This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Acta histochemica 112 (2010) 508—517

Differential Shh, Bmp and Wnt gene expressionsduring craniofacial development in mice

Katiucia Batista Silva Paivaa,�, Maria das Grac-as Silva-Valenzuelaa,Silvia Maria Gomes Massironib, Gui Mi Koc, Filipe Modolo Siqueirad,Fabio Daumas Nunesa

aDepartment of Oral Pathology, Dental School, University of Sao Paulo, Sao Paulo, BrazilbDepartment of Immunology, Biomedical Sciences Institute, University of Sao Paulo, BrazilcCenter for Development of Animal Models, Federal University of Sao Paulo, BrazildDepartment of Pathology, Center of Health Sciences, Federal University of Santa Catarina, Brazil

Received 9 March 2009; received in revised form 5 May 2009; accepted 19 May 2009

KEYWORDSBMP-4;SHH;Wnt-5a;Craniofacialdevelopment;In situ hybridization

SummaryIn this study, Bmp-4, Wnt-5a and Shh gene expressions were compared during earlycraniofacial development in mice by comparative non-isotopic in situ hybridization.Wild-type C57BL/6J mice were studied at various stages of embryonic development(from 8.5- to 13.5-day-old embryos – E8.5–13.5). During early odontogenesis,transcripts for Bmp-4, Shh and Wnt-5a were co-localised at the tooth initiationstage. At E8.5, Shh mRNA expression was restricted to diencephalon and pharyngealendoderm. Before maxillae and mandible ossification, Bmp-4 and Wnt-5a signalswere detected in the mesenchymal cells and around Meckel’s cartilage. Duringpalatogenesis, Shh was expressed only in the epithelium and Wnt-5a only in themesenchyme of the elevating palatal shelves. During tongue development, Shhexpression was found in mesenchyme, probably contributing to tongue miogenesis,while Wnt-5a signal was in the epithelium, possibly during placode development andpapillae formation. Taken together, these findings suggest that Bmp-4, Shh andWnt-5a gene expressions may act together on the epithelial–mesenchymalinteractions occurring in several aspects of the early mouse craniofacial develop-ment, such as odontogenesis, neuronal development, maxillae and mandibleossification, palatogenesis and tongue formation.& 2009 Elsevier GmbH. All rights reserved.

www.elsevier.de/acthis

0065-1281/$ - see front matter & 2009 Elsevier GmbH. All rights reserved.doi:10.1016/j.acthis.2009.05.007

�Corresponding to: Laboratorio de Patologia Molecular, Departamento de Patologia Bucal, Faculdade de Odontologia, Universidadede Sao Paulo, Avenida Lineu Prestes, 2227, Cidade Universitaria, Sao Paulo, SP, Brazil. Tel.: +55 1130918046; fax: +55 1130917894.

E-mail address: katipaiva@yahoo.com.br (K.B.S. Paiva).

Author's personal copy

Introduction

Bone morphogenetic proteins (BMPs) are multi-functional growth factors belonging to the trans-forming growth factor-beta (TGF-b) superfamily.BMPs were originally isolated by their ability toinduce ectopic bone and cartilage formation in vivo(Urist, 1965), but it became rapidly evident thatBMPs also act as multifunctional regulators inmorphogenesis during development in vertebratesand invertebrates. Family members are expressedon various cell types, including monocytes, epithe-lial cells, mesenchymal cells and neuronal cellsduring limb development, endochondral ossifica-tion, early fracture, cartilage repair, odontogenesisand other biological processes (Balemans, 2002;Granjeiro et al., 2005). BMPs are present in boneand dentin, capable of stimulating the formation ofnew bone and transmitting inductive signals duringinteractions between epithelial and mesenchymaltissues in developing organs (Thesleff, 1995). A nullmutation of the Bmp-4 gene in mice results indefects in the extraembryonic and posterior/ventral mesoderm formation and mutant embryosdie between embryonic days (E) 6.5 and E9.5(Winnier et al., 1995).

The hedgehog (Hh) family of signaling moleculesis conserved throughout evolution with threesecreted proteins that have been identified inmammals: Sonic hedgehog (Shh), Indian hedgehog(Ihh) and Desert hedgehog (Dhh) (Ingham andMcMahon, 2001). Hh signaling is crucial for thedevelopment of the anterior face and skull. Loss offunction of a key Hh molecule (Shh) causes a widevariety of craniofacial defects, and it is clear thatShh is involved in multiple steps of craniofacialdevelopment (Cordero et al., 2004).

Wnt proteins are a family of extracellularsignaling molecules that participate in a remark-able variety of cellular processes, such as cellfate, cell proliferation and differentiation,cell survival and apoptosis, cell behavior andmigration during morphogenesis (Logan and Nusse,2004). Wnt members can be separated intotwo classes: the Wnt-1 class (e.g. Wnt-1, -3a, -7a,-8, and -10), which activates the canonicalWnt/b-catenin pathway, and the Wnt-5a class(e.g. Wnt-4, -5a, -5b, and -11), which activatethe noncanonical Wnt pathway (Church andFrancis-West, 2002).

Bmp, HH and Wnt family members can mediateautocrine and paracrine signaling over short- orlong-range distances, and they are consideredimportant molecules in epithelial–mesenchymalinteraction (Uusitalo et al., 1999). This study co-localizes for the first time Bmp-4, Wnt-5a and Shh

transcripts during early craniofacial developmentin mice.

Material and methods

Animals and tissue processing

Pregnant female wild-type C57BL/6J mice weresacrificed at various stages of embryonic develop-ment – embryonic day (E) 8.5 to E13.5. The first dayof embryonic development was identified by thepresence of a vaginal plug in the mother. Theanimals received food and water ad libitum andwere sacrificed by a lethal dose of anesthetic. Allanimal procedures conducted in this study receivedprior approval from the local Animal ProtectionCommittee (Dental School, University of Sao Paulo)and were in agreement with the COBEA (ColegioBrasileiro de Experimentac-ao Animal). The embryos(n ¼ 5/period) were dissected out, quickly washedin phosphate buffered saline (PBS) and fixed in 4%paraformaldehyde in PBS, then the heads weredissected and classified according to Theiler (1972).Samples were then transferred to fresh fixativesolution for 24 h, washed in PBS and immersed in30% sucrose for 18 h at 4 1C. Finally, they wereembedded in a 1:1 optimal cutting temperature(OCT, Tissue Tek, Sakura, USA) medium/30% sucrosesolution. Specimens were then transferred tocryomolds containing OCT and stored at �80 1Cuntil study. Serial cryostat sections (coronal orien-tation) of 10 mm thickness were transferred tosilane-coated RNAse-free slides (Sigma-Aldrich,St. Louis, MO), air-dried and stored frozen until insitu hybridization (ISH) reaction.

Probes

All ISH riboprobes were generated by in vitrotranscription labelling with digoxygenin-UTP ac-cording to the manufacturer’s manual (Roche,Basel, Switzerland) and designed using the Gene-Tool 2.0 software (BioTools Incorporated, Alberta,Canada, http://www.biotools.com). Probe size andyield were determined by electrophoresis on a 1.5%agarose gel with an RNA standard (Table 1).Hybridization with transcripts derived from thesense orientation of each probe resulted in nosignal above background levels (Figure 1).

Synthesis of DNA template

Total RNA was extracted from E9-E10 miceheads using Trizol (Invitrogen, Carlsbad, CA, USA)

Differential gene expression in mouse craniofacial development 509

Author's personal copy

according to the manufacturer’s instructions.Reverse transcription was performed withSuperscriptTM (Invitrogen) and oligoDT primers.cDNA was synthesized with 1 mg of total RNAtreated with DNAse I in a volume of 20 ml. Wnt-5aprimers were designed spanning intron–exonboundary using the human and mouse mRNAsequence by GeneTool 2.0 Software (Table 2). PCRwas conducted in 25 ml reactions using 50 pM ofeach primer, 1.5mM MgCl2, 1� PCR buffer, 100 mMof each dNTP, 20 ng cDNA and 0.02U/ml Taq

(Invitrogen). The thermal profile consisted of aninitial denaturation step for 4min at 93 1C, fol-lowed by 25 cycles of amplification. Each roundconsisted of denaturation for 45 s/94 1C, annealingfor 1min 30 s/55 1C, extension for 2min/72 1C andan additional 7min/72 1C for terminal elongation.Amplification products were analyzed on a 1%agarose gel with ethidium bromide, where asingle band of 381 bp was visualized. Specificity ofthe amplicons was confirmed by cloning andsequencing.

Table 1. ISH probes according to orientation, RNA polymerase and restriction endonuclease.

Probes Sense Antisense References

RNA polymerase Restrictionendonuclease

RNA polymerase Restrictionendonuclease

Bmp-4 T7 AccI SP6 EcoRI Jones et al. (1991)Shh SP6 SmaI T3 HindIII Epstein et al. (1999)Wnt-5a T7 PstI T3 NotI Jones et al. (1991

Figure 1. Expression of Bmp-4, Shh and Wnt-5a. The signal was observed in anti-sense probes for (A) Bmp-4, (C) Shhand (E) Wnt-5a. No signal was seen in sense probes (B, D and F-Bmp-4, Shh and Wnt-5a, respectively). Scalebar ¼ 200 mm.

K.B.S. Paiva et al.510

Author's personal copy

ISH of frozen sections, alkaline phosphataselabelling

Frozen sections of mice embryos from E11.5 toE14.5, prepared as described above, were pro-

cessed for ISH as described previously (Cole et al.,2000) with some changes described below (Nuneset al., 2007). Sections were permeabilized with10 mg/mL proteinase K (Sigma-Aldrich) for 2min.Hybridizations were carried out in ‘‘seal-a-meal’’

Figure 2. Transcripts for Bmp-4 during mouse craniofacial development. (A and B) No signal was observed at E8.5 andE9. (C) At E11.5, the signal was detected in tooth formation area (black arrows) in both maxillae and mandible, as wellas in the middle of the mandible and in the tongue epithelium. (D) At E12, transcripts were found in oral epitheliumclose to odontogenesis area (black arrows), in tongue mesenchyme and epithelium. (E) At E12.5, a signal was seen inthe mesenchymal tissue around the frontonasal process and in lateral maxillar proeminence. (F) At maxillae, a strongsignal was observed at the ossification site and, at the mandible, transcripts present close to Meckel’s cartilage, in theodontogenesis area and in tongue. Scale bar ¼ 200 mm.

Table 2. Primer sequence for ISH probes.

Gene target Primer (50-30) Amplifiedfragment size (bp)

Bmp-4 F: CCT GGT AAC CGA ATG CTG AT 260R: AGC CGG TAA AGA TCC CTC AT

Shh F: AGG TGC AAA GAC AAG TTA AA 274R: CCA CGG AGT TCT CTG CTT TC

Wnt-5a F: GGA GAA GGT GCG AAG ACA G 319R: GGG CGT CCA CGA ACT CCT T

Differential gene expression in mouse craniofacial development 511

Author's personal copy

bags, overnight at 70 1C, in 5mL of hybridizationsolution (50% formamide, 5� SSC (pH 4.5),heparin 50 mg/mL, yeast RNA 50 mg/mL, 1% SDS)(Sigma-Aldrich) with a probe concentration of�0.2 mg/mL. Washes were as follows: three 15minchanges of 50% formamide, 30% 20� SSC (pH 4.5)and 10% 10� SDS at 70 1C, and three 15minchanges of 50% formamide and 12% 20� SSC(pH 4.5) at 65 1C. Detection of bound probe wasperformed using anti-digoxigenin antibody andNBT/BCIP as color-generating substrate (Roche,Basel, Switzerland). Slides were examined using aNikon SMZ-2 T microscope and digital pictures werecaptured using an Axiophot 2 Zeiss microscope(Carl Zeiss MicroImaging, Thornwood, NY, USA)and a 3CCD MTI camera (Dage-MTI, Michigan City,IN, USA).

Results

Bmp4

No signal was seen from E8.5 to E9 embryos(Figure 2A and B). From E11.5 to E12.5, transcriptsfor Bmp-4 were observed close to the early toothformation area and oral epithelium (Figure 2C–E) inboth upper and lower jaws. At the tongue, thesignal increased in both epithelium and mesench-yme from E11.5 to E12.5 (Figure 2C–F). At E12.5,an intense signal was detected in mesenchymaltissue around the frontonasal process, probably anossification site, lateral to the maxillar proemi-nence and in the mesenchyme close to Meckel’scartilage (Figure 2F).

Shh

Transcripts for Shh were only seen in epitheliallayers at all stages studied. At E8.5, a weak signalwas detected in faringeal endoderm and a strongone in the diencephalon (Figure 3A). At E10, thesame profile was seen and the signal was moreintense than at later stages (Figure 3B). From E12to E12.5, the tongue epithelium showed Shhtranscripts during the papillae and placode devel-opment (Figure 3C–F), being more intense in thelater stage (Figure 3F). At E12, a signal wasobserved close to the early tooth formation area(Figure 3C and D). The maxillar oral epithelium waslabelled uniformly from E12 to E12.5 (Figure 3Eand F). At E13.5, transcripts were restricted to theoral epithelium during elevation of the palatalshelves (Figure 3G and H).

Wnt-5a

No signal was seen from E8.5 to E11 embryos(Figure 4A and B). In the tongue mesenchyme, thestrong signal was seen (from E12 to E12.5) and wasweaker at E13.5 during placode development(Figure 4C–I). The maxillae mesenchyme (maxillarand frontonasal processes) was widely labelledfrom E12 to E12.5 (Figure 4C–F). At E13.5,transcripts were restricted to the mesenchymearound the tooth germs in both upper and lowerjaws (Figure 4H), in the mandibular mesenchymeand around Meckel’s cartilage (Figure 4I) and themesenchyme during elevation of the palatal shelves(Figure 4J).

Discussion

The findings reported in this work suggest thatBmp-4, Shh and Wnt-5a gene expressions maycontribute to the main epithelial–mesenchymalinteractions occurring in the early mouse craniofa-cial development: odontogenesis, neuronal deve-lopment, maxillae and mandible ossification, pala-togenesis and tongue formation.

The developing mouse molar is one currentmodel for studying the regulation of organogenesis.Specific epithelial–mesenchymal interactions con-trol odontogenesis, which comprises tooth initia-tion, morphogenesis, epithelial histogenesis andcytodifferentiation. These interactions are regu-lated by several conserved signaling molecules,including BMPs and FGFs. In this study, Bmp-4, ShhandWnt-5a were detected during tooth initiation inagreement with the results reported recently byour group (Nunes et al., 2007).

A reduction in Shh signaling in mice, humans andchick can result in a spectrum of craniofacialabnormalities manifesting as holoprosencephaly,cyclopia and other milder abnormalities, includingatrophy of the first pharyngeal arch (Cordero et al.,2004; Lazaro et al., 2004; Nagase et al., 2005). Inthe Shh mutant, the first arch forms, but is reducedin size. Loss of Shh results in cell death in themesenchyme and migrating neural crest cells(Ahlgren and Bronner-Fraser, 1999; Moore-Scottand Manley, 2005; Yamagishi et al., 2006). Thedefect in early arch development might thereforebe due the role of Shh in neural crest development.When the Shh pathway is knocked out specifically inthe neural crest, however, the pharyngeal archesinitially form normally (Jeong et al., 2004).Recently, Eberhart et al. (2006) and Haworthet al. (2007) reported that a hedgehog signal

K.B.S. Paiva et al.512

Author's personal copy

provides instructions for the lower jaw oralepithelium in zebrafish and mice, respectively.However, in the first case, the source for thehedgehog signal is the neural tube and, in the lattercase, the source is the pharyngeal endoderm. OnlyShh has been shown to play a role in neural tubepatterning among higher vertebrates (Bitgood andMcMahon, 1995; Zhang et al., 2001). Shh starts tobe expressed in the basal plate of the diencephalonand mesencephalon, as well as in the floor plate ofthe rhombencephalon (Echevarria et al., 2003;Vieira et al., 2005). As described above, weobserved a strong signal in the diencephalon and

a moderate one in the pharyngeal endoderm onlyby Shh.

The maxillae and mandible ossifications occur bytwo distinct ways. In the maxillae, the bonedevelops by intramembranous ossification and, inthe mandible, by the multistep process of endo-chondral ossification (Nagata et al., 1991). Inendochondral ossification, a cartilage templateforms when mesenchymal cells condense anddifferentiate into chondrocytes. These cells thenundergo a program of proliferation, maturation,hypertrophy, calcification and cell death (Erlebacheret al., 1995). Later, osteoblasts replace the

Figure 3. Transcripts for Shh during mouse craniofacial development. (A) At E8.5, a weak signal was detected inpharyngeal endoderm (black arrow) and a strong signal in diencephalon (arrow heads). (B) At E10, the signal was moreintense than at later stages. (C and D) At E12, signal was observed close to the early tooth formation area and in thetongue epithelium. (E and F) At E12.5, the tongue and oral epithelia were labelled. (G and H). At the E13.5, transcriptswere restricted to the oral epithelium during elevation of palatal shelves. Scale bar ¼ 200 mm (A–G) and 50 mm (H).

Differential gene expression in mouse craniofacial development 513

Author's personal copy

Figure 4. Transcripts for Wnt-5a during mouse craniofacial development. (A and B) No signal was observed at E8.5 andE11. (C and D) At E12, the maxillar mesenchyme (maxillar and frontonasal processess) as well as the tongue were widelylabelled. (E and F) At E12.5, the signal was more intense than at later stages. At E13.5, (H) transcripts were restrictedto the mesenchyme around the tooth germs in both upper and lower jaws, (I) in the mandibular mesenchyme andaround Meckel’s cartilage and (J) the mesenchyme during elevation of palatal shelves. Scale bar ¼ 200 mm (A–I) and50 mm (J).

K.B.S. Paiva et al.514

Author's personal copy

cartilage matrix by bone matrix. Thus, this processrequires active remodeling of cartilage and bonematrices and the concerted differentiation andactivity of several cell types. Initially (E12), weobserved that Wnt-5a expression was widelydetected in mesenchymal tissue in the maxillarand frontonasal processes. At a later stage (E13.5),the expression was more restricted to the chon-drogenic and odontogenic areas at the mandible.Wnt-5a and Wnt-5b are reported to promotechondrocyte differentiation and to inhibit chondro-cyte maturation (Church et al., 2002). Wnt-5a isknown to regulate chondrogenesis, limb develop-ment and longitudinal skeletal outgrowth (Dealyet al., 1993; Kawakami et al., 1999; Yang et al.,2003). In addition, forced expression of Wnt-5adelays chondrocyte maturation by inhibiting type Xcollagen expression, which is a marker for hyper-trophic chondrocytes (Tufan and Tuan, 2001) andinhibits type II collagen (Ryu and Chun, 2006).Curiously, Wnt-5a transcripts were localized in bothmaxillae and mandible, even though these havedifferent ossification processes. At E12, a strongphosphatase alkaline activity was detected in themesenchyme of the maxillae and frontonasalprocesses (Chimal-Monroy et al., 2002). Then, it ispossible that labelled mesenchymal cells observedcan be differentiating bone cells. Furthermore, atthe mandible, the mRNA found may contribute tothe early endochondral ossification. In the sameway, Bmp-4 transcripts were co-localized at max-illae and mandible, in agreement with Chimal-Monroy et al. (2002) and Kim et al. (1998),respectively. Taken together, these results suggestthat the interaction between Bmp-4 and Wnt-5amay act in both intramembranous and endochon-dral ossifications.

It is interesting that at the maxillae, only themesenchymal cells present in the elevation of thepalatal shelves presented a representative signalfor Wnt-5a, while only epithelial cells expressedShh. Mutation of the low-density lipoprotein re-ceptor-related protein 5 (LRP5), which acts in theWnt signaling pathway, causes high bone density,with a thickened mandible and torus palatinus, byimpairing the action of a normal antagonist of theWnt pathway and thus increasing Wnt signaling(Boyden et al., 2002). Shh in the mouse ispredominantly expressed in epithelia at the sitesof epithelial–mesenchymal interactions (Bitgoodand McMahon, 1995) occuring during palate devel-opment (Ferguson et al., 1984). In palatogenesis,Shh participates in the rugal development (Sasakiet al., 2007) and during fusion of the palatalshelves (Zhang et al., 2002). We evaluated theelevation of the palatal shelves process before

fusion of the palatal shelves. At this stage, the Shhsignal was predominantly in the epithelium, whilethe Wnt-5a signal was in the mesenchyme. Thus,we suggest that Shh and Wnt-5a may be importantto epithelial–mesenchymal interactions during pa-latogenesis.

Taste tissue development in mice starts aroundE11.5, after the emergence of the tongue swellingon the floor of the developing mandible (Paulsonet al., 1985). This is followed by the formation ofthe taste placode (E12.5), gustatory papillae(E13.5) and taste buds (around birth). As withdevelopment of other epithelial tissues, formationand patterning of taste papillae are thought to beinduced by epithelial–mesenchymal interactions(Mistretta, 1998). Expression of Shh, Bmp, Noggin,FGF and EGF are associated with taste papillaeinitiation and patterning, indicating the potentialrole of these factors in taste papillae development(Hall et al., 1999; Jung et al., 1999; Mistrettaet al., 2005; Miura et al., 2001; Zhou et al., 2006).Disruption of Shh signaling during the prenatalperiod caused aberrant development and patternsof fungiform papillae (Hall et al., 2003; Mistrettaet al., 2003). A recent study indicated that Wnt-10b signaling and interactions between the Wnt andShh pathways play essential roles in the develop-ment of fungiform papillae (Iwatsuki et al., 2007).Moreover, Tajbakhsh et al. (1998) demonstratedthat Wnt members, including Wnt-5a, can be activein tongue miogenesis. We observed that Shh signalwas predominantly in the mesenchyme, while Wnt-5a signal was in the epithelium. These findingssuggest that Bmp-4, Shh and Wnt-5a participate inthe epithelial–mesenchymal interactions duringtongue development.

Acknowledgement

We are grateful to FAPESP (The State of Sao PauloResearch) for Grant no. 1997/13228-5.

References

Ahlgren SC, Bronner-Fraser M. Inhibition of sonic hedge-hog signaling in vivo results in craniofacial neural crestcell death. Curr Biol 1999;9:1304–14.

Balemans W, Van HW. Extracellular regulation of BMPsignaling in vertebrates: a cocktail of modulators. DevBiol 2002;250:231–50.

Bitgood MJ, McMahon AP. Hedgehog and Bmp genes arecoexpressed at many diverse sites of cell–cell inter-action in the mouse embryo. Dev Biol 1995;172:126–38.

Differential gene expression in mouse craniofacial development 515

Author's personal copy

Boyden LM, Mao J, Belsky J, Mitzner L, Farhi A, MitnickMA, Wu D, Insogna K, Lifton RP. High bone density dueto a mutation in LDL-receptor-related protein 5. NEngl J Med 2002;346:1513–21.

Chimal-Monroy J, Montero JA, Ganan Y, Macias D, Garcia-Porrero JA, Hurle JM. Comparative analysis of theexpression and regulation of Wnt5a, Fz4, and Frzb1during digit formation and in micromass cultures. DevDyn 2002;224:314–20.

Church V, Nohno T, Linker C, Marcelle C, Francis-West P.Wnt regulation of chondrocyte differentiation. J CellSci 2002;115:4809–18.

Church VL, Francis-West P. Wnt signalling during limbdevelopment. Int J Dev Biol 2002;46:927–36.

Cole LK, Le R, Nunes I, F, Laufer E, Lewis J, Wu DK.Sensory organ generation in the chicken inner ear:contributions of bone morphogenetic protein 4,serrate1, and lunatic fringe. J Comp Neurol 2000;424:509–20.

Cordero D, Marcucio R, Hu D, Gaffield W, Tapadia M,Helms JA. Temporal perturbations in sonic hedgehogsignaling elicit the spectrum of holoprosencephalyphenotypes. J Clin Invest 2004;114:485–94.

Dealy CN, Roth A, Ferrari D, Brown AM, Kosher RA. Wnt-5a and Wnt-7a are expressed in the developing chicklimb bud in a manner suggesting roles in patternformation along the proximodistal and dorsoventralaxes. Mech Dev 1993;43:175–86.

Eberhart JK, Swartz ME, Crump JG, Kimmel CB. EarlyHedgehog signaling from neural to oral epitheliumorganizes anterior craniofacial development. Devel-opment 2006;133:1069–77.

Echevarria D, Vieira C, Gimeno L, Martinez S. Neuroe-pithelial secondary organizers and cell fate specifica-tion in the developing brain. Brain Res Brain Res Rev2003;43:179–91.

Epstein DJ, McMahon AP, Joyner AL. Regionalization ofSonic hedgehog transcription along the anteroposter-ior axis of the mouse central nervous system isregulated by Hnf3-dependent and -independent me-chanisms. Development 1999;126:281–92.

Erlebacher A, Filvaroff EH, Gitelman SE, Derynck R.Toward a molecular understanding of skeletal devel-opment. Cell 1995;80:371–8.

Ferguson MW, Honig LS, Slavkin HC. Differentiation ofcultured palatal shelves from alligator, chick, andmouse embryos. Anat Rec 1984;209:231–49.

Granjeiro JM, Oliveira RC, Bustos-Valenzuela JC, SogayarMC, Taga R. Bone morphogenetic proteins: fromstructure to clinical use. Braz J Med Biol Res 2005;38:1463–73.

Hall JM, Bell ML, Finger TE. Disruption of sonic hedgehogsignaling alters growth and patterning of lingual tastepapillae. Dev Biol 2003;255:263–77.

Hall JM, Hooper JE, Finger TE. Expression of sonichedgehog, patched, and Gli1 in developing tastepapillae of the mouse. J Comp Neurol 1999;406:143–55.

Haworth KE, Wilson JM, Grevellec A, Cobourne MT, HealyC, Helms JA, et al. Sonic hedgehog in the pharyngeal

endoderm controls arch pattern via regulation of Fgf8in head ectoderm. Dev Biol 2007;303:244–58.

Ingham PW, McMahon AP. Hedgehog signaling in animaldevelopment: paradigms and principles. Genes Dev2001;15:3059–87.

Iwatsuki K, Liu HX, Gronder A, Singer MA, Lane TF,Grosschedl R, et al. Wnt signaling interacts with Shh toregulate taste papilla development. Proc Natl Acad SciUSA 2007;104:2253–8.

Jeong J, Mao J, Tenzen T, Kottmann AH, McMahon AP.Hedgehog signaling in the neural crest cells regulatesthe patterning and growth of facial primordia. GenesDev 2004;18:937–51.

Jones CM, Lyons KM, Hogan BL. Involvement of bonemorphogenetic protein-4 (BMP-4) and Vgr-1 in mor-phogenesis and neurogenesis in the mouse. Develop-ment 1991;111:531–42.

Jung HS, Oropeza V, Thesleff I. Shh, Bmp-2, Bmp-4 andFgf-8 are associated with initiation and patterning ofmouse tongue papillae. Mech Dev 1999;81:179–82.

Kawakami Y, Wada N, Nishimatsu SI, Ishikawa T, Noji S,Nohno T. Involvement of Wnt-5a in chondrogenicpattern formation in the chick limb bud. Dev GrowthDiffer 1999;41:29–40.

Kim J, Ault KT, Chen HD, Xu RH, Roh DH, Lin MC, et al.Transcriptional regulation of BMP-4 in the Xenopusembryo: analysis of genomic BMP-4 and its promoter.Biochem Biophys Res Commun 1998;250:516–30.

Lazaro L, Dubourg C, Pasquier L, Le DF, Blayau M, DurouMR, et al. Phenotypic and molecular variability of theholoprosencephalic spectrum. Am J Med Genet A2004;129A:21–4.

Logan CY, Nusse R. The Wnt signaling pathway indevelopment and disease. Annu Rev Cell Dev Biol.2004;20:781–810.

Mistretta CM. The role of innervation in induction anddifferentiation of taste organs: introduction andbackground. Ann NY Acad Sci 1998;855:1–13.

Mistretta CM, Grigaliunas A, Liu HX. Development ofgustatory organs and innervating sensory ganglia.Chem Senses 2005;30(Suppl. 1):i52–3.

Mistretta CM, Liu HX, Gaffield W, MacCallum DK.Cyclopamine and jervine in embryonic rat tonguecultures demonstrate a role for Shh signaling intaste papilla development and patterning: fungiformpapillae double in number and form in novel locationsin dorsal lingual epithelium. Dev Biol 2003;254:1–18.

Miura H, Kusakabe Y, Sugiyama C, Kawamatsu M,Ninomiya Y, Motoyama J, et al. Shh and Ptc areassociated with taste bud maintenance in the adultmouse. Mech Dev 2001;106:143–5.

Moore-Scott BA, Manley NR. Differential expression ofSonic hedgehog along the anterior–posterior axisregulates patterning of pharyngeal pouch endodermand pharyngeal endoderm-derived organs. Dev Biol2005;278:323–35.

Nagase T, Nagase M, Osumi N, Fukuda S, Nakamura S,Ohsaki K, et al. Craniofacial anomalies of the culturedmouse embryo induced by inhibition of sonic hedgehog

K.B.S. Paiva et al.516

Author's personal copy

signaling: an animal model of holoprosencephaly.J Craniofac Surg 2005;16:80–8.

Nagata M, Ohashi Y, Ozawa H. A histochemical study ofthe development of premaxilla and maxilla duringsecondary palate formation in the mouse embryo.Arch Histol Cytol 1991;54:267–78.

Nunes FD, Valenzuela MG, Rodini CO, Massironi SM, KoGM. Localization of Bmp-4, Shh and Wnt-5a transcriptsduring early mice tooth development by in situhybridization. BrazOral Res 2007;21:127–33.

Paulson RB, Hayes TG, Sucheston ME. Scanning electronmicroscope study of tongue development in the CD-1mouse fetus. J Craniofac Genet Dev Biol 1985;5:59–73.

Ryu JH, Chun JS. Opposing roles of WNT-5A and WNT-11 ininterleukin-1beta regulation of type II collagen ex-pression in articular chondrocytes. J Biol Chem2006;281:22039–47.

Sasaki Y, O’Kane S, Dixon J, Dixon MJ, Ferguson MW.Temporal and spatial expression of Pax9 and Sonichedgehog during development of normal mousepalates and cleft palates in TGF-beta3 null embryos.Arch Oral Biol 2007;52:260–7.

Tajbakhsh S, Borello U, Vivarelli E, Kelly R, Papkoff J,Duprez D, et al. Differential activation of Myf5 andMyoD by different Wnts in explants of mouse paraxialmesoderm and the later activation of myogenesisin the absence of Myf5. Development 1998;125:4155–62.

Theiler K. In: The house mouse development and normalstages from fertilization to 4 weeks of age. New York:Springer; 1972 168p.

Thesleff I. Homeobox genes and growth factors inregulation of craniofacial and tooth morphogenesis.Acta Odontol Scand 1995;53:129–34.

Tufan AC, Tuan RS. Wnt regulation of limb mesenchymalchondrogenesis is accompanied by altered N-cadherin-related functions. FASEB J 2001;15:1436–8.

Urist MR. Bone: formation by autoinduction. Science1965;150:893–9.

Uusitalo M, Heikkila M, Vainio S. Molecular geneticstudies of Wnt signaling in the mouse. Exp Cell Res1999;253:336–48.

Vieira C, Garda AL, Shimamura K, Martinez S. Thalamicdevelopment induced by Shh in the chick embryo. DevBiol 2005;284:351–63.

Winnier G, Blessing M, Labosky PA, Hogan BL. Bonemorphogenetic protein-4 is required for mesodermformation and patterning in the mouse. Genes Dev1995;9:2105–16.

Yamagishi C, Yamagishi H, Maeda J, Tsuchihashi T, Ivey K,Hu T, et al. Sonic hedgehog is essential for firstpharyngeal arch development. Pediatr Res 2006;59:349–54.

Yang BL, Yang BB, Erwin M, Ang LC, Finkelstein J, Yee AJ.Versican G3 domain enhances cellular adhesion andproliferation of bovine intervertebral disc cells cul-tured in vitro. Life Sci 2003;73:3399–413.

Zhang XM, Ramalho-Santos M, McMahon AP. Smoothenedmutants reveal redundant roles for Shh and Ihhsignaling including regulation of L/R symmetry bythe mouse node. Cell 2001;106:781–92.

Zhang Z, Song Y, Zhao X, Zhang X, Fermin C, Chen Y.Rescue of cleft palate in Msx1-deficient mice bytransgenic Bmp4 reveals a network of BMP and Shhsignaling in the regulation of mammalian palatogen-esis. Development 2002;129:4135–46.

Zhou Y, Liu HX, Mistretta CM. Bone morphogenetic proteinsand noggin: inhibiting and inducing fungiform tastepapilla development. Dev Biol 2006;297:198–213.

Differential gene expression in mouse craniofacial development 517