crest-inducing function.
Core Conservation of a NC-GRN at the Base
Traditionally the neural crest isconsidered an evolutionary
innovation of vertebrates, since protochordates lack bona
like BMP, Notch, and Wnt are expressed in a pattern
closely resembling that of vertebrates, consistent with
their conserved role in patterning the early ectoderm. Fur-
thermore, in both Amphioxus and ascidians, homologs of
neural plate border specifiers Msx, Zic, and Pax3/7 are
present within the neural plate border territory in late gas-
trula/early neurula, suggesting that the initial steps of bor-
der patterning and specification are already present in
protochordates. In contrast, no neural crest specifiers,
with the exception of Snail, are deployed at the neural
plate border (reviewed in Meulemans and Bronner-Fraser
). Recently, a large number of gene interactions
were tested in the ascidian Ciona intestinalis using mor-
pholino-mediated gene knockdown (Imai et al., 2006).
system formation appear to be conserved between uro-
chordates and vertebrates, neural-crest-specific links
are absent in Ciona, and only the activation of Snail by
Zic is reminiscent of the vertebrate NC-GRN.
Evolution of neural crest was likely driven by changes at
the gene-regulatory level that may include co-option of
ancestral gene batteries to a new purpose (Meulemans
plementary transcription factor or factors into the regula-
tory cascade. While the proximal gene regulatory ele-
ments are highly conserved between lamprey and
gnathostomes, the neural crest specifier portion can
clearly be subdivided into two temporally separated sub-
sets. More distal regulatory modules that involve deploy-
ment of intracellular and extracellular signaling cues and
gene batteries responsible for migration and differentia-
tion of neural crest cells are present, implying a high de-
gree of evolutionary constraint. Differences between gna-
thostome models are likely to reflect lineage-specific
alterations in expression of paralogous genes or slight
alterations in degrees of cis-regulatory robustness.
tion of neural crest are a vertebrate synapomorphy. As
such, this conserved network fits the proposed criteria
for defining gene regulatory networks functioning during
development of animal body plans. The NC-GRN is com-
posed of one or more ‘‘kernels’’ (Davidson and Erwin,
lutionarily inflexible unit that plays an essential upstream
function in establishing the identity of the neural crest pro-
genitor territory and is also found in protochordates that
of the neural crest specifier module into the network led to
ing of two interconnected parts—the neural plate border
and neural crest specifier modules. Other ‘‘plug-ins’’ and
‘‘switches’’ may have been co-opted into the circuit from
vide signaling inputs (Wnts) or guidance cues (Npn/Sema
ligand-receptor couple), whereas switches like Myc/Id, in-
tegrated at the specification level of the network, provide
a mechanism of cell cycle control that alternates between
neural crest cell proliferation and cell death.
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 417
Lamprey Neural Crest Gene Regulatory Network
Addition of neural crest modules to the network
occurred prior to separation of jawed and jawless verte-
brates, likely during the transition from protochordates to
vertebrates. This reflects an ancient origin of the NC-
200 million years between the divergence of cephalochor-
dates and vertebrates. Furthermore, it is likely that ‘‘differ-
entiation’’ subcircuits may have been incorporated and
co-opted to more proximal use in revising the NC-GRN
from agnathans to gnathostomes. As an example, we
find that Ets1 and Twist are deployed late in migratory
and postmigratory lamprey neural crest, but exist more
proximally in the gnathostome network. Thus, though
a neural crest gene network was largely fixed at the base
of vertebrates, there appears to be remodeling of individ-
ual subcircuits that may be responsible for species-spe-
cific traits. It is interesting to note that a recent paper re-
ported the successful isolation of embryos from another
agnathan, hagfish, for the first time after 100 years of
known attempts throughout the literature. Intriguingly,
the gene expression patterns for the neural crest markers
reported in this study appear highly reminiscent of those
of lamprey (Ota et al., 2007).
Our findings are all the more significant when taking into
account recent fossil finds suggesting that modern lam-
preys are ‘‘living fossils,’’ with similar characteristics to
the common ancestor with jawed vertebrates (Gess
et al., 2006), thus reflecting the primitive vertebrate condi-
tion and occupying an important ancestral position. Prior
to this study, only two genes were studied thoroughly in
the context of early events in neural crest formation in
observations couple the formation of the neural crest
proper with the establishment of a NC-GRN at the dawn
of vertebrates, pushing back the date that such a gene
regulatory network was invented by at least 200 million
years, and thus giving deep insight into the steps neces-
sary for the creation of defining vertebrate features.
Screening of an Arrayed cDNA Library
A directional lamprey cDNA library waspreparedfrom 2-to12-day-old
embryos and arrayed using Genetix Q-bot. The filters were screened
at varying stringency using heterospecific probes (see Supplemental
Embryo Collection and Maintenance
Embryo collection and maintenance was performed in the laboratory
as previously described (McCauley and Bronner-Fraser, 2006; Sup-
plemental Experimental Procedures).
FITC-labeled antisense MOs (Gene Tools, Philomath, OR) were de-
signed to target the translation initiation site as follows:
Pax3/7 MO1; 50–ACTCTGCTGCATAAACACCGTGCCG-30. Pax3/7
5-mismatch MO1; 50–ACTgTGCTcCATAAAgACCcTGgCG-30(mis-
match base pairs in lowercase). Pax3/7 MO2; 50-TGTCCTGGTGCCG
GGCGCATCATCC-30. MsxA MO; 50-GACCGCGAAGCGAAATGCGTT
CATG-30. ZicA MO; 50-CGCCTCCAGAAGCATCGCGTGCGGT-30.
n-Myc MO; 50-GCTGCACCCGGCATCGAAGTTTATC-30. Id MO1; 50–
TTAACAAGAGGTGGCTGTCGGT–30. Id MO2; 50–ACTCACGGCCTTC
ATGGCTGTAATG–30. AP2 MO; 50–CCTGTAATTTCAAAAGCATGACT
CC–30. FoxD-A MO1; 50–GAGTGACGGTTTTTGTTTATTTGGC–30.
FoxD-A MO2; 50–TATCGTCCCCCTCTGGCACGTCAGT–30. SoxE1
MO; 50–GAGTGACGGTTTTTGTTTATTTGGC–30. Control MO (CoMO);
The corresponding coding sequences are available in GenBank
EU086588, EU086589, and EU086590). Ten to forty nanograms of
MO was injected into a single blastomere at the two- to four-cell stage.
In Situ Hybridization
In situ hybridization was performed in accordance with Sauka-Spen-
gler et al. (2003) and the Supplemental Experimental Procedures.
The Supplemental Data for this article, including methods, two figures,
We thank Dr. J. Langeland for providing material and expertise during
cDNA library construction; C. Krontiris and P. Fraser for excellent
assistance; Drs. S. Fraser and N. Nikitina for critical reading of the
manuscript; Dr. Y. Kee for sharing reagents and expertise; Drs. S.
Bhattacharyya and M. Barembaum for helpful discussions; Drs. A.
Monsoro-Burq and J.-P. Saint-Jeannet for providing Xenopus Msx1
and Zic1 clones; R. Bergstedt and the staff at Hammond Bay Biolog-
ical Station for kindly providing animals; and N. Johnson for sending
spawning adult lampreys. This work was supported by DE17911 to
Received: December 14, 2006
Revised: March 9, 2007
Accepted: August 15, 2007
Published: September 4, 2007
Aoki, Y., Saint-Germain, N., Gyda, M., Magner-Fink, E., Lee, Y.H.,
Credidio, C., and Saint-Jeannet, J.P. (2003). Sox10 regulates the
development of neural crest-derived melanocytes in Xenopus. Dev.
Biol. 259, 19–33.
Aruga, J., Tohmonda, T., Homma, S., and Mikoshiba, K. (2002). Zic1
promotes the expansion of dorsal neural progenitors in spinal cord
by inhibiting neuronal differentiation. Dev. Biol. 244, 329–341.
Aybar, M.J., Nieto, M.A., and Mayor, R. (2003). Snail precedes slug in
the genetic cascade required for the specification and migration of the
Xenopus neural crest. Development 130, 483–494.
Bang, A.G., Papalopulu, N., Goulding, M.D., and Kintner, C. (1999).
Expression of Pax-3 in the lateral neural plate is dependent on
a Wnt-mediated signal from posterior nonaxial mesoderm. Dev. Biol.
Basch, M.L., Bronner-Fraser, M., and Garcia-Castro, M.I. (2006).
Specification of the neural crest occurs during gastrulation and
requires Pax7. Nature 441, 218–222.
Bellmeyer,A.,Krase,J., Lindgren, J.,andLaBonne,C.(2003).Thepro-
tooncogene c-myc is an essential regulator of neural crest formation in
xenopus. Dev. Cell 4, 827–839.
Borchers, A., David, R., and Wedlich, D. (2001). Xenopus cadherin-11
restrains cranial neural crest migration and influences neural crest
specification. Development 128, 3049–3060.
Butler, A.B., and Hodos, W. (1996). Comparative Vertebrate Neuro-
anatomy: Evolution and Adaptation (New York: John Wiley & Sons).
Davidson, E.H., and Erwin, D.H. (2006). Gene regulatory networks and
the evolution of animal body plans. Science 311, 796–800.
418 Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc.
Lamprey Neural Crest Gene Regulatory Network
Dutton, K.A., Pauliny, A., Lopes, S.S., Elworthy, S., Carney, T.J.,
ourless encodes sox10 and specifies non-ectomesenchymal neural
crest fates. Development 128, 4113–4125.
Gammill, L.S., Gonzalez, C., Gu, C., and Bronner-Fraser, M. (2006).
Guidance of trunk neural crest migration requires neuropilin 2/sema-
phorin 3F signaling. Development 133, 99–106.
Gans, C., and Northcutt, R.G. (1983). Neural Crest and the Origin of
Vertebrates: A New Head. Science 220, 268–273.
dermal Wnt function as a neural crest inducer. Science 297, 848–851.
Devonian period of South Africa. Nature 443, 981–984.
Glavic, A., Silva, F., Aybar, M.J., Bastidas, F., and Mayor, R. (2004). In-
terplay between Notch signaling and the homeoprotein Xiro1 is
required for neural crest induction in Xenopus embryos. Development
Hollyday, M., McMahon, J.A., and McMahon, A.P. (1995). Wnt expres-
sion patterns in chick embryo nervous system. Mech. Dev. 52, 9–25.
Honore, S.M., Aybar, M.J., and Mayor, R. (2003). Sox10 is required for
the early development of the prospective neural crest in Xenopus
embryos. Dev. Biol. 260, 79–96.
Hopwood, N.D., Pluck, A., and Gurdon, J.B. (1989). A Xenopus mRNA
related to Drosophila twist is expressed in response to induction in the
mesoderm and the neural crest. Cell 59, 893–903.
S. (1999). Development of cephalic neural crest cells in embryos of
Lampetra japonica, with special reference to the evolution of the jaw.
Dev. Biol. 207, 287–308.
print for a chordate embryo. Science 312, 1183–1187.
Jeffery, W.R., Strickler, A.G.,and Yamamoto, Y.(2004). Migratory neu-
ral crest-like cells form body pigmentation in a urochordate embryo.
Nature 431, 696–699.
Jia, L., Cheng, L., and Raper, J. (2005). Slit/Robo signaling is neces-
sary to confine early neural crest cells to the ventral migratory pathway
in the trunk. Dev. Biol. 282, 411–421.
Johnels, A.G. (1956). On the peripheral autonomic nervous system of
the trunk region of Lampetra planeri. Acta. Zool. (Stockholm) 37,
Kee, Y., and Bronner-Fraser, M. (2001a). Id4 expression and its rela-
tionship to other Id genes during avian embryonic development.
Mech. Dev. 109, 341–345.
Kee, Y., and Bronner-Fraser, M. (2001b). Temporally and spatially re-
stricted expression of the helix-loop-helix transcriptional regulator
Id1 during avian embryogenesis. Mech. Dev. 109, 331–335.
Kee, Y., and Bronner-Fraser, M. (2001c). The transcriptional regulator
Id3 is expressed in cranial sensory placodes during early avian embry-
onic development. Mech. Dev. 109, 337–340.
Kee, Y., and Bronner-Fraser, M. (2005). To proliferate or to die: role of
Id3 in cell cycle progression and survival of neural crest progenitors.
Genes Dev. 19, 744–755.
Khadka, D., Luo, T., and Sargent, T.D. (2006). Msx1 and Msx2 have
shared essential functions in neural crest but may be dispensable in
Knecht, A.K., and Bronner-Fraser, M. (2002). Induction of the neural
crest: a multigene process. Nat. Rev. Genet. 3, 453–461.
Knight, R.D., Nair, S., Nelson, S.S., Afshar, A., Javidan, Y., Geisler, R.,
Rauch, G.J., and Schilling, T.F. (2003). lockjaw encodes a zebrafish
tfap2a required for early neural crest development. Development
LaBonne, C., and Bronner-Fraser, M. (1998). Neural crest induction in
Xenopus: evidence for a two-signal model. Development 125, 2403–
LaBonne, C., and Bronner-Fraser, M. (2000). Snail-related transcrip-
tional repressors are required in Xenopus for both the induction of
theneural crest and itssubsequent migration. Dev. Biol. 221,195–205.
Lee, Y.H., Aoki, Y., Hong, C.S., Saint-Germain, N., Credidio, C., and
Saint-Jeannet, J.P. (2004). Early requirement of the transcriptional
activator Sox9 for neural crest specification in Xenopus. Dev. Biol.
Liem,K.F.,Jr.,Tremml, G.,Roelink,H.,and Jessell, T.M.(1995).Dorsal
differentiation of neural plate cells induced by BMP-mediated signals
from epidermal ectoderm. Cell 82, 969–979.
Light, W., Vernon, A.E., Lasorella, A., Iavarone, A., and LaBonne, C.
(2005). Xenopus Id3 is required downstream of Myc for the formation
of multipotent neural crest progenitor cells. Development 132, 1831–
Linker, C., Bronner-Fraser, M., and Mayor, R. (2000). Relationship be-
tween gene expression domains of Xsnail, Xslug, and Xtwist and cell
movement in the prospective neural crest of Xenopus. Dev. Biol.
Liu, Q., Duff, R.J., Liu, B., Wilson, A.L., Babb-Clendenon, S.G., Francl,
J., and Marrs, J.A. (2006). Expression of cadherin10, a type II classic
cadherin gene, in the nervous system of the embryonic zebrafish.
Gene Expr. Patterns 6, 703–710.
Luo, T., Matsuo-Takasaki, M., and Sargent, T.D. (2001). Distinct roles
for Distal-less genes Dlx3 and Dlx5 in regulating ectodermal develop-
ment in Xenopus. Mol. Reprod. Dev. 60, 331–337.
tion of neural crest in Xenopus by transcription factor AP2alpha. Proc.
Natl. Acad. Sci. USA 100, 532–537.
Maroulakou, I.G., and Bowe, D.B. (2000). Expression and function of
Ets transcription factors in mammalian development: a regulatory net-
work. Oncogene 19, 6432–6442.
Maroulakou, I.G., Papas, T.S., and Green, J.E. (1994). Differential
genesis. Oncogene 9, 1551–1565.
tions to the lamprey head. Development 130, 2317–2327.
McCauley, D.W., and Bronner-Fraser, M. (2006). Importance of SoxE
in neural crest development and the evolution of the pharynx. Nature
Meulemans, D., and Bronner-Fraser, M. (2002). Amphioxus and lam-
prey AP-2 genes: implications for neural crest evolution and migration
patterns. Development 129, 4953–4962.
Meulemans, D., and Bronner-Fraser, M. (2004). Gene-regulatory inter-
actions in neural crest evolution and development. Dev. Cell 7, 291–
Meulemans, D., and Bronner-Fraser, M. (2005). Central role of gene
cooption in neural crest evolution. J. Exp. Zoolog. B Mol. Dev. Evol.
Meulemans, D., McCauley, D., and Bronner-Fraser, M. (2003). Id ex-
pression in amphioxus and lamprey highlights the role of gene coop-
tion during neural crest evolution. Dev. Biol. 264, 430–442.
cooperate to mediate FGF8 and WNT signals during Xenopus neural
crest induction. Dev. Cell 8, 167–178.
Montero-Balaguer, M., Lang, M.R., Sachdev, S.W., Knappmeyer, C.,
Stewart, R.A., De La Guardia, A., Hatzopoulos, A.K., and Knapik,
E.W. (2006). The mother superior mutation ablates foxd3 activity in
neural crest progenitor cells and depletes neural crest derivatives in
zebrafish. Dev. Dyn. 235, 3199–3212.
Myojin, M., Ueki, T., Sugahara, F.,Murakami,Y.,Shigetani, Y.,Aizawa,
S., Hirano, S., and Kuratani, S. (2001). Isolation of Dlx and Emx gene
Developmental Cell 13, 405–420, September 2007 ª2007 Elsevier Inc. 419
Lamprey Neural Crest Gene Regulatory Network