Cleft lip with or without cleft palate(CLP) has an occurrence of 1 in 500 to2,500 live births worldwide, whichrepresents the most common craniofa-cial birth defect in humans (Vanderas,1987; Schutte and Murray, 1999; Gor-lin et al., 2001). Clinically, cleft lip is aunilateral or bilateral gap betweenthe philtrum and the lateral upper lip,often extending through the upper lipand jaw into the nostril and is some-
times accompanied by cleft of the sec-ondary palate—the roof of the oralcavity. Another common form of orofa-cial clefting is cleft palate (CP), whichappears as a gap in the secondary pal-ate while the upper lip appears intact.Epidemiological and embryologicalstudies suggest that CLP and CP havedistinct etiology, although these twophenotypes sometimes appear in thesame family (Fraser, 1970; Vanderas,1987; Gorlin et al., 2001). Both CLP
and CP have syndromic and nonsyn-dromic forms with the syndromicclefting often caused by single genemutations, chromosomal abnormali-ties, or teratogenic exposure (Gorlin etal., 2001; Wyszynski, 2002). Approxi-mately 70% of CLP cases are nonsyn-dromic for which the etiology andpathogenesis are complex and poorlyunderstood.
To understand the etiology of CLP,it is necessary to understand the de-
1Center for Oral Biology and Department of Biomedical Genetics, University of Rochester School of Medicine and Dentistry, Rochester, NewYork2Department of Orthodontics & Dows Institute for Dental Research, University of Iowa School of Dentistry, Iowa City, IowaGrant Sponsor: NIH; Grant numbers: DE013681; DE015207; DE016215; DE07202; DE015291; DE014667.*Correspondence to: Rulang Jiang, Ph.D., Center for Oral Biology, University of Rochester Medical Center, 601 ElmwoodAvenue, Box 611, Rochester, NY 14642. E-mail: [email protected]
DOI 10.1002/dvdy.20646Published online 16 November 2005 in Wiley InterScience (www.interscience.wiley.com).
velopmental processes leading to theformation of the intact upper lip, atboth the morphogenetic and molecu-lar levels. However, elucidating thecauses of CLP on even the morpholog-ical level has been hindered by a pau-city of understanding of the funda-mental processes of lip formation.Confusion exists in the literature withregard to the morphological processesleading to the formation of the intactupper lip. Whereas several studies de-scribe that the upper lip forms fromfusion between the maxillary and themedial nasal processes (e.g., Sun etal., 2000; Ashique et al., 2002; Sper-ber, 2002; Cox, 2004), others statethat a cleft lip results when the epi-thelia of the opposing medial and lat-eral nasal processes fail to make con-tact (Trasler, 1968; Gaare andLangman, 1977a; Gong and Guo,2003). The confusion may have arisendue in part to species differences (e.g.,chick vs. mouse and human) in facialmorphogenesis and in part to lack ofsynthesis of the fragmentary and of-ten incomplete information gainedfrom individual studies. Moreover,whereas it has been widely acceptedthat epithelial–mesenchymal trans-formation (EMT) of the epithelialseam is the major mechanism for bothlip and palate fusion (Fitchett andHay, 1989; Shuler et al., 1991, 1992;Griffith and Hay, 1992; Hay, 1995,2005; Sun et al., 2000; Cox, 2004;Nawshad et al., 2004), recent studieshave challenged this theory and dem-onstrated that the palatal epithelialseam gradually regresses by pro-grammed cell death rather than byEMT (Cuervo and Covarrubias, 2004;Vaziri Sani et al., 2005). At the molec-ular level, recent studies in chick andmice have identified specific roles forseveral major signaling pathways, in-cluding Bmp, Fgf, and Shh signalingpathways in midfacial morphogenesis(Hu and Helms, 1999; Trumpp et al.,1999; Ashique et al., 2002; Trokovic etal., 2003; Jeong et al., 2004; Liu et al.,2005b). In addition, genetic studies inhuman and mice have also identifiedtwo Wnt genes involved in CLP patho-genesis (Juriloff et al., 2004, 2005; Ni-emann et al., 2004; Carroll et al.,2005). These data provide new insightinto the molecular mechanisms un-derlying midfacial morphogenesis andCLP formation. This review will at-
tempt to clarify the morphogeneticprocesses leading to formation of theintact upper lip and discuss the newadvances in the understanding of thesignaling pathways regulating upperlip development.
MORPHOGENESIS OF THEUPPER LIP
In 1985, Klaus Hinrichsen publisheda detailed scanning electron micros-copy (SEM) study of a collection of var-ious stage human embryos, focusingon the morphology and pattern of thedeveloping face (Hinrichsen, 1985). Re-cently, Senders et al. (2003) presentedhigh resolution SEM pictures of devel-oping cynomolgus monkey embryonicfaces. Comparing these with other his-tological and SEM studies of facial de-velopment in mouse and chick (Trasler,1968; Gaare and Langman, 1977a,b;Yee and Abbott, 1978; Millicovsky andJohnston, 1981; Millicovsky et al., 1982;Trasler and Ohannessian, 1983; Cox,2004) provides an accurate understand-ing of the morphological processes in-volved in facial development.
Development of the human face be-gins in the fourth week of embryogen-esis (stage 10 according the Carnegiestaging system for human embryos,O’Rahilly, 1972), with migrating neu-ral crest cells that combine with thecore mesoderm and the epithelialcover to establish the facial primordia.The neural crest-derived facial mesen-chyme will give rise to the facial skel-eton, whereas mesoderm-derived cellswill form facial muscles (Noden, 1978,1983, 1988; Couly et al., 1992, 1993).At stage 11 (approximately 24 days ofgestation and corresponding to embry-onic day [E] 9.0 of mouse embryogen-esis), the primitive mouth, or stomo-deum, is bound rostrally by thedeveloping forebrain and caudally bythe swelling mandibular arches (thefirst pharyngeal arch), whereas struc-tures associated with the formation ofthe upper lip are not distinguishableyet at this stage (Yoon et al., 2000). Bystage 12 (approximately 26 days ofgestation, corresponding to E9.5 ofmouse embryogenesis), the facial pri-mordia consist of five separate promi-nences surrounding the stomodeum(Hinrichsen, 1985; Fig. 1A). At therostral side of the stomodeum is asymmetrical, unpaired frontonasal
prominence, which is fitted ventrolat-erally to the forebrain and populatedby mesenchymal cells derived fromthe fore- and mid-brain neural crest.The stomodeum is bound laterally bya pair of maxillary processes and cau-dally by the pair of mandibular pro-cesses, which are populated by neuralcrest cells originating from the firsttwo rhombomeres of the hindbrain.
From stage 13 to stage 15 (fourth tofifth week) of human embryogenesis,the frontonasal prominence widens asthe forebrain gives rise to the pairedtelencephalic vesicles (primordia of ce-rebral hemispheres), while the medialends of the mandibular processesgradually merge in a caudal to rostraldirection to form the mandible (lowerlip and jaw; Hinrichsen, 1985; Yoon etal., 2000). At stage 14 (approximately32 days of gestation and correspond-ing to E10.0 of mouse embryogenesis),thickening of surface ectoderm occursbilaterally on the ventrolateral part ofthe frontonasal prominence, givingrise to the nasal placodes. The fronto-nasal process grows and bulgesaround the nasal placodes, resultingin the formation of nasal pits and theswelling horseshoe-shaped lateral andmedial nasal processes (Hinrichsen,1985; Sperber, 2002). In adaptation tothe development of the telencephalicvesicles, the rostral end of the embryoforms a paired configuration with amedian groove extending in betweenthe paired medial nasal processes andinto the stomodeum. The nasal pitsare also in continuity with the stomo-deum at this stage (Hinrichsen, 1985).
By stage 15 (approximately 35 daysof gestation, corresponding to E10.5 ofmouse embryogenesis), rapid growthof the mesenchyme in the maxillaryprocesses have pushed the nasal pitsmedially, while the medial nasal pro-cesses have grown ventrolaterally,converting the nasal pits from rounddepressions into dorsally pointed slits(Fig. 1B). At this stage, the upper lipconsists of the maxillary processes lat-erally and the medial nasal processesmedially with the lateral nasal pro-cesses wedged in between the medialnasal and maxillary processes (Fig.1C). Fusion between the medial andlateral nasal processes has initiated,while maxillary processes lie belowthe lateral nasal processes (Fig. 1C).By stage 16 (approximately 38 days of
gestation in human, corresponding toE11.0 of mouse embryogenesis), rapidgrowth of the maxillary and medialnasal processes have pushed the lat-eral nasal processes further rostrallyin relative position and brought thedistal ends of maxillary and medialnasal processes into direct contact(Fig. 1D). Lateral view of the humanembryonic face at this stage gives theimpression that the maxillary pro-cesses are wedged in between the me-dial and lateral nasal processes (Fig.1D). High-resolution SEM micro-graphs of the cynomolgus monkey em-bryonic face at a similar stage alsoclearly demonstrated active fusion be-tween the lateral nasal and medialnasal processes as well as betweenmaxillary and medial nasal processes(Fig. 4 in Senders et al., 2003). Studiesin mouse embryos showed that fusionbetween the nasal processes occurredinitially at the posterior part of thenasal pits and proceeded in an ante-rior direction (Trasler, 1968; Gaareand Langman, 1977a), similar to whatHinrichsen described for human em-bryonic face development (Hinrichsen,1985).
Facial morphogenesis in chick isslightly different from that in mam-mals, because the medial nasal pro-
cess appears as a single entity some-times referred as the frontal orfrontonasal process, and the entireembryonic chick face appears in asquare configuration before lip fusion(Yee and Abbott, 1978; Young et al.,2000; Cox, 2004). Despite the differ-ences, close examination of SEM mi-crographs of the early fusion stagechick face showed that the initial con-tact and initiation of active cellularprocesses of fusion also begins be-tween the lateral and medial nasalprocesses (Fig. 2 in Cox, 2004).
Trasler (1968) emphasized the im-portance of fusion between medial andlateral nasal processes and postulatedthat lateral cleft lip results when thisfusion process does not occur. Ohba-yashi and Eto (1986) carried out a mi-crosurgical assay of relative contribu-tions of the different facial processesin facial morphogenesis in rat em-bryos and found that surgical removalof either a lateral nasal or a maxillaryprocess from one side of the face didnot prevent fusion of the other processwith the medial nasal process,whereas removal of the distal part of amedial nasal process resulted in cleftlip on the surgical side. These resultsindicate that contact and fusion be-tween maxillary and medial nasal
processes are not dependent on theprior fusion between the lateral andmedial nasal processes. Once upperlip morphogenesis is complete (de-scribed below), the lateral nasal pro-cesses form the sides (alae) of thenose, whereas the intact upper lip iscomposed of tissues derived from themedial nasal and maxillary processes.Although the lateral nasal processesdo not contribute to the final upper lip,the type of cleft lip in which the cleftextends into the nostril is clearly in-dicative of failure of fusion of the me-dial nasal processes with both maxil-lary and lateral nasal processesduring upper lip development.
Whereas the union between thefreely projected maxillary, lateral na-sal, and medial nasal processesclearly involves active epithelial fu-sion, closure of the median groove be-tween the paired medial nasal pro-cesses in mammals does not (Trasler,1968; Millicovsky and Johnston, 1981;Millicovsky et al., 1982; Trasler andOhannessian, 1983; Hinrichsen, 1985;Senders et al., 2003; Cox, 2004). Asthe epithelial fusion between maxil-lary, lateral nasal, and medial nasalprocesses continues from stage 16 tostage 18 (toward the beginning of theseventh week of gestation in human,
Fig. 1. Morphogenesis of the human upper lip. A: Scanning electron microscopy (SEM) facial view of a stage 13 human embryonic head. B: SEMmicrograph of the right nasal pit of a late stage 15 human embryo. C: Enlarged detail of the lower nasal pit shown in B. The boundary between themaxillary and lateral nasal processes is clearly marked by the rounded cells at the surface. Rounded cells also appear at the contact site between themedial and lateral nasal processes. D: Lateral view of a stage 17 human embryonic head. The maxillary process is puffed laterally and wedges betweenthe medial and lateral nasal processes. E: SEM micrograph of a stage 18 human embryonic head (facial view). F: Enlarged detail view of the left nostrilof the embryo shown in E. Arrowhead points to distinct epithelial bridges in the lower part of the slit-shaped nostril, which continue to fuse and reducethe nostril. All panels are from Hinrichsen (1985; original figure numbers 4, 15, 17, 27, 46, and 52, copyright of Springer-Verlag Berlin Heidelberg 1985),with kind permission of Springer Science and Business Media. fnp, frontonasal prominence; lnp, lateral nasal process; man, mandibular process; max,maxillary process; mnp, medial nasal process. Scale bars � 100 �m in B–D, 1 mm in E, 10 �m in F.
corresponding to E11.5 to E12.0 ofmouse embryogenesis), the maxillaryprocesses continue to grow rapidlyand push the nasal pits and medialnasal processes mediofrontally (Hin-richsen, 1985). The groove betweenthe medial nasal processes becomesgradually shallow and eventuallysmooth as a result of continuedgrowth and confluence of medial nasaland maxillary mesenchyme (Fig. 1E).These morphogenetic processes alsogradually convert the nasal pits tonose chambers and to nasal ducts asthe fusion between the medial and lat-eral nasal processes is completed. Thechoanal membranes at the dorsal endsof the nose chambers, however, arenot perforated until stage 18 to con-nect the nostrils to the posterior oralcavity. During the final stages of up-per lip formation, the nostrils are trans-formed to small slits and their loweredge remodeled by the fusion betweenthe medial nasal and maxillary pro-cesses (Hinrichsen, 1985; Fig. 1F).
By stage 19 (approximately 48 daysof gestation in human, correspondingto E12.5 of mouse embryogenesis), af-ter disintegration of the epithelialseams and mesenchymal confluencebetween medial nasal and maxillaryprocesses, formation of the upper lip iscomplete, with the intermaxillary seg-ment derived from the distal part ofthe medial nasal processes formingthe central lip. The medialization ofthe nose chambers and the filling ofthe median groove by mesenchymeare followed by outgrowth of the inter-maxillary segment into the oral cavityto form the anterior part of the palate(Hinrichsen, 1985). Some authors re-ferred to this anterior, intermaxillarypalate as the “primary palate,”whereas others used “primary palate”to describe the tissues formed by fu-sion between the maxillary and me-dial nasal processes (Diewert andWang, 1992; Wang et al., 1995; Sper-ber, 2002; Cobourne, 2004). The ante-rior palate derived from the intermax-illary process later fuses with thesecondary palate derived from themaxillary processes.
Development of the secondary pal-ate has been reviewed extensively(e.g., Ferguson, 1988; Murray andShutte, 2004; Nawshad et al., 2004).Because fusion between the secondarypalatal shelves, which arise bilater-
ally from the maxillary processes(Ferguson, 1988), and fusion betweenthe primary and secondary palates oc-cur much later in embryogenesis thanthe fusions between maxillary, lat-eral, and medial nasal processes dur-ing lip formation, failure of proper lipfusion often affects palatal contactsecondarily. Therefore, cleft lip is of-ten accompanied by cleft palate.
Normal lip fusion involves a seriesof remarkable cellular transforma-tions as the freely projected medialnasal, lateral nasal, and maxillaryprocesses are brought into proximityby proliferation of the neural crest-derived mesenchyme. In chick em-bryos, as the maxillary and medial na-sal processes near each other andprepare for fusion, the periderm cov-ering these processes undergo region-restricted apoptosis, resulting in theirsloughing off (Sun et al., 2000). SEManalysis of human embryos at the be-ginning of lip fusion (stage 16) showedmany rounded cells appearing to de-tach from the surface of the furrowbetween the maxillary and lateral na-sal processes as well as at the caudalend of the nasal pits where the medialand lateral nasal processes are in di-rect contact (Hinrichsen, 1985; Fig.1B,C). These rounded cells probablyrepresent dead cells extruded duringthe fusion between the maxillary andlateral nasal processes and betweenthe lateral and medial nasal pro-cesses. It has been hypothesized thatdeath of periderm cells promote epi-thelial adherence by exposing basallayers of the opposed epithelia andpermitting adherence junctions suchas desmosomes to form between them(Sun et al., 2000). The death of peri-derm cells before contact of the prefu-sion epithelia of facial processes hasalso been observed in hamster, mouse,and rat embryos and has been pro-posed to play an important role in sec-ondary palatal fusion (Lejour, 1970;Chaudhry and Shah, 1973; Hinrich-sen and Stevens, 1974; Gaare andLangman, 1977b; Fitchett and Hay,1989; Holtgrave et al., 2002).
As the free ends of the facial pro-cesses are brought into proximity, ep-ithelial filopodia in highly localizedprimary fusion areas begin to spanand establish bridges between thesefacial processes (Gaare and Langman,1977b; Millicovsky and Johnston,
1981; Millicovsky et al., 1982; Hin-richsen, 1985; Senders et al., 2003;Cox, 2004). These filopodia anchorinto the surface of the opposing prom-inences by penetrating between sur-face cells and are reinforced by theaccumulation of larger cellular exten-sions and adhering junctions (Milli-covsky and Johnston, 1981; Sun et al.,2000). Filopodial attachments aregreatly reduced in A/WySn and CL/Frmouse embryos, two strains with highfrequency of spontaneous CLP (Milli-covsky et al., 1982; Forbes et al.,1989). Similarly, filamentous projec-tions have been observed in chick em-bryos between the fusing facial prom-inences and are notably missing fromthe cleft primary palate chick mutantembryos (Yee and Abbott, 1978, Cox,2004). These observations, therefore,correlate the presence of filopodialprocesses spanning the prefusion pri-mordia with an ability to fuse.
Comparisons of embryonic faces ofcleft-predisposing and noncleft mousestrains indicated that facial geometryalso plays an important role in lip de-velopment (Trasler, 1968; Millicovskyet al., 1982). It was demonstrated thatembryos of both the A/J and CL/Frstrains, which have high frequency ofspontaneous cleft lip, have moreprominent and more medially conver-gent medial nasal processes thanthose of the C57BL/6 strain, whichhas a negligible spontaneous inci-dence of cleft lip (Millicovsky et al.,1982; Trasler and Ohannessian,1983). It was postulated that the spon-taneous cleft lip in the A/J and CL/Frstrains is a threshold character wherea slight change in the divergence ofthe medial and lateral nasal processesleads to their partial or complete lackof fusion. Thus, the fusion process re-quires temporal coordination of sur-face changes in the prefusion epitheliaand proper facial geometry for approx-imation of the facial prominences(Johnston and Millicovsky, 1985).
IS PROGRAMMED CELLDEATH, EMT, OR BOTHTHE MECHANISMINVOLVED IN LIP FUSION?
Fusion of the medial and lateral nasalprocesses generates an interveningepithelial seam known as the nasalfin, which is subsequently broken
down and replaced by continuous mes-enchyme between the processes(Trasler, 1968; Gaare and Langman,1977b). Similarly, fusion betweenmaxillary and medial nasal processesalso generates an epithelial seam thatis subsequently replaced by mesen-chymal tissue (Wang et al., 1995; Sunet al., 2000). The fate of the epithelialseam cells during lip fusion primarilyhas been analyzed using transmissionelectron microscopy (TEM) and li-pophilic dye cell labeling, whereas ter-minal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphatenick end-labeling (TUNEL) assay wasused to detect apoptotic cells. Gaareand Langman (1977b) investigatednasal fin regression during lip fusionin mouse embryos using TEM and re-ported that degenerating epithelialcells, characterized by an electron-dense nucleus and cytoplasm, were aprominent feature in the fusion of thenasal swellings. They showed that thenumber of degenerating cells in thecontacting epithelial linings was con-siderably higher than in the nonfusingepithelia and surrounding mesen-chyme and considered the fusing epi-thelia “cell-death zones.” However,they also reported that most of theepithelial cells appeared healthy butdid not mix with the mesenchyme atthe stage of nasal fin regression andsuggested that the surviving seamcells were probably incorporated intothe neighboring epithelial liningsrather than transformed into mesen-chyme (Gaare and Langman, 1977b).Sun et al. (2000) examined lip fusionin chick embryos using TEM andTUNEL assays and also found thatthe epithelial seam cells were healthylooking and very few were TUNEL-positive. They then used 5,6-carboxy-2,7-dichlorofluoresscein diacetate suc-cinimidyl ester, a lipophilic dye, tolabel the entire surface epithelia ofchick embryos before lip fusion andfound, after 24 hr, that there werelabeled mesenchyme-like cells in thefacial region after breakdown of thefusing epithelial seam between themedial nasal and maxillary processes.Thus, Sun et al. (2000) concluded thatthe epithelial seam cells transforminto mesenchyme during lip fusion.However, questions remain about thefate of the epithelial seam cells. Couldthe few labeled cells be due to dye
transfer into internal mesenchymalcells or to phagocytosis of dead labeledepithelial cells by macrophages? Evenif the seam cells indeed transdifferen-tiate into mesenchyme, do they con-tribute to mesenchyme-derived struc-tures later or do they die shortly afterEMT?
With regard to the fate of the fusingepithelial seam, whether apoptosis orEMT, it is believed that similar mech-anisms are involved in lip fusion andsecondary palate fusion (Gaare andLangman, 1977b; Sun et al., 2000;Cox, 2004). In mammals, the second-ary palate arises as bilateral palatalshelves that initially grow verticallyand later elevate to the horizontal po-sition above the tongue and fuse witheach other at the midline to form theroof of the oral cavity (Ferguson, 1988;Murray and Schutte, 2004). In con-trast to the few studies of the lip fu-sion process, the fate of the medialedge epithelial (MEE) cells of the sec-ondary palatal shelves, which formthe midline epithelial seam upon pal-atal shelf adhesion, has been studiedextensively although considerable dis-agreement still exists. TEM and cellbiological studies have provided clearevidence of apoptosis of at least a por-tion of the MEE cells (Glucksmann,1965; Saunders, 1966; DeAngelis andNalbandian, 1968; Smiley and Dixon,1968; Shapiro and Sweney, 1969; Smi-ley and Koch, 1975; Mori et al., 1994;Taniguchi et al., 1995; Cuervo et al.,2002; Cuervo and Covarrubias, 2004).Others, however, reported that themidline epithelial seam cells lookedhealthy at the TEM level and foundevidence of transdifferentiation ofMEE cells into mesenchymal cells byusing various cell labeling techniques(Fitchett and Hay, 1989; Shuler et al.,1991, 1992; Griffith and Hay, 1992;Sun et al., 1998; Martinez-Alvarez etal., 2000; Nawshad et al., 2004). Be-cause large numbers of apoptotic cellsin the fusing epithelial seam wereonly observed in palatal explant cul-tures (Martinez-Alvarez et al., 2000;Cuervo et al., 2002; Cuervo and Co-varrubias, 2004), Nawshad et al.(2004) and Hay (2005) suggested thatthe observed dying cells in the seamwere trapped dying periderm cells andargued in favor of EMT of palatalMEE cells. To answer definitivelywhether the MEE cells contribute to
the palatal mesenchyme in vivo,Vaziri Sani et al. (2005) used the Cre/loxP-mediated genetic labeling ap-proach to trace the MEE cells duringmouse palate development. In theirexperiments, mice carrying the shh-GFPCre or K14-Cre transgene werecrossed to mice carrying the loxP-STOP-loxP-lacZ cassette targeted intothe ROSA26 locus (R26R). The Ro-sa26 gene promoter normally drivesubiquitous gene expression (Zam-browicz et al., 1997). However, theloxP-flanked transcription STOP cas-sette prevents the lacZ gene from be-ing transcribed in the R26R mice(Soriano, 1999). Crossing the shh-GFPCre transgenic mice with theR26R mice results in the Cre recom-binase specifically removing the STOPcassette from 5� of the lacZ gene byexcising sequences in between theloxP sites in the double transgenicmice, which activates �-galactosidaseexpression permanently from the lacZgene at the ROSA26 locus in all cellsderived from ShhGFPCre-expressingcells. Because the ShhGFPCre fusiongene is expressed in the palatal epi-thelium but not in the palatal mesen-chyme, any �-galactosidase–express-ing palatal mesenchyme cell in theShhGFPCre;R26R double transgenicmice would have to be derived fromthe palatal epithelium during palatalfusion. Similarly, the K14-Cre trans-genic mice express Cre under the ker-atin-14 promoter, which is activatedin all epithelial cells after E11.75.Vaziri Sani et al. (2005) found well-labeled palatal epithelial cells, includ-ing palatal MEE cells before their de-velopmental disappearance from thepalatal midline, in both ShhGFPCre;R26R and K14-Cre;R26R embryos butnever saw any evidence of palatalmesenchymal cells displaying specific�-galactosidase activity even after to-tal disappearance of the �-galactosi-dase–positive midline epithelial seam.Furthermore, Vaziri Sani et al. (2005)reported that the regressing midlineepithelial seam cells and epithelial is-lands formed during palatal fusion ex-pressed activated Caspase-3, an earlymarker for apoptosis. These data indi-cate that MEE cells undergo pro-grammed cell death rather thantransdifferentiate into palatal mesen-chyme during palatal fusion in vivo.
programmed cell death as the majormechanism for palatal fusion, we an-alyzed programmed cell death inmouse embryos during fusion of themedial and lateral nasal processes byusing immunostaining for activatedCaspase-3. As shown in Figure 2, wefound that a lot of the epithelial seamcells between the fusing medial andlateral nasal processes express acti-vated Caspase-3, indicating thatmany epithelial seam cells are fated todegenerate by apoptosis. These datasuggest, like in secondary palatal fu-sion, that programmed cell deathplays an important role in lip fusion.
It is conceivable that some epithe-lial cells of the fusing seam may re-main viable and become incorporatedinto the facial epithelium as the facialmesenchyme rapidly expands. Epithe-lial seam cells in the secondary palatehave been observed to migrate alongthe midline to contribute to the oraland nasal epithelia of the fused palatein some species (Carette and Fergu-son, 1992). Further studies will benecessary to address whether any ep-ithelial cells transdifferentiate andcontribute to mesenchymal structuresof the face or what was called EMTduring lip fusion was just the cellularprocesses of shape changes, filopodialinteractions, and intercalation of theepithelial seam cells before they de-generate.
GENES AND MOLECULARPATHWAYS CRITICAL FORUPPER LIP DEVELOPMENT
It is clear that growth and morpho-genesis of the facial primordia have tobe exquisitely coordinated to developthe intact face. Because most of thecraniofacial mesenchyme is derivedfrom neural crest cells, genes and mo-lecular pathways regulating neuralcrest formation, migration, pattern-ing, proliferation, and apoptosis, areall important for craniofacial develop-ment. Various aspects of cranial neu-ral crest development and the roles ofneural crest in craniofacial develop-ment have been reviewed recently byothers (e.g., Wilkie and Morris-Kay,2001; Chambers and McGonnell,2002; Basch et al., 2004; Cox, 2004;Huang and Saint-Jeannet, 2004; Gra-ham et al., 2004; Kulesa et al., 2004;Marazita and Mooney, 2004; Helms et
al., 2005). We will focus on discussingthe genes and molecular pathwayscritical for upper lip morphogenesisafter the five facial prominences haveformed.
Whereas rapid proliferation of theneural crest derived mesenchyme isthe driving force of facial morphogen-esis, fate mapping and tissue recombi-nation experiments in chick showedthat proliferation and directed expan-sion of the facial mesenchyme dependon signals from the facial epithelia(Wedden, 1987; Richman and Tickle,1989; McGonnell et al., 1998). At thesame time, signals from the mesen-chyme also influence development ofthe facial ectoderm (reviewed in Fran-cis-West et al., 1998; Jernvall andThesleff, 2000). The reciprocal inter-actions involve many intercellular sig-naling pathways. We will discuss be-low the current understanding of themajor molecular pathways critical formidfacial growth and upper lip mor-phogenesis.
The Bmp Pathway
Bmps (bone morphogenetic proteins)are a group of secreted signaling mol-ecules of the transforming growth fac-tor beta (Tgf�) superfamily (Wozneyet al., 1988). This family of ligandsinitiates signaling by binding andbringing together two types of recep-tor serine/threonine kinases on thecell surface (reviewed in Shi and Mas-sague, 2003; Nohe et al., 2004). Uponligand binding, the type II receptorphosphorylates and activates the typeI receptor, which in turn phosphory-lates a set of transcriptional coactiva-tors called Smads and leads to theirnuclear translocation and transcrip-tional activation of downstream targetgenes. The Bmp signaling pathwayhas been shown to regulate diversedevelopmental processes, includingcell proliferation, differentiation, apo-ptosis, and tissue morphogenesis (re-viewed in Wan and Cao, 2005). Fran-cis-West et al. (1994) first showed thatBmp2 and Bmp4 mRNAs were ex-pressed in dynamic, spatiotemporallyregulated patterns in the developingchick facial primordia, with Bmp4having highly restricted expression inthe distal epithelia of the medial na-sal, lateral nasal, maxillary and man-dibular processes. Ectopic application
of Bmp2 or Bmp4 protein inducedovergrowth and changed the pattern-ing of the chick facial primordia (Bar-low and Francis-West, 1997). On theother hand, inhibiting Bmp signalingby application of Noggin, a specificBmp antagonist, in the chick facialprimordia caused reduced mesenchy-mal proliferation and outgrowth (Ash-ique et al., 2002; Wu et al., 2004).Moreover, recent expression and func-tional assays in fish and birds alsosuggested that Bmp signaling playsan important role in the evolution offacial shape and size (reviewed inHelms et al., 2005, and referencestherein).
Interestingly, expression patternsof Bmp2 and Bmp4 in the facial ecto-derm correlated with the largely over-lapping mesenchymal expression do-mains of the homeobox genes Msx1and Msx2 in the developing facial pri-mordia. Moreover, ectopic Bmp2 orBmp4 activated Msx1 and Msx2 geneexpression in the facial mesenchyme(Barlow and Francis-West, 1997), sug-gesting the Msx1 and Msx2 are down-stream transcription factors of theBmp pathway. Bmp4 is also expressedin the distal ectoderm of the facial pri-mordia surrounding the stomodeumbefore and during lip fusion in mouseembryos (Gong and Guo, 2003; Fig.3A), whereas Msx1 and Msx2 are ex-pressed in the adjacent facial mesen-chyme (Fig. 3B,C). Heterozygous lossof function of the MSX1 gene has beenassociated with CLP and tooth agene-sis in humans (van den Boogaard etal., 2000). Furthermore, missense mu-tations and variants in the MSX1gene have been associated with non-syndromic CLP (Lidral et al., 1998;Jezewski et al., 2003). Although micedeficient in Msx1 have cleft palate butnot CLP (Satokata and Maas, 1994),mice lacking both Msx1 and Msx2gene function exhibit bilateral CLP(Y. Chai, personal communication).Msx1 and Msx2 likely play criticalroles in facial mesenchymal prolifera-tion, as Msx1�/� mutant mice haveshortened maxilla and mandibles aswell as defects in palatal mesenchymeproliferation (Satokata and Maas,1994; Zhang et al., 2002). Introductionof a Bmp4 transgene under the controlof Msx1 promoter rescued the palatalgrowth defect in Msx1�/� mutantmice (Zhang et al., 2002). These data
MECHANISMS UNDERLYING CLEFT LIP 1157
indicate that Bmp4 and Msx1/Msx2function in a common molecular path-way essential for facial growth andupper lip morphogenesis.
Recently, Liu et al. (2005b) reportedthat tissue-specific inactivation of ei-ther Bmp4 or a Bmp type I receptor(Bmpr1a) gene in the facial primordiacaused cleft lip. Interestingly, inacti-vation of Bmpr1a caused elevated ap-optosis in both the prefusion epithe-lium and the distal medial nasalmesenchyme (Liu et al., 2005b),whereas inhibition of BMP signalingin the chick facial primordia with Nog-gin increased epithelial survival (Ash-ique et al., 2002). Another interestingfinding by Liu et al. (2005b) was thatmany of the mouse embryos with fa-
Fig. 2. Apoptosis plays an important role in breakdown of the epithelial seam during lip fusion.A: Frontal section of an embryonic day (E) 11.0 mouse embryo through the telencephalon and thefusing medial and lateral nasal processes. Red signal marks specific anti-active Caspase-3 anti-body staining. B: High-magnification view of the fusing epithelial seam between the medial andlateral nasal processes shown in A. Many of the fusing epithelial cells express active Caspase-3,while very few nasal mesenchyme cells and epithelial cells in other regions express activeCaspase-3, indicating specific programmed cell death of the fusing epithelial cells. lnp, lateral nasalprocess; mnp, medial nasal process.
Fig. 3. Selected gene expression patterns in the developing facial primordia of embryonic day (E) 10.5 mouse embryos. A: Whole-mount in situ hybridizationshowing specific expression of Bmp4 mRNA (blue/purple staining) in the distal ectoderm of the lateral nasal, medial nasal, maxillary, and mandibularprocesses. B,C: Msx1 (B) and Msx2 (C) mRNAs are expressed in overlapping patterns in the distal lateral nasal, medial nasal, maxillary, and mandibularmesenchyme. D: Fgf8 mRNA is expressed dynamically in the ectoderm around the nasal pits as well as in the proximal maxillary and mandibular ectoderm.E: Wnt3 mRNA is expressed in the maxillary and rostral mandibular ectoderm as well as in the distal medial nasal ectoderm. F: X-gal staining of an E10.5hemizygous TOPGAL transgenic mouse embryo showing �-galactosidase activity in the distal ectoderm of the lateral nasal, medial nasal, maxillary, andmandibular processes. lnp, lateral nasal process; man, mandibular process; max, maxillary process; mnp, medial nasal process.
1158 JIANG ET AL.
cial epithelial inactivation of Bmp4had delayed lip fusion, but the initialcleft lip was repaired by E14.5 in mostmutants, perhaps due to functionalcomplementation by or cross-regula-tion of other Bmp family genes. In ad-dition, Ashique et al. (2002) showedthat either inhibition or enhancementof BMP signaling in the facial primor-dia caused defective lip fusion. Thesedata indicate that Bmp signaling istightly regulated during upper lip de-velopment. Whereas defects in maxil-lary mesenchyme proliferation in theBmpr1a conditional mutants is con-sistent with a role for Bmp signalingin promoting facial primordial out-growth (Liu et al., 2005b), the role ofBmp signaling in facial ectoderm sur-vival and in the lip fusion processneeds to be further investigated.
The Fgf Pathway
Fgfs (fibroblast growth factors) andtheir cell surface receptors (Fgfr)make up a large and complex family ofsignaling molecules that play impor-tant roles in a variety of processes ofembryogenesis and tissue homeosta-sis (for recent reviews, see Itoh andOrnitz, 2004; Chen and Deng, 2005;Dailey et al., 2005; Eswarakumar etal., 2005). There are 22 Fgf genes inhumans and mice, several of whichare expressed in partially overlappingand dynamic patterns in the develop-ing mouse facial primordia (Francis-West et al., 1998; Colvin et al., 1999;Bachler and Neubuser, 2001). In par-ticular, Fgf8 is expressed broadly inthe frontonasal and mandibular epi-thelia before outgrowth of the nasalprocesses and its expression becomeshighly localized to around the nasalpits as well as in the maxillary andmandibular epithelia during active fa-cial primordial outgrowth (Bachlerand Neubuser, 2001; Fig. 3D). Studiesusing mandibular and nasal explantcultures showed that Fgf8 protein cansubstitute for the facial ectoderm tostimulate mesenchymal proliferationand maintain mesenchymal gene ex-pression (Neubuser et al., 1997; Firn-berg and Neubuser, 2002), suggestingthat Fgf signaling regulates facial pri-mordial outgrowth. Direct geneticanalysis of the roles of Fgf genes infacial morphogenesis, however, hasbeen complicated by early embryonic
lethality and functional redundancy(reviewed in Dailey et al., 2005). Nev-ertheless, analysis of mouse mutantscarrying hypomorphic alleles of Fgf8demonstrated that it is required forsurvival of the neural crest derivedfacial mesenchyme (Abbu-Issa et al.,2002; Frank et al., 2002). Moreover,tissue-specific inactivation of Fgf8 inthe mandibular epithelium showedthat it is required for mandibularmesenchymal survival as well asproximodistal patterning (Trumpp etal., 1999), whereas specific inactiva-tion of Fgf8 in the forebrain and facialectoderm led to severe facial defects,including midfacial cleft (Firnbergand Neubuser, 2002). In addition, de-spite broad overlapping expression ofFgfr1 and Fgfr2 in the developing fa-cial primordia, analysis of various mu-tations in these genes in mice havedemonstrated essential roles of Fgfsignaling in neural crest migration,survival, proliferation, and patterningof both the facial epithelia and mesen-chyme (Trokovic et al., 2003; Rice etal., 2004). These, together with therecent findings that nonsense muta-tions and deletions in the FGFR1 genein humans cause Kallmann syndrome,an autosomal dominant disorder char-acterized by infertility and anosomiabut in which 5% of patients have CLP(Dode et al., 2003; Kim et al., 2005),indicate that Fgf signaling plays es-sential roles in midfacial growth andupper lip development.
The Shh Pathway
Shh is a member of the Hedgehog fam-ily of secreted proteins and possessesremarkable morphogenetic patterningactivity (reviewed in Ingham and Mc-Mahon, 2001). It is involved in numer-ous key developmental events duringembryogenesis, including left–rightaxis establishment, dorsoventral pat-terning of the neural tube, endodermdevelopment, limb and craniofacialdevelopment, brain and pituitary de-velopment, among others (reviewed inIngham and McMahon, 2001; McMa-hon et al., 2003; Roessler and Muenke,2003, and references therein). TheShh signaling pathway is also in-volved in many human diseases, par-ticularly holoprosencephaly and can-cer (reviewed in Mullor et al., 2002;Roessler and Muenke, 2003). Shh sig-
nals to cells by binding to its cell sur-face receptor Patched1 (Ptch1) to re-lieve its inhibition of Smoothened(Smo), a seven-transmembrane pro-tein obligatory for the activation ofdownstream targets of the Shh path-way. Through a series of steps thatare currently not entirely understood,Smo activation leads to conversion ofmembers of the Gli family of tran-scription factors from repressors totranscriptional activators and to acti-vation of downstream gene expres-sion. One of the downstream targetgenes of Shh signaling is Ptch1, thusestablishing a feedback regulatoryloop (reviewed in Ingham and McMa-hon, 2001; McMahon et al., 2003).
During facial outgrowth, Shh is ex-pressed in the ectoderm of the facialprimordia (Echelard et al., 1993; Huand Helms, 1999; Jeong et al., 2004).Whereas a targeted null mutation inShh caused severe cranial deficienciesthat initially precluded direct assess-ment of the role of Shh in facial mor-phogenesis (Chiang et al., 1996), inhi-bition of Shh signaling in theoutgrowing chick frontonasal processwith a function blocking antibody in-hibited facial outgrowth and causedcleft lip (Hu and Helms, 1999). Ahl-gren and Bronner-Fraser (1999)showed that inhibition of Shh in thecranial mesenchyme also caused neu-ral crest mesenchymal cell death.Moreover, Ahlgren et al. (2002) dem-onstrated that application of Shh pro-tein rescued cranial mesenchymaldeath in chick embryos induced byethanol treatment. These data indi-cate that Shh signaling is required forfacial mesenchyme survival. In addi-tion, Hu and Helms (1999) demon-strated that Shh might also regulatefacial mesenchyme proliferation as ec-topic application of Shh protein to thefrontonasal process caused mediolat-eral expansion of that tissue. Tissuespecific inactivation of Smo in the cra-nial neural crest further confirms thatShh signaling is required for both sur-vival and proliferation of the facialmesenchyme (Jeong et al., 2004). Cra-nial neural crest cells lacking Smo mi-grated and formed facial primordianormally in mouse embryos but exhib-ited high levels of apoptosis from E9.5to E10.5 and reduced cell proliferationat E11.5, indicating that Shh expres-sion in the facial ectoderm specifically
MECHANISMS UNDERLYING CLEFT LIP 1159
supports cell survival during earlystages and promotes proliferation atlater stages to control the size of thefacial primordia (Jeong et al., 2004).Interestingly, whereas overactivationof Shh signaling by loss of the inhibi-tor Gli3 or constitutive activation ofSmo in the neural crest causes slightovergrowth of the facial primordia,some patients with mutations inPTCH1 have bilateral CLP (Hahn etal., 1996; Aoto et al., 2002; Jeong etal., 2004), suggesting that Shh signal-ing is regulated at multiple levels dur-ing facial morphogenesis.
The Wnt Pathway
The Wnt family of secreted glycopro-teins bind cell surface receptors of theFrizzled (Fzd) family and signalthrough several different intracellularsignal transduction pathways to regu-late diverse developmental processes,including cell proliferation, cell fatedetermination and differentiation,and cell survival (reviewed in Cadiganand Nusse, 1997; Wodarz and Nusse,1998; Eastman and Grosschedl, 1999;Huelsken and Birchmeier, 2001). Thebest characterized Wnt signalingpathway, termed the canonical Wntpathway, signals through �-catenin, adual functional protein involved in celladhesion and signaling (reviewed inBienz, 2005). In cells without Wnt sig-naling, cytoplasmic �-catenin is rap-idly degraded through the ubiquitin–proteosome pathway. In cellsresponding to canonical Wnt signal-ing, �-catenin is stabilized and entersthe nucleus to activate the Tcf/Leffamily transcription factors and regu-late transcription of downstreamgenes. Although several Wnt genes aswell as Tcf1 and Lef1 are known to beexpressed in the developing facial pri-mordia (Gavin et al., 1990; Oosterwe-gel et al., 1993; Parr et al., 1993;Christiansen et al., 1995; Wang andShackleford, 1996), a direct role forWnt signaling in facial morphogenesiswas not known until recently. Insearch for genes conferring suscepti-bility to spontaneous CLP in the Astrains of mice, Juriloff and colleaguesgenetically mapped an essentialcausal recessive mutation, clf1, to asmall region of mouse chromosome 11containing the closely linked Wnt3and Wnt9b genes (Juriloff and Mah,
1995; Juriloff et al., 1996, 2001). Re-cently, Niemann et al. (2004) reportedassociations of a nonsense mutation inthe WNT3 gene with tetra-amelia, arare recessive genetic disorder in hu-mans characterized by complete ab-sence of all four limbs and otheranomalies, including CLP. Carroll etal. (2005) reported that a targeted mu-tation in the Wnt9b gene in micecaused severe kidney developmentaldefects and an incomplete penetranceof CLP. Although the clf1 locus did notcontain any coding mutation in theWnt3 and Wnt9b genes, direct se-quence analysis showed that clf1 isassociated with a retrotransposon in-sertion at 6.6 kb downstream of theWnt9b gene (Juriloff et al., 2004,2005). These data indicate that bothWnt3 and Wnt9b play important rolesin midfacial morphogenesis.
To understand what roles Wnt3 andWnt9b may have during facial devel-opment, we analyzed their expressionpatterns during mouse embryogene-sis. We found that both Wnt3 andWnt9b mRNAs are expressed in theectoderm of the developing facial pri-mordia (Ryan et al., manuscript sub-mitted for publication; Fig. 3E). Fur-thermore, we found that canonicalWnt signaling is specifically activatedin the prefusion epithelia and in theunderlying mesenchyme in the medialnasal, lateral nasal, and maxillaryprocesses, as demonstrated by expres-sion of the specifically responsiveTOPGAL transgene (DasGupta andFuchs, 1999; Merrill et al., 2004; Ryanet al., manuscript submitted, Fig. 3F).These data, together with the CLPphenotype in WNT3�/� humans andWnt9b�/� mutant mice, suggest thatthe canonical Wnt signaling pathwaydirectly regulates facial mesenchymalgrowth and lip fusion. Of interest, thedomains of active canonical Wnt sig-naling in the developing facial primor-dia overlap significantly with the do-mains of Bmp4 gene expression (Gongand Guo, 2003; Liu et al., 2005b; com-pare Fig. 3A with F). Previously, it hasbeen demonstrated that Wnt/�-cate-nin signaling acts upstream of Bmp4expression during limb and lung de-velopment and that in cell transfec-tion assays Wnt/�-catenin signalingcan activate the mouse Bmp4 pro-moter directly through evolutionarilyconserved Tcf/Lef binding sites (Bar-
row et al., 2003; Soshnikova et al.,2003; Shu et al., 2005). Thus, it ispossible that Wnt signaling acts up-stream or interacts with the Bmp4pathway to regulate midfacial mor-phogenesis.
Other Genes and Pathways
Many other genes have been impli-cated in upper lip development. Over300 Mendelian syndromes in humansinclude CLP as part of the phenotype(Gorlin et al., 2001). Genes for severalof these have been identified, includ-ing PVRL1 in CLP-ectodermal dyspla-sia syndrome (CLPED1), P63 in dom-inant ectrodactyly with ectodermaldysplasia and CLP (EEC) and relatedsyndromes, and IRF6 in Van derWoude syndrome (recently reviewedin Cobourne, 2004; Cox, 2004; Ma-razita and Mooney, 2004). In addition,mutations in E-cadherin (CDH1) wererecently found in two families withhereditary diffuse gastric cancer asso-ciated with CLP (Frebourg et al.,2005) and mutations in EFNB1 incraniofrontonasal syndrome (CFNS;Twigg et al., 2004; Wieland et al.,2004). Interestingly, PVRL1, P63,IRF6, and CDH1 are all predomi-nantly expressed in epithelial tissues,indicating that proper epithelial dif-ferentiation, organization, or pattern-ing play important roles in lip devel-opment.
Whereas CLP is common in hu-mans, CLP is rare in mice, althoughmany mutant mouse strains exhibitCP. In addition to the A strains ofmice described above, mice homozy-gous for either of two spontaneousmutations, Dancer and Twirler, ex-hibit high penetrance of CLP (Lyon,1958; Deol and Lane, 1966, Gong etal., 2000). Whereas the Twirler generemains to be identified, Bush et al.(2004) recently positionally cloned theDancer mutation and showed that theCLP phenotype in the Dancer homozy-gous mutants results from widespreadmisexpression of the Tbx10 gene dueto insertion of a heterologous pro-moter. How Tbx10 misexpression dis-rupts the normal molecular and cellu-lar programs of facial morphogenesisremains to be determined.
Components of several other signal-ing pathways, including Tgf�/Egf,Pdgf, and retinoic acid pathways are
1160 JIANG ET AL.
expressed during craniofacial develop-ment and gene knockout studies inmice have confirmed the involvementof these pathways in upper lip mor-phogenesis (reviewed in Francis-Westet al., 1998, 2003). Mice lacking Egfrexhibit a low penetrance of CLP (Mi-ettinen et al., 1999), whereas theTGF� locus has been associated withnonsyndromic CLP in some humanpopulations (reviewed in Schutte andMurray, 1999; Cobourne, 2004). Micecarrying a null mutation in Pdgfr�and mice homozygous for mutations inboth the Pdgfr� and Pdgfc genes havea median cleft (Soriano, 1997; Ding etal., 2004). Pdgfr function is appar-ently autonomous to the neural crest,because conditional disruption ofPdgf� in neural-crest cells results in asimilar facial cleft (Tallquist et al.,2003). Mice harboring mutations inboth the retinoic acid receptor genesRAR� and RAR� also display a severemedian cleft and defects in other neu-ral crest-derived structures (Lohnes etal., 1994; Johnston and Bronsky,1995).
Many transcription factors of differ-ent classes are expressed in spatio-temporally regulated patterns in thedeveloping facial primordia (reviewedin Francis-West et al., 1998, 2003). Asubset of the Aristaless-like family ofhomeobox transcription factors appar-ently plays an important role in regu-lating morphogenesis of the frontona-sal processes (Meijlink, 1999; Qu etal., 1999; Beverdam et al., 2001). Al-though single mutations in any of theAlx3/Alx4/Cart1 genes do not displayorofacial clefting, Alx3�/�Alx4�/� orAlx4�/�Cart1�/� double mutantsdisplay median cleft lip and cleft pal-ate, indicating a degree of redundancyin this subfamily of transcription fac-tors (Qu et al., 1999; Beverdam et al.,2001). In the case of Alx3�/�Alx4�/�
double mutants, the median cleft phe-notype has been attributed to defectsin survival of the frontonasal mesen-chyme and failure of the medial nasalprocesses to merge properly (Bever-dam et al., 2001). The AP2� gene alsoplays an important role in midfacialmorphogenesis, because mice chi-meric for a null mutation in AP2�exhibited CLP (Nottoli et al., 1998).Further compound mutant and con-ditional gene inactivation studies willhelp elucidate how interactions of dif-
ferent transcription factors integratevarious signals from the facial ecto-derm to regulate facial primordial out-growth and upper lip morphogenesis.
SUMMARY ANDPERSPECTIVES
In summary, upper lip developmentinvolves a series of highly coordinated,genetically programmed morphoge-netic events that include directedgrowth and expansion of the facialprominences, programmed cell death,active fusion, and subsequent break-down of the epithelial seam betweenthe initially freely projected maxil-lary, medial nasal, and lateral nasalprocesses. Even subtle abnormalitiesin any one of these events may lead toa CLP phenotype. These developmen-tal weak points along with the signif-icant number of genes and signalingpathways involved in the morphoge-netic processes provide an explana-tion for the frequent occurrence andgenetic heterogeneity of CLP in hu-mans.
The complete sequencing of the hu-man genome brought development ofincreasingly high throughput geno-typing capabilities, which has led torapid identification of genes involvedin Mendelian syndromes as well ascandidate genes for complex geneticdiseases such as CLP (reviewed inLidral and Murray, 2004). At thesame time, more and more sophisti-cated approaches are being developedto efficiently analyze gene function inspecific developmental and cellularprocesses in animal model systems,which have significantly advanced ourunderstanding of genes and molecularpathways involved in craniofacial de-velopment. Whereas continued geneidentification will certainly improveour understanding of the molecularmechanisms of craniofacial develop-ment and malformations, the majorchallenges are (1) to understand thecomplex interactions between and in-tegration of various signaling path-ways, (2) to understand gene–envi-ronment interactions and epigeneticcontrol of craniofacial development,and (3) to understand the relationshipbetween genetic variation and suscep-tibility to craniofacial malformations.
There is clear genetic evidence thatthe major signaling pathways, includ-
ing Bmp, Fgf, Shh, and Wnt pathways,interact synergistically or antagonisti-cally during many developmental pro-cesses. The best characterized develop-mental system where these signalinginteractions occur extensively is the de-veloping limb (reviewed in Niswander,2002). Limb bud formation is initiatedby Wnt molecules (Wnt2b and Wnt8)expressed in the lateral plate meso-derm, which signal through �-cateninto restrict Fgf10 expression to the pre-sumptive limb mesoderm (Kawakamiet al., 2001). Fgf10 then induces expres-sion of another Wnt gene (Wnt3a inchick and Wnt3 in mice) in the limbectoderm, which in turn signalsthrough �-catenin and acts in conjunc-tion with Bmp signaling to induce andrestrict Fgf8 expression in the apicalectodermal ridge (AER; Kawakami etal., 2001; Barrow et al., 2003; Soshni-kova et al., 2003). The Wnt3/�-cateninsignaling in the limb ectoderm appearsto be regulated by Bmp signaling by anunidentified ligand but through theBmpr1a receptor (Soshnikova et al.,2003). Wnt3/�-catenin signaling also di-rectly regulates Bmp4 expression in thelimb ectoderm, generating a positivefeedback loop to pattern the proximal–distal axis of the limb (Barrow et al.,2003; Soshnikova et al., 2003). More-over, during limb outgrowth, Fgf signal-ing from the AER interacts with Wnt7asignaling from the dorsal ectoderm toinduce Shh expression in the posterior–distal limb mesenchyme (reviewed inNiswander, 2002). Shh induces expres-sion of Gremlin, an antagonist of Bmpsignaling, which in turn regulates Fgf4expression in the posterior AER, andFgf signaling from the AER maintainsShh expression in the posterior–distalmesenchyme, forming a signaling loop(reviewed in Niswander, 2002). Some ofthese signaling interactions havebeen found in other developmentalprocesses, including craniofacial de-velopment (Neubuser et al., 1997; St.Amand et al., 2000; Liu et al., 2005a;Shu et al., 2005). For example, Fgf8and Bmp4 are expressed in comple-mentary proximal–distal patterns inthe rostral mandibular ectoderm andBmp4 signaling appears to regulateFgf8 expression in a dose-dependentmanner (Liu et al., 2005a). Bmp4 andFgf10 have been shown to regulate ex-pression of Shh in the palatal ecto-derm, which, in turn, regulates Bmp2
MECHANISMS UNDERLYING CLEFT LIP 1161
expression in the palatal mesenchyme(Zhang et al., 2002; Rice et al., 2004;reviewed in Murray and Schutte,2004). As discussed above, the canon-ical Wnt signaling activity overlapswith Bmp4 expression in the distalectoderm of the facial primordia dur-ing facial outgrowth and lip fusion.Fgf8 is expressed dynamically in thefacial ectoderm and exhibits bothoverlapping and complementary do-mains with Bmp4 during facial out-growth. In addition to cross-regula-tion at the transcriptional level, thesesignaling pathways also converge andcrosstalk through interactions of theintracellular signaling components.Bmp4 and Fgf8 have been shown tointeract antagonistically to regulateexpression of downstream transcrip-tion factors involved in proximal–dis-tal patterning of the mandible andteeth (Neubuser et al, 1997; St.Amand et al., 2000). The Smad pro-teins in the Tgf�/Bmp signalingpathway have been found to directlyinteract with Tcf/Lef proteins, tran-scription factors of the Wnt/�-cateninpathway (Nishita et al., 2000). Fgf sig-naling has been shown to induce phos-phorylation of GSK3� and influencethe stability and nuclear entry of�-catenin in a cell-type dependentmanner (Torres et al., 1999; Israsenaet al., 2004). That the same major sig-naling pathways are involved in regu-lating cell proliferation and survivalin various developmental contexts topattern different tissues and organshighlights the complexity and impor-tance of understanding the interac-tions and integration of these signal-ing pathways at the molecular andcellular levels.
In both humans and mice, it isknown that environmental and epige-netic factors affect CLP susceptibility(reviewed in Murray, 2002; Finnell etal., 2002). Folate supplementation hasbeen shown to decrease the preva-lence of CLP in the A/WySn mousestrain (Angela Paros, 1999), and somestudies have shown a protective effectin humans as well (reviewed in Pres-cott and Malcolm, 2002). Presumablythese environmental factors act onboth the maternal and embryonic ge-notype; however, the molecular mech-anisms have not been discerned. Fur-thermore, genetic variation at someloci likely sensitizes the embryo to
other genetic and environmental in-sults. For example, modifications ofBmp4 expression or activity have beenimplicated in the evolution of facialshape in fish and birds (reviewed inHelms et al., 2005). Bmp4 is an essen-tial regulator of facial primordial out-growth and lip fusion, as discussedabove. Differences in facial shape,such as slight changes in the shape ofthe medial and lateral nasal processesduring facial development, has beenproposed to be a threshold factor un-derlying CLP in the A/WySn andCL/Fr strains of mice (Millicovsky etal., 1982; Trasler and Ohannessian,1983) and may account for the differ-ent frequencies of CLP in different hu-man populations (Fraser and Pa-shayan, 1970). Considering thecomplexity involving the interactionsand integration of signaling pathwaysand complex cellular processes in-volved in facial morphogenesis, ge-netic variation causing subtle changesof activity in one molecular pathwaymay tip the balance and result inhigher susceptibility to developmentalmalformations such as CLP. Thus, fa-cial morphogenesis is truly a quanti-tative genetic trait and an excellentmodel for understanding the molecu-lar mechanisms of organogenesis andcomplex diseases.
ACKNOWLEDGMENTSWe apologize to the many authorswhose excellent work was not citeddue to space limitations. We thank YuLan for extensive discussions and crit-ical reading of the manuscript. Wethank Yang Chai, Jeff Murray, andBrian Schutte for discussions andYang Chai for sharing data beforepublication. We thank Springer-Ver-lag GmbH for permission to reproducethe figure panels used in Figure 1.
REFERENCES
Abu-Issa R, Smyth G, Smoak I, YamamuraK, Meyers EN. 2002. Fgf8 is required forpharyngeal arch and cardiovascular de-velopment in the mouse. Development129:4613–4625.
Ahlgren SC, Bronner-Fraser M. 1999. In-hibition of sonic hedgehog signaling invivo results in craniofacial neural crestcell death. Curr Biol 9:1304–1314.
Ahlgren SC, Thakur V, Bronner-Fraser M.2002. Sonic hedgehog rescues cranialneural crest from cell death induced by
ethanol exposure. Proc Natl Acad SciU S A 99:10476–10481.
Aoto K, Nishimura T, Eto K, Motoyama J.2002. Mouse GLI3 regulates Fgf8 ex-pression and apoptosis in the developingneural tube, face, and limb bud. Dev Biol251:320–332.
Ashique AM, Fu K, Richman JM. 2002.Endogenous bone morphogenetic pro-teins regulate outgrowth and epithelialsurvival during avian lip fusion. Devel-opment 129:4647–4660.
Bachler M, Neubuser A. 2001. Expressionof members of the Fgf family and theirreceptors during midfacial development.Mech Dev 100:313–316.
Barlow AJ, Francis-West PH. 1997. Ec-topic application of recombinant BMP-2and BMP-4 can change patterning of de-veloping chick facial primordia. Develop-ment 124:391–398.
Barrow JR, Thomas KR, Boussadia-ZahuiO, Moore R, Kemler R, Capecchi MR,McMahon AP. 2003. Ectodermal Wnt3/beta -catenin signaling is required forthe establishment and maintenance ofthe apical ectodermal ridge. Genes Dev17:394–409.
Basch ML, Garcia-Castro MI, Bronner-Fraser M. 2004. Molecular mechanismsof neural crest induction. Birth DefectsRes C Embryo Today 72:109–123.
Beverdam A, Brouwer A, Reijnen M, Korv-ing J, Meijlink F. 2001. Severe nasalclefting and abnormal embryonic apopto-sis in Alx3/Alx4 double mutant mice. De-velopment 128:3975–3986.
Bienz M. 2005. beta-Catenin: a pivot be-tween cell adhesion and Wnt signalling.Curr Biol 15:R64–R67.
Bush JO, Lan Y, Jiang R. 2004. The cleftlip and palate defects in Dancer mutantmice result from gain of function of theTbx10 gene. Proc Natl Acad Sci U S A101:7022–7027.
Cadigan KM, Nusse R. 1997. Wnt signal-ing: a common theme in animal develop-ment. Genes Dev 11:3286–3305.
Carette MJ, Ferguson MW. 1992. The fateof medial edge epithelial cells during pal-atal shelf fusion in vitro: an analysis byDiI labeling and confocal microscopy. De-velopment 114:379–388.
Carroll TJ, Park JS, Hayashi S, MajumdarA, McMahon AP. 2005. Wnt9b plays acentral role in the regulation of mesen-chymal to epithelial transitions underly-ing organogenesis of the Mammalianurogenital system. Dev Cell 9:283–292.
Chambers D, McGonnell IM. 2002. Neuralcrest: facing the facts of head develop-ment. Trends Genet 18:381–384.
Chaudhry AP, Shah RM. 1973. Palatogen-esis in hamster. II. Ultrastructural ob-servations on the closure of palate. JMorphol 139:329–350.
Chen L, Deng CX. 2005. Roles of FGF sig-naling in skeletal development and hu-man genetic diseases. Front Biosci 10:1961–1976.
1162 JIANG ET AL.
Chiang C, Litingtung Y, Lee E, Young KE,Corden JL, Westphal H, Beachy PA.1996. Cyclopia and defective axial pat-terning in mice lacking Sonic hedgehoggene function. Nature 383:407–413.
Christiansen JH, Dennis CL, Wicking CA,Monkley SJ, Wilkinson DG, WainwrightBJ. 1995. Murine Wnt-11 and Wnt-12have temporally and spatially restrictedexpression patterns during embryonicdevelopment. Mech Dev 51:341–350.
Cobourne MT. 2004. The complex geneticsof cleft lip and palate. Eur J Orthod 26:7–16.
Colvin JS, Feldman B, Nadeau JH, Gold-farb M, Ornitz DM. 1999. Genomic orga-nization and embryonic expression of themouse fibroblast growth factor 9 gene.Dev Dyn 216:72–88.
Couly GF, Coltey PM, Le Douarin NM.1992. The developmental fate of the ce-phalic mesoderm in quail-chick chime-ras. Development 114:1–15.
Couly GF, Coltey PM, Le Douarin NM.1993. The triple origin of skull in highervertebrates: a study in quail-chick chi-meras. Development 117:409–429.
Cox TC. 2004. Taking it to the max: thegenetic and developmental mechanismscoordinating midfacial morphogenesisand dysmorphology. Clin Genet 65:163–176.
Cuervo R, Covarrubias L. 2004. Death isthe major fate of medial edge epithelialcells and the cause of basal lamina deg-radation during palatogenesis. Develop-ment 131:15–24.
Cuervo R, Valencia C, Chandraratna RA,Covarrubias L. 2002. Programmed celldeath is required for palate shelf fusionand is regulated by retinoic acid. DevBiol 245:145–156.
Dailey L, Ambrosetti D, Mansukhani A,Basilico C. 2005. Mechanisms underly-ing differential responses to FGF signal-ing. Cytokine Growth Factor Rev 16:233–247.
DasGupta R, Fuchs E. 1999. Multiple rolesfor activated LEF/TCF transcriptioncomplexes during hair follicle develop-ment and differentiation. Development126:4557–4568.
DeAngelis V, Nalbandian J. 1968. Ultra-structure of mouse and rat palatal pro-cesses prior to and during secondary pal-ate formation. Arch Oral Biol 13:601–608.
Deol MS, Lane PW. 1966. A new gene af-fecting the morphogenesis of the vestib-ular part of the inner ear in the mouse. JEmbryol Exp Morphol 16:543–558.
Diewert VM, Wang KY. 1992. Recent ad-vances in primary palate and midfacemorphogenesis research. Crit Rev OralBiol Med 4:111–130.
Ding H, Wu X, Bostrom H, Kim I, Wong N,Tsoi B, O’Rourke M, Koh GY, Soriano P,Betsholtz C, Hart TC, Marazita ML,Field LL, Tam PP, Nagy A. 2004. A spe-cific requirement for PDGF-C in palateformation and PDGFR-alpha signaling.Nat Genet 36:1111–1116.
Dode C, Levilliers J, Dupont JM, De PaepeA, Le Du N, Soussi-Yanicostas N, Coim-
bra RS, Delmaghani S, Compain-Nouaille S, Baverel F, Pecheux C, LeTessier D, Cruaud C, Delpech M, Spele-man F, Vermeulen S, Amalfitano A,Bachelot Y, Bouchard P, Cabrol S, CarelJC, Delemarre-van de Waal H, Goulet-Salmon B, Kottler ML, Richard O,Sanchez-Franco F, Saura R, Young J,Petit C, Hardelin JP. 2003. Loss-of-func-tion mutations in FGFR1 cause autoso-mal dominant Kallmann syndrome. NatGenet 33:463–465.
Eastman Q, Grosschedl R. 1999.Regula-tion of LEF-1/TCF transcription factorsby Wnt and other signals. Curr Opin CellBiol 11:233–240.
Echelard Y, Epstein DJ, St-Jacques B,Shen L, Mohler J, McMahon JA, McMa-hon AP. 1993. Sonic hedgehog, a memberof a family of putative signaling mole-cules, is implicated in the regulation ofCNS polarity. Cell 75:1417–1430.
Eswarakumar VP, Lax I, Schlessinger J.2005. Cellular signaling by fibroblastgrowth factor receptors. CytokineGrowth Factor Rev 16:139–149.
Firnberg N, Neubuser A. 2002. FGF signal-ing regulates expression of Tbx2, Erm,Pea3, and Pax3 in the early nasal region.Dev Biol 247:237–250.
Fitchett JE, Hay ED. 1989. Medial edgeepithelium transforms to mesenchymeafter embryonic palatal shelves fuse.Dev Biol 131:455–474.
Forbes DP, Steffek AJ, Klepacki M. 1989.Reduced epithelial surface activity is re-lated to a higher incidence of facial cleft-ing in A/WySn mice. J Craniofac GenetDev Biol 9:271–283.
Francis-West PH, Tatla T, Brickell PM.1994. Expression patterns of the bonemorphogenetic protein genes Bmp-4 andBmp-2 in the developing chick face sug-gest a role in outgrowth of the primordia.Dev Dyn 201:168–178.
Francis-West P, Ladher R, Barlow A,Graveson A. 1998. Signalling interac-tions during facial development. MechDev 75:3–28.
Francis-West PH, Robson L, Evans DJ.2003. Craniofacial development: the tis-sue and molecular interactions that con-trol development of the head. Adv AnatEmbryol Cell Biol 169:III–VI, 1–138.
Frank DU, Fotheringham LK, Brewer JA,Muglia LJ, Tristani-Firouzi M, CapecchiMR, Moon AM. 2002. An Fgf8 mouse mu-tant phenocopies human 22q11 deletionsyndrome. Development 129:4591–4603.
Fraser FC. 1970. The genetics of cleft lipand cleft palate. Am J Hum Genet 22:336–352.
Fraser FC, Pashayan H. 1970. Relation offace shape to susceptibility to congenitalcleft lip: a preliminary report. J MedGenet 7:112–117.
mans M, Hartmann A, Baert-DesurmontS, Alexandre C, Lejeune Dumoulin S,Marroni C, Martin C, Castedo S, LovettM, Winston J, Machado JC, Attie T, JabsEW, Cai J, Pellerin P, Triboulet JP,Scotte M, Le Pessot F, Hedouin A, Car-neiro F, Blayau M, Seruca R. 2005. Cleftlip/palate and CDH1/E-cadherin muta-tions in families with hereditary diffusegastric cancer. J Med Genet (in press[DOI: 10.1136/jmg.2005.03185]).
Gaare JD, Langman J. 1977a. Fusion ofnasal swellings in the mouse embryo:surface coat and initial contact. Am JAnat 150:461–475.
Gaare JD, Langman J. 1977b. Fusion ofnasal swellings in the mouse embryo: re-gression of the nasal fin. Am J Anat 150:477–499.
Gavin BJ, McMahon JA, McMahon AP.1990. Expression of multiple novel Wnt-1/int-1-related genes during fetal andadult mouse development. Genes Dev 4:2319–2332.
Glucksmann A. 1965. Cell death in normaldevelopment. Arch Biol (Liege) 76:419–437.
Gong SG, Guo C. 2003. Bmp4 gene is ex-pressed at the putative site of fusion inthe midfacial region. Differentiation 71:228–236.
Gong SG, White NJ, Sakasegawa AY.2000. The Twirler mouse, a model for thestudy of cleft lip and palate. Arch OralBiol 45:87–94.
Gorlin RJ, Cohen MM Jr, Hennekam RCM.2001. Syndromes of the head and neck.New York: Oxford University Press.
Graham A, Begbie J, McGonnell I. 2004.Significance of the cranial neural crest.Dev Dyn 229:5–13.
Griffith CM, Hay ED. 1992. Epithelial-mesenchymal transformation duringpalatal fusion: carboxyfluorescein tracescells at light and electron microscopiclevels. Development 116:1087–1099.
Hahn H, Wicking C, Zaphiropoulous PG,Gailani MR, Shanley S, ChidambaramA, Vorechovsky I, Holmberg E, UndenAB, Gillies S, Negus K, Smyth I, Press-man C, Leffell DJ, Gerrard B, GoldsteinAM, Dean M, Toftgard R, Chenevix-Trench G, Wainwright B, Bale AE. 1996.Mutations of the human homolog of Dro-sophila patched in the nevoid basal cellcarcinoma syndrome. Cell 85:841–851.
Hay ED. 1995. An overview of epithelio-mesenchymal transformation. Acta Anat(Basel) 154:8–20.
Hay ED. 2005. The mesenchymal cell, itsrole in the embryo, and the remarkablesignaling mechanisms that create it. DevDyn 233:706–720.
Helms JA, Cordero D, Tapadia MD. 2005.New insights into craniofacial morpho-genesis. Development 132:851–861.
Hinrichsen K. 1985. The early develop-ment of morphology and patterns of theface in the human embryo. Adv AnatEmbryol Cell Biol 98:1–79.
Hinrichsen CF, Stevens GS. 1974. Epithe-lial morphology during closure of the sec-ondary palate in the rat. Arch Oral Biol19:969–980.
MECHANISMS UNDERLYING CLEFT LIP 1163
Holtgrave EA, Stoltenburg-Didinger G.2002. Apoptotic epithelial cell death: aprerequisite for palatal fusion. An in vivostudy in rabbits. J CraniomaxillofacSurg 30:329–336.
Hu D, Helms JA. 1999. The role of sonichedgehog in normal and abnormalcraniofacial morphogenesis. Develop-ment 126:4873–4884.
Huang X, Saint-Jeannet JP. 2004. Induc-tion of the neural crest and the opportu-nities of life on the edge. Dev Biol 275:1–11.
Huelsken J, Birchmeier W. 2001. New as-pects of Wnt signaling pathways inhigher vertebrates. Curr Opin GenetDev 11:547–553.
Ingham PW, McMahon AP. 2001. Hedge-hog signaling in animal development:paradigms and principles. Genes Dev 15:3059–3087.
Israsena N, Hu M, Fu W, Kan L, KesslerJA. 2004. The presence of FGF2 signal-ing determines whether beta-catenin ex-erts effects on proliferation or neuronaldifferentiation of neural stem cells. DevBiol 268:220–231.
Itoh N, Ornitz DM. 2004. Evolution of theFgf and Fgfr gene families. Trends Genet20:563–569.
Jeong J, Mao J, Tenzen T, Kottmann AH,McMahon AP. 2004. Hedgehog signalingin the neural crest cells regulates thepatterning and growth of facial primor-dia. Genes Dev 18:937–951.
Jernvall J, Thesleff I. 2000. Reiterative sig-naling and patterning during mamma-lian tooth morphogenesis. Mech Dev 92:19–29.
Jezewski PA, Vieira AR, Nishimura C,Ludwig B, Johnson M, O’Brien SE,Daack-Hirsch S, Schultz RE, Weber A,Nepomucena B, Romitti PA, ChristensenK, Orioli IM, Castilla EE, Machida J,Natsume N, Murray JC. 2003. Completesequencing shows a role for MSX1 innon-syndromic cleft lip and palate.J Med Genet 40:399–407.
Johnston MC, Bronsky PT. 1995. Prenatalcraniofacial development: new insightson normal and abnormal mechanisms.Crit Rev Oral Biol Med 6:368–422.
Johnston MC, Millicovsky G. 1985. Normaland abnormal development of the lip andpalate. Clin Plast Surg 12:521–532.
Juriloff DM, Mah DG. 1995. The major lo-cus for multifactorial nonsyndromic cleftlip maps to mouse chromosome 11.Mamm Genome 6:63–69.
Juriloff DM, Harris MJ, Mah DG. 1996.The clf1 gene maps to a 2- to 3-cM regionof distal mouse chromosome 11. MammGenome 7:789.
Juriloff DM, Harris MJ, Brown CJ. 2001.Unravelling the complex genetics of cleftlip in the mouse model. Mamm Genome12:426–435.
Juriloff DM, Harris MJ, Dewell SL. 2004.A digenic cause of cleft lip in A-strainmice and definition of candidate genesfor the two loci. Birth Defects Res A ClinMol Teratol 70:509–518.
2005. Investigations of the genomic re-gion that contains the clf1 mutation, acausal gene in multifactorial cleft lip andpalate in mice. Birth Defects Res A ClinMol Teratol 73:103–113.
Kawakami Y, Capdevila J, Buscher D, ItohT, Rodriguez Esteban C, Izpisua Bel-monte JC. 2001. WNT signals controlFGF-dependent limb initiation and AERinduction in the chick embryo. Cell 104:891–900.
Kim HG, Herrick SR, Lemyre E, Kish-ikawa S, Salisz JA, Seminara S, Mac-Donald ME, Bruns GA, Morton CC,Quade BJ, Gusella JF. 2005. Hypogona-dotropic hypogonadism and cleft lip andpalate caused by a balanced transloca-tion producing haploinsufficiency forFGFR1. J Med Genet 42:666–672.
Kulesa P, Ellies DL, Trainor PA. 2004.Comparative analysis of neural crest celldeath, migration, and function duringvertebrate embryogenesis. Dev Dyn 229:14–29.
Lejour M. 1970. Cleft lip induced in the rat.Cleft Palate J 7:169–186.
Lidral AC, Murray JC. 2004. Genetic ap-proaches to identify disease genes forbirth defects with cleft lip/palate as amodel. Birth Defects Res A Clin MolTeratol 70:893–901.
Lidral AC, Romitti PA, Basart AM,Doetschman T, Leysens NJ, Daack-Hir-sch S, Semina EV, Johnson LR, MachidaJ, Burds A, Parnell TJ, Rubenstein JL,Murray JC. 1998. Association of MSX1and TGFB3 with nonsyndromic cleftingin humans. Am J Hum Genet 63:557–568.
Liu W, Selever J, Murali D, Sun X, BruggerSM, Ma L, Schwartz RJ, Maxson R, Fu-ruta Y, Martin JF. 2005a. Threshold-specific requirements for Bmp4 in man-dibular development. Dev Biol 283:282–293.
Liu W, Sun X, Braut A, Mishina Y, Be-hringer RR, Mina M, Martin JF. 2005b.Distinct functions for Bmp signaling inlip and palate fusion in mice. Develop-ment 132:1453–1461.
Lohnes D, Mark M, Mendelsohn C, Dolle P,Dierich A, Gorry P, Gansmuller A,Chambon P. 1994. Function of the reti-noic acid receptors (RARs) during devel-opment (I). Craniofacial and skeletal ab-normalities in RAR double mutants.Development 120:2723–2748.
Lyon MF. 1958. Twirler: a mutant affect-ing the inner ear of the house mouse. JEmbryol Exp Morphol 6:105–116.
Marazita ML, Mooney MP. 2004. Currentconcepts in the embryology and geneticsof cleft lip and cleft palate. Clin PlastSurg 31:125–140.
McGonnell IM, Clarke JD, Tickle C. 1998.Fate map of the developing chick face:analysis of expansion of facial primordiaand establishment of the primary palate.Dev Dyn 212:102–118.
McMahon AP, Ingham PW, Tabin CJ.2003. Developmental roles and clinicalsignificance of hedgehog signaling. CurrTop Dev Biol 53:1–114.
Meijlink F, Beverdam A, Brouwer A, Oost-erveen TC, Berge DT. 1999. Vertebratearistaless-related genes. Int J Dev Biol43:651–663.
Merrill BJ, Pasolli HA, Polak L, Rendl M,Garcia-Garcia MJ, Anderson KV, FuchsE. 2004. Tcf3: a transcriptional regulatorof axis induction in the early embryo.Development 131:263–274.
Miettinen PJ, Chin JR, Shum L, SlavkinHC, Shuler CF, Derynck R, Werb Z.1999. Epidermal growth factor receptorfunction is necessary for normal cranio-facial development and palate closure.Nat Genet 22:69–73.
Millicovsky G, Johnston MC. 1981. Activerole of embryonic facial epithelium: newevidence of cellular events in morpho-genesis. J Embryol Exp Morphol 63:53–66.
Millicovsky G, Ambrose LJ, Johnston MC.1982. Developmental alterations associ-ated with spontaneous cleft lip and pal-ate in CL/Fr mice. Am J Anat 164:29–44.
Mori C, Nakamura N, Okamoto Y, OsawaM, Shiota K. 1994. Cytochemical identi-fication of programmed cell death in thefusing fetal mouse palate by specific la-belling of DNA fragmentation. Anat Em-bryol (Berl) 190:21–28.
Mullor JL, Sanchez P, Altaba AR. 2002.Pathways and consequences: Hedgehogsignaling in human disease. Trends CellBiol 12:562–569.
Nawshad A, LaGamba D, Hay ED. 2004.Transforming growth factor beta (TGF-beta) signalling in palatal growth, apo-ptosis and epithelial mesenchymaltransformation (EMT). Arch Oral Biol 49:675–689.
Neubuser A, Peters H, Balling R, MartinGR. 1997. Antagonistic interactions be-tween FGF and BMP signaling path-ways: a mechanism for positioning thesites of tooth formation. Cell 90:247–255.
Niemann S, Zhao C, Pascu F, Stahl U,Aulepp U, Niswander L, Weber JL, Mul-ler U. 2004. Homozygous WNT3 muta-tion causes tetra-amelia in a large con-sanguineous family. Am J Hum Genet74:558–563.
Nishita M, Hashimoto MK, Ogata S, Lau-rent MN, Ueno N, Shibuya H, Cho KW.2000. Interaction between Wnt andTGF-beta signalling pathways duringformation of Spemann’s organizer. Na-ture 403:781–785.
Niswander L. 2002. Interplay between themolecular signals that control vertebratedevelopment. Int J Dev Biol 46:877–881.
Noden DM. 1978. The control of avian ce-phalic neural crest cytodifferentiation. I.Skeletal and connective tissues. Dev Biol67:296–312.
1164 JIANG ET AL.
Noden DM. 1983. The embryonic origins ofavian cephalic and cervical muscles andassociated connective tissues. Am J Anat168:257–276.
Nohe A, Keating E, Knaus P, Petersen NO.2004. Signal transduction of bone mor-phogenetic protein receptors. Cell Signal16:291–299.
Nottoli T, Hagopian-Donaldson S, Zhang J,Perkins A, Williams T. 1998. AP-2-nullcells disrupt morphogenesis of the eye,face, and limbs in chimeric mice. ProcNatl Acad Sci U S A 95:13714–13719.
Ohbayashi N, Eto K. 1986. Relative contri-butions of the facial processes to facialdevelopment: a microsurgical assay. JCraniofac Genet Dev Biol Suppl 2:41–44.
Oosterwegel M, van de Wetering M, Tim-merman J, Kruisbeek A, Destree O, Mei-jlink F, Clevers H. 1993. Differential ex-pression of the HMG box factors TCF-1and LEF-1 during murine embryogene-sis. Development 118:439–448.
O’Rahilly R. 1972. Guide to the staging ofhuman embryos. Anat Anz 130:556–559.
Parr BA, Shea MJ, Vassileva G, McMahonAP. 1993. Mouse Wnt genes exhibit dis-crete domains of expression in the earlyembryonic CNS and limb buds. Develop-ment 119:247–261.
Prescott NJ, Malcolm S. 2002. Folate andthe face: evaluating the evidence for theinfluence of folate genes on craniofacialdevelopment. Cleft Palate Craniofac J 39:327–331.
Qu S, Tucker SC, Zhao Q, deCrombruggheB, Wisdom R. 1999. Physical and geneticinteractions between Alx4 and Cart1.Development 126:359–369.
Richman JM, Tickle C. 1989. Epithelia areinterchangeable between facial primor-dia of chick embryos and morphogenesisis controlled by the mesenchyme. DevBiol 136:201–210.
Roessler E, Muenke M. 2003. How aHedgehog might see holoprosencephaly.Hum Mol Genet 12 Spec No 1:R15–25.
Satokata I, Maas R. 1994. Msx1 deficientmice exhibit cleft palate and abnormali-ties of craniofacial and tooth develop-ment. Nat Genet 6:348–356.
Saunders JW Jr. 1966. Death in embryonicsystems. Science 154:604–612.
Schutte BC, Murray JC. 1999. The manyfaces and factors of orofacial clefts. HumMol Genet 8:1853–1859.
Senders CW, Peterson EC, Hendrickx AG,Cukierski MA. 2003. Development of theupper lip. Arch Facial Plast Surg 5:16–25.
Shapiro BL, Sweney L. 1969. Electron mi-croscopic and histochemical examinationof oral epithelial-mesenchymal interac-tion (programmed cell death). J Dent Res48:652–660.
Shi Y, Massague J. 2003. Mechanisms ofTGF-beta signaling from cell membraneto the nucleus. Cell 113:685–700.
Shu W, Guttentag S, Wang Z, Andl T, Bal-lard P, Lu MM, Piccolo S, Birchmeier W,Whitsett JA, Millar SE, Morrisey EE.2005. Wnt/b-catenin signaling acts up-stream of N-myc, BMP4, and FGF sig-naling to regulate proximal-distal pat-terning in the lung. Dev Biol 283:226–239.
Shuler CF, Guo Y, Majumder A, Luo RY.1991. Molecular and morphologicchanges during the epithelial-mesenchy-mal transformation of palatal shelf me-dial edge epithelium in vitro. Int J DevBiol 35:463–472.
Shuler CF, Halpern DE, Guo Y, Sank AC.1992. Medial edge epithelium fate tracedby cell lineage analysis during epithelial-mesenchymal transformation in vivo.Dev Biol 154:318–330.
Smiley GR, Dixon AD. 1968. Fine structureof midline epithelium in the developingpalate of the mouse. Anat Rec 161:293–310.
Smiley GR, Koch WE. 1975. A comparisonof secondary palate development withdifferent in vitro techniques. Anat Rec181:711–723.
Soriano P. 1997. The PDGF alpha receptoris required for neural crest cell develop-ment and for normal patterning of thesomites. Development 124:2691–2700.
Soriano P. 1999. Generalized lacZ expres-sion with the ROSA26 Cre reporterstrain. Nat Genet 21:70–71.
Soshnikova N, Zechner D, Huelsken J,Mishina Y, Behringer RR, Taketo MM,Crenshaw EB III, Birchmeier W. 2003.Genetic interaction between Wnt/beta-catenin and BMP receptor signaling dur-ing formation of the AER and the dorsal-ventral axis in the limb. Genes Dev. 17:1963–1968.
Sperber GH. 2002. Formation of the pri-mary palate. In: Wyszynski DF, editor.Cleft lip and palate: from origin to treat-ment. New York: Oxford UniversityPress. p 5–13.
St Amand TR, Zhang Y, Semina EV, ZhaoX, Hu Y, Nguyen L, Murray JC, Chen Y.2000. Antagonistic signals betweenBMP4 and FGF8 define the expression ofPitx1 and Pitx2 in mouse tooth-forminganlage. Dev Biol 217:323–332.
Sun D, Vanderburg CR, Odierna GS, HayED. 1998. TGFbeta3 promotes transfor-mation of chicken palate medial edge ep-ithelium to mesenchyme in vitro. Devel-opment 125:95–105.
Sun D, Baur S, Hay ED. 2000. Epithelial-mesenchymal transformation is themechanism for fusion of the craniofacialprimordia involved in morphogenesis ofthe chicken lip. Dev Biol 228:337–349.
Tallquist MD, French WJ, Soriano P. 2003.Additive effects of PDGF receptor betasignaling pathways in vascular smoothmuscle cell development. PLoS Biol1:E52.
Taniguchi K, Sato N, Uchiyama Y. 1995.Apoptosis and heterophagy of medialedge epithelial cells of the secondary pal-
atine shelves during fusion. Arch HistolCytol 58:191–203.
Torres MA, Eldar-Finkelman H, Krebs EG,Moon RT. 1999. Regulation of ribosomalS6 protein kinase-p90(rsk), glycogensynthase kinase 3, and beta-catenin inearly Xenopus development. Mol CellBiol 19:1427–1437.
Trasler DG. 1968. Pathogenesis of cleft lipand its relation to embryonic face shapein A-J and C57BL mice. Teratology 1:33–49.
Trasler DG, Ohannessian L. 1983. Ultra-structure of initial nasal process cell fu-sion in spontaneous and 6-aminonicotin-amide-induced mouse embryo cleft lip.Teratology 28:91–101.
Trokovic N, Trokovic R, Mai P, Partanen J.2003. Fgfr1 regulates patterning of thepharyngeal region. Genes Dev 17:141–153.
Trumpp A, Depew MJ, Rubenstein JL,Bishop JM, Martin GR. 1999. Cre-medi-ated gene inactivation demonstratesthat FGF8 is required for cell survivaland patterning of the first branchialarch. Genes Dev 13:3136–3148.
Twigg SR, Kan R, Babbs C, Bochukova EG,Robertson SP, Wall SA, Morriss-KayGM, Wilkie AO. 2004. Mutations of eph-rin-B1 (EFNB1), a marker of tissueboundary formation, cause craniofronto-nasal syndrome. Proc Natl Acad Sci U SA 101:8652–8657.
van den Boogaard MJ, Dorland M, BeemerFA, van Amstel HK. 2000. MSX1 muta-tion is associated with orofacial cleftingand tooth agenesis in humans. NatGenet 24:342–343.
Vanderas AP. 1987. Incidence of cleft lip,cleft palate, and cleft lip and palateamong races: a review. Cleft Palate J24:216–225.
Vaziri Sani F, Hallberg K, Harfe BD, Mc-Mahon AP, Linde A, Gritli-Linde A.2005. Fate-mapping of the epithelialseam during palatal fusion rules out ep-ithelial-mesenchymal transformation.Dev Biol 285:490–495.
Wan M, Cao X. 2005. BMP signaling inskeletal development. Biochem BiophysRes Commun 328:651–657.
Wang J, Shackleford GM. 1996. MurineWnt10a and Wnt10b: cloning and expres-sion in developing limbs, face and skin ofembryosandinadults.Oncogene13:1537–1544.
Wang KY, Juriloff DM, Diewert VM. 1995.Deficient and delayed primary palatalfusion and mesenchymal bridge forma-tion in cleft lip-liable strains of mice. JCraniofac Genet Dev Biol 15:99–116.
Wedden SE. 1987. Epithelial-mesenchy-mal interactions in the development ofchick facial primordia and the target ofretinoid action. Development 99:341–351.
Wieland I, Jakubiczka S, Muschke P, Co-hen M, Thiele H, Gerlach KL, AdamsRH, Wieacker P. 2004. Mutations of theephrin-B1 gene cause craniofrontonasalsyndrome. Am J Hum Genet 74:1209–1215.
MECHANISMS UNDERLYING CLEFT LIP 1165
Wilkie AO, Morriss-Kay GM. 2001. Genet-ics of craniofacial development and mal-formation. Nat Rev Genet 2:458–468.
Wodarz A, Nusse R. 1998. Mechanisms ofWnt signaling in development. Annu RevCell Dev Biol 14:59–88.
Wu P, Jiang TX, Suksaweang S, WidelitzRB, Chuong CM. 2004. Molecular shap-ing of the beak. Science 305:1465–1466.
Wyszynski DF. 2002. Cleft lip and palate.New York: Oxford University Press.
Yee GW, Abbott UK. 1978. Facial develop-ment in normal and mutant chick em-bryos. I. Scanning electron microscopy ofprimary palate formation. J Exp Zool 206:307–321.
Yoon H, Chung IS, Seol EY, Park BY, ParkHW. 2000. Development of the lip andpalate in staged human embryos andearly fetuses. Yonsei Med J 41:477–484.
Young DL, Schneider RA, Hu D, Helms JA.2000. Genetic and teratogenic ap-proaches to craniofacial development.Crit Rev Oral Biol Med 11:304–317.
Zambrowicz BP, Imamoto A, Fiering S,Herzenberg LA, Kerr WG, Soriano P.1997. Disruption of overlapping tran-scripts in the ROSA beta geo 26 genetrap strain leads to widespread expres-sion of beta-galactosidase in mouse em-bryos and hematopoietic cells. Proc NatlAcad Sci U S A 94:3789–3794.
Zhang Z, Song Y, Zhao X, Zhang X, FerminC, Chen Y. 2002. Rescue of cleft palate inMsx1-deficient mice by transgenic Bmp4reveals a network of BMP and Shh sig-naling in the regulation of mammalianpalatogenesis. Development 129:4135–4146.
