neural crest cells that enter the mesenchyme of the foregut and
migrate in a rostrocaudal wave toward the cloaca. The sacral
NCC migrates in a caudorostral direction within the intestinal
mesenchyme,contributinga minority of the neural crest-derived
cells of the distal hindgut (Jones, 1942; Le Douarin and Teillet,
Endothelin-3 regulates neural crest cell proliferation and differentiation in the
hindgut enteric nervous system
Nandor Nagya,b, Allan M. Goldsteina,⁎
aDepartment of Pediatric Surgery, Massachusetts General Hospital, Harvard Medical School, Warren 1153, Boston, MA 02114, USA
bDepartment of Human Morphology and Developmental Biology, Faculty of Medicine, Semmelweis University, Budapest, Hungary
Received for publication 24 October 2005; revised 30 January 2006; accepted 31 January 2006
Neural crest cells (NCC) migrate, proliferate, and differentiate within the wall of the gastrointestinal tract to give rise to the neurons and glial
cells of the enteric nervous system (ENS). The intestinal microenvironment is critical in this process and endothelin-3 (ET3) is known to have an
essential role. Mutations of this gene cause distal intestinal aganglionosis in rodents, but its mechanism of action is poorly understood. We find
that inhibition of ET3 signaling in cultured avian intestine also leads to hindgut aganglionosis. The aim of this study was to determine the role of
ET3 during formation of the avian hindgut ENS. To answer this question, we created chick–quail intestinal chimeras by transplanting
preganglionic quail hindguts into the coelomic cavity of chick embryos. The quail grafts develop two ganglionated plexuses of differentiated
neurons and glial cells originating entirely from the host neural crest. The presence of excess ET3 in the grafts results in a significant increase in
ganglion cell number, while inhibition of endothelin receptor-B (EDNRB) leads to severe hypoganglionosis. The ET3-induced hyperganglionosis
is associated with an increase in enteric crest cell proliferation. Using hindgut explants cultured in collagen gel, we find that ET3 also inhibits
neuronal differentiation in the ENS. Finally, ET3, which is strongly expressed in the ceca, inhibits the chemoattraction of NCC to glial-derived
neurotrophic factor (GDNF). Our results demonstrate multiple roles for ET3 signaling during ENS development in the avian hindgut, where it
influences NCC proliferation, differentiation, and migration.
© 2006 Published by Elsevier Inc.
Keywords: Enteric nervous system; Endothelin-3 (ET3); Endothelin receptor B (EDNRB); GDNF; Chick–quail chimeras; Coelomic transplantation; Hirschsprung's
Neural crest cells (NCC) migrate, proliferate, and differen-
tiate within the gut wall to give rise to the neurons and glial cells
of the enteric nervous system (ENS). Precursor cells of the ENS
the ENSis contributed bythe vagal neural crest (LeDouarinand
Teillet, 1973). The hindgut ENS is derived mainly from vagal
1973; Pomeranz et al., 1991; Serbedzija et al., 1991; Burns and
Le Douarin, 1998). Avians have an additional sacral crest-
derived structure, the ganglionated nerve of Remak, which ex-
tends within the mesentery from the cloaca tothe umbilical level
(Teillet, 1978; Catala et al., 1995; Doyle et al., 2004). Although
the molecular mechanisms directing enteric neural crest for-
mation are largely unknown, several factors that influence these
processes have been identified.
Enteric NCC migration, proliferation, survival, and diffe-
rentiation are primarily regulated by factors present in the gut
mesenchyme. These proteins, together with their respective
receptors expressed on the enteric NCC, are necessary for
normalENSformation.Theseessential factors include: glialcell
line-derived neurotrophic factor (GDNF) and its receptors Ret
Developmental Biology xx (2006) xxx–xxx
YDBIO-02420; No. of pages: 15; 4C: 4, 5, 7, 10, 12
⁎Corresponding author. Fax: +1 617 726 2167.
E-mail address: email@example.com (A.M. Goldstein).
0012-1606/$ - see front matter © 2006 Published by Elsevier Inc.
ARTICLE IN PRESS
101 lopment using the avian embryo that complements these other
102 systems. In this assay, preganglionic quail hindgut is trans-
103 planted into the coelomic cavity of a chick embryo. We find that
104 neural crest cells from the chick host populate the graft and give
105 rise to two ganglionated plexuses containing differentiated neu-
106 rons and glial cells. This method allows us to manipulate the
107 molecular environment of the developing gut and to study the
109 for studying both normal ENS formation and the role of ET3
110 during this process. Using coelomic transplants and intestinal
100 animal. We have developed an assay for studying ENS deve-
and the BMP receptors (Sukegawa et al., 2000; Kruger et al.,
2003; Goldstein et al., 2005); netrin and deleted in colorectal
cancer (Jiang et al., 2003); sonic hedgehog (Sukegawa et al.,
2000; Fu et al., 2004); Indian hedgehog (Ramalho-Santos et al.,
2000); endothelin-3 (ET3) and endothelin receptor-B (EDNRB)
(Baynash et al., 1994; Hosoda et al., 1994; Kruger et al., 2003;
Barlow et al., 2003). Mutations affecting many of these genes
lead to a variety of developmental intestinal disorders in mam-
mals, the most common of which is Hirschsprung's disease
(HSCR), a severe intestinal motility disorder affecting 1 in
5000 human births and characterized by the absence of ENS
along a variable length of colon. The majority of HSCR can be
accounted for by mutations in Ret (Chakravarti, 2001), which is
required for proliferation, migration, and differentiation of
enteric NCC. Null mutations in GDNF and Ret in mice prevent
enteric neural crest migration and result in aganglionosis of the
entire midgut and hindgut (Schuchardt et al., 1994; Moore et al.,
1996; Pichel et al., 1996; Sanchez et al., 1996).
Mutations of the ET3-EDNRB signaling pathway have also
been implicated in HSCR (Chakravarti, 2001). Absence of these
genes in mice leads to aganglionosis limited to the distal colon
(Baynash et al., 1994; Hosoda et al., 1994). While this suggests
an essential role for ET3 in hindgut colonization, its mechanism
of action remains controversial. The leading hypothesis is that
ET3 signaling functions to maintain the pool of enteric NCC
precursors by inhibiting neuronal differentiation and promoting
proliferation of those cells. Loss of ET3 activity would thereby
decrease cell proliferation and lead to premature neuronal diffe-
rentiation, leaving an inadequate number of progenitor cells to
populate the distal gut (Gershon, 1999). While several groups
have confirmed the capacity of ET3 to inhibit neurogenesis in
cultured NCC (Stone et al., 1997; Hearn et al., 1998; Wu et al.,
1999), Kruger et al. (2003) recently showed in enteric neural
crest stem cells that the absence of EDNRB does not increase
neurogenesis. Similarly, the ability of ET3 to promote NCC
proliferation has been supported by some (Stone et al., 1997;
Lahav et al., 1996; Barlow et al., 2003) and refuted by others
(Wu et al., 1999; Kruger et al., 2003; Woodward et al., 2003).
Much of what is currently known about the molecular regu-
lation of ENS development has been learned from transgenic
rodents and from cultured neural crest cells, each of which has
some limitations. Mice, for example, are not easily amenable to
experimental manipulation during embryogenesis. In vitro cell
culture techniques lackthemesenchymalenvironmentof thegut
and are therefore limited in their applicability to the living
organ culture, we find that ET3-EDNRB signaling has an essen-
tial role in promoting enteric crest cell proliferation, inhibiting
neurogenesis, and modulating enteric NCC migration during
ENS development in the avian hindgut.
Materials and methods
Fertilized White Leghorn chicken and quail (Coturnix coturnix japonica)
eggs were obtained from commercial breeders and maintained at 37°C in a
humidified incubator. Embryos were staged according to the Hamburger and
Hamilton (HH) tables (Hamburger and Hamilton, 1951) or the number of
embryonic days (E).
Organ culture assays
To study the chemoattractive role of exogenous GDNF and ET3, the intes-
ml type I rat tail collagen (BD Biosciences) to DMEM medium (Gibco) supple-
mented with 1% penicillin–streptomycin. GDNF (10 ng/ml; R&D Systems),
as previously described (Goldstein et al., 2005).
To studythe role of ET3 on NCC migration and differentiation along the gut,
E5 chicken embryonic midgut–hindgut was dissected from the umbilicus to the
cloaca and embedded in collagen gel. The nerve of Remak was removed from
the explants using tungsten needles. The collagen gel was supplemented with
ET3 (250 ng/ml) or BQ788 (5 μM). After 3 days, the gut was removed and
processed for immunocytochemistry.
Chick–quail intestinal chimeras
Hindgut was removed from E5 quail embryos. The ceca, cloaca, and nerve
of Remak were removed and the hindgut was embedded in collagen gel. The
collagen gel was supplemented with ET3 (250 ng/ml), GDNF (100 ng/ml), or
BQ788 (20 μM). Controls consisted of media alone, without added proteins.
After 24 h in culture, the hindgut was removed from the gel and processed for
transplantation. For bead implantation, Affigel blue beads (BioRad) or heparin-
acrylic beads (70–150 μm diameter; Sigma) were rinsed in phosphate-buffered
saline (PBS) and then soaked in 30 μl of 10 μg/ml GDNF protein, 10 μg/ml ET3
protein, or 200 μM BQ788 at 37°C for 1 h. The mesenchyme of isolated E5
quail hindgut was punctured with electrolytically sharpened tungsten needles to
create a space for the bead. The bead was inserted into the mesenchyme using a
blunt-end glass needle. Hindguts were then transplanted into chick coelomic
cavities. Both untreated guts and PBS-soaked beads served as controls.
Preganglionic hindguts from E5 quail embryos treated with ET3, GDNF, or
BQ788 (either in organ culture or by bead implantation) were transplanted into
the coelomic cavity of E3 chicks. Grafting experiments were performed as
chick embryo was immobilized and the surface ectoderm removed with a
to the right coelomic cavity of chicken embryos. Grafts were positioned longi-
tudinally within the coelom, along the long axis of the embryo. Host chicken
embryos were allowed to develop for 7 days following transplantation.
Chorioallantoic membrane transplants
The hindgut between the ceca and cloaca was dissected from E5 and E6
embryos and transplanted on the chorioallantoic membrane (CAM) of E8 chick
embryos. Before transplantation, a small portion of the CAM was gently
traumatized by laying a strip of sterile lens paper onto the surface of the
epithelium and then removing it immediately. The dissected hindgut was placed
over the junction of blood vessels on the CAM and incubated for 9 days. The
graft, together with the surrounding CAM, was excised, fixed in 4% buffered
formaldehyde, and embedded in gelatin.
2N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
formaldehyde, dehydrated in methanol, and stored at −20°C until ready for
processing. Riboprobe synthesis and whole-mount RNA in situ hybridization
were performed as previously described (Jowett, 1999). Digoxigenin-labeled
riboprobe was generated from a partial clone from the quail, lacking the 5′
noncoding region and the translation initiation site. Both ET3 and EDNRB
clones were kind gifts of Andrew Lassar. Antisense GDNF RNA probe was
generated by amplifying a fragment of GDNF sequence from total RNA
prepared from E9 chick gut using Trizol Reagent (Invitrogen). The following
oligonucleotide primers were used: upstream primer, 5′-tgccagaggattacccagat-
3′; downstream primer, 5′-aggtcatcgtcaaaggctgt-3′. The amplified product was
cloned into pCRII-TOPO using TOPO TA Cloning (Invitrogen).
For cryostat sections, guts were fixed in 4% formaldehyde in PBS for 1 h,
rinsed with PBS, and infiltrated with 15% sucrose/PBS overnight at 4°C. The
medium was changed for 7.5% gelatin containing 15% sucrose at 37°C for 1–
2 h, and the tissues rapidly frozen at −60°C in isopentane (Sigma). Frozen
sections were cut at 10 μm and collected on poly-L-lysine coated slides (Sigma).
Cryostat sections were stained by immunocytochemistry according to standard
techniques as described below. The following primary antibodies were used:
HNK-1 (NeoMarkers, CA), which recognizes an epitope on the surface of
migratory and postmigratory neural crest-derived cells; anti-HuC/D (Molecular
Probes, OR), which recognizes a neuron-specific RNA-binding protein; Bfabp
(brain fatty acid binding protein); 8F3, a chicken cell marker (Developmental
Studies Hybridoma Bank, DSHB); QCPN, a quail cell nuclear marker (DSHB);
4H6, aneurofilament marker(DSHB);andQH1, aquail-specificendothelialcell
marker (DSHB). After rehydration in PBS, sections were incubated with
primary antibodies for 45 min, followed by biotinylated goat anti-mouse IgG,
goat anti-mouse IgM, or goat anti-rabbit IgG (Vector Laboratories, Burlingame,
CA) and avidin-biotinylated peroxidase complex (Vectastain Elite ABC kit,
Vector Laboratories) at room temperature. Endogenous peroxidase activity was
quenched by incubation for 10 min with 3% hydrogen-peroxide (Sigma) in
PBS. The binding sites of the primary antibodies were visualized by 4-chloro-
1-naphtol (Sigma). Sections were examined under a Nikon Microphot FXA
microscope and digital images captured with a Spot camera and software
version 3.3.1 (Diagnostic Instruments). Images were compiled using Adobe
The histochemical method for the demonstration of nicotineamide adenine
dinucleotide phosphate (NADPH) diaphorase activity was performed with
modification as described previously by Burns and Le Douarin (1998). Briefly,
frozen sections were incubated for 30–50 min at 37°C in PBS containing 1 mg/
ml reduced β-NADPH (Sigma), 0.5 mg/ml nitroblue tetrazolium (Roche), and
3 μl/ml Triton X-100 (Fisher). Reaction product appeared as a blue granular
staining and the reaction was stopped by rinsing sections with PBS.
201 BrdU labeling, apoptosis, and double immunofluorescence
For analysis of cell proliferation, ET3- or BQ788-treated guts were isolated
from the collagen gel and incubated for 3 h in BrdU solution (DMEM medium
containing 5 mg/ml BrdU; Roche). BrdU staining was performed on 10 μm
cryostat sections. DNA was denaturated by incubating slides with 2 N HCl in
H2O at 37°C for 30 min. After neutralization with 0.1 M boric acid (pH 8.5) for
10 min, sections were stained with primary antibodies (QCPN, HNK-1, Hu, and
fluorescein-conjugated anti-BrdU antibody; Roche). Fluorescent secondary
antibodies included Alexa Fluor 594 goat anti-mouse IgG, Alexa Fluor 594
goat anti-mouse IgM, and Alexa Fluor 488 goat anti-rabbit IgG (Molecular
Probes). To detect apoptotic cells, sections were examined by double immu-
nofluorescence using anti-Hu and anti-activated caspase-3 antibodies (Cell
214 In situ hybridization
Whole embryos or dissected gastrointestinal tracts were fixed in 4%
ET3, EDNRB, and GDNF expression during avian gut
Although ET3, EDNRB, and GDNF are known to be ex-
pressed in the avian gut (Nataf et al., 1996, 1998; Homma et al.,
2000), a detailed analysis of their expression during ENS
development has not been described. We examined the expres-
sion patterns of these genes by whole-mount in situ hybridiza-
mesenchymal environment where crest-derived cells migrate.
Between HH 18 and 21, ET3 is present in the subectodermal
mesenchyme (Fig. 1A, arrows), where melanoblasts are
migrating. At E4.5, ET3 is expressed specifically in the cecal
primordia and cloaca (Fig. 1C), whereas it is expressed in the
outer mesenchyme along the entire midgut and hindgut starting
at E5 (Figs. 1B–F). EDNRB is expressed in the neural crest-
limited tothestomachandproximalduodenum (Fig.1G).ByE7
(Fig. 1H) and E9 (Figs. 1I–K), EDNRB-positive cells are
present throughout the hindgut.
GDNF expression was present in the nephrogenic mesen-
chyme at HH 18 (Figs. 2A, B). Expression in the distal gut was
evident atE5, where GDNFis present in the cecal primordia and
cloaca (Fig. 2C). Longitudinal sections through the cecum show
GDNF localized to the outer mesenchyme (Fig. 2D). Beginning
at E7, GDNF expression was observed throughout the intestine,
again restricted to the peripheral mesenchyme (Figs. 2E–H),
similar to the region of ET3 expression (Figs. 1D–F).
Neural crest-derived cells reach the proximal hindgut at E6
To determine the relationship between ET3 and GDNF ex-
pression in the ceca and the timing of NCC arrival to that region
of the gut, we established the time at which the migratory
wavefront first reaches the proximal hindgut. A midsagittal
section of an E6 embryo stained with Hu shows the presence of
enteric crest cells in the distal midgut and ceca, with an occa-
sional cell in the proximal hindgut (Fig. 3A, magnified view in
panel B). The postcecal hindgut was removed at this stage and
grown on the chorioallantoic membrane (CAM) of a host
Fig. 3C, the CAM graft became completely populated by a
normal-appearing ENS, suggesting that those rare cells in
the proximal E6 hindgut are sufficient to populate the entire
hindgut. CAM grafts were performed using E5 hindguts and
these remain completely aganglionic (Fig. 3D), confirming that
the host is not serving as a potential source of crest cells to the
EDNRB signaling is necessary for hindgut colonization
To establish that EDNRB activity is required for hindgut
colonization, E5 chicken or quail guts (midgut + hindgut) were
cultured in a three-dimensional collagen gel matrix for 72 h in
the presence of ET3 or the EDNRB antagonist, BQ788. Hu
3N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
shows EDNRB expression in the myenteric and submucosal plexuses. Crest cell
expression is also evident at E9 in a whole-mount (I), midgut section (J), and
longitudinal section of the distal gut (K). ep, epithelium; hg, hindgut; mg,
midgut; mp, myenteric plexus; nt, neural tube; smp, submucosal plexus.
279 antibody was used as a marker of neuronal differentiation and
280 HNK-1 as a marker of all enteric neural crest cells. In wild-type
281 E5 embryonic gut, Hu-expressing crest-derived cells were pre-
282 sent in the rostral midgut and absent from the caudal region of
283 the midgut and the entire hindgut (Fig. 4A). By E8, Hu-positive
crest-derived cells have reached the terminal hindgut (Fig. 4B).
In control medium containing no added proteins (n = 24),
complete colonization of the hindgut was observed after 72 h in
culture (Fig. 4C). The same results were obtained in the pre-
sence of purified ET3 (Fig. 4D, n = 35). Explants grown in
the presence of BQ788 (n = 35) demonstrated migratory arrest
at the level of the ceca, leading to complete hindgut
aganglionosis (Fig. 4E). HNK-1 immunostaining (Fig. 4F)
confirmed the results obtained with Hu antibody, demonstrating
that there is an arrest of enteric crest cell migration, not simply a
delay in neuronal differentiation. These findings demonstrate
that EDNRB signaling is necessary for NCC to colonize the
Chick–quail intestinal chimeras
To investigate the role of ET-3 in development of the avian
hindgut ENS, we developed a new application for avian co-
elomic transplantation. E5 quail hindguts, between ceca and
cloaca, were microsurgically removed and transplanted into the
coelomic cavity of E3 chick embryos (Figs. 5A, E) for 7 days
(n = 32). At the time of transplantation, the E5 quail hindgut has
not yet been colonized by crest-derived cells, as shown in a
section of an E5 hindgut (Fig. 5C). In all chick–quail transplant
serving as a source of enteric crest cells (Fig. 5D). The trans-
planted hindgut developed in the coelomic cavity of the host, as
demonstrated in Fig. 5F. Fig. 5G shows the hindgut graft before
(boxed area on right) and after 7 days of transplantation (left),
demonstrating significant growth of the graft during the trans-
plantation period. At the completion of the 7-day coelomic
incubation, the graft is found attached to the liver and right lung.
Although the graft is initially positioned adjacent to the foregut,
at the time of harvesting the graft is found a distance from the
esophagus and never attached to it.
Grafts were analyzed by immunohistochemistry using spe-
cies-specific and cell-specific antibodies to look at the origin of
nuclear marker, QCPN, heavily stained the epithelium and
mesenchyme of the graft, while it did not stain the presumptive
neural plexuses (Fig. 5H, arrows). Those plexuses, both sub-
mucosal and myenteric, were strongly stained with the 8F3
antibody, which exclusively recognizes cells of chicken origin
(Fig. 5I), suggesting that neural crest cells from the chick host
colonized the quail hindgut. Immunostaining on serial sections
cell marker HNK-1 (Fig. 5J), neuron-specific Hu (Fig. 5K) and
neurofilament (Fig. 5L), and glial-specific BFABP (Fig. 5M).
We conclude that, in this chimeric system, NCC from the chick
host colonize the quail hindgut graft and form two ganglionated
plexuses containing neurons and glial cells. In addition to neural
crest-derived cells, host endothelial cells are also found in the
graft. Using QH1, a quail-specific endothelial cell maker, we
find that the intestinal grafts have a chimeric vasculature,
composed of both quail-derived (QH1+) and chick-derived
(QH1−, 8F3+) endothelial cells (not shown).
Fig. 1. Expression of ET3 and EDNRB during hindgut development. Cross-
section of an E3 embryo demonstrates ET3 transcript in the dorsal
subectodermal mesenchyme (A, arrows). Whole-mount in situ hybridization
of an E4.5 gut shows ET3 expression concentrated in the ceca and cloaca (C).
Inset shows a longitudinal section through the ceca, demonstrating mesodermal
expression. From E5 through E9, ET3 is expressed in the outer mesenchyme
along the entire midgutand hindgut,shownin longitudinal(D–F) and transverse
(B) sections. Whole-mount of an E5 gut shows EDNRB-expressing neural crest
cells in the gizzard, with the wavefront of EDNRB-expressing cells in the
duodenum (G, arrowheads). Transverse section of an E7 proximal hindgut (H)
4N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
Fig. 2. GDNF expression during hindgut development.Dorsal view of a whole-mountin situ hybridization at HH 18 stained for GDNF (A). The line indicatesthe level
of the transverse section shown in panel B, where GDNF expression is seen in the nephrogenic mesenchyme (arrows). E5 intestinal whole-mount shows expression in
the cloaca and cecal primordial (C), where expression is localized to the mesenchymal cells (D). From E6 onward, staining is observed throughout the intestine, as
shown at E7 (E) and E9 (F–H) in longitudinal (E–G) and transverse (H) sections. ep, epithelium; hg, hindgut; nt, neural tube.
Fig. 3. The neural crest cell wavefront reaches the proximal hindgut at E6. Sagittal section of the caudal end of an E6 quail embryo stained with Hu (A, boxed area
magnified in panel B), demonstrating a few NCC in the ceca and proximal hindgut (B, arrows). Removal of the postcecal hindgut at this stage and culture on the
chorioallantoic membrane (CAM) for 9 days leads to complete colonization of the ENS from those few starting cells (C). CAM culture using an E5 hindgut, where
NCC have not yet reached the hindgut, results in aganglionosis (D). hg, hindgut; mg, midgut; mp, myenteric plexus; noR, nerve of Remak; smp, submucosal plexus.
5N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
363 BQ788-treated hindguts and the data are summarized in Fig. 7.
364 ET3 treatment produced significant hyperganglionosis, while
365 inhibition of EDNRB activity yielded severe hypoganglionosis.
366 The effect on ganglion cell number was seen equally in both
367 submucosal and myenteric plexuses.
To locally inhibit EDNRB signaling and verify our findings,
369 BQ788-coated beads were implanted into the wall of E5
370 hindguts prior to coelomic transplantation (Fig. 6H). After 7
371 days of transplantation, guts were sectioned and analyzed for
372 the presence of NCC. Enteric ganglia adjacent to ET3 beads
339 Inhibition of EDNRB signaling leads to severe
340 hypoganglionosis in the hindgut
342 that NCC cannot migrate past the ceca in the absence of ET3
343 signaling. Therefore, in order to study the role of ET3 within the
344 postcecal hindgut, we used chick–quail intestinal transplants.
345 Preganglionic hindguts were isolated from E5 quail embryos.
346 The ceca, cloaca, and nerve of Remak were removed to eli-
347 minate any intrinsic sources of NCC. Hindguts were then cul-
348 tured overnight in collagen matrix containing media alone
349 (controls, n = 15; Fig. 6A), 250 ng/ml ET3 (n= 36; Figs. 6B–E),
350 or 20 μM BQ788 (n = 42; Figs. 6F, G). Following removal from
351 the collagen gel, hindguts were transplanted into the coelomic
352 cavity of E3 chicken embryos as described above. Analysis of
353 the hindguts after 7 days of incubation revealed that chicken-
354 derived neural crest cells, identified by their immunoreactivity
355 to 8F3, HNK-1, and Hu, are present in the ET3-treated quail
357 (Figs. 6B–E). These ganglia appear markedly increased in
358 number and size as compared to untreated controls (Fig. 6A).
359 Hindguts treated with BQ788, on the other hand, demonstrate
360 marked hypoganglionosis, with the development of few and
361 very small enteric ganglia (Figs. 6F, G). The total number of
362 ganglion cells was counted in untreated, ET3-treated, and
The organ culture experiments shown in Fig. 4 demonstrate
(Figs. 6I, J; n = 25) were larger than control ganglia (Fig. 6H,
n = 10) and significantly larger than ganglia adjacent to BQ788-
coated beads (Figs. 6K, L; n = 25). These results demonstrate
that EDNRB signaling, independent of its role within the ceca,
promotes the development of enteric NCC within the hindgut.
ET3 promotes proliferation of crest-derived cells in the hindgut
The effectof ET3 on ganglion size could result from an effect
on NCC proliferation, survival, or differentiation. In order to
study the effect on cell proliferation, E5 hindguts were cultured
for 3 days in the presence of BQ788 or ET3. BrdU was added to
the culture for the final 3 h. Longitudinal sections were then
analyzed by double immunofluorescence using QCPN, Hu,
HNK-1 and anti-BrdU antibodies. Numerous BrdU-positive
epithelialcells,anda smaller number of proliferating mesenchy-
mal cells, were detected in wild-type intestine (Fig. 8B).
Occasional BrdU+ crest-derived cells were identified in the
ENS(not shown).InBQ788-treated guts,onlyrare entericcrest-
derived cells expressed BrdU (Figs. 8C, D). In contrast, guts
cultured in the presence of ET3 contained numerous cells that
coexpressed BrdU and HNK-1/Hu (Figs. 8E, F). Fig. 8A sum-
marizes the results obtained from 15 hindguts, showing that the
percentage of ganglia containing BrdU+ cells is 26% in wild-
type hindguts, 10% in the presence of BQ788, and 70% in the
presence of added ET3, suggesting that ET3 promotes pro-
liferation of crest-derived cells in the avian hindgut.
The data also suggest that BQ788 treatment reduces the
density of BrdU+ cells in enteric ganglia relative to controls. To
ensure that this effect is specific to the ganglia and not a non-
specific cellular toxicity,wecountedthenumber of BrDU+ cells
in the epithelium and non-ganglionic mesenchyme per high-
power (40×) field. In ET3-treated, BQ788-treated, and control
guts, respectively, there were 24 ± 14, 22 ± 5, and 25 ± 4 BrDU+
Fig. 4. ET3-EDNRB signaling is required for hindgut colonization. E5 gut was cultured in collagen gel for 72 h in the presence of ET3 protein or BQ788. Longitudinal
sections stained with Hu antibody are shown. In wild-type E5 gut, NCC are present only in the proximal midgut (A, arrow) and absent distally. E8 wild-type intestine
contains Hu-immunoreactive cells along its entire length (B). An E5 gut cultured for 72 h with no added proteins (C) looks indistinguishable from a wild-type E8 gut
(B). When cultured in the presence of excess ET3 protein, no effect on migration is seen (D), while in the presence of BQ788, crest cell migration is arrested at the ceca
and the hindgut is aganglionic, as shown by the absence of Hu+ (E) or HNK+ (F) cells. hg, hindgut; mg, midgut; noR, nerve of Remak.
6 N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
Fig. 5. Chick–quail intestinal transplants recapitulate normal ENS development. The E5 quail aneural hindgut was removed and transplanted into the coelomic cavity
of an E3 chick embryo (A). Sagittal section of an E5 quail embryo stained with Hu (B) shows expression in the nerve of Remak (arrowheads), neural tube (nt), and
dorsal root ganglia (DRG). Cross-sections from the isolated E5 quail hindguts before (C) and after (D) removal of the nerve of Remak (noR) are shown. Immediately
followingtransplantationinto the chick coelom, the carbon-labeledquail intestinal graft is visualizedjust ventral to the aorta (E, arrow). Two daysafter transplantation,
QCPN-immunoreactive quail gut is seen adjacentto the host coelomic epithelium (F, arrow). The grafted intestine grows significantlyafter 7 days of incubation (G), as
compared to the original size of the graft at the time of transplantation (G, inset). Seven days after transplantation, the quail hindguts were analyzed by
immunocytochemistry. QCPNantibody demonstrates quail-derived cellsthroughoutthe gut, exceptfor the neural plexuses,whichremainunstained(H,arrows). These
plexuses are comprised of 8F3-immunoreactive chicken cells (I), which also express the NCC marker HNK-1 (J), neuron-specific Hu (K) and neurofilament (L), and
glial-specific BFABP (M). ep, epithelium; DRG, dorsal root ganglia; hg, hindgut; mp, myenteric plexus; noR, nerve of Remak; nt, neural tube; smp, submucosal
7N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
embryos.Ahindguttreated witha control bead(H, asterisk) isshown.Insetdemonstratesthe intestinal graft withbeadat time of isolationfrom the chickcoelom. In the
presence of an ET3-coated bead (I, J), enteric ganglia are significantly increased in size as compared to those exposed to BQ788-coated beads (K, L). ep, epithelium;
mp, myenteric plexus; smp, submucosal plexus.
406 in the mesenchyme. None of these differences were statistically
ET3 could also increase ganglion size by preventing apop-
409 tosis of crest-derived cells. Double immunofluorescence was
410 performed using an antibody to activated caspase-3, a marker of
411 apoptosis, and HNK-1. Apoptotic enteric crest cells were not
apoptotic NCC, is shown as a positive control (Fig. 8H). The
inhibition of enteric NCC proliferation observed with BQ788
treatment does not appear to be associated with increased cell
Fig. 6. The presence of excess ET3 leads to hyperganglionosis. E5 quail hindguts were treated with ET3 protein (B–E) or BQ788 (F, G) for 24 h, transplanted into the
coelomic cavity of E3 chick embryos, and allowed to develop for 7 days. A longitudinal section from a control transplanted gut without additive demonstrates normal
ganglia (A). In the presence of excess ET3, ganglia are significantly increased in size and number, as shown by QCPN (B), 8F3 (C), Hu (D), and HNK-1 (E) staining.
Inhibition of ET3 signaling with BQ788 leads to a marked reduction in the size and density of ganglia, shown by immunoreactivity to Hu (F) and HNK-1 (G).
Acrylamidebeadscoated witheither ET3(I, J) or BQ788(K, L)were implantedinto preganglionic E5 quailhindgutsthat weresubsequentlytransplanted into E3chick
8N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
447 that ET3 inhibits neuronal differentiation of enteric crest-
448 derived cells.
418 ET3 inhibits neuronal differentiation in the hindgut
420 BQ788 may result not only from inhibition of crest-derived cell
421 proliferation but also from premature differentiation of those
422 cells, preventing them from further migration. To assess for
423 neuronal differentiation, we determined the presence of nico-
424 tinamide adenine dinucleotide phosphate (NADPH) diaphorase
425 activity in hindgut explants. NADPH-diaphorase activity co-
426 localizes with nitric oxide synthase (NOS) in the central and
427 peripheral nervous system (Branchek and Gershon, 1989;
428 Dawson etal.,1991;Balaskasetal.,1995;BurnsandDelalande,
429 2005). In the ENS, neuronal NOS (nNOS) marks a subset of
430 terminally differentiated enteric neurons (Young et al., 2005),
431 unlike Hu and neurofilament, which are earlier markers of neu-
432 ronal commitment (Payette et al., 1984; Fairman et al., 1995).
NADPH-diaphorase activity is present along the entire hind-
434 gut in E8 wild-type embryos (Fig. 9A). Following a 3-day
435 incubation of an E5 hindgut with purified ET3 protein in organ
436 culture, neurofilament is expressed along the length of the gut
437 (Fig. 9B), similar to the expression of Hu (Fig. 4D). However,
438 NADPH activity is only present in the midgut and ceca, and is
439 absent from the entire hindgut (Fig. 9C), indicating a delay in
440 neuronal differentiation. Glial differentiation, as detected by
441 BFABP-immunoreactivity, is not inhibited (Fig. 9D). Guts ex-
442 posed to BQ788 show expression of neurofilament, NADPH,
443 and BFABP in the midgut and ceca (Figs. 9E–H), confirming
444 the absence of ENS colonization of the hindgut and demon-
445 strating that the most terminal cells in the ceca have undergone
446 neuronal and glial differentiation. We conclude from these data
The absence of hindgut colonization in the presence of
449 ET3 inhibits the migratory response of neural crest-derived
450 cells to GDNF
452 mediated enteric neural crest cell migration is modulated by
453 ET3 (Kruger et al., 2003; Barlow et al., 2003). To test this in the
454 avian embryo, we cultured E8 guts in collagen gel (n = 45) in
Recent studies in rodents have suggested that GDNF-
the absence (Fig. 10A) or presence of GDNF (Figs. 10B, C).
Addition of GDNF to the culture medium resulted in significant
emigration of crest-derived cells out of the midgut and ceca into
the surrounding collagen matrix. Interestingly, GDNF did not
promote any cell migration from the hindgut (Fig. 10C). The
addition of ET3 alone did not promote enteric crest cell mig-
ration (Fig. 10D). However, when ET3 and GDNF were added
together, migration was significantly decreased (Fig. 10E). ET3
inhibition of GDNF-induced migration was abolished after
addition of BQ788, which resulted in enhanced emigration of
HNK-1+ cells (Figs. 10F, G). These results indicate that ET3
inhibits the migration of midgut crest-derived cells in response
The importance of ET3 signaling in ENS formation is well
established (Baynash et al., 1994; Hosoda et al., 1994; Barlow
et al., 2003; Kruger et al., 2003; Woodward et al., 2003),
although its mechanism of action remains unclear. The most
commonly held hypothesis is that ET3 activity maintains the
pool of enteric NCC precursors by promoting their proliferation
and inhibiting their differentiation into neurons (Gershon,
1999). In the absence of ET3, therefore, premature neuronal
differentiation exhausts the pool before intestinal colonization is
complete. The observation that the aganglionosis in ET3-
deficient mice is limited to the hindgut suggests that in the
absence of ET3 signalingthere are only enough NCC precursors
to populate the intestine to the level of the cecum. This suggests
that ET3 may have a specific effect on migrating crest-derived
cells as they reach the cecum, functioning to expand the popu-
lation of precursors at that point so that they can populate the
remaining intestine. Our findings in avians, together with pre-
vious experimental work in rodents, confirm that the cecum is
central to ET3 function. For example, ET3 signaling is only
required between E10.5 and E12.5 in mice, the time at which
migrating crest cells are crossing the ileocecal junction (Shin et
al., 1999; Woodward et al., 2000; Sidebotham et al., 2002; Lee
et al., 2003). Moreover, NCC migration is normal in ET3 and
EDNRB mutant mice until those cells reach the cecum, where
there is a transient migratory arrest (Kapur et al., 1992, 1995;
Lee et al., 2003). Similarly, we find that inhibition of EDNRB
activity in the avian gut leads to arrested migration precisely at
the level of the ceca (Fig. 4E).
We examined the expression of ET3 and GDNF during avian
gut development and find both genes initially expressed in the
ceca (Figs. 1 and 2), as has been demonstrated in mice (Leibl et
al., 1999; Young et al., 2001). The cecal expression of ET3,
however, is very transient, occurring just before E5 (Fig. 1C),
and this may explain why it was not previously described in
chicks (Nataf et al., 1998). GDNF is strongly expressed in the
ceca at E5 (Figs. 2C, D). The timing of expression of both genes
is notable in that it occurs just prior to the arrival of crest-
derived cells at the cecal buds at E5.5 (Newgreen et al., 1996),
suggesting that these factors may locally influence, directly or
indirectly, enteric NCC at the migratory wavefront. Interesting-
ly, using CAM grafts, we find that the first few crest-derived
Fig. 7. Inhibition of EDNRB signaling produces severe hypoganglionosis. Guts
treated with ET3 or BQ788 for 24 h, followed by coelomic transplantation, were
stained with HNK-1 and analyzed quantitatively. The total number of ganglion
cells per high-power (200×) field was counted in untreated, ET3-treated, and
BQ788-treated guts (n = 5 from each group). Error bars represent standard error
of the mean. Statistical analysis was performed by one-way ANOVA using
Tukey's Multiple Comparison Test (PRISM software). A P value <0.05 was
considered significant. The difference between each of the three groups was
statistically significant at P < 0.01.
9 N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
positive for both BrdU and HNK-1 (E, arrows) or Hu (F, arrows). Sections from an E8 BQ788-treated gut (G) double-stained with an anti-activated caspase-3 antibody
to detect apoptotic cells and anti-Hu to label neurons. Apoptotic epithelial and mesenchymal cells were present, but there was no evidence of colocalization of Hu with
activated caspase-3 (G). The section shown in panel G is from the cecum of a treated gut. Apoptosis was detected in the dorsal root ganglion (H), used as a positive
control. drg, dorsal root ganglion; ep, epithelium; mp, myenteric plexus; smp, submucosal plexus.
510 cells that reach the postcecal intestine, sometime between E5
511 and E6, are sufficient to populate the entire hindgut (Fig. 3).
512 These results suggest two important aspects of the leading edge
513 of migrating crest-derived cells when they arrive at the ceca:
514 first, that they are exposed to a unique molecular environment
515 rich in ET3 and GDNF and, second, that those few leading cells
have a high proliferative capacity and can independently give
rise to the entire hindgut ENS.
Since the cecal expression of both ET3 and GDNF occurs at
least 0.5 days before NCC arrival, these factors may influence
arriving crest-derived cells indirectly by altering the mesen-
chymal environment of the ceca in anticipation of NCC arrival.
Fig.8.ET3promotesentericneural crestcell proliferation inthe hindgut.E5hindgutswereculturedfor 3daysin the presenceof BQ788(C,D)or ET3(E,F) andBrdU
labeling performed. The presence of added ET3 led to a significantly higher percentage of BrdU-positive enteric ganglia as compared to controls and BQ788-treated
guts (A). Double immunohistochemistry was performed on longitudinal sections through the midgut and hindgut. Numerous BrdU-positive epithelial cells, and
scattered mesenchymal cells, were identified throughout the hindgut in an untreated control (B). Guts treated with BQ788 show very rare BrdU incorporation into
HNK-1+ (C) or Hu+ (D) enteric neural crest cells in the midgut and ceca. In contrast, guts cultured in the presence of ET3 contain many ganglia in the hindgut that are
10N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
531 ceca and on the arriving crest-derived cells.
In order to investigate the role of ET3-EDNRB signaling in
533 the avian hindgut ENS, we developed a method for studying
534 ENS development using coelomic transplantation. Grafting of
535 tissue into the avian coelom is a well-established technique
536 that has been used to study development of the limb bud
537 (Hamburger, 1939), lymphoid organs (Nagy et al., 2004), and
540 mesenchymal interactions during gut development (Kedinger et
522 This idea has been proposed in mice, where EDNRB activation
523 inhibits laminin α1 production (Wu et al., 1999). Since laminin
524 α1 normally promotes neurogenesis (Chalazonitis et al., 1997),
525 the loss of EDNRB activity would lead to an environment
526 promoting premature neuronal differentiation. This model may
527 not be relevant in avians, however, since EDNRB expression
528 does not appear to be present in the non-NCC mesenchyme
529 (Fig. 1; Nataf et al., 1996). ET3 and GDNF may have another
530 effect, yet to be defined, on the molecular environment of the
al., 1981; Bolcato-Bellemin et al., 2003). However, ENS
development has not been studied using this system. Pregangli-
onic quail hindgut was transplanted into the chick coelom and
allowed to develop for 7 days. We find that NCC from the chick
host colonize the quail hindgut and form two ganglionated
plexuses that are grossly indistinguishable from a normal ENS
is not surprising, given that others have previously created
chimeras by coculturing the neural tube and intestine from each
species onto the CAM (Teillet, 1978; Rothman et al., 1986;
Hearn and Newgreen, 2000). By adapting the technique of
coelomic grafting to create our chimeras, we have developed a
useful model system in which we are able to manipulate the
molecular environment of the aneural hindgut prior to grafting
advantages of this technique is that coelomic grafting can be
performed on a 3-day-old chick embryo, unlike CAM grafting,
which cannot be done prior to E8, allowing longer incubation
times in the coelom. We are currently determining the origin of
the donor cells to understand what molecular environment they
arise in, and travel through, to get to the transplanted gut.
Preliminary evidence using DiI labeling of the host neural tube
suggests that the graft is colonized by vagal NCC from the host
embryo (unpublished results).
These coelomic grafts are different from wild-type hindgut in
several respects. Sacral crest cells from the host chick do not
appear to contribute to the grafted hindgut, nor do quail sacral-
derived cells, since the nerve of Remak is removed prior to
grafting. While the nerve of Remak normally extends nerve
fibers into the gut as early as E7.5, sacral NCC do not become
incorporated into myenteric and submucosal ganglia until E10–
E12 (Burns and Le Douarin, 1998, 2001), the stage at which our
coelomic grafts are harvested (E5 graft +7-day coelomic
<2% of ganglion cells in the submucosal plexus of the colo-
rectum, while in the myenteric plexus they make up 0.3% of
ganglion cells in the proximal colorectum and 17% distally
(Burns and Le Douarin, 1998). The ENS effects we observed in
our experiments were present equally throughout both plexuses,
proximally and distally. We therefore believe that the absence of
the nerve of Remak is unlikely to have a significant effecton our
observations, although a better understanding of the role of this
avian-specific structure is needed.
The ceca are also removed from the grafts prior to trans-
plantation in order to eliminate quail-derived NCC. Migrating
crest-derived cells thus enter the hindgut directly, without
passing through the cecal environment. Whether this alters their
development is unknown. Cocultures of ganglionated mouse
in grossly normal colonization of the hindgut, suggesting that
migration through the cecum is not essential for hindgut ENS
formation (Young et al., 2002). Our untreated coelomic grafts
also appear to develop a normal ENS (Fig. 6A), although we
cannot exclude the possibility that a molecular “modification”
normally occurs during migration through the ceca. Thus, while
coelomic grafts are a useful method for studying ENS deve-
lopment, the potential consequences of removing the nerve of
Fig. 9. ET3 inhibits neuronal differentiation in the ENS. NADPH-diaphorase
activity is present along the entire hindgut in E8 wild-type embryos (A). E5
hindgut cultured for 3 days in the presence of excess ET3 shows neurofilament-
positive cells along the length of the hindgut (B). However, NADPH activity is
only present in the midgut and ceca, and is absent from the hindgut (C). Glial
differentiation, as detected by BFABP-immunoreactivity, is present throughout
the hindgut (D). Exposure to BQ788 led to expression of neurofilament (E),
NADPH (F, boxed area magnified in panel G), and BFABP (H) in the midgut
and ceca, consistent with normal differentiation of the most distal cells in these
11N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
615 NCC (Stone et al., 1997; Lahav et al., 1996) and of undiffe-
616 rentiated enteric crest-derived cells (Barlow et al., 2003; Hearn
617 et al., 1998). We also find no evidence of increased apoptosis
618 when EDNRB signaling is inhibited (Fig. 8G), in agreement
619 with Lee et al. (2003), who did not observe increased apoptotic
620 cell death in the ENS of EDNRB-deficient mice.
In addition to promoting proliferation of enteric crest-derived
622 cells, ET3 also inhibits neuronal differentiation, as demonstrat-
623 ed by the absence of nNOS-expressing cells in ET3-treated
624 hindguts (Fig. 9C). NOS is the first neurotransmitter synthetic
598 Remak and ceca should be considered. In general, when using
599 avian embryos for studying ENS development, one should
600 consider that the nerve of Remak and the cecal appendages in
601 avians represent potentially important differences from mam-
602 mals that could affect the results obtained.
Using coelomic grafts, we find that inhibition of EDNRB
604 signaling in the hindgut leads to severe hypoganglionosis, while
606 This impact on ganglion cell number can result from effects on
607 NCC proliferation, survival, and/or differentiation. We find sig-
608 nificantly more BrdU-positive ganglia in the presence of excess
610 derived cells in the presence of BQ788 (Fig. 8). The effect is
611 specific to enteric ganglion cells, as we observe no difference in
613 NCC mesenchyme. This result is consistent with previous in
614 vitro work showing that ET3 promotes proliferation of cultured
enzyme expressed by mature enteric neurons (Branchek and
Gershon, 1989). These terminally differentiated crest-derived
cells are present in the ceca of BQ788-treated guts (Figs. 9F, G),
reflecting their premature differentiation and accounting for
their inability to migrate further along the gut, as initially
suggested by Gershon (1999). Conversely, the absence of ter-
minally differentiated neurons in ET3-treated hindguts (Fig. 9C)
may account for the continued proliferation and migration of
those undifferentiated crest-derived cells.
The importance of the ceca to ET3 function is further sup-
ported by the effect of ET3 on NCC migration. GDNF is known
to be chemoattractive to enteric crest cells (Young et al., 2001).
However, its strong expression in the ceca raises the question of
how enteric NCC are able to migrate past that GDNF-rich
segment of the gut. Recent work in rodents suggests that ET3
modulates the chemoattractive response of enteric NCC to
GDNF (Barlow et al., 2003; Kruger et al., 2003). We find that
this is also true in the avian gut, where ET3 inhibits the
chemoattraction to GDNF (Fig. 10), and by so doing may allow
crest-derived cells to migrate past the ceca and into the hindgut.
BMP (Goldstein et al., 2005) and Shh (Fu et al., 2004) signaling
has also been shown to modulate the migratory response to
GDNF, suggesting potential interactions among these various
pathways in the regulation of ENS migration. Once NCC are
past the ceca and in the hindgut, GDNF no longer has a chemo-
attractive role (Fig. 10), suggesting the presence of other che-
moattractive factors in the distal gut.
Fig. 10. ET3 inhibits the chemoattractive effect of GDNF. E8 gut was grown in collagen gel in the absence (A) or presence of 10 ng/ml GDNF (B, C), 250 ng/ml ET3
(D), GDNF+ET3 (E), or GDNF+ET3+ 5 μM BQ788 (F). While GDNF normally stimulates migration of neural crest cells from the midgut into the media (B, C), the
presence of ET3 inhibits this GDNF-induced cell migration (E). The addition of BQ788 relieves the inhibitory effect of ET3 (F). The migrating cells are
immunoreactive to HNK-1 antibody, confirming that they represent NCC (G). hg, hindgut; mg, midgut.
12 N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
We are grateful to Drucilla Roberts for her mentorship,
698 enthusiasm, and valuable insight. We thank Katie Brewer and
699 Olive Mwizerwa for excellent technical assistance, Herve
700 Kempf and Andrew Lassar for providing ET3 and EDNRB
701 probes, and Carmen Birchmeier for Bfabp antibody. The
702 monoclonal antibodies QCPN, 8F3, QH1, and 4H6 (anti-
703 neurofilament) were obtained from the Developmental Studies
704 Hybridoma Bank developed under the auspices of the NICHD
705 and maintained by The University of Iowa, Department of
706 Biological Sciences, Iowa City, IA 52242. This work was
653 on enteric crest-derived cell proliferation, differentiation, and
654 migration corroborate previous observations in cultured NCC
655 and enteric crest-derived cells from mice, we recognize that
656 ET3 activity may exert slightly different effects in avians
657 and mammals. The absence of EDNRB expression in the
658 intestinal non-NCC mesenchyme of avians (Fig. 1; Nataf et
659 al., 1996), for example, illustrates one such difference from
660 mammals (Okabe et al., 1995; Pla and Larue, 2003) and
661 may impact on the mechanism of action of this signaling
662 pathway. Identifying other avian-specific differences will
663 be important to interpreting results obtained from these and
664 similar experiments.
Our results suggest a role for ET3-EDNRB signaling, not
666 only in the ceca, but also in the hindgut, where ET3 becomes
667 expressed as NCC are arriving (Figs. 1D, E). As shown in Figs.
668 4E, F, in the presence of ceca, EDNRB inhibition leads to
669 hindgut aganglionosis. However, when ceca are excluded, as in
670 the coelomic grafts, NCC enter the EDNRB-deficient avian
671 hindgut, but only in very small numbers (Figs. 6F, G; Fig. 7).
672 This hypoganglionosis suggests that EDNRB signaling may be
673 necessary to ensure the development of a normal complement of
674 enteric ganglion cells in the hindgut. The requirement for
675 EDNRB signaling within the hindgut is in agreement with the
676 findings of Wu et al. (1999), who demonstrated in explant
677 cultures that the terminal colon of ET3-deficient mice remains
678 aganglionic until exogenous ET3 is added to the culture
679 medium, promoting the migration of crest-derived cells into
680 the aganglionic region. The notion that ET3 activity not only
681 maintains a pool of enteric crest cell precursors but also has a
682 critical role within the hindgut itself may help to explain why
683 sacral crest-derived cells do not contribute to the distal colo-
684 rectum in ET3- and EDNRB-deficient mice (Baynash et al.,
685 1994; Hosoda et al., 1994). If the defect was entirely due to
686 premature neuronal differentiation and failure to complete mig-
687 ration, one would expect sacral crest-derived cells to contribute
688 to the distal bowel in those mutants. Our results suggest that
689 deficient ET3 signaling may create a hindgut environment that
690 does not allow normal colonization by enteric crest-derived
691 cells. The presence of a small number of ganglion cells in our
692 grafts may be due to incomplete inhibition of EDNRB activity.
693 In conclusion, ET3 signaling promotes enteric crest-derived cell
694 proliferation, inhibits neuronal differentiation, and modulates
695 migration in the avian hindgut ENS.
While our findings on the effect of ET3-EDNRB signaling
supported by NIH K08HD46655-01 (AMG) and a grant from
the Charles H. Hood Foundation (AMG).
Appendix A. Supplementary data
Supplementary data associated with this article can be found
in the online version at doi:10.1016/j.ydbio.2006.01.032.
Balaskas, C., Saffrey, M.J., Burnstock, G., 1995. Distribution of NADPH-
diaphorase activity in the embryonic chicken gut. Anat. Embryol. 192,
Barlow, A., de Graaff, E., Pachnis, V., 2003. Enteric nervous system progenitors
are coordinately controlled by the G protein-coupled receptor EDNRB and
the receptor tyrosine kinase RET. Neuron 40, 905–916.
Baynash, A.G., Hosoda, K., Giaid, A., Richardson, J.A., Emoto, N., Hammer,
R.E., Yanagisawa, M., 1994. Interaction of endothelin-3 with endothelin-B
receptor is essential for development of epidermal melanocytes and enteric
neurons. Cell 79, 1277–1285.
Bolcato-Bellemin, A., Lefebvre, O., Arnold, C., Sorokin, L., Miner, J.H.,
Kedinger, M., Simon-Assmann, P., 2003. Laminin α5 chain is required for
intestinal smooth muscle development. Dev. Biol. 260, 376–390.
Branchek, T.A., Gershon, M.D., 1989. Time course of expression of
neuropeptide Y, calcitonin gene-related peptide, and NADPH diaphorase
activity in neurons of the developing murine bowel and the appearance of 5-
hydroxytryptamine in mucosal enterochromaffin cells. J. Comp. Neurol.
Burns, A.J., Delalande, J.M., 2005. Neural crest cell origin for intrinsic ganglia
of the developing chicken lung. Dev. Biol. 277, 63–79.
Burns, A.J.,Le Douarin, N.M., 1998.The sacral neural crest contributesneurons
and glia to the post-umbilical gut: spatiotemporal analysis of the
development of the enteric nervous system. Development 125, 4335–4347.
Burns, A.J., Le Douarin, N.M., 2001. Enteric nervous system development:
analysis of the selective developmental potentialities of vagal and sacral
neural crest cells using quail–chick chimeras. Anat. Rec. 262, 16–28.
Caprioli, A., Jaffredo, T., Gautier, R., Dubourg, C., Dieterlen-Lièvre, F., 1997.
Blood-borne seeding by hematopoietic and endothelial precursors from the
allantois. Proc. Natl. Acad. Sci. U. S. A. 95, 1641–1646.
Catala, M., Teillet, M.A., Le Douarin, N.M., 1995. Organization and
development of the tail bud analyzed with the quail–chick chimaera system.
Mech. Dev. 51, 51–65.
Chakravarti, A.L., 2001. Hirschsprung's disease. In: Scriver, C.R., Beaudet,
A.L., Valle, D., Sly, W.S., Childs, B., Kinzler, K., Vogelstein, B. (Eds.),
The Metabolic and Molecular Bases of Inherited Diseases. McGraw-Hill,
New York, pp. 6231–6255.
Chalazonitis, A., Rothman, T.P., Chen, J., Lamballe, F., Barbacid, M., Gershon,
M.D., 1994. Neurotrophin-3 induces neural crest-derived cells from fetal rat
gut to develop in vitro as neurons or glia. J. Neurosci. 14, 6571–6584.
Chalazonitis, A.,Tennyson,V.M., Kibbey, M.C.,Rothman,T.P.,Gershon,M.D.,
1997.The alpha1 subunitof laminin-1 promotesthe developmentof neurons
by interacting with LBP110 expressed by neural crest-derived cells
immunoselected from the fetal mouse gut. J. Neurobiol. 33, 118–138.
Dawson, T.M., Bredt, D.S., Fotuhi, M., Hwang, P.M., Snyder, S.H., 1991. Nitric
oxide synthase and neuronal NADPH diaphorase are identical in brain and
peripheral tissues. Proc. Natl. Acad. Sci. U. S. A. 88, 7797–7801.
Doyle, A.M., Roberts, D.J., Goldstein, A.M., 2004. Enteric nervous system
patterning in the avian hindgut. Dev. Dyn. 229, 708–712.
Fairman, C.L., Clagett-Dame, M., Lennon, V.A., Epstein, M.L., 1995.
Fu, M., Lui, V.C., Sham, M.H., Pachnis, V., Tam, P.K., 2004. Sonic hedgehog
regulates the proliferation, differentiation, and migration of enteric neural
crest cells in gut. J. Cell Biol. 166, 673–684.
Gershon, M.D., 1999. Endothelin and the development of the enteric nervous
system. Clin. Exp. Pharmacol. Physiol. 26, 985–988.
Goldstein, A.M., Brewer, K.C., Doyle, A.M., Nagy, N., Roberts, D.J., 2005.
13N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
Nagy, N., Magyar, A., Toth, M., Olah, I., 2004. Origin of the bursal secretory
dendritic cell. Anat. Embryol. 208, 97–107.
Nagy, N., Biro, E., Takacs, A., Polos, M., Magyar, A., Olah, I., 2005. Peripheral
blood fibrocytes contribute to the formation of the avian spleen. Dev. Dyn.
Nataf, L., Lecoin, V., Eichmann, A., Le Douarin, N.M., 1996. Endothelin-B
receptor is expressed by neural crest cells in the avian embryo. Proc. Natl.
Acad. Sci. U. S. A. 93, 9645–9650.
Nataf, V., Amemiya, A., Yanagisawa, M., Le Douarin, N.M., 1998. The expres-
sion pattern of endothelin 3 in the avian embryo. Mech. Dev. 73, 217–220.
Natarajan, D., Marcos-Gutierrez, C., Pachnis, V., de Graaff, E., 2002.
BMP signaling is necessary for neural crest cell migration and ganglion
formation in the enteric nervous system. Mech. Dev. 122, 821–833.
Hamburger, V., 1939. The development and innervation of transplanted limb
primordia of chick embryos. J. Exp. Zool. 80, 347–389.
Hamburger, V., Hamilton, H.L., 1951. A series of normal stages in the deve-
lopment of the chick embryo. J. Morphol. 88, 49–92.
Hearn, C., Newgreen, D., 2000. Lumbo-sacral neural crest contributes to the
avian enteric nervous system independently of vagal neural crest. Dev. Dyn.
Hearn, C.J., Murphy, M., Newgreen, D., 1998. GDNF and ET3 differentially
modulate the numbers of avian enteric neural crest cells and enteric neurons
in vitro. Dev. Biol. 197, 93–105.
Homma, S., Oppenheim, R.W., Yaginuma, H., Kimura, S., 2000. Expression
pattern of GDNF, c-ret, and GFRalphas suggests novel roles for GDNF
ligands during early organogenesis in the chick embryo. Dev. Biol. 217,
Hosoda, K., Hammer, R.E., Richardson, J.A., Baynash, A.G., Cheung, J.C.,
Giaid, A., Yanagisawa, M., 1994. Targeted and natural (piebald-lethal)
mutations of endothelin-B receptor gene produce megacolon associated with
spotted coat color in mice. Cell 79, 1267–1276.
Jiang, Y., Liu, M.T., Gershon, M.D., 2003. Netrins and DCC in the guidance of
migrating neural crest-derived cells in the developing bowel and pancreas.
Dev. Biol. 258, 364–384.
Jones, D.S., 1942. The origin of vagi and the parasympathetic ganglion cells of
the viscera of the chick. Anat. Rec. 82, 185–197.
Jowett, T., 1999. Analysis of protein and gene expression. Methods Cell Biol.
Kapur, R.P., Yost, C., Palmiter, R.D., 1992. A transgenic model for studying
development of the enteric nervous system in normal and aganglionic mice.
Development 116, 167–175.
Kapur, R.P., Sweetser, D.A., Doggett, B., Siebert, J.R., Palmiter, R.D., 1995.
Intercellular signals downstream of endothelin receptor-B mediate coloni-
zation of the large intestine by enteric neuroblasts. Development 121,
Kedinger, M., Simon, P.M., Grenier, J.F., Haffen, K., 1981. Role of epithelial–
mesenchymal interactions in the ontogenesis of intestinal brush-border
enzymes. Dev. Biol. 86, 339–347.
Kruger, G.M., Mosher, J.T., Tsai, Y.H., Yeager, K.J., Iwashita, T., Gariepy, C.E.,
Morrison, S.J., 2003. Temporally distinct requirements for endothelin
receptor B in the generation and migration of gut neural crest stem cells.
Neuron 40, 917–929.
Lahav, R., Ziller, C., Dupin, E., Le Douarin, N.M., 1996. Endothelin 3 promotes
neural crest cell proliferation and mediates a vast increase in melanocyte
number in culture. Proc. Natl. Acad. Sci. U. S. A. 93, 3892–3897.
Le Douarin, N.M., Teillet, M.A., 1973. The migration of neural crest cells to the
wall of the digestive tract in avian embryo. J. Embryol. Exp. Morphol. 30,
Lee, H.O., Levorse, J.M., Shin, M.K., 2003. The endothelin receptor-B is
required for the migration of neural crest-derived melanocyte and enteric
neuron precursors. Dev. Biol. 259, 162–175.
Leibl, M.A., Ota, T., Woodward,M.N., Kenny, S.E.,Lloyd,D.A.,Vaillant,C.R.,
Edgar, D.H., 1999. Expression of endothelin 3 by mesenchymal cells of
embryonic mouse caecum. Gut 44, 246–252.
Moore, M.W., Klein, R.D., Farinas, I., Sauer, H., Armanini, M., Phillips, H.,
Reichardt, L.F., Ryan, A.M., Carver-Moore, K., Rosenthal, A., 1996.
Renal and neuronal abnormalities in mice lacking GDNF. Nature 382,
Requirement of signalling by receptor tyrosine kinase RET for the directed
migration of enteric nervous system progenitor cells during mammalian
embryogenesis. Development 129, 5151–5160.
Newgreen, D.F., Southwell, B., Hartley, L., Allan, I.J., 1996. Migration of
enteric neural crest cells in relation to growth of the gut in avian embryos.
Acta Anat. 157, 105–115.
Okabe, H., Chijiiwa, Y., Nakamura, K., Yoshinaga, M., Akiho, H., Harada, N.,
Nawata, H., 1995. Two endothelin receptors (ETA and ETB) expressed on
circular smooth muscle cells of guinea pig cecum. Gastroenterology 108,
Payette, R.F., Bennett, G.S., Gershon, M.D., 1984. Neurofilament expression in
vagal neural crest-derived precursors of enteric neurons. Dev. Biol. 105,
Pichel, J.G., Shen, L., Sheng, H.Z., Granholm, A.C., Drago, J., Grinberg, A.,
Lee, E.J., Huang, S.P., Saarma, M., Hoffer, B.J., Sariola, H., Westphal, H.,
1996. Defects in enteric innervation and kidney development in mice
lacking GDNF. Nature 382, 73–76.
Pla, P., Larue, L., 2003. Involvement of endothelin receptors in normal and
pathological development of neural crest cells. Int. J. Dev. Biol. 47,
Pomeranz, H.D., Rothman, T.P., Gershon, M.D., 1991. Colonization of the post-
umbilical bowel by cells derived from the sacral neural crest: direct tracing
of cell migration using an intercalating probe and a replication-deficient
retrovirus. Development 111, 647–655.
Ramalho-Santos, M., Melton, D.A., McMahon, A.P., 2000. Hedgehog signals
regulate multipleaspects of gastrointestinal development.Development 127,
Rothman,T.P., Sherman,D.,Cochard,P., Gershon, M.D.,1986.Development of
the monoaminergic innervation of the avian gut: transient and permanent
expression of phenotypic markers. Dev. Biol. 116, 357–380.
Sanchez, M.P., Silos-Santiago, I., Frisen, J., He, B., Lira, S.A., Barbacid, M.,
1996. Renal agenesis and the absence of enteric neurons in mice lacking
GDNF. Nature 382, 70–73.
Schiltz, C.A., Benjamin, J., Epstein, M.L., 1999. Expression of the GDNF
receptors ret and GFRalpha1 in the developingavian enteric nervoussystem.
J. Comp. Neurol. 414, 193–211.
Schuchardt, A., D'Agati, V., Larsson-Blomberg, L., Costantini, F., Pachnis, V.,
1994. Defects in the kidney and enteric nervous system of mice lacking the
tyrosine kinase receptor Ret. Nature 367, 380–383.
Serbedzija, G.N., Burgan, S., Fraser, S.E., Bronner-Fraser, M., 1991. Vital dye
labelling demonstrates a sacral neural crest contribution to the enteric
nervous system of chick and mouse embryos. Development 111, 857–866.
Shin, M.K., Levorse, J.M., Ingram, R.S., Tilghman, S.M., 1999. The temporal
requirement for endothelin receptor-B signalling during neural crest
development. Nature 402, 496–501.
Sidebotham, E.L., Woodward, M.N., Kenny, S.E., Lloyd, D.A., Vaillant, C.R.,
Edgar, D.H., 2002. Localization and endothelin-3 dependence of stem cells
of the enteric nervous system in the embryonic colon. J. Pediatr. Surg. 37,
Stone, J.G., Spirling, L.I., Richardson, M.K., 1997. The neural crest population
responding to endothelin-3 in vitro includes multipotent cells. J. Cell Sci.
Sukegawa, A., Narita, T., Kameda, T., Saitoh, K., Nohno, T., Iba, H., Yasugi, S.,
Fukuda,K.,2000.Theconcentricstructureof thedevelopinggutis regulated
by Sonic hedgehog derived from endodermal epithelium. Development 127,
Teillet, M.A., 1978. Evolution of the lumbo-sacral neural crest in the avian
embryo: origin and differentiation of the ganglionated nerve of Remak
studied in interspecific quail–chick chimerae. Roux's Arch. Dev. Biol. 184,
Woodward, M.N., Kenny, S.E., Vaillant, C.R., Lloyd, D.A., Edgar, D.H., 2000.
Time-dependent effects of endothelin-3 on enteric nervous system
development in an organ culture model of Hirschsprung's disease. J. Pediatr.
Surg. 35, 25–29.
Woodward, M.N., Sidebotham, E.L., Connell, M.G., Kenny, S.E., Vaillant,
C.R., Lloyd, D.A., Edgar, D.H., 2003. Analysis of the effects of
endothelin-3 on the development of neural crest cells in the embryonic
mouse gut. J. Pediatr. Surg. 38, 1322–1328.
14N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS
Wu, J.J., Chen, J.X., Rothman, T.P., Gershon, M.D., 1999. Inhibition of in vitro
enteric neuronal development by endothelin-3: mediation by endothelin B
receptors. Development 126, 1161–1173.
Young, H.M., Hearn, C.J., Farlie, P.G., Canty, A.J., Thomas, P.Q., Newgreen,
D.F., 2001. GDNF is a chemoattractant for enteric neural cells. Dev. Biol.
Young, H.M., Jones, B.R., McKeown, S.J., 2002. The projections of early
enteric neurons are influenced by the direction of neural crest cell migration.
J. Neurosci. 22, 6005–6018.
Young, H.M., Turner, K.N., Bergner, A.J., 2005. The location and phenotype of
proliferating neural-crest-derived cells in the developing mouse gut. Cell
Tissue Res. 320, 1–9.
15N. Nagy, A.M. Goldstein / Developmental Biology xx (2006) xxx–xxx
ARTICLE IN PRESS