untitled


Noradrenergic neuronal development is
impaired by mutation of the proneural
HASH-1 gene in congenital central
hypoventilation syndrome (Ondine’s curse)

Loı̈c de Pontual
1,{

, Virginie Népote
2,{

, Tania Attié-Bitach
1,{

, Hassan Al Halabiah
2
,

Ha Trang
2
, Vincent Elghouzzi

1
, Béatrice Levacher

2
, Karim Benihoud

3
, Joëlle Augé

1
,

Christophe Faure2, Béatrice Laudier1, Michel Vekemans1, Arnold Munnich1,

Michel Perricaudet
3
, François Guillemot

4
, Claude Gaultier

2
, Stanislas Lyonnet

1
,

Michel Simonneau
2

and Jeanne Amiel
1,*

1
Unité de Recherches sur les Handicaps Génétiques de l’Enfant INSERM U-393, and Département de Génétique,

Hôpital Necker-Enfants Malades, Paris, France,
2
Service de Physiologie, INSERM E9935, and CIC INSERM 9202,

Hôpital Robert Debré, Paris, France,
3
UMR 1582 CNRS, IGR, Villejuif, France and

4
NIMR, Mill Hill, UK

Received August 28, 2003; Revised and Accepted September 30, 2003

Congenital central hypoventilation syndrome (CCHS, Ondine’s curse) is a rare disorder of the chemical
control of breathing. It is frequently associated with a broad spectrum of dysautonomic symptoms,
suggesting the involvement of genes widely expressed in the autonomic nervous system. In particular, the
HASH-1–PHOX2A–PHOX2B developmental cascade was proposed as a candidate pathway because it
controls the development of neurons with a definitive or transient noradrenergic phenotype, upstream from
the RET receptor tyrosine kinase and tyrosine hydroxylase. We recently showed that PHOX2B is the major
CCHS locus, whose mutation accounts for 60% of cases. We also studied the proneural HASH-1 gene and
identified a heterozygous nucleotide substitution in three CCHS patients. To analyze the functional
consequences of HASH-1 mutations, we developed an in vitro model of noradrenergic differentiation in
neuronal progenitors derived from the mouse vagal neural crest, reproducing in vitro the HASH–PHOX–RET
pathway. All HASH-1 mutant alleles impaired noradrenergic neuronal development, when overexpressed
from adenoviral constructs. Thus, HASH-1 mutations may contribute to the CCHS phenotype in rare cases,
consistent with the view that the abnormal chemical control of breathing observed in CCHS patients is due to
the impairment of noradrenergic neurons during early steps of brainstem development.

INTRODUCTION

Congenital central hypoventilation syndrome (CCHS, Ondine’s
curse, MIM 209880) is a life-threatening disorder characterized
by persistent hypoventilation during sleep, beginning in the
neonatal period and requiring life-long ventilatory assistance
(1). While the pathophysiologic mechanisms underlying CCHS
remain elusive, several lines of evidence suggest that genetic
factors are involved (occurrence in siblings, concordant
monozygotic twins and rare vertical transmission) (2).

Changes in the integration of afferent inputs from central and
peripheral chemoreceptors in the brainstem are the most likely
disease mechanisms. In addition, most CCHS patients present a
broader defect of the autonomic nervous system (3), including
Hirschsprung disease (25–30% of cases, Haddad syndrome,
MIM 209880) (2), a malformation of the enteric nervous
system that has been ascribed to a mutation at the RET receptor
tyrosine kinase locus. In mice, the Ret signaling pathway,
which involves the sequential expression of the Mash1, Phox,
Ret and TH genes, is responsible for the development of all

*To whom correspondence should be addressed at: Département de Génétique, Hôpital Necker-Enfants Malades, 149, rue de Sèvres, 75743 Paris cedex
15, France. Tel: þ33 144495648; Fax: þ33 144495150; Email: amiel@necker.fr
{
The authors wish it to be known that, in their opinion, the first three authors should be regarded as joint First Authors.

Human Molecular Genetics, 2003, Vol. 12, No. 23 3173–3180
DOI: 10.1093/hmg/ddg339

Human Molecular Genetics, Vol. 12, No. 23 # Oxford University Press 2003; all rights reserved

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


transient or permanent noradrenergic derivatives (4,5). We
recently showed that PHOX2B is the major disease-causing
gene in both CCHS and Haddad syndromes, with an autosomal
dominant mode of inheritance and de novo mutations at the
first generation (2). Cross-regulation of the Phox2 and Mash-1
genes has been shown in mice (6,7); moreover, newborn Mash-
1
þ/�

heterozygous mice show an impaired ventilatory
responses to hypercarbia (8). We therefore regarded the human
ortholog of Mash1 (HASH-1), which encodes a tissue-specific
basic helix–loop–helix transcription factor, as an additional
candidate gene for CCHS. We identified a heterozygous
nucleotide substitution of the HASH-1 gene in three patients.
The functional consequences of HASH-1 mutations were
studied using a novel in vitro model of noradrenergic
differentiation based on neuronal progenitors derived from
the mouse vagal neural crest and reproducing in vitro the
HASH–PHOX–RET pathway. Forced expression of each of the
mutant HASH-1 alleles induced the impairment of noradrener-
gic neuronal development. We conclude that, at least in some
cases, heterozygous HASH-1 mutation may contribute to the
CCHS phenotype by modifying noradrenergic neurons during
early steps of brainstem development, and that the HASH-1
gene is tightly involved in the formation of the neuronal
network for autonomous control of ventilation in human.

RESULTS

HASH-1 heterozygous nucleotide substitutions

We screened the single coding exon of HASH-1 for
nucleotide variations in a series of 30 CCHS patients
(Fig. 1). We obtained an abnormal SSCP pattern in three
cases. Direct DNA sequencing showed a C to A transversion
at nucleotide 52, resulting in a proline to threonine substi-
tution at amino acid position 18 (P18T), in a female
patient with isolated CCHS. Two in-frame deletions of five
(111–125del15nt, A37–A41del) and eight (108–131del24nt,
A36–A43del) alanine codons in a 13-residue polyalanine
tract, were identified in female patients with CCHS and
Haddad syndrome respectively (Fig. 1A). Both the P18T and
the A37–A41del mutations were inherited from the healthy
father (Fig. 1A), while the mother of the patient with the
A36–A43del mutation was not available for study. The
HASH-1 mutations are located in regions highly conserved
across mammalian ASH genes as shown by Clustal W
analysis (Fig. 1B). These data suggest that these protein
domains are functionally relevant. Accordingly, none of the
HASH-1 gene mutations was detected in a panel of 120
control chromosomes. Interestingly, a de novo mutation of
PHOX2B (polyalanine expansion of seven alanines) was
found in the patient harboring the P18T mutation of the
HASH-1 gene, whereas direct sequencing of the three exons
of PHOX2B showed no variation in the other two cases.
Indeed, all CCHS patients described in this series have been
tested for the RET, GDNF and PHOX2B genes (2).

HASH-1 expression in early human development

We further investigated the possible involvement of HASH-1
mutations in the CCHS phenotype by studying the pattern of

expression of HASH-1 in early human development (Fig. 2).
Consistent with the wide spectrum of afferent nervous system
(ANS) dysfunction observed in CCHS patients and the pattern
of expression in mice (9,10), HASH-1 is expressed in both the
central nervous system (mesencephalon, rhombencephalon,
spinal cord; Fig. 2A, B and E–H), and the peripheral nervous
system (cells lining the dorsal aorta, presumed to be neural
crest-derived precursors of the sympathetic ganglia, Fig. 2C
and D) from Carnegie stage 14. At Carnegie stage 18, HASH-1
expression was detected in the presumptive enteric nervous
system, from the esophagus to the rectum (Fig. 2E–H) and in
the adrenal medulla (Fig. 2I and J).

Functional analysis of HASH-1 mutant alleles

We investigated the effects of the HASH-1 mutated alleles on
their transcriptional activity by means of a luciferase assay. P19
cells were co-transfected with HASH-1 alleles (þ/� E47) and
the luciferase gene under the control of the Delta1 promoter.
The transcriptional activities of the P18T and A36-A43del
alleles were similar to that of the control, regardless of whether
the E47 cofactor was present (data not shown, available on
request).

We then overexpressed the wild-type and mutant HASH-1
alleles, and assessed the functional consequences for the
differentiation of the noradrenergic lineage, using the PHOX–
RET signaling pathway and the tyrosine hydroxylase (TH)
phenotype as markers (11,12). We demonstrated that undiffer-
entiated vagal neural crest progenitor cells, isolated from E9
mouse embryos, may be engaged in noradrenergic differentia-
tion, and then sequentially produce Phox2a, Ret, and TH,
thereby reproducing the mouse Mash-1 molecular cascade
(Fig. 3A). Interestingly, no alternative lineage was obtained, as
almost all THþ cells were derived from PHOX2A-expressing
neurons. We then showed that it was possible to reproduce the
catecholaminergic differentiation pathway, 72 h after the forced
expression of HASH-1 by recombinant adenoviruses (Fig. 3B
and C). All the infected cells (100%) were transduced with the
HASH-1 adenoviral constructs (Fig. 4), as shown by GFP
production (Fig. 4B). The overexpression of wild-type HASH-1
resulted in the differentiation of 52.3�3% (n¼12) of neural
crest cells into postmitotic noradrenergic derivatives, as shown
by Phox2a production (data not shown). These values are
similar to those previously reported for retroviral forced
expression (11). We investigated the functional consequences
of HASH-1 gene mutations for the noradrenergic lineage
(Fig. 4B). All mutated forms of HASH-1 tested led to a
significant decrease (20–30%) in the number of neuronal
derivatives of the noradrenergic lineage, as shown by Phox2a
production (Fig. 4B). In particular, only 34.4�0.7%
(p < 0.001) of neural crest progenitors differentiated to give a
noradrenergic phenotype using the A37–A41del HASH-1
mutant construct, whereas 39.3�0.6% and 38.7�0.6% of
cells produced Phox2a when the P18T and A36–A43del HASH-
1 mutant alleles, respectively, were used (P < 0.01; Fig. 4B).

DISCUSSION

Taking advantage of the well-characterized specification of
catecholaminergic neurons derived from the vagal neural crest

3174 Human Molecular Genetics, 2003, Vol. 12, No. 23

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


(5), we recapitulated in vitro the molecular cascade involved in
the early steps of noradrenergic determination. In that system,
we used a forced expression of wild-type and mutant HASH-1
to test the functional relevance of mutated HASH-1 alleles
identified in CCHS patients. We showed a significant decrease
in the number of noradrenergic neurons generated after
expression of mutant HASH-1 alleles. However, this is a
limited effect considering that patients are heterozygotes for the
HASH-1 mutated allele as opposed to our experimental model.
Nevertheless, subtle consequences on the noradrenergic
neuronal development could be expected with HASH-1 gene
mutations lying outside of the bHLH domain of the protein.
Moreover, the HASH-1 mutations were located in highly
conserved regions of the protein, although not involved in the
basic helix–loop–helix consensus motives (13). Together with

the expression pattern of HASH-1 in the developing human
embryo, these results cope well with the phenotype of
heterozygous Mash-1

þ/�
mice (10). These data also suggest

that variant HASH-1 alleles may contribute to the CCHS
phenotype. However, HASH-1 gene mutations are neither
necessary (most patients do not have a HASH-1 gene variant),
nor sufficient for CCHS to occur (carriers have no phenotypic
expression). This observation is reminiscent of the findings in
isolated Hirschsprung disease in the Mennonite population,
where endothelin B receptor (EDNRB) mutations are neither
necessary nor sufficient for the disease to occur and act in
conjunction with a yet unknown hypomorphic RET allele (14).
The question as to whether HASH-1 acts as a modifier of
PHOX2B, and/or a disease-causing autosomal dominant gene
with incomplete penetrance remains unresolved.

Figure 1. HASH-1 gene mutations and phylogenetic analysis of HASH-1 proteins. (A) HASH-1 gene mutations. SSCP results and mutated sequences are illustrated
for the three individuals whose pedigrees are shown directly above. (B) Clustal W alignment of the human HASH-1 protein and its mouse, Drosophila and
C. elegans orthologs. The residues mutated in CCHS patients are indicated by arrows. Note the conservation of the basic, helix 1, loop and helix2 amino acid
domains between the four species studied.

Human Molecular Genetics, 2003, Vol. 12, No. 23 3175

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


Polyalanine repeats are common in transcription factors
but their normal function is largely unknown. Expansions
of polyalanine tracts in transcription factors, including de novo
PHOX2B polyalanine expansions in CCHS, have been shown
to be responsible for disease in several malformation
syndromes in humans (15,16), whereas contractions have not

been reported thus far. Polyalanine contractions may therefore
act as hypomorphic alleles.

As the CCHS phenotypes of patients with PHOX2B and
HASH-1 mutations are undistinguishable, our data suggest that
any mutation affecting the HASH-1–PHOX developmental
pathway is likely to impair the development of neurons of the

Figure 2. HASH-1 gene expression in developing human embryos. Slides stained with hematoxylin-eosin (A, C, E, G, I) and dark-field illumination of the hybri-
dized adjacent sections (B, D, F, H, J) on day 32 (Carnegie 14, A–D), and day 44 (Carnegie 18, E–J). Parasagittal sections showing weak HASH-1 expression at
Carnegie stage 14 and strong expression of this gene at Carnegie stage 18 in mesencephalon (mes), rhombencephalon (rh), and the dorsal area of the spinal cord
(sc) (arrowheads, A, B and E–H, respectively). From Carnegie stage 14, expression was detected in cells to the side of the dorsal aorta (arrow, D). HASH-1 expres-
sion was also detected in the intestine, from the foregut (arrow, F) to the hindgut (arrow, H), and in the adrenal medulla (arrow, J) at Carnegie stage 18. ad, Adrenal
gland; ao, aorta; go, gonad; meta, metanephros; mes, mesencephalon; meso, mesonephros; e, esophagus; re, rectum; rh, rhombencephalon; sc, spinal cord.

3176 Human Molecular Genetics, 2003, Vol. 12, No. 23

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


Figure 3. In vitro culture model of mouse vagal neural crest cells to analyze the HASH-1–PHOX2B/PHOX2A cascade controlling. (A) Experimental design;
NCSCs were isolated from E9.5 mouse embryo neural tubes (NT). Ec, ectoderm; Som, somite; End, endoderm; NC, neural crest. (B) Transcription-factor cascade
controlling the RET and tyrosine hydroxylase (TH) lineage of neural crest cells (NCSCs). Cells cultured in the presence of diffusible embryonic mesodermic factors
retain the capacity to generate catecholaminergic derivatives, as shown by neurofilament (NF160) and TH production, recapitulating the MASH1–PHOX2B–
PHOX2A–RET pathway as in vivo (lower arrow). Conversely, NCSCs cultured in medium containing serum (FCS medium) remain multipotent and retain their
proliferative abilities (upper arrow). (C) Sequential expression of the Mash1–Phox2a–Ret–TH gene pathway, studied in cultured NCSCs by immunohistochemistry
(first row), peripherin staining (Per, second row), merged immunohistochemistry and peripherin staining (merged, third row) and in situ hybridization (phase, fourth
row), from day 1 to day 5 (d1–d5). The Mash1 (d1), Phox2a (d2), Ret (d3) and TH (d5) are shown in red fluorescence; the peripherin (d1–d3) and neurofilament
(NF160, NF, d5) are shown in green fluorescence. (D) Temporal expression of the Mash1–Phox2a–Ret–TH gene pathway during NCSCs culture, from day 0 (d0) to
day 6 (d6). The expression timing of Mash1, Phox2a, Ret and TH are indicated by solid bars. Peripherin and NF (NF160) are consistently expressed from d1 to d6.

Human Molecular Genetics, 2003, Vol. 12, No. 23 3177

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


noradrenergic lineage and the resulting defect may modify the
respiratory network in the brainstem and periphery.

MATERIALS AND METHODS

Patients

A series of 30 patients fulfilled the inclusion criteria for
diagnosis of CCHS namely: (i) hypoventilation, hypoxemia and
hypercarbia during quiet sleep on polygraphic respiratory
recording; with (ii) no cardiac, pulmonary, neuromuscular,

EEG or cerebral MRI abnormality (17). Haddad syndrome was
diagnosed in 11 cases (36%, five long segment HSCR, two
short segment HSCR, four unknown). The histological criteria
for HSCR were: (i) absence of enteric plexuses with
histological evaluation of the aganglionic tract; and (ii) strong
histochemical staining of acetylcholinesterase in nerve fibers.
Blood samples were obtained with informed consent and DNA
was extracted according to standard protocols.

We screened the coding sequence of the HASH-1 gene by
single-strand conformation polymorphism (SSCP) analysis
and/or direct DNA sequencing. The PCR products were heated
for 10 min at 95�C, loaded onto a Hydrolink MDE gel

Figure 4. Functional analysis, indicating a decrease in the number of noradrenergic neuronal derivatives obtained from NCSCs after forced expression of the three
mutated forms of HASH-1, with respect to the wild-type. (A) HASH-1 adenovirus constructs for the production of both green fluorescent protein (GFP) and wild-
type or mutated forms of HASH1 under cytomegalovirus (CMV) promoter control. (B) Left panel: representative samples of infected NCSCs 24 h after infection
with adenoviruses producing the wild-type (WT) and mutant (P18T, A36–A43del, A37–A41del) forms of HASH-1. GFP was detected by its green fluorescence
and Phox2a was detected by staining with Fast Red. A typical Phox2a-positive cell is indicated by a white arrow; a typical Phox2a-negative cell for is indicated by a
white arrowhead. Right panel: percentage of Phox2a positive cells, illustrating the decrease in the number of noradrenergic derivatives obtained from NCSCs after
infection with the mutant HASH-1 retroviral constructs (P18T, A36–A43del and A37–A41del).

3178 Human Molecular Genetics, 2003, Vol. 12, No. 23

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


(Bioprobe), and subjected to electrophoresis at 4 W. The gel
was then dried and placed against a X-ray film for 48 h. If
an abnormal SSCP pattern was observed, direct DNA
sequencing was performed by the fluorometric method, for
both strands (Big DyeTerminator Cycle Sequencing kit,
Applied Biosystems, Warrington, UK).

In situ hybridization in human embryos

Human embryos and fetal tissues were collected from legally
terminated pregnancies in agreement with French law and
ethics committee recommendations. Tissues were prepared as
previously described (18). A 222 bp PCR fragment correspon-
ding to HASH-1 exon 1 was amplified from human genomic
DNA with a T7 extension (TAATACGACTCACTAGGGAGA)
added to both primers to generate the sense and antisense
probes. Probes were labeled with [

35
S]UTP and purified as pre-

viously described (2). Tissues on slides were dehydrated, placed
against BIOMAX MR X-ray film (Amersham) for 3 days, and
immersed in Kodak NTB2 emulsion for 3 weeks at þ4�C.
Developed and toluidine blue-counterstained slides were
analyzed with dark- and bright-field illumination. No hybridiza-
tion signal was detected with the sense probe (data not shown).

Construction of adenoviral plasmids

Adenoviral plasmids were constructed using the Adeasy
system (19). Polyalanine tract contractions and the control
HASH-1 coding sequence were amplified by polymerase
chain reaction with the 50 primer, TCTAGACGCATGGAAA-
GCTCTGCCAA, and the 30 primer, GTCGACTCAGAACC-
AGTTGGTGAAGTCGA, using genomic DNA from controls
or patients. PCR-derived fragments were checked by sequen-
cing and subcloned into the pAd-track CMV vector, using
the XbaI–SalI restriction sites. PAd-track CMV vectors were
used to produce viruses that could be tracked by following
GFP fluorescence. Adenoviral plasmids were generated
by homologous recombination in electrocompetent E. coli
BJ5183, according to the manufacturer’s instructions
(Stratagene). Adenoviral plasmids were then cut with PacI
and viral stocks were produced in QBI-293A cells.

Mouse neural crest culture

Neural crest cells were cultured as previously described (20).
Neural tubes were microdissected from Swiss mouse
(Iffa Credo) embryos on day 9 of gestation. Somites and
surrounding tissues were removed to obtain neural tube
segments corresponding to the first seven somites of the vagal
region. These segments were then plated on fibronectin-coated
glass coverslips (10 mg/ml, Boehringer Roche) and cultured for
2 days in DMEM (Life Technologies) supplemented with 10%
fetal calf serum (AbCys) and 100 IU/ml penicillin/streptomycin
(Life Technologies), at 37�C, in a humidified atmosphere
containing 5% CO2.

Infection of primary cultures of neural crest cells

We tested various levels of multiplicity of infection (MOI)
and decided to use 2000 plaque-forming unit (pfu), which
gave 100% infected neural crest cells. Neural crest cells

(NCCs) were cultured for 2 days, infected on day 2 and fixed
24 h later.

ACKNOWLEDGEMENTS

We thank Bert Vogelstein for adenoviral constructs, and
Frances Goodman and Jean-François Brunet for helpful
discussions. H.A. was recipient of an IFRO fellowship. V.N.
was partly supported by a Fondation pour la Recherche
Médicale fellowship. This work was supported by grants from
EURExpress, HMR (Hoechst-Marion-Roussel), the European
Community (2001-01646), Fondation pour la Recherche
Médicale, and Association Française contre les Myopathies-
INSERM (Maladies Rares).

REFERENCES

1. Gozal, D. (1998) Congenital central hypoventilation syndrome: an update.
Pediatr. Pulmonol., 26, 273–282.

2. Amiel, J., Laudier, B., Attié-Bitach, T., Trang, H., de Pontual, L., Gener, B.,
Trochet, D., Etchevers, H., Ray, P., Simonneau, M. et al. (2003) Polyalanine
expansion and frameshift mutations of the paired-like homeobox gene
PHOX2B in congenital central hypoventilation syndrome. Nat. Genet.,
33, 459–461.

3. Croaker, G.D., Shi, E., Simpson, E., Cartmill, T. and Cass, D.T. (1998)
Congenital central hypoventilation syndrome and Hirschsprung’s disease.
Arch. Dis. Child., 78, 316–322.

4. Brunet, J.F. and Pattyn, A. (2002) Phox2 genes—from patterning to
connectivity. Curr. Opin. Genet. Dev., 12, 435–440.

5. Goridis, C. and Rohrer, H. (2002) Specification of catecholaminergic
and serotonergic neurons. Nat. Rev. Neurosci., 3, 531–541.

6. Pattyn, A., Morin, X., Cremer, H., Goridis, C. and Brunet, J.F. (1999) The
homeobox gene Phox2b is essential for the development of autonomic
neural crest derivatives. Nature, 399, 366–370.

7. Pattyn, A., Goridis, C. and Brunet, J.F. (2000) Specification of the central
noradrenergic phenotype by the homeobox gene Phox2b. Mol. Cell.
Neurosci., 15, 235–243.

8. Dauger, S., Renolleau, S., Vardon, G., Népote, V., Mas, C., Simonneau, M.,
Gaultier, C. and Gallego, J. (1999) Ventilatory responses to hypercapnia
and hypoxia in Mash-1 heterozygous newborn and adult mice.
Pediatr. Res., 46, 535–542.

9. Guillemot, F. and Joyner, A.L. (1993) Dynamic expression of the murine
Achaete–Scute homologue Mash-1 in the developing nervous system.
Mech. Dev., 42, 171–185.

10. Guillemot, F., Lo, L.C., Johnson, J.E., Auerbach, A., Anderson, D.J. and
Joyner, A.L. (1993) Mammalian achaete-scute homolog 1 is required
for the early development of olfactory and autonomic neurons. Cell,
75, 463–476.

11. Lo, L., Tiveron, M.C. and Anderson, D.J. (1998) MASH1 activates
expression of the paired homeodomain transcription factor Phox2a, and
couples pan-neuronal and subtype-specific components of autonomic
neuronal identity. Development, 125, 609–620.

12. Lo, L., Morin, X., Brunet, J.F. and Anderson, D.J. (1999) Specification of
neurotransmitter identity by Phox2 proteins in neural crest stem cells.
Neuron, 22, 693–705.

13. Bertrand, N., Castro, D.S. and Guillemot, F. (2002) Proneural genes
and the specification of neural cell types. Nat. Rev. Neurosci., 3, 517–530.

14. Puffenberger, E.G., Hosoda, K., Washington, S., Nakao, K., deWit, D.,
Yanagisawa, M. and Chakravart, A. (1994) A missense mutation of the
endothelin-B receptor gene in multigenic Hirschsprung’s disease. Cell,
79, 1257–1266.

15. Goodman, F.R. and Scambler, P.J. (2001) Human HOX gene mutations.
Clin. Genet., 59, 1–11.

16. Stromme, P., Mangelsdorf, M.E., Shaw, M.A., Lower, K.M.,
Lewis, S.M., Bruyere, H., Lutcherath, V., Gedeon, A.K., Wallace, R.H.,
Scheffer, I.E. et al. (2002) Mutations in the human ortholog of
Aristaless cause X-linked mental retardation and epilepsy. Nat. Genet.,
30, 441–445.

Human Molecular Genetics, 2003, Vol. 12, No. 23 3179

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org


17. Weese-Mayer, D.E., Shannon, D.C., Keens, T.G. and Silvestri, J.M.
(1999) Idiopathic congenital central hypoventilation syndrome: diagnosis
and management. Am. J. Respir. Crit. Care Med., 160, 368–73.

18. Odent, S., Attié-Bitach, T., Blayau, M., Mathieu, M., Augé,J.,
Delezoide, A.L., Gall, J.Y., Le Marec, B., Munnich, A., David, V. et al.
(1999) Expression of the Sonic Hedgehog (SHH) gene during early human
development and phenotypic expression of new mutations causing
holoprosencephaly. Hum. Mol. Genet., 8, 1683–1689.

19. He, T.C., Zhou, S., da Costa, L.T., Yu, J., Kinzler, K.W. and
Vogelstein, B. (1998) A simplified system for generating
recombinant adenoviruses. Proc. Natl Acad. Sci. USA, 95,
2509–2514.

20. Boisseau, S. and Simonneau, M. (1989). Mammalian neuronal
differentiation: early expression of a neuronal phenotype from mouse
neural crest cells in a chemically defined cultured medium. Development,
106, 665–674.

3180 Human Molecular Genetics, 2003, Vol. 12, No. 23

 a
t U

n
ive

rsite
 d

e
 M

o
n
tre

a
l - B

ib
lio

th
e
q
u
e
s - A

cq
u
isitio

n
s (P

e
rio

d
iq

u
e
s) o

n
 A

u
g

u
st 2

2
, 2

0
1

0
 

h
ttp

://h
m

g
.o

xfo
rd

jo
u

rn
a
ls.o

rg
D

o
w

n
lo

a
d

e
d

 fro
m

 

http://hmg.oxfordjournals.org