MOLECULAR PROFILING INDICATES AVIAN BRANCHIOMOTOR

MN

MAa

Db

cc

J

AhpTlcatmci7cbtTerlinaAs

micrP

Km

*JTEAscHmsTn

Neuroscience 128 (2004) 785–796

0d

OLECULAR PROFILING INDICATES AVIAN BRANCHIOMOTORUCLEI INVADE THE HINDBRAIN ALAR PLATE

Inhbhre[usitpaCS(2lhmbfid12

a1hfiptrglibTpipCttmp11

. J. JU,a P. AROCA,b J. LUO,c L. PUELLESb

ND C. REDIESa,c*

Institute of Anatomy, University of Duisburg-Essen Medical School,-45122 Essen, Germany

Department of Human Anatomy, University of Murcia, E-30100 Mur-ia, Spain

Institute of Anatomy I, University of Jena School of Medicine, D-07743ena, Germany

bstract—It is generally believed that the spinal cord andindbrain consist of a motor basal plate and a sensory alarlate. We now have molecular markers for these territories.he relationship of migrating branchiomotor neurons to mo-

ecularly defined alar and basal domains was examined in thehicken embryo by mapping the expression of cadherin-7nd cadherin-6B, in comparison to genetic markers for ven-rodorsal patterning (Otp, Pax6, Pax7, Nkx2.2, and Shh) andotoneuron subpopulations (Phox2b and Isl1). We show

adherin-7 is expressed in a complete radial domain occupy-ng a lateral region of the hindbrain basal plate. The cadherin-

domain abuts the medial border of Pax7 expression; thisommon limit defines, or at least approximates, the basal/alaroundary. The hindbrain branchiomotor neurons originate inhe medial part of the basal plate, close to the floor plate.heir cadherin-7-positive axons grow into the alar plate andxit the hindbrain close to the corresponding afferent nerveoot. The cadherin-7-positive neuronal cell bodies later trans-ocate laterally, following this axonal trajectory, thereby pass-ng through the cadherin-7-positive basal plate domain. Fi-ally, the cell bodies traverse the molecularly defined basal/lar boundary and move into positions within the alar plate.fter the migration has ended, the branchiomotor neuronswitch expression from cadherin-7 to cadherin-6B.

These findings demonstrate that a specific subset of pri-ary motor neurons, the branchiomotor neurons, migrate

nto the alar plate of the chicken embryo. Consequently, theentury-old concept that all primary motor neurons come toeside in the basal plate should be revised. © 2004 IBRO.ublished by Elsevier Ltd. All rights reserved.

ey words: spinal cord, branchiomotor neuron, cadherin,igration, cell adhesion, chicken.

Correspondence to: C. Redies, Institute of Anatomy I, University ofena School of Medicine, Teichgraben 7, D-07743 Jena, Germany.el: �49-3641-93-8511; fax: �49-3641-93-8512.-mail address: [email protected] (C. Redies).bbreviations: BCIP, 5-bromo-4-chloro-3-indolyl phosphate-toluidinealt; bm, branchiomotor; BMP, bone morphogenetic protein; cad6B,adherin-6B; cad7, cadherin-7; HBSS, HEPES-buffered salt solution;H, stage according to Hamburger and Hamilton (1951); Isl1, islet-1;bm, migrating branchiomotor neurons; NBT, nitroblue tetrazolium

alt; p0-p3, ventral progenitor cell domain; Shh, Sonic hedgehog; TBS,ris-buffered saline; VMN, ventral motor nucleus of the trigeminal

Terve.

306-4522/04$30.00�0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reseroi:10.1016/j.neuroscience.2004.06.063

785

n the developing spinal cord and hindbrain, the lateral wall ofeural tube can be divided into alar and basal plates. (In theindbrain, the paired lateral walls are flapped open like aook so that the alar plates come to lie laterally [like wings,ence “alar”] and the basal plate appears medially. For thiseason, we will use the conventional descriptive terms “lat-ral” and “medial” for dorsoventral topology in the hindbrain

dorsal�lateral; ventral�medial]). With the advent of molec-lar markers for dorsoventral patterning in the hindbrain andpinal cord, some genetic mechanisms inducing this pattern-

ng have been elucidated. The spinal cord is patterned alonghe dorsoventral axis by the secretion of bone morphogeneticroteins (BMPs) and dorsalin from the roof plate (Basler etl., 1993; Lee et al., 2000; Liem et al., 2000; Litingtung andhiang, 2000; Timmer et al., 2002) and by the secretion ofonic hedgehog (Shh) from the notochord and the floor plate

Marti et al., 1995; Ericson et al., 1996, 1997a; Briscoe et al.,000; Poh et al., 2002). Pax7 signal, which is present in a

arge dorsal domain extending from spinal cord to the rostralindbrain and beyond, is generally regarded as an alar platearker (Jostes et al., 1990; Kawakami et al., 1997). In theasal plate of the spinal cord, molecular markers distinguishve classes of motoneurons and interneurons, arranged inorsoventral subzones (Simeone et al., 1994; Pattyn et al.,997; Briscoe et al., 2000; Garel et al., 2000; Marthiens et al.,002).

It is generally assumed that all primary motor neuronsre located in the basal plate (His, 1888; Kuhlenbeck,975; for review, see Nieuwenhuys et al., 1998). In theindbrain, research on cranial nerve components has de-ned somatomotor and visceromotor columns in the basallate, and viscerosensory and somatosensory columns inhe alar plate (Gaskell, 1886, 1889; Johnston, 1905; Her-ick, 1922). This columnar scheme did not consider tan-ential migrations of motoneurons, which were discovered

ater. A specialized group of visceromotor neurons, whichnnervate directly striated muscles associated withranchial arches, are the “branchiomotor” (bm) neurons.hese motoneurons are born on either side of the floorlate, from where they migrate by translocating their bod-

es along their leading axons that exit laterally from the alarlate (Ramon y Cajal, 1911; Heaton and Moody, 1980;ovell and Noden, 1989; Fritzsch, 1998). A similar migra-

ion is performed by most preganglionic parasympa-hetic neurons. Upon reaching a more lateral territory,ost bm neurons migrate radially pialward to their finalosition, now with trailing axons (Heaton and Moody,980; Arens and Straznicky, 1986; Covell and Noden,989; Lumsden and Keynes, 1989; Guthrie et al., 1991).
he trigeminal, facial, retrofacial and ambiguus bm nu-
ved.

FecIsbm

FecfsFScale bar�100 �m (in G) for A–G.

1ABbBBdEFflggggHIIlm

M. J. Ju et al. / Neuroscience 128 (2004) 785–796786

ig. 1. Expression of cad7 (A) in the spinal cord of a stage HH19mbryo, compared with the expression of other markers (Pax7, B;adherin-6B, cad6B, C; Pax6, D; Nkx2.2, E; Lim3/Lhx4, F; and Isl1, G).mmunostains are shown for a series of consecutive sections. (H) Nissltain (Thio) of an adjacent section. The arrows in A, B and I point to theasal/alar boundary. (I) Schematic diagram of the expression do-
ains. Scale bar�100 �m (in G) for A–H.
medial band of Phox2b expression

ig. 2. Expression of cad7 (A) in the spinal cord of a stage HH23mbryo, compared with the expression of other markers (Pax7, B;ad6B, C; Pax6, D; Shh, E and Nkx2.2, F). Immunostains are shownor a series of consecutive sections. G, Nissl stain (Thio) of an adjacentection. (H) Schematic diagram of the expression domains (see alsoig. 3A). The arrows in A, B and H point to the basal/alar boundary.

Abbreviations used in the figures

-6 rhombomeres 1-6alar platebasal plate

/a basal/alar plate boundaryl lateral part of basal platem lmedial part of basal plate

dorsal band of Phox2b expressionembryonic day (say of incubation)floor plate

m medial longitudinal fascicleV trigeminal ganglionVII facial ganglionIX glossopharyngeal ganglionX vagal ganglionb hindbrain

II oculomotor complexV trochlear complex

lateral band of Phox2b expression

mz mantle zonenc notochordpMN motor neuron progenitor domainSc Schwann cellsTect tectumThio thioninev4 fourth ventricleVm laterally/dorsally migrated branchiomotor nucleus of the tri-

geminal nerveVIn abducens nerveVIIn facial nerveVIIm laterally/dorsally migrated branchiomotor nucleus of the facial

nervevz ventricular zoneXII hypoglossal nucleusXIIn hypoglossal nerveXm laterally/dorsally migrated branchiomotor nucleus of the va-

gus nerve

cbfow

n1woa

FcpmE(H

M. J. Ju et al. / Neuroscience 128 (2004) 785–796 787

lei thus produced were assumed to be in the lateralasal plate (Kuhlenbeck, 1975; Nieuwenhuys, 1998). A

ew laterally migrated bm neurons and most pregangli-nic parasympathetic elements remain periventricularly,here they form the “dorsal” bm or the preganglionic

ig. 3. (A, B) Schematic representation of the expression domains of coding, see right side of the figure) in the spinal cord (A) and in the hath of bm neurons. Note that the bm neurons (dark blue in B) come toedially the cad7 expression domain. (C–E) Bm neurons of the ventra) at stage 26. An overlay of the double-label immunostaining for this fr

in C) for C–E. (F–H) Expression of Phox2b and Otp in flat-mount spec
H28 (H). Specimens were hybridized with riboprobes for Phox2b (reddish bro
uclei (Heaton and Moody, 1980; Moody and Heaton,983a; Covell and Noden, 1989). In chicken embryos,e investigated the relationship between the migrationf bm neurons and the position of the basal/alar bound-ry, as determined by molecular markers. Expression of

ther genes (cad6B, Pax7, Pax6, Lim3/Lhx4, Nkx2.2 and Shh; for colorB) at stage HH23. The blue dotted line in B represents the migratoryalar plate. The exit point of the somatomotor neurons (red in B) abutsal motor nucleus coexpress cad6B (red in C, D) and Isl1 (green in D,

tion is shown in E (nuclear stain [Hoechst] in blue). Scale bar�100 �mhindbrain from embryonic stage HH20 (F), stage HH24 (G), and stage

ad7 and oindbrain (lie in thel trigeminontal secimens of

wn) and Otp (purple). Scale bars�600 �m for F–H.

cb

A

Fw

hHnnElenwcHerltttE5sisH

A

TpaakDPmec6rMm

Farb

FcPoStG


adherin-7 (cad7), Pax7 and Otp was used to define theasal/alar boundary.

EXPERIMENTAL PROCEDURES

nimals

ertilized eggs from White Leghorn chicken (Gallus domesticus)ere incubated in a forced-draft incubator at 37 °C and 65%

ig. 4. Whole-mount immunostaining for cad7 of the hindbrain at stagt stage HH20 (C). A, B show dorsal views and C shows a lateral viewhombomeres and between midbrain and hindbrain. The arrows pars�200 �m (in A–C).

ig. 5. Expression of cad7 (A) in the hindbrain of a stage 23 embryo,ompared with the expression of other markers (Pax7, B; cad6B, C;ax6, D; Nkx2.2, E; and Shh, F). Immunostains are shown for a seriesf consecutive sections. G, Nissl (Thio) of an adjacent section. (H)chematic diagram of the expression domains. The arrows in A point

o migrating bm neurons of the trigeminal nerve. Scale bar�100 �m (in
N) for A–G.
umidity until the desired stage (HH; according to Hamburger andamilton, 1951). The experiments conformed to institutional andational guidelines on the ethical use of animals in research. Theumber of embryos used and their suffering were minimized.mbryos were transferred into ice-cold HEPES-buffered salt so-

ution supplemented with 1 mM CaCl2 (HBSS; pH 7.6). Late stagembryos were killed by decapitation according to institutional andational guidelines for the use of animals in research. Specimensere fixed in 4% formaldehyde dissolved in HBSS for 2 h. Forryoprotection, heads were immersed in an ascending series ofBSS-buffered sucrose solutions (12% [w/v], 15%, 18%) andmbedded in Tissue Tec O.C.T. compound (Sakura Finetek Eu-ope, Zoeterwoude, The Netherlands). Specimens were frozen iniquid nitrogen and stored at �80 °C. Sections of 18–20 �mhickness were cut on a refrigerated microtome and directlyhawed onto gelatin-coated slide glasses for immunohistochemis-ry and onto silane-coated slide glasses for in situ hybridization.mbryos at the following stages were used: stage HH15 (ca.0–55 h of incubation); stage HH19 (ca. 68–72 h of incubation);tage HH20 (ca. 70–72 h of incubation); stage HH21 (3.5 days ofncubation; E3.5); stage HH22 (E3.5); stage HH23 (E3.5-E4);tage HH24 (E4); stage HH26 (E4.5-E5); stage HH27 (E5), stageH28 (E5.5) and stage HH30 (E6.5).

ntibodies

he following antibodies were used for immunostaining: rabbitolyclonal antiserum raised against chicken cadherin-6B (cad6B)nd mouse monoclonal antibodies CC6B-1 and CC7-1 raisedgainst chicken cad6B and cad7 (Nakagawa and Takeichi, 1998;ind gift of Drs. S. Nakagawa and M. Takeichi, RIKEN Institute forevelopmental Biology, Kobe); mouse monoclonal antibodyAX6 raised against chicken Pax6 (Ericson et al., 1997b); mouseonoclonal antibody PAX7 raised against chicken Pax7 (Ericsont al., 1996); mouse monoclonal antibody 5E1 raised againsthicken Shh (Ericson et al., 1996); mouse monoclonal antibody7.4E12 raised against Lim3 (Ericson et al., 1997b) which alsoecognizes Lhx4 (Dr. Johan Ericson, Department of Cell andolecular Biology, Karolinska Institute, personal communication);ouse monoclonal antibody 74.5A5 raised against chicken

(A) and stage HH21 (B) and of the hindbrain/midbrain transition areaowheads point to cad7-negative boundary regions between individualilateral stripes of cad7 expression in the lateral floor plate. Scale

e HH15. The arroint to b

kx2.2 (Ericson et al., 1997b); and mouse monoclonal antibody

3ittet

I

PvdibEtbsabdrstAdG

prab(brtb3ivZ(

uf

psdbttrea

I

TcCdtkSicU

ppa(wm(hcnd

ad1osa

PSmwwrabbtal4a

P

ActF

Ftelwal


9.4D5 raised against Islet-1 (Isl1; Ericson et al., 1992). Antibod-es PAX6, PAX7, 5E1, 67.4E12, 74.5A5 and 39.4D5 were ob-ained from the Developmental Studies Hybridoma Bank main-ained by the University of Iowa, Department of Biological Sci-nces, Iowa City, IA, USA, under contract NO1-HD-7-3263 fromhe NICHD.

mmunohistochemistry

rocedures for immunohistochemistry have been published pre-iously (Redies et al., 1993; Gänzler and Redies, 1995). Briefly,ried sections were rehydrated in HBSS. Sections were postfixed

n 4% formaldehyde in HBSS for 30 min and washed in Tris-uffered saline supplemented with 1 mM CaCl2 (pH 7.5; TBS).ndogenous peroxidase activity was suppressed by immersing

he sections in 0.3% H2O2 in methanol for 30 min at �20 °C. Tolock unspecific binding and to dilute antibodies, 1.5% horseerum in TBS was used. Primary antibodies were applied for 1.5 ht room temperature or overnight at 4 °C. After sequential incu-ation with appropriate biotinylated secondary antibodies and avi-in-coupled peroxidase complex (ABC Elite kit; Vector Laborato-ies, Burlingame, CA, USA), sections were treated with a sub-trate solution containing 0.7% 3–3= diaminobenzidineetrachloride, 0.5% nickel chloride, and 0.1% peroxide in TBS.fter enough reaction product had formed, sections were washed,ehydrated, and embedded in Histomount (Shandon, Frankfurt,ermany).

Double-labeling immunostaining procedures were describedreviously (Redies et al., 1992, 1993). Slides were stained usingabbit polyclonal antiserum against cad6B and mouse monoclonalntibody against Isl1 or cad7. Appropriate Cy3-conjugated anti-odies and FITC- or Alexa 488-conjugated secondary antibodiesDianova, Hamburg, Germany; or Molecular Probes, Poortge-ouw, The Netherlands) were used to label the antigens sepa-ately. Primary antibodies were applied for 1 h at room tempera-ure. After sequential incubation with appropriate secondary anti-odies, sections were counterstained with nuclear dye Hoechst3258 (Sigma, Deisenhofen, Germany). Sections were mounted

n Vectashield medium (Vector Laboratories). Fluorescence wasisualized with a confocal laser-scanning microscope (LSM 510;eiss, Oberkochen, Germany) or an epifluorescence microscope

ig. 6. Immunostains for cad7 (A) and for Pax7 (B), in situ hybridiza-ion for Otp (C) and Nissl stain (Thio, D) in the hindbrain of a stage 24mbryo are shown for a series of consecutive sections. Note that the

ateral border of the Otp expression domain coincides approximatelyith the lateral border of the cad7 expression domain in mantle layernd with the medial boundary of Pax7 expression in the ventricular

ayer. Scale bar�200 �m (in C) for A–D.

BX40; Olympus, Hamburg, Germany). t

For neuroanatomic orientation, sections adjacent to thosesed for immunohistochemistry were stained with thionine acetateor Nissl substance, as described previously (Redies et al., 1993).

Wholemount immunostaining was performed as describedreviously (Gänzler and Redies, 1995). Briefly, embryonic braintems from stages 9–27 (Hamburger and Hamilton, 1951) wereissected and fixed in 4% formaldehyde in HBSS on ice. Afterlocking endogenous peroxidase, specimens were incubated withhe monoclonal antibody against chicken cad7. Following incuba-ion with appropriate biotinylated secondary antibody and ABCeagent, specimens were reacted with substrate solution untilnough reaction product had formed. Specimens were viewednd photographed under a stereomicroscope (Stemi SV; Zeiss).

n situ hybridization

o detect Otp mRNA, a plasmid containing a 180 bp insert ofhicken Otp cDNA was used (kind gift of Dr. A. Simeone, MRCentre for Developmental Neurobiology, King’s College Lon-on, London, UK). To detect Phox2b mRNA, a plasmid con-aining chicken Phox2b cDNA was used (Pattyn et al., 1997;ind gift of Dr. J. F. Brunet, CNRS UMR 8542, École Normaleupérieure, Paris, France). Digoxigenin-labeled and fluoresce-

n-labeled, sense and anti-sense RNAs were prepared by usingommercially available kits (Stratagene, La Jolla, CA,SA).

The in situ hybridization procedure for sections was describedreviously (Redies et al., 1993). Briefly, the digoxigenin-labeledrobe for Otp, which was hybridized in situ, was detected withnti-digoxigenin alkaline-phosphatase-conjugated Fab fragmentsBoehringer, Mannheim, Germany). The sections were reactedith a solution containing nitroblue tetrazolium salt (NBT; 75 mg/l) and 5-bromo-4-chloro-3-indolyl phosphate-toluidine salt

BCIP; 50 mg/ml) in alkaline buffer until enough colored precipitatead formed. Sections were embedded in Entellan (Merck). Adja-ent sections to those used for in situ hybridization were immu-ostained for cad7 and Pax7, and stained for Nissl substance, asescribed above.

The combined immunostaining of sections with antibodygainst cad7 and in situ hybridization with Phox2b probe wasescribed previously (Brahic and Ozden, 1992; Wöhrn et al.,999). Briefly, peroxidase immunohistochemistry was carriedut as described above, but with solutions containing 10 mg/mlodium heparin, followed by in situ hybridization (seebove).

Flat-mount double RNA in situ hybridization for Otp andhox2b was performed basically following the protocols ofhimamura et al. (1994) and Jowett and Lettice (1994), addinginor modifications. Chicken embryos of stages 20, 24 and 28ere hybridized with 2 �g/ml of each probe for 12 h. Embryosere then washed in the hybridization buffer to remove the

emaining probe. Digoxigenin-labeled probe was detected withlkaline-phosphatase-conjugated anti-Dig-antibody, followedy a reaction with NBT/BCIP (blue precipitate). To decrease theackground in the tissue, NBT was used at a lower concentra-ion (33 mg/ml). Fluorescein-labeled probe was detected withlkaline-phosphatase-conjugated anti-fluorescein antibody, fol-

owed by a reaction with a solution containing 248 mg/ml 2--iodophenyl)-3-(4-nitrophenyl)-5-phenyl-tetrazolium chloridend 33 mg/ml BCIP (reddish-brown precipitate).

hotomicrograph production

ll stains shown in the figures were digitized and adjusted inontrast and brightness with the Photoshop software (Adobe Sys-ems, Mountain, CA, USA). Labeling of figures was done with thereehand software (Macromedia, San Francisco, CA, USA) and

he Photoshop software (Adobe Systems).

C

Ipdt2ceo1a

cWR(Ptbpevvarnatsdpi

(TdcenIosp

sPtswl2seu

E

Lrd(spWdti

Ftib


RESULTS

ad7 expression in the spinal cord

n both the spinal cord and the hindbrain, Cad7 is ex-ressed in a well-demarcated domain at early stages ofevelopment, confirming previous preliminary data fromhe spinal cord (Nakagawa and Takeichi, 1998; Redies,000). We studied this expression domain first in the spinalord because numerous well-established molecular mark-rs for the ventrodorsal domains are available in this partf the CNS (Ericson et al., 1997b; Briscoe and Ericson,999; Briscoe et al., 2000; Liem et al., 2000; Takahashind Osumi, 2002).

At stage HH19, the Cad7 expression domain of the spinalord marks a distinct radial domain of the neural tube wall.e used transcription factors to characterize its boundaries.esults show that the dorsal boundary of cad7 expression

arrow in Fig. 1A, I) coincides with the ventral boundary ofax7 expression (arrow in Fig. 1B, I), which, in turn, defines

he border between the ventral and dorsal spinal cord (orasal/alar boundary [We assume that the alar and basallates are the major dorsoventral domains of the neural tube,ach with characteristic fate and function (i.e. mostly sensoryersus motor). In the alar and basal plates, dorsalizing andentralizing patterning effects predominate, respectively. Thepproximate position of the basal/alar boundary would thusepresent a molecularly-determined turning point of the alter-ative fates.]; see Discussion; Ericson et al., 1996; Briscoend Ericson, 1999; Liem et al., 2000). This result indicateshat the cad7 expression domain lies dorsally in the ventralpinal cord (i.e. the basal plate). The basal plate has beenivided into five progenitor cell (p) domains termed p0, p1, p2,MN and p3. These domains each give rise to a correspond-

ig. 7. Immunostains for cad7 (A, D), cad6B (B, E) and a Nissl stain (The exit of the somatomotor nerves (VIn; A, B; and XIIn, D, E) are locaten A points to cad7-positive cells at the exit point of the VIIn. The arrowars�100 �m (in A for A–C, and in D for D–F).

ng group of neurons termed V0, V1, V2, VMN and V3 F

Ericson et al., 1997a; Briscoe et al., 2000; Muhr et al., 2001).he ventral boundary of the cad7 expression domain liesorsal to the Nkx2.2/ Pax6 border (Fig. 1D, E, I), but coin-ides in the mantle zone with the dorsal border of Lim3/Lhx4xpression (Fig. 1F). Lim3/Lhx4 is a marker for VMN and V2eurons (Sharma et al., 1998). The VMN neurons also showsl1 (Fig. 1G) and cad6B (Fig. 1C) signals, which do notverlap with cad7 at this stage. Thus the cad7 signal corre-ponds to the dorsal part of the basal plate and includes the0 and p1 domains (Takahashi and Osumi, 2002).

At stage HH23, the radial cad7 expression domain istill present and shares the same dorsal boundary with theax7 expression domain (arrows in Fig. 2A,B,H). In addi-

ion, the V2 and VMN neurons now express cad7, and aubset expresses cad6B (Fig. 2C; Marthiens et al., 2002),hile the Nkx2.2-positive p3 domain is cad7 negative. The

ateral part of the floor plate, which expresses Shh (Fig.E), is cad7 immunoreactive also. Results are summarizedchematically in Figs. 2H and 3A. The domain of cad7xpression persists in the spinal ventricular zone at leastntil stage HH28 (data not shown).

xpression of cad7 in the hindbrain

ike the spinal cord, the hindbrain contains a completeadial domain, which expresses cad7 (Figs. 4–8). Thisomain appears in the hindbrain around stages 11–12around 45 h of incubation) and it has become prominent attage HH15 (Fig. 4A). In the ventricular layer, the domainersists at least until stage HH30 (E6.5–7; Fig. 7G).hole-mount immunostaining of the hindbrain (Fig. 4)

emonstrates that the cad7-positive domain extends fromhe spinal cord to the rostral limit of the hindbrain. It isnterrupted at the rhombomere boundaries (arrowheads in

) for a series of consecutive frontal sections at stage HH24. Note thatedial border of the cad7 expression domain (arrows). The arrowhead

B points to cad6B-positive cells in the more distal facial nerve. Scale

hio; C, Fd at the mhead in

ig. 4A, B) and at the midbrain/hindbrain boundary (arrow-

hcifp

iecdomcmb5atsp5F

C

Wr

otcltmcdbgctepo2rgENeiHcm

FHci in G for G


ead in Fig. 4C). The floor plate of the hindbrain expressesad7 weakly. The median epichordal strip (Puelles, 1978a)

n the caudal hindbrain, and the paramedian floor domainsrom about r3 to the midbrain/hindbrain boundary are cad7ositive (arrows in Fig. 4A, B).

To define the limits of the cad7-positive lateral domainn the hindbrain, we used the same panel of genetic mark-rs as in the spinal cord, but added Otp. Like in the spinalord, the lateral (dorsal) border of the cad7 expressionomain (Fig. 5A) coincides with the medial (ventral) borderf Pax7 expression in the ventricular layer (Fig. 5B). In theantle layer, the lateral border of the cad7 domain coin-

ides with the lateral border of Otp expression (Fig. 6). Theedial border of the cad7 domain coincides with theoundary between Pax6 and Nkx2.2 expression (Figs. 3B;D, E). The roots of the somatomotor nerves (abducensnd hypoglossal nerves, arrows in Fig. 7A, D) are found athe medial border of the cad7 domain. The paramediantripes of cad7 expression in the floor plate coincide ap-roximately with intense cad6B and Shh expression (Fig.C, F). These results are summarized schematically inigs. 3B and 5H.

adherin expression by bm neurons

e next studied the migration of the bm neurons with

ig. 8. Cadherin expression during and after the migration of trigemH27, D–F; and stage HH30, G–I). Immunostains for cad7 (A, D, Gonsecutive sections. The arrowheads in A and G point at cad7-positn the section). Scale bars�100 �m (in A for A–C, in D for D–F, and

espect to the cad7 domain of the hindbrain. At their place p

f birth, adjacent to the floor plate, the bm neurons of therigeminal nerve begin to express cad7, but do not expressad6B (Fig. 8). As they extend axons and then migrate

aterally through the cad7 domain, both the cell bodies andhe axons express cad7 (Fig. 8A). Fig. 9 shows a higheragnification of the cad7-positive neurons migrating along

ad7-positive neurites at a position lateral to the cad7omain. Similar observations were made for the migratingm (mbm) neurons of the facial nerve (Fig. 7A) and of thelossopharyngeal/vagus nerves (Fig. 7D). The migratingad7-positive cells co-express Phox2b, confirming thathey are bm neurons (Fig. 10A, B). The Phox2b genencodes a homeodomain transcription factor that is ex-ressed in progenitors and mature bm neurons, apart of inther cell types (Pattyn et al., 1997, 2000; Garel et al.,000). Cad7 expression persists until the bm neurons haveeached their final lateroventral location. Here, the postmi-ratory bm neurons begin to co-express cad6B (Figs. 8D,; 10C). Later, cad7 is slowly downregulated (Fig. 8G).ote that the migrating neurons and their axons do notxpress cad6B (Figs. 8B, E; 10C). Migration of the trigem-

nal motoneurons ends at around stage HH28 (E5.5–6;eaton and Moody, 1980; Covell and Noden, 1989), andad7 expression is no longer observed in the trigeminalotor nucleus at stage HH30, while cad6B expression

eurons at different stages of development (stage HH22, A–C; stage(B, E, H) and a Nissl stain (Thio; C, F, I) are shown for series of

in the trigeminal nerve. The asterisks in D–F indicate an artifact (fold–I).

inal bm n), cad6B

ive cells

ersists until at least stage HH37 (E11; C. Redies and L.

Pcrd(a1f

P

Tnalboort3ttl1

t(P1(

Incpm(1mO2ttwb(nhicocsca1dsvcCPsdclattea

Cph

IiBtb“

Fdtctban


uelles, unpublished observations). To confirm that thead6B-positive neurons in the alar plate are motor neu-ons, double immunostaining was performed. Resultsemonstrate that all cad6B-positive neurons express Isl-1Figs. 3C–E; Varela-Echavarria et al., 1996; Marthiens etl., 2002) and most of them also express Phox2b (Fig.0A) and cad7 (Fig. 10C). Similar observations were madeor the other bm cell groups (data not shown).

ostmigratory localization of bm neurons

he results described above demonstrate that the bmeurons of the chicken hindbrain, which migrate first later-lly and then radially pialward, assume a final position

ateral to the cad7 domain. As described above, the lateralorder of the cad7 domain coincides with the lateral borderf the Otp domain in the mantle layer (Fig. 6). We corrob-rated also in flat mounts that the main bm nuclei come toeside lateral to this border, using double in situ hybridiza-ion for Otp and Phox2b at stages 20, 24 and 28 (Fig.F–H). Results suggest that the majority of Phox2b-posi-ive motoneurons eventually assume positions lateral tohe Otp expression domain and ventromedial to the lateralongitudinal band of Phox2b expression (l in Figs. 3H;

ig. 9. Cad7 expression by mbm of the trigeminal nerve at 4 days ofevelopment (stage 23). A transverse section through the hindbrain athe level of the trigeminal nerve is shown. B shows a higher magnifi-ation of the migrating bm neurons. The arrows in B point to cell bodieshat are migrating along cad7-positive neurites. Note that the migratingm neurons display cad7 immunoreactivity on their cell surface. Therrowheads point to the cad7-positive motor root of the trigeminalerve. Scale bars�50 �m (in A), and 20 �m (in B).

0A). This also applies to the “dorsal” bm neuron popula- 1

ions that stop their migration and stabilize periventricularlydata not shown). Note that the lateral border of the medialhox2b domain in the ventricular zone (large arrow in Fig.0A) coincides with the lateral border of the cad7 domainarrowheads in Fig. 10A).

DISCUSSION

n the present study, we investigated the relation of bmeuronal migration to mediolateral patterning of thehicken embryonic hindbrain. Although a number of ex-ression markers have been available for different do-ains of the proliferative (ventricular) zone of the hindbrain

Simeone et al., 1994; Myat et al., 1996; Osumi et al.,997; Takahashi and Osumi, 2002) and for the differentotoneuron populations (Varela-Echavarria et al., 1996;sumi et al., 1997; Pattyn et al., 1997, 2000; Garel et al.,000; Puelles et al., 2001; Takahashi and Osumi, 2002),he present study is the first to examine in which domain ofhe mantle layer the bm neurons assume their final positionith respect to the basal/alar boundary. This issue haseen obscured by both the “sulcus limitans of His” conceptHis, 1888) and the four functional columns of cranial nerveuclei defined in the adult hindbrain (reviewed in Nieuwen-uys, 1998). The sulcus limitans is a variable morpholog-

cal feature (Hugosson, 1955) that does not correlatelearly with any relevant molecular boundary. The conceptf adult functional columns, which are long known not toorrelate with primary histogenetic phenomena (Hugos-on, 1955, 1957), disregards the existence of tangentialell migrations and dorsoventral genetic patterning mech-nisms (Puelles and Rubenstein, 1993; Shimamura et al.,995; Briscoe and Ericson, 2001). There is sound evi-ence on complexly regulated alternative “ventral” or “dor-al” molecular fates of the neural tube lateral wall, thusalidating a molecular redefinition of the old alar-basaloncept (see Results; Briscoe et al., 2000; Litingtung andhiang, 2000; Muhr et al., 2001; Vallstedt et al., 2001;ersson et al., 2002). Our analysis of recent literatureuggests that the ventral limit of the Pax7 expressionomain across the spinal cord and hindbrain is a credibleandidate for the postulated basal/alar boundary (see be-

ow). We think the terms “alar”/“basal” are semanticallydvantageous, mainly because “dorsal”/“ventral” are rela-ive terms (i.e. there is a ventral part of the “dorsal” neuralube) and, in the hindbrain, dorsal/ventral changes to lat-ral/medial as the two halves of the neural tube open likebook (see Introduction).

ad7 expression marks the most dorsal (lateral)ortion of the basal plate in the spinal cord andindbrain

n the spinal cord, ventrodorsal patterning has been wellnvestigated with several markers (Ericson et al., 1997b;riscoe et al., 2000; Marthiens et al., 2002). Specifically,

he ventral border of the Pax7 expression domain haseen used to define the boundary between “dorsal” andventral” spinal cord (Jostes et al., 1990; Ericson et al.,
997b; Kawakami et al., 1997; Briscoe and Ericson, 1999;

BtbdaRVpbdfimwRdgctcre

(1

pbllwdfbvdi

Bp

Tm1

Fitcfnaae(


riscoe et al., 2000; Mizuguchi et al., 2001). Other markershat approximate this boundary at both spinal and hind-rain levels are Dbx1 and Evx1, the determinants of theorsal/p0 progenitor domain and V0 interneurons (Bastiannd Gruss, 1990; Puelles and Rubenstein, 1993; Moran-ivard et al., 2001; Muhr et al., 2001; Pierani et al., 2001;allstedt et al., 2001; Takahashi and Osumi, 2002). It isresently a matter of convention whether the basal/alaroundary is defined at the dorsal/p0 limit, across the p0omain itself, or at the p0/p1 limit. Though we chose therst option in this report, our conclusions with regard to theigration of the bm neurons into the alar plate (see below)ould not change by selecting one of the alternatives.esults from the present study show that the radial cad7omain in the spinal cord represents the p0 and p1 pro-enitor regions of the basal plate. The cad7 domain in-ludes the VO and V1 interneurons in the mantle layer. Ithus marks the most dorsal region of the basal plate. Theomplete radial labeling in this region is likely due to theegional expression of this cadherin by radial glia thatxtends processes from the ventricular to the pial surface

ig. 10. Double-labeling of transverse sections through the hindbrainmmunostained with antibodies against cad7 (purple) and hybridized ino the floor plate (F) and translocate to the Vm. The large arrow in Aoincides with the lateral border of the cad7 domain (arrowheads). Throm the Phox2b-positive Vm nucleus and are also associated with theurons that co-express Phox2b (cytoplasmic staining; turquoise) andhigher magnification in the insert. Scale bars�100 �m (in A), 50 �m

gainst cad7 (red) and cad6B (green). A transverse section throughxpress cad7 whereas most of the postmigratory VMN neurons co-ex
in C).
Shimamura and Takeichi, 1992; Gänzler and Redies,995; Yoon et al., 2000).

Like in the spinal cord, the lateral border of cad7 ex-ression in the hindbrain coincides closely with the medialorder of Pax7 expression in the ventricular zone, with the

ateral border of the Otp-positive domain in the mantleayer (Figs. 5, 6; Simeone et al., 1994) and, in addition,ith the lateral border of a medial Phox2b expressionomain in the ventricular layer (Fig. 10A). Together, theour markers (cad7, Pax7, Otp and Phox2b) visualize theasal/alar boundary along its entire radial extent from theentricular to the pial surface. In conclusion, the cad7omain represents a dorsal domain of the basal plate also

n the hindbrain.

m neurons assume their final positions in the alarlate

he bm neurons are born adjacent to the floor plate and laterigrate laterally (Windle and Austin, 1936; Hugosson, 1955,957; Windle, 1970; Millet and Alvarado-Mallart, 1995;

evel of the gV. (A, B) The sections from stage HH27 embryos werea RNA probe for Phox2b (brown). Mbm originate at a position adjacent

the lateral border of Phox2b expression in the ventricular layer thatositive motor root fibers (small arrows in A; arrowheads in B) emergeng bm neurons. The arrow in B points to cell bodies of migrating bmll surface staining; brown). The same neurons (arrows) are shown atd 20 �m (inset in B). (C) Double-labeling of the VMN with antibodies

rain of an E5 embryo is shown. Note that the migrating bm neuronsh molecules (yellow) at this stage of development. Scale bar�50 �m

at the lsitu withindicatese cad7-pe migraticad7 (ce(in B), an

the hindbpress bot

VemaHstmbcclagtsame

ccc1Nn(pswpMtammm1prve11thlnbaiaSbrflmsi

Cm

IsertpelspdttbsmnpbfatdFtme

AD4sSr

A

B

B

B

B

B

B

C


arela-Echavarria et al., 1996; Pattyn et al., 1997, 2000; Qiut al., 1998; Puelles et al., 2001; Marthiens et al., 2002). Theigration is initially guided by their axons exiting the hindbrainlar plate close to the respective afferent root (Moody andeaton, 1983a,b). Results from the present study demon-trate that the final location of the bm cell bodies is lateral tohe overlapping cad7 and Otp expression domains in theantle layer (Fig. 6). Because this limit marks the basal/alaroundary (see above), we conclude that bm neurons in thehicken take up their final positions in the alar plate. Thisonclusion applies to all dorsally migrating bm neuron popu-ations: the trigeminal main and dorsal nuclei (Figs. 3C–E, 8,nd data not shown) and the bm nuclei of the facial andlossopharyngeal/vagus nerves (Fig. 7). In the present study,he bm neurons and their progenitors were visualized by initu hybridization for Phox2b (Figs. 3F–H, 10A, B; Pattyn etl., 1997, 2000). In addition, Isl1 (Fig. 3C–E) served as aarker for the postmigratory bm neurons (Varela-Echavarriat al., 1996; Marthiens et al., 2002).

Our results make it necessary to revise the century-oldoncept that all primary motor neurons, including the vis-eromotor column, are to be found in the basal plateolumns of the vertebrate brain (Gaskell, 1886, 1889; His,888; Johnston, 1905; Herrick, 1933; Kuhlenbeck, 1975;ieuwenhuys et al., 1998). The visceromotor column ofeurons is generally thought to contain the bm neurons“special visceromotor” neurons) and the parasympatheticreganglionic (“general visceromotor”) neurons. Note thatympathetic preganglionic neurons migrate also dorsal-ard in the thoracic spinal cord, settling in an intermediateosition between the dorsal and ventral horns (Levi-ontalcini, 1950). At least for the chicken brain, we show

hat during their migration, the bm neurons cross the basal/lar boundary (Fig. 6) and settle within a medial subdo-ain of the alar plate, either periventricularly or at inter-ediate/superficial radial levels, after an additional radialigration step with trailing axons (Windle and Austin,936). This particular radial domain is characterized by theresence of catecholaminergic neurons near the bm neu-ons (Puelles, 1978b; Puelles and Verney, 1998) and liesentromedially adjacent to the “lateral” longitudinal bandxpressing Phox2b in the ventricular zone (l in Figs. 3F–H,0A), a possible source of such neurons (Pattyn et al.,997, 2000). In mammals, a chemoarchitectonically dis-

inct “intermediate” sector (or “lateral paracore”; Nieuwen-uys, 1985) of the reticular formation develops in this alar

ocation (Foster, 1998; Jacobowitz and Abbott, 1998; Paxi-os and Watson, 1998; Paxinos and Franklin, 2001). Them neurons migrating into this alar domain thus encounterdistinct cellular and molecular environment. This result

mplies that the radial domain apparently is able to eitherttract or at least stabilize specific migrating elements.ite-specific guidance by radial glia may be a potentialasis for the pialward migration of subgroups of bm neu-ons within the alar plate. As a result of the partial reshuf-ing of the diverse cell populations via tangential and radialigrations, a neuronal complex is formed that might pos-

ibly carry out novel functions, which may depend on the
nteraction of the neurons of hybrid developmental origin.
hange of cadherin expression during bm neuronigration and differentiation

t has been reported that facial bm neurons in the mouseuccessively change their expression of cell surface mol-cules (TAG-1, Ret and cadherin-8) as they migrate from4 to r6 (Garel et al., 2000). An examination of relevantransgenic mice suggested that this change of gene ex-ression is an adaptation to the different rhombomericnvironments and may play a role in the selection of the

ocal migratory pathway. In the present study, we report aimilar switch from cad7 expressed during the migratoryhase to cad6B in the post-migratory phase of bm neuronifferentiation. It is conceivable that the bm neurons stopheir tangential translocation in response to a signal fromhe cad7-negative medial alar plate. Moreover, it is possi-le that the bm neurons change their expression of cellurface molecules in response to the different environ-ent, as demonstrated in the mouse for the facial bmeurons (Garel et al., 2000; Studer, 2001). The cad7-ositive axons of the migrating bm neurons exit the hind-rain at a point where cad7-positive Schwann cells areound (arrowheads in Figs. 7A, 8A, G; Sc in Figs. 9A, 10B),lso at early stages of development when bm axons growhrough this area (Nakagawa and Takeichi, 1995). Moreistal Schwann cells express cad6B strongly (arrowhead inig. 7B; Nakagawa et al., 1995). The two cadherins may

hus provide homotypic substrates at early stages (proxi-al growth) and late stages (distal growth) of bm axonlongation, respectively.

cknowledgments—This work was supported by a grant from theeutsche Forschungsgemeinschaft to C.R. (Re 616/4–3 and–4) and an MCYT grant to L.P. (BFI2002-03668). P.A. wasupported by a “Ramon y Cajal” contract. We thank J.-F. Brunet,. Nakagawa, A. Simeone, M. Takeichi for their generous gifts of

eagents, and U. Jonetat and U. Laub for technical assistance.

REFERENCES

rens M, Straznicky C (1986) The development of the trigeminal (V)motor nucleus in normal and tubocurare treated chick embryos.Anat Embryol 174:67–72.

asler K, Edlund T, Jessell TM, Yamada T (1993) Control of cellpattern in the neural tube: regulation of cell differentiation by dor-salin-1, a novel TGF beta family member. Cell 73:687–702.

astian H, Gruss P (1990) A murine even-skipped homologue, Evx 1,is expressed during early embryogenesis and neurogenesis in abiphasic manner. EMBO J 9:1839–1852.

rahic M, Ozden S (1992) Simultaneous detection of cellular RNA andproteins. In: In situ hybridization (Wilkinson DG, eds), pp 85–104.Oxford: IRL Press.

riscoe J, Ericson J (1999) The specification of neuronal identity bygraded Sonic Hedgehog signalling. Semin Cell Dev Biol10:353–362.

riscoe J, Ericson J (2001) Specification of neuronal fates in theventral neural tube. Curr Opin Neurobiol 11:43–49.

riscoe J, Pierani A, Jessell TM, Ericson J (2000) A homeodomainprotein code specifies progenitor cell identity and neuronal fate inthe ventral neural tube. Cell 101:435–445.

ovell D Jr, Noden DM (1989) Embryonic development of the chickprimary trigeminal sensory-motor complex. J Comp Neurol
286:488–503.

E

E

E

E

F

F

G

G

G

G

G

H

H

H

HH

H

H

J

J

J

J

K

K

L

L

L

L

L

M

M

M

M

M

M

M

M

M

N

N

N

N

O

P

P


ricson J, Briscoe J, Rashbass P, van Heyningen V, Jessell TM(1997a) Graded sonic hedgehog signaling and the specification ofcell fate in the ventral neural tube. Cold Spring Harbor Symp QuantBiol 62:451–466.

ricson J, Morton S, Kawakami A, Roelink H, Jessell TM (1996) Twocritical periods of Sonic Hedgehog signaling required for the spec-ification of motor neuron identity. Cell 87:661–673.

ricson J, Rashbass P, Schedl A, Brenner-Morton S, Kawakami A,van Heyningen V, Jessell TM, Briscoe J (1997b) Pax6 controlsprogenitor cell identity and neuronal fate in response to graded Shhsignaling. Cell 90:169–180.

ricson J, Thor S, Edlund T, Jessell TM, Yamada T (1992) Earlystages of motor neuron differentiation revealed by expression ofhomeobox gene Islet-1. Science 256:1555–1560.

oster GA (1998) Chemical neuroanatomy of the prenatal rat brain.Oxford: Oxford University Press.

ritzsch B (1998) Of mice and genes: evolution of vertebrate braindevelopment. Brain Behav Evol 52:207–217.

änzler SII, Redies C (1995) R-cadherin expression during nucleusformation in chicken forebrain neuromeres. J Neurosci15:4157–4172.

arel S, Garcia-Dominguez M, Charnay P (2000) Control of the mi-gratory pathway of facial branchiomotor neurones. Development127:5297–5307.

askell W (1886) On the structure, distribution and function of thenerves which innervate the visceral and vascular systems.J Physiol 7:1–81.

askell W (1889) On the relation between the structure, function,distribution and origin of the cranial nerves; together with an theoryof the origin of the nervous system of vertebrata. J Physiol10:153–211.

uthrie S, Butcher M, Lumsden A (1991) Patterns of cell division andinterkinetic nuclear migration in the chick embryo hindbrain. J Neu-robiol 22:742–754.

amburger V, Hamilton H (1951) A series of normal stages in thedevelopment of the chick embryo. J Morphol 88:49–92.

eaton MB, Moody SA (1980) Early development and migration of thetrigeminal motor nucleus in the chick embryo. J Comp Neurol189:61–99.

errick CJ (1922) Functional factors in the morphology of the forebrainof fishes. In: Libro en honor de D. Santiago Ramón y Cajal, pp142–204. Madrid: Jiméney Molina.

errick CJ (1933) Morphogenesis of the brain. J Morphol 54:233–258.is W (1888) Zur Geschichte des Gehirns sowie der centralen und

periferischen Nervenbahnen beim menschlichen Embryo. AbhMath Phys Kl Kgl Sächs Ges Wiss 14:339–393.

ugosson R (1955) Studien über die Entwicklung der longitudinalenZellsäulen und der Anlagen der Gehirnnervenkerne in der Medullaoblongata bei verschiedenen Vertebraten. Z Anat Entwickl-Gesch118:543–566.

ugosson, R (1957) Morphologic and experimental studies on thedevelopment and significance of the rhombencephalic longitudinalcolumns. Dissertation, University of Lund.

acobowitz DM, Abbott LC (1998) Chemoarchitectonic atlas of thedeveloping mouse brain. Boca Raton: CRC Press.

ohnston JB (1905) The cranial nerve components of Petromyzon.Morphologisches Jahrbuch 34:149–203.

ostes B, Walther C, Gruss P (1990) The murine paired box gene,Pax7, is expressed specifically during the development of thenervous and muscular system. Mech Dev 33:27–37.

owett T, Lettice L (1994) Whole-mount in situ hybridizations on ze-brafish embryos using a mixture of digoxigenin- and fluorescein-labelled probes. Trends Genetics 10:73–74.

awakami A, Kimura-Kawakami M, Nomura T, Fujisawa H (1997)Distributions of PAX6 and PAX7 proteins suggest their involve-ment in both early and late phases of chick brain development.
Mech Dev 66:119–130.
uhlenbeck H (1975) The central nervous system of vertebrates, Vol.4: spinal cord and deuterencephalon. Basel: Karger.

ee KJ, Dietrich P, Jessell TM (2000) Genetic ablation reveals that theroof plate is essential for dorsal interneuron specification. Nature403:734–740.

evi-Montalcini R (1950) The origin and development of the visceralsystem in the spinal cord of the chick embryo. J Morphol86:253–283.

iem KF Jr, Jessell TM, Briscoe J (2000) Regulation of the neural pat-terning activity of sonic hedgehog by secreted BMP inhibitors ex-pressed by notochord and somites. Development 127:4855–4866.

itingtung Y, Chiang C (2000) Control of Shh activity and signaling inthe neural tube. Dev Dynam 219:143–154.

umsden A, Keynes R (1989) Segmental pattern of neuronal devel-opment in the chick hindbrain. Nature 337:424–428.

arthiens V, Padilla F, Lambert M, Mege RM (2002) Complementaryexpression and regulation of cadherins 6 and 11 during specificsteps of motoneuron differentiation. Mol Cell Neurosci20:458–475.

arti E, Takada R, Bumcrot DA, Sasaki H, McMahon AP (1995)Distribution of Sonic hedgehog peptides in the developing chickand mouse embryo. Development 121:2537–2547.

illet S, Alvarado-Mallart RM (1995) Expression of the homeobox-containing gene En-2 during the development of the chick centralnervous system. Eur J Neurosci 7:777–791.

izuguchi R, Sugimori M, Takebayashi H, Kosako H, Nagao M, Yo-shida S, Nabeshima Y, Shimamura K, Nakafuku M (2001) Combi-natorial roles of olig2 and neurogenin2 in the coordinated inductionof pan-neuronal and subtype-specific properties of motoneurons.Neuron 31:757–771.

oody SA, Heaton MB (1983a) Developmental relationships betweentrigeminal ganglia and trigeminal motoneurons in chick embryos: I.Ganglion development is necessary for motoneuron migration.J Comp Neurol 213:327–343.

oody SA, Heaton MB (1983b) Developmental relationships betweentrigeminal ganglia and trigeminal motoneurons in chick embryos: II.Ganglion axon ingrowth guides motoneuron migration. J CompNeurol 213:344–349.

oran-Rivard L, Kagawa T, Saueressig H, Gross MK, Burrill J, Goul-ding M (2001) Evx1 is a postmitotic determinant of v0 interneuronidentity in the spinal cord. Neuron 29:385–399.

uhr J, Andersson E, Persson M, Jessell TM, Ericson J (2001) Grou-cho-mediated transcriptional repression establishes progenitor cellpattern and neuronal fate in the ventral neural tube. Cell104:861–873.

yat A, Henrique D, Ish-Horowicz D, Lewis J (1996) A chick homo-logue of Serrate and its relationship with Notch and Delta homo-logues during central neurogenesis. Dev Biol 174:233–247.

akagawa S, Takeichi M (1995) Neural crest cell-cell adhesion con-trolled by sequential and subpopulation-specific expression ofnovel cadherins. Development 121:1321–1332.

akagawa S, Takeichi M (1998) Neural crest emigration from theneural tube depends on regulated cadherin expression. Develop-ment 125:2963–2971.

ieuwenhuys R (1985) Chemoarchitecture of the brain. Berlin:Springer.

ieuwenhuys R, ten Donkelaar HJ, Nicholson C (1998) The centralnervous system of vertebrates. Heidelberg: Springer.

sumi N, Hirota A, Ohuchi H, Nakafuku M, Iimura T, Kuratani S,Fujiwara M, Noji S, Eto K (1997) Pax-6 is involved in the specifi-cation of hindbrain motor neuron subtype. Development124:2961–2972.

attyn A, Hirsch M, Goridis C, Brunet JF (2000) Control of hindbrainmotor neuron differentiation by the homeobox gene Phox2b. De-velopment 127:1349–1358.

attyn A, Morin X, Cremer H, Goridis C, Brunet JF (1997) Expression
and interactions of the two closely related homeobox genes

P

P

P

P

P

P

P

P

P

P

Q

R

R

R

R

S

S

S

S

S

S

T

T

V

V

W

W

W

Y


Phox2a and Phox2b during neurogenesis. Development124:4065–4075.

axinos G, Franklin KBJ (2001) The mouse brain in stereotaxic coor-dinates. San Diego: Academic Press.

axinos G, Watson C (1998) The rat brain in stereotaxic coordinates.San Diego: Academic Press.

ersson M, Stamataki D, te Welscher P, Andersson E, Bose J, RutherU, Ericson J, Briscoe J (2002) Dorsal-ventral patterning of thespinal cord requires Gli3 transcriptional repressor activity. GenesDev 16:2865–2878.

ierani A, Moran-Rivard L, Sunshine MJ, Littman DR, Goulding M,Jessell TM (2001) Control of interneuron fate in the developingspinal cord by the progenitor homeodomain protein Dbx1. Neuron29:367–384.

oh A, Karunaratne A, Kolle G, Huang N, Smith E, Starkey J, Wen D,Wilson I, Yamada T, Hargrave M (2002) Patterning of the verte-brate ventral spinal cord. Int J Dev Neurosci 46:597–608.

uelles E, Rubenstein JL, Puelles L (2001) Chicken Nkx6.1 expres-sion at advanced stages of development identifies distinct brainnuclei derived from the basal plate. Mech Dev 102:279–282.

uelles L (1978a) A Golgi-study of oculomotor neuroblasts migratingacross the midline in chick embryos. Anat Embryol (Berl)152:205–215.

uelles L (1978b) Velate glioblasts in the developing chick optictectum: probable immature forms of oligodendroglia. Neuroscience3:41–47.

uelles L, Rubenstein JLR (1993) Expression patterns of homeoboxand other putative regulatory genes in the embryonic mouse fore-brain suggest a neuromeric organization. Trends Neurosci16:472–479.

uelles L, Verney C (1998) Early neuromeric distribution of tyrosine-hydroxylase-immunoreactive neurons in human embryos. J CompNeurol 394:283–308.

iu M, Shimamura K, Sussel L, Chen S, Rubenstein JL (1998) Controlof anteroposterior and dorsoventral domains of Nkx-6.1 gene ex-pression relative to other Nkx genes during vertebrate CNS devel-opment. Mech Dev 72:77–88.

amon y Cajal S (1911) Histologie du système nerveux de l=homme etdes vertébrés. Paris. Reprint, 1972. Madrid: Maloine.

edies C (2000) Cadherins in the central nervous system. Prog Neu-robiol 61:611–648.

edies C, Engelhart K, Takeichi M (1993) Differential expression of N-and R-cadherin in functional neuronal systems and other struc-tures of the developing chicken brain. J Comp Neurol333:398–416.

edies C, Inuzuka H, Takeichi M (1992) Restricted expression of N-and R-cadherin on neurites of the developing chicken CNS. J Neu-rosci 12:3525–3534.

harma K, Sheng HZ, Lettieri K, Li H, Karavanov A, Potter S, West-phal H, Pfaff SL (1998) LIM homeodomain factors Lhx3 and Lhx4assign subtype identities for motor neurons. Cell 95:817–828.

himamura K, Hartigan DJ, Martı́nez S, Puelles L, Rubenstein JLR(1995) Longitudinal organization of the anterior neural plate andneural tube. Development 121:3923–3933.

himamura K, Hirano S, McMahon AP, Takeichi M (1994) Wnt-1-dependent regulation of local E-cadherin and �N-catenin expres-sion in the embryonic mouse brain. Development 120:2225–2234.

himamura K, Takeichi M (1992) Local and transient expression ofE-cadherin involved in mouse embryonic brain morphogenesis.Development 116:1011–1019.

imeone A, D’Apice MR, Nigro V, Casanova J, Graziani F, Acampora D,Avantaggiato V (1994) Orthopedia, a novel homeobox-containinggene expressed in the developing CNS of both mouse and Drosoph-ila. Neuron 13:83–101.

tuder M (2001) Initiation of facial motoneurone migration is depen-dent on rhombomeres 5 and 6. Development 128:3707–3716.

akahashi M, Osumi N (2002) Pax6 regulates specification of ventralneurone subtypes in the hindbrain by establishing progenitor do-mains. Development 129:1327–1338.

immer JR, Wang C, Niswander L (2002) BMP signaling patterns thedorsal and intermediate neural tube via regulation of homeoboxand helix-loop-helix transcription factors. Development129:2459–2472.

allstedt A, Muhr J, Pattyn A, Pierani A, Mendelsohn M, Sander M,Jessell TM, Ericson J (2001) Different levels of repressor activityassign redundant and specific roles to Nkx6 genes in motor neuronand interneuron specification. Neuron 31:743–755.

arela-Echavarria A, Pfaff SL, Guthrie S (1996) Differential expressionof LIM homeobox genes among motor neuron subpopulations inthe developing chick brain stem. Mol Cell Neurosci 8:242–257.

indle WF (1970) Development of neural elements in human embryosof four to seven weeks gestation. Exp Neurol 28:44–83.

indle WF, Austin MF (1936) Neurofibrillar development in the centralnervous system of chick embryos up to 5 days’ incubation. J CompNeurol 63:431–463.

öhrn JCP, Nakagawa S, Ast M, Takeichi M, Redies C (1999) Com-binatorial expression of cadherins and the sorting of neurites in thetectofugal pathways of the chicken embryo. Neuroscience90:985–1000.

oon MS, Puelles L, Redies C (2000) Formation of cadherin-express-ing brain nuclei in diencephalic alar plate subdivisions. J CompNeurol 421:461–480.

(Accepted 16 June 2004)

Documents

MOLECULAR PROFILING INDICATES AVIAN BRANCHIOMOTOR