Brief Communication 1481
Pharyngeal arch patterning in the absence of neural crest
Emma Veitch*†, Jo Begbie*, Thomas F. Schilling‡, Moya M. Smith†
and Anthony Graham*
Pharyngeal arches are a prominent and critical feature
of the developing vertebrate head. They constitute a
series of bulges within which musculature and skeletal
elements form; importantly, these tissues derive from
different embryonic cell types [1]. Numerous studies
have emphasised the role of the cranial neural crest,
from which the skeletal components derive, in
patterning the pharyngeal arches [2–4]. It has never
been clear, however, whether all arch patterning is
completely dependent on this cell type. Here, we show
that pharyngeal arch formation is not coupled to the
process of crest migration and, furthermore, that
pharyngeal arches form, are regionalised and have a
sense of identity even in the absence of the neural
crest. Thus, vertebrate head morphogenesis can now be
seen to be a more complex process than was previously
believed and must result from an integration of both
neural-crest-dependent and -independent patterning
mechanisms. Our results also reflect the fact that the
evolutionary origin of pharyngeal segmentation
predates that of the neural crest, which is an exclusively
vertebrate characteristic.
Addresses: *Molecular Neurobiology Group, Guy’s, King’s, and St
Thomas’s Schools of Medicine, Dentistry and Biomedical Sciences,
King’s College, Guy’s Campus, London Bridge, London SE1 9RT, UK.
†Department of Craniofacial Development, Guy’s, King’s, and St
Thomas’s Schools of Medicine, Dentistry and Biomedical Sciences,
King’s College, Guy’s Campus, London Bridge, London SE1 9RT, UK.
‡Department of Anatomy and Developmental Biology, University
College London, London WC1E 6BT, UK.
Correspondence: Anthony Graham
E-mail: anthony.graham@kcl.ac.uk
Received: 18 October 1999
Accepted: 10 November 1999
Published: 6 December 1999
Current Biology 1999, 9:1481–1484
0960-9822/99/$ – see front matter
© 1999 Elsevier Science Ltd. All rights reserved.
Results and discussion
Relationship between neural crest migration and
pharyngeal arch formation
The timing of emergence and contribution of neural
crest from distinct axial levels of the neural tube to the
pharyngeal arches has been well studied [5–7]. The rela-
tionship between the timing of the arrival of the crest in
the periphery with the formation of the endodermal
pouches, which delineate the pharyngeal arches, has not
been closely scrutinised, however. We therefore investi-
gated the expression patterns in chick embryos of Bmp-7,
which marks the pharyngeal pouches [8,9], and AP-2α,
which labels the migrating cranial neural crest [10].
Interestingly, we found that there was no strict relation-
ship between pharyngeal pouch formation and the arrival
of the neural crest. Thus, although the first and third
pouches form just after crest has arrived in the periphery
(Figure 1a,d), the timing of the formation of the second
pouch does not seem to relate to arrival of crest in the
same fashion. At the 15 somite stage, Bmp-7-expressing
pharyngeal endoderm is forming the second pouch
before contact with the crest (Figure 1a–c), and it is only
at later stages that this pouch contacts the migrating crest
(Figure 1d). Consequently, the formation of the pouches
seems to be temporally decoupled from the process of
crest migration.
Regional markers of pharyngeal epithelia
To assess pharyngeal arch regionalisation, we have exam-
ined genes expressed within localised domains of the pha-
ryngeal epithelia. One of the earliest genes expressed in
the pharyngeal pouch endoderm is Bmp-7, which labels
the posterior endodermal margin of each arch from Ham-
burger and Hamilton (HH) stage 15 (25 somites) onwards
(Figure 2a,f) [8]. Notably, this gene is also more strongly
expressed in the posterior endodermal margin of the
second arch than the other arches (Figure 2a). By contrast,
although Fgf-8 expression also marks the pharyngeal
endoderm [8,11], it is localised to the anterior endodermal
margin of each arch as well as the overlying ectoderm
(Figure 2b,g). Pax1 expression provides a marker of proxi-
modistal pattern in that it is specifically expressed in the
most proximal (dorsal) endoderm of each pouch
(Figure 2c,h) [12]. Finally, we also used Shh as a marker
because it labels a distinct stripe at the posterior endoder-
mal border of the second arch at HH15 [8] and later in the
posterior endoderm of the third arch (Figure 2d,i).
Effect of neural crest ablation on pharyngeal arch patterning
To determine whether pharyngeal arches form and are
correctly patterned in the absence of the neural crest, we
ablated the neural tube before neural crest formation
from the mid-mesencephalon to the level of somite 1–4
(shown in Figure 3j) and then re-incubated for 24 h.
Unfortunately, we were not able to incubate the embryos
much beyond this stage, as the limit of viability after
these ablations is stage HH17/18. To minimise the likeli-
hood of regeneration, the entire dorsoventral extent of
the neural tube was removed before crest emigration
1482 Current Biology Vol 9 No 24

Figure 1

Timing of pharyngeal pouch formation and

(a) (b) neural crest migration. In situ hybridisation
ov with probes for Ap2-α (developed with Fast
Red) and Bmp-7 (developed with NBT and
BCIP, seen in brown). Ov, otic vesicle; pp1
and pp2 (black arrows), first and second
pp1 pharyngeal pouches, respectively; white
pp2
arrow, limits of the second arch crest stream.
(a) Lateral view of a 15 somite (15s) embryo,
with anterior to the right. (b) Section through
15s 15s the rhombomere 4 level of a 15s embryo; the
second arch migratory crest is located
(c) (d) subectodermally and is distant from the
pharyngeal endoderm. (c) Adjacent transverse
ov
section through the level of the otic vesicle,
which forms in a crest-free region, showing
the formation of the second pharyngeal pouch
at this axial level. (d) Lateral view of a 19s
pp2
embryo with anterior to the right, showing the
pp2 pp1 crest streams and the first and second
pouches. Scale bars represent 100 µm.

15s 19s
Current Biology

(Figure 3j), and the surgery was performed between the 7
and 10 somite stages when the capacity for regeneration
has significantly decreased [13,14]. The embryos were then
analysed by double in situ hybridisation with each of the
pharyngeal arch markers described above and Dlx-2 (which
marks the branchial arch neural crest; Figure 2e,j) [9].
These analyses allowed us to determine which branchial
arches the ablation has rendered devoid of crest and the
extent to which these arches are regionalised.
Ablated embryos (Figure 3a–d) expressed Dlx2 (shown in
red) in the first pharyngeal arch. This probably indicates
the contribution of crest from the rostral midbrain
regions, or the emigration of some prospective first arch

Figure 2

Bmp-7 Fgf-8 Pax1 Shh Dlx-2

(a) ov (b) (c) (d) (e) (k) HB
ov
pp3 pp2 pp1
4 3 2 1

(f) (g) (h) (i) (j) (l)

pp1
2
pp2
3
pp3
4
Current Biology

Expression of pharyngeal arch markers. In situ hybridisation of stage
HH15 embryos with Bmp-7, Fgf-8, Pax1, Shh and Dlx-2 probes.
(a–e) Whole-mount embryos with anterior to the right and dorsal to the
top. (f–i) Horizontal sections with anterior to the top. (j) A transverse
section through the first arch. (a,f) Bmp-7 is expressed in the posterior
endoderm of each arch (arrows). (b,g) Fgf-8 is expressed in the anterior
endoderm of each arch (arrows). (c,h) Pax-1 is expressed in each
dorsal pouch (arrows in (c)). (d,i) Shh is expressed in the posterior
endoderm of arches 2 and 3 (arrowheads). (e,j) Dlx-2 is expressed in
the neural crest cells of the arches (white arrows). (k,l) Schematic
representations of the anatomy of this region of the embryo. (k) A side
view of pharyngeal arch region. (l) A horizontal section through the
pharyngeal arches. OV, otic vesicle; HB, hindbrain; pp1–pp3,
endodermal pouches; 1–4, arches. All scale bars represent 100 µm.

Figure 3
Bmp-7 Fgf-8
(a) (b) (c)
(e) (f) (g)
Development of the pharyngeal arches in neural-crest-ablated
embryos. In total, 55 embryos were analysed by double in situ
hybridisation and immunohistochemistry. The stages at time of
operation were as follows: 7s, n = 10; 8s, n = 15; 9s, n = 20; 10s,
n = 10. Dark brown colour indicates (a,e,i) Bmp-7 RNA, (b,f) Fgf-8
RNA, (c,g) Pax1 RNA, (d,h) Shh RNA. (a–h) Red colour indicates
Dlx-2 RNA; (i) light brown indicates staining with an anti-HNK-1
antibody. (a–d,i) Lateral views with anterior to the right; (e–h) horizontal
sections through the arches with anterior to the top. It is clear that the
crest before ablation. The more caudal pharyngeal
region, arches 2 and 3, were entirely crest-free, however,
and surprisingly the pharyngeal arches and endodermal
pouches formed. Bmp-7 was clearly expressed in the pha-
ryngeal endoderm and, importantly, it was localised to
the posterior margin, as is seen in normal embryos
(Figure 3a,e). Similarly, Fgf-8 expression in the ablated
embryos was unperturbed and was evident in the ecto-
derm overlying arch 2 and 3 and, particularly, in the ante-
rior endodermal margin of the second arch (Figure 3b,f).
Furthermore, Pax1 expression defined dorsal pouch
endoderm even in the absence of crest (Figure 3c,g).
Finally, Shh expression was also unaltered by the ablation
of the crest, and two patches of expression in the poste-
rior arch margins of arches two and three were still clearly
evident (Figure 3d,h). Finally, to be certain that the crest
ablations did generate arches devoid of crest, we also, by
combining immunohistochemistry with in situ hybridisa-
tion, analysed the ablated embryos with the crest marker
HNK-1 [15] and our arch epithelial markers. As shown in
Figure 3i, we again found that our neural tube ablations
generated arches that were devoid of HNK-1 expressing
cells, shown in light brown, but that still exhibited normal
Bmp-7 expression. Thus, the expression of the regionally
restricted arch epithelial markers in the ablated embryos
demonstrates that pharyngeal pouches and arches form
correctly, and that both proximodistal and anteroposterior
polarity arise independently of neural crest.
Brief Communication 1483
Pax1 Shh Bmp-7/HNK-1
(d) (i) ov
(h) (j)
Current Biology
ablations result in arches that are devoid of crest, as shown by the
absence of expression of the crest markers Dlx-2 and HNK-1, but the
four marker genes tested are expressed normally in these arches
(arrowheads in (e,f,h)). (i) Although HNK-1 expression cannot be seen
in the second arch (arrowhead), it is still present in the otic vesicle (ov)
and in the first arch crest and that around the third pouch. (j)
Schematic representation of the ablations of the neural tube that were
performed. The red dotted line indicates limits of ablated tissue. All
scale bars represent 100 µm.
The importance of this work is that it demonstrates that
normal arch morphogenesis must result from an integra-
tion between those patterning cues that are derived from
the neural crest and those that are independent of this
cell type. Indeed, support for this assertion can be found
in previous neural crest grafting studies, which demon-
strated that rostral crest, not normally fated to form
mandibular elements, could do so when transplanted to
the second arch [2]. Clearly, this work implies that neural
crest cells are not rigidly predetermined and can read
peripheral cues.
Further, our work demonstrates that the development of
pharyngeal segmentation is not dependent upon the
neural crest, and this reflects the evolutionary history of
the pharyngeal arches. Specifically, a rostrocaudal series of
pharyngeal perforations originated in evolution prior to
the neural-crest-derived mesenchyme that, in vertebrates,
fills the arches [16–18]. Recently, studies on the Pax genes
AmphiPax1/9 and AmphiPax2/5/8 of amphioxus, the nearest
extant relative of the vertebrates, have demonstrated that
pharyngeal pouches form in this animal and that Pax
genes are deployed in a similar fashion to their vertebrate
homologues [19,20]. Thus, the mechanism for generating
pharyngeal pouches does seem to have predated the evo-
lution of the neural crest, and as such, the evolution of the
vertebrate head is likely to have resulted from an integra-
tion between these two patterning systems.
1484 Current Biology Vol 9 No 24

Materials and methods 13. Hunt P, Ferretti P, Krumlauf R, Thorogood P: Restoration of normal
Hox code and branchial arch morphogenesis after extensive
Chick embryo manipulation deletion of hindbrain neural crest. Dev Biol 1995, 168:584-597.
Fertile Rhode Island Red hens’ eggs were obtained from Needle Farm, 14. Scherson T, Serbedzija G, Fraser S, Bronner-Fraser M: Regulative
Enfield, UK. These were incubated at 38°C in a humid atmosphere to capacity of the neural tube to form neural crest. Development,
the required stage. Neural-tube ablations were performed with a flame- 1993, 118:1049-1061.
sharpened tungsten needle. The entire neural tube was removed (see 15. Tucker GC, Aoyama H, Lipinski M, Tursz T, Thiery, JP: Identical
Figure 3; red lines show limits of ablation) from the middle of the mid- reactivity of monoclonal antibodies HNK-1 and NC-1: conservation
brain region to the levels of somites 1–4. This was performed on in vertebrates on cells derived from the neural primordium and on
some leukocytes. Cell Differ 1984, 14:223-230.
embryos at HH stages 9–10 inclusive [21]. Embryos were moistened
16. Northcutt RG, Gans C: The genesis of neural crest and epidermal
with Ringer solution and reincubated for 24 h, then fixed in 4% placodes: a reinterpretation of vertebrate origins. Q Rev Biol
paraformaldehyde in PBS. 1983, 58:1-28.
17. Maisey JG: Heads and tails: a chordate phylogeny. Cladistics 1986,
In situ hybridisation and immunohistochemistry 2:201-256.
In situ hybridisation was performed as described by Henrique et al. 18. Schaeffer B: Deuterostome monophyly and phylogeny. Evol Biol
1987, 21:179-234.
[22]. This protocol was modified for two-colour double in situ hybridis-
19. Holland LZ, Holland ND: Expression of AmphiHox-1 and
ation by developing the digoxygenin probe as described [22], then AmphiPax-1 in amphioxus embryos treated with retinoic acid:
fixing the embryos in 4% paraformaldehyde, followed by detection of insights into evolution and patterning of the chordate nerve cord
the FITC-labelled Dlx-2 probe with an anti-FITC antibody (Boehringer) and pharynx. Development 1996, 122:1829-1838.
diluted to 1:8000, then using Fast Red (Sigma) as the colour sub- 20. Kozmik Z, Holland ND, Kalousova A, Paces J, Schubert M, Holland
strate. The probes used have all been described previously and are as LZ: Characterisation of an amphioxus paired box gene,
follows: Ap-2α [11]; Bmp-7 [23]; Dlx2 [9]; Fgf-8 [11]; Pax1 [24]; Shh AmphiPax2/5/8: developmental expression patterns in optic
[25]. Following in situ hybridisation, embryos were embedded in support cells, nephridium, thyroid-like structures and pharyngeal
gelatin–albumen, fixed in glutaraldehyde and sectioned to 30 µm on a gill slits, but not in the midbrain-hindbrain boundary region.
Development 1999, 126:1295-1304.
vibratome. Following in situ hybridisation, those embryos that were
21. Hamburger V, Hamilton HL: A series of normal stages in the
analysed with HNK-1 were post-fixed in MEMFA (4% fomaldehyde in development of the chick embryo. J Morphol 1951, 88:49-92.
MEM), and processed for HNK-1 staining using the CD-57 antibody 22. Henrique D, Adam J, Myat A, Chitnis A, Lewis J, Ish-Horowicz D:
(Cambridge Biosciences; used at a 1:100 dilution) as described [26]. Expression of a Delta homologue in the chick. Nature 1995,
375:787-790.
Acknowledgements 23. Houston B, Thorp BH, Burt DW: Molecular cloning and expression
We would like to thank the following for the gift of clones used in this paper: of bone morphogenetic protein-7 in the chick epiphyseal growth-
Ap-2α, Joy Richman; Bmp-7, Brian Houston; Fgf-8, Ivor Mason; Pax1, Rudi plate. J Mol Endocrinol 1994, 13:289-301.
Balling; Shh, Tom Jessell. This work was funded by the Special Trustees of 24. Ebensperger C, Wilting J, Brand-Saberi B, Mizutani Y, Christ B,
Guy’s Hospital (E.V., M.M.S.), the Wellcome Trust (T.F.S., A.G.) and the Balling R, et al.: Pax1, a regulator of sclerotome development is
Medical Research Council (J.B., A.G.). induced by notochord and floor plate signals in avian embryos.
Anat Embryol (Berl) 1995, 191:297-310.
25. Roelink H, Augsburger A, Heemskerk J, Korzh V, Norlin S, Ruiz i
References Altaba A, et al.: Floor plate and motor neuron induction by vhh-1, a
1 Noden DM: Interactions and fates of avian craniofacial vertebrate homologue of hedgehog expressed by the notochord.
mesenchyme. Development 1988, 103 (suppl):121-140. Cell 1994, 76:761-775.
2. Noden DM: The role of the neural crest in patterning of avian 26. MacKenzie S, Walsh FS, Graham A: Migration of hypoglossal
cranial skeletal, connective, and muscle tissues. Dev Biol 1983, myoblast precursors. Dev Dyn 1998, 213:349-358.
96:144-165.
3. Noden DM: Patterning of avian craniofacial muscles. Dev Biol
1986, 116:347-356.
4. Hunt P, Guilisano M, Cook M, Sham M, Faiella A, Wilkinson D, et al.:
A distinct Hox code for the branchial region of the vertebrate
head. Nature 1991, 358:861-864.
5. Tosney KW: The segregation and early migration of cranial neural
crest cells in the avian embryo. Dev Biol 1982, 89:13-24.
6. Lumsden A, Sprawson N, Graham A: Segmental origin and
migration of neural crest cells in the hindbrain region of the chick
embryo. Development 1991, 113:1281-1291.
7. Koentges G, Lumsden A: Rhombencephalic neural crest
segmentation is preserved throughout craniofacial ontogeny.
Development 1996, 122:3229-3242.
8. Wall NA, Hogan BLM: Expression of bone morphogenetic protein-4
(BMP-4), bone morphogenetic protein-7 (BMP-7), fibroblast growth
factor-8 (FGF-8) and sonic hedgehog (SHH) during branchial arch
development in the chick. Mech Dev 1995, 53:383-392.
9. Begbie J, Brunet J-F, Rubenstein JLR, Graham A: Induction of the
epibranchial placodes. Development 1999, 126:895-902.
10. Shen H, Wilke T, Ashique AM, Narvey M, Zerucha T, Savino E, et al.:
Chicken transcription factor AP-2: cloning, expression and its role
in outgrowth of facial prominences and limb buds. Dev Biol 1997,
188:248-266.
11. Mahmood R, Bresnick J, Hornbruch A, Mahony C, Morton N,
Colquhoun K, et al.: A role for FGF-8 in the initiation and
maintenance of vertebrate limb bud outgrowth. Curr Biol 1995,
5:797-806.
12. Muller TS, Ebensperger C, Neubuser A, Koseki H, Balling R, Christ B,
et al.: Expression of avian Pax1 and Pax9 is intrinsically
regulated in the pharyngeal endoderm, but depends on
environmental influences in the paraxial mesoderm. Dev Biol
1996, 178:403-417.