Altered mandibular development precedes the time of palate closure in mice homozygous for disproportionate micromelia: an oral clefting model supporting the Pierre-Robin sequence

BACKGROUND: Development of the human craniofacial anatomy involves a number of interrelated, genetically controlled components. The complexity of the interactions between these components suggests that interference with the spaciotemporal interaction of the expanding tongue and elongating Meckel's cartilage correlates with the appearance of cleft palate. Mice homozygous for the semi-dominant Col2a1 mutation Disproportionate micromelia (Dmm), presenting at birth with both cleft palate and micrognathia, provide the opportunity to test the hypothesis that mandibular growth retardation coincides with formation of the secondary palate as predicted from our understanding of the Pierre Robin sequence. The present study was conducted in embryonic day 14 (E14) mice, 1 day before palate closure, to describe the relationship between growth of the lower jaw/tongue complex versus genotype of the embryo. METHODS: Whole heads, isolated from E14.25, E14.5 and E14.75 wild-type and homozygous mutant embryos, were fixed in Bouin's solution, embedded in paraffin, and serially sectioned. Mid-sagittal sections, stained with toluidine blue, were used to estimate growth of both tongue and lower jaw (Meckel's cartilage length) during a 12-hr period preceding palate closure. RESULTS: In control embryos, the largest increase in Meckel's cartilage length occurred between E14.5 and E14.75. Compared to control, the mean Meckel's cartilage length in the mutant was similar at E14.25, but was significantly less at E14.5 and E14.75. Absolute tongue size in control embryos increased linearly during this period of E14.25 to E14.75. Relative to the rapidly growing Meckel's cartilage, however, relative tongue size in control embryos actually decreased over time. Absolute tongue size in the mutant was not significantly different from that of control at any of the embryonic stages examined, however, relative tongue size in the mutant was significantly greater at E14.75 compared to control. CONCLUSION: Mandibular growth retardation, coupled with relative macroglossia in E14 Dmm/Dmm mice, suggests that the concerted development of the palate and lower jaw complex in the mutant is aberrant. Detection of micrognathia and pseudomacroglossia in homozygotes, before the time of palate closure, supports the hypothesis that a relationship exists between growth retardation of Meckel's cartilage and malformation of the secondary palate, as predicted by the Pierre-Robin sequence.     

Study of the mechanisms of BUdR-induced cleft palate in the mouse

This study was designed to examine the pathogenesis of bromodeoxyuridine (BUdR)-induced clefts of the secondary palate in the LACA mouse. Intraperitoneal injections of BUdR (500 mg/kg body weight) were given at various days and combinations of days between E11 and E15 (plug day = E1). Treatment on E11 alone resulted in approximately 22% of fetuses with cleft palate when the latter were examined either on E16 or E19. Treatment on E11 and E12 approximately doubled the above incidence, and treatment on E11, 12 and 13 raised it to 100%. However, no treatment, either single or multiple, caused cleft palate when given later than E11. This suggests that the cellular changes caused by BUdR that lead to cleft palate must be inflicted during E11 and that such damage can be repaired in about 80% of embryos. All fetuses with cleft palate had severe micrognathia on E16 and E19, which skeletal staining showed to be the result of a bilateral sigmoid buckling of Meckel's cartilage. Studies with the scanning electron microscope (SEM) on E15, 16, and 19 suggested strongly that the micrognathia caused a relative macroglossia and hence mechanical interference with palatal shelf reorientation. Histological studies with the light microscope showed that BUdR caused cellular necrosis in many embryonic tissues during the 24 hours after its administration. This necrosis was strikingly more severe in the mandibular rudiment of the first branchial arch than in the maxillary. The latter observation accords well with findings by other workers that cell proliferation is more rapid in the mandibular blastema than in the maxillary. Transmission electron microscope (TEM) studies of the buckled region of Meckel's cartilage failed to reveal any ultrastructural differences from control Meckel's cartilage. Hence BUdR had only interfered with the shape of the cartilage but not with its histiogenesis. We conclude that BUdR, by its cytotoxicity or antidifferentiative effects, interfered with the formation of the anterior end of Meckel's cartilage, initiating a chain of events leading through micrognathia and relative macroglossia to failure of palatal shelf reorientation and cleft palate.     

Is Far a Hox mutation?

The mouse First arch mutation, Far, causes a severe syndrome of craniofacial defects described previously. All of the known defects are derived from the anterior first arch, and to a very small extent, the dorsal second arch. Recently Far has been shown to be closely linked to Ulnaless on chromosome 2, and therefore in the vicinity of the Hox-4 gene cluster. This paper reports the results of several studies focused on the development origin of the most consistently expressed dominant effect caused by Far, an abnormal major bifurcation of the maxillary nerve. Nerve-stained whole-mount preparations of day 12 embryos showed that in Far mutants the maxillary nerve appears to have a central wedge missing from the normal single-stalked fan shape, and that the nerve defect in Far/Far and +/Far may be equally severe. The effect of retinoic acid on the development of the maxillary nerve was tested. Maternal treatment with 5 mg/kg retinoic acid on day 9 of gestation had no detectable effect on the maxillary nerve of +/Far embryos, and similar treatment with a teratogenic dosage (20 mg/kg) on day 8 or 9 produced no Far-like maxillary nerve defects in genetically normal embryos. The neural crest cells that give rise to nerves and mesenchyme of the first arch originate from specific rhombomeres, discrete segments of the developing head. The rhombomeres of 15 embryos at the 14-23 somite stages, of which 75% are expected to be +/Far or Far/Far, were examined. There was no detectable defect in segmentation or morphology of the rhombomeres compared with controls. The significance of ectopic cartilage in the palate of Far/Far mutants in relation to nerve bifurcation was explored. In histological studies, five out of six Far/Far day-15 fetuses had a rod of ectopic cartilage lateral to the posterior palate, running parallel to, and morphologically similar to, Meckel's cartilage, and lying between the two trunks of the abnormally bifurcated maxillary nerve. None of six +/Far day-15 fetuses examined had detectable ectopic cartilage in this region. We hypothesize that the maxillary nerve defects in Far mutants may be explained by the presence of an ectopic precartilaginous blastema that does not always further develop into detectable cartilage. The ectopic cartilage found in Far/Far resembles the epibranchial cartilage expressed in more posterior branchial arches and in the first arch of lower organisms, and therefore may represent an atavistic posteriorization of the anterior first arch in Far mutants.(ABSTRACT TRUNCATED AT 400 WORDS)     

Craniofacial and central nervous system malformations induced by triamcinolone acetonide in nonhuman primates: II. Craniofacial pathogenesis

This study further defines the craniofacial malformations induced by triamcinolone acetonide in the rhesus monkey. Ten timed-mated pregnant rhesus monkeys (Macaca mulatta) received intramuscular injections of 10 mg/kg TAC on days 23, 25, 27, 29, and 31 of gestation. Results of previous experiments with rhesus and bonnet monkeys and baboons indicated that specific craniofacial and brain malformations could be induced with TAC during this period of pregnancy (Hendrickx et al., '80). Stage-matched TAC-treated and control embryos (stages 17-18 and 22) and age-matched TAC-treated and control fetuses (50, 60, and 70 days gestation) were removed by hysterotomy. Stage 17-18 TAC embryos appeared grossly normal but histologic evaluation revealed a shortened anlage of the posterior cranial base. Stage 22 TAC embryos and all TAC fetuses exhibited craniofacial dysmorphia and encephalocele. The developing sphenoid was the earliest affected and most severely malformed bone. Its defects included reduced anterioposterior and transverse dimensions, reduced orbitosphenoid and alisphenoid, abnormal pituitary fossa, and reduced dorsum and tuberculum sellae. In addition, shortening of the posterior cranial base and decreased cranial base angle was a consistent finding in the treated embryos and fetuses. Decreased ossification and remodeling in the facial bones and abnormal position due to the malformed sphenoid occurred.     

Fuz regulates craniofacial development through tissue specific responses to signaling factors

The planar cell polarity effector gene Fuz regulates ciliogenesis and Fuz loss of function studies reveal an array of embryonic phenotypes. However, cilia defects can affect many signaling pathways and, in humans, cilia defects underlie several craniofacial anomalies. To address this, we analyzed the craniofacial phenotype and signaling responses of the Fuz(-/-) mice. We demonstrate a unique role for Fuz in regulating both Hedgehog (Hh) and Wnt/β-catenin signaling during craniofacial development. Fuz expression first appears in the dorsal tissues and later in ventral tissues and craniofacial regions during embryonic development coincident with cilia development. The Fuz(-/-) mice exhibit severe craniofacial deformities including anophthalmia, agenesis of the tongue and incisors, a hypoplastic mandible, cleft palate, ossification/skeletal defects and hyperplastic malformed Meckel's cartilage. Hh signaling is down-regulated in the Fuz null mice, while canonical Wnt signaling is up-regulated revealing the antagonistic relationship of these two pathways. Meckel's cartilage is expanded in the Fuz(-/-) mice due to increased cell proliferation associated with the up-regulation of Wnt canonical target genes and decreased non-canonical pathway genes. Interestingly, cilia development was decreased in the mandible mesenchyme of Fuz null mice, suggesting that cilia may antagonize Wnt signaling in this tissue. Furthermore, expression of Fuz decreased expression of Wnt pathway genes as well as a Wnt-dependent reporter. Finally, chromatin IP experiments demonstrate that β-catenin/TCF-binding directly regulates Fuz expression. These data demonstrate a new model for coordination of Hh and Wnt signaling and reveal a Fuz-dependent negative feedback loop controlling Wnt/β-catenin signaling.     

Does the tongue play a role in the initial development of vertical palatal shelf in hamster?

A study was undertaken to examine whether the tongue plays any role in determining the primordial development of palatal shelves in a vertical direction in mammals. Control and 6-mercaptopurine-treated embryos from Golden Syrian hamsters were examined by scanning electron microscopic and histological techniques for the spatio-temporal relationship of primordial development of the palate, tongue, and mandible. DNA synthesis, measured by 3H-thymidine incorporation, was used as an index of growth. The data indicated that in controls, vertical palate development began in the anterior half from the roof of the oronasal cavity, whereas the tongue bulges and the mandibular process developed in the posterior half of the oronasal cavity. A burst in DNA synthesis occurred in the palate and mandible, but not in the tongue. In 6-mercaptopurine-treated fetuses, although the chronological appearance of primordia of all three structures was normal, DNA synthesis was inhibited in all three structures. The recovery in DNA synthesis, albeit partial, was faster in the palate and mandible than in the tongue. On the basis of observations from the present study, along with those from other vertebrates, it is suggested that the developing tongue may not play any role in determining the direction of development of the palatal primordia.     

Developmental origins and evolution of jaws: new interpretation of "maxillary" and "mandibular"

Cartilage of the vertebrate jaw is derived from cranial neural crest cells that migrate to the first pharyngeal arch and form a dorsal "maxillary" and a ventral "mandibular" condensation. It has been assumed that the former gives rise to palatoquadrate and the latter to Meckel's (mandibular) cartilage. In anamniotes, these condensations were thought to form the framework for the bones of the adult jaw and, in amniotes, appear to prefigure the maxillary and mandibular facial prominences. Here, we directly test the contributions of these neural crest condensations in axolotl and chick embryos, as representatives of anamniote and amniote vertebrate groups, using molecular and morphological markers in combination with vital dye labeling of late-migrating cranial neural crest cells. Surprisingly, we find that both palatoquadrate and Meckel's cartilage derive solely from the ventral "mandibular" condensation. In contrast, the dorsal "maxillary" condensation contributes to trabecular cartilage of the neurocranium and forms part of the frontonasal process but does not contribute to jaw joints as previously assumed. These studies reveal the morphogenetic processes by which cranial neural crest cells within the first arch build the primordia for jaw cartilages and anterior cranium.     

A comparison of the craniofacial morphology in 2-month-old unoperated infants with unilateral complete cleft lip and palate, and unilateral incomplete cleft lip.

This paper reports a cephalometric analysis of the craniofacial morphology in infants with unoperated unilateral complete cleft lip and palate (UCCLP) and unoperated unilateral incomplete cleft lip (UICL). The purpose of the study was to determine the nature and extent of the craniofacial deviations in UCCLP as compared to the morphology in UICL, which has previously been shown to be close to normal. The samples comprised 82 infants with UCCLP (58 males and 24 females) and 75 with UICL (48 males and 27 females). The mean age was about 2 months in both groups. The cephalometric analysis of craniofacial morphology included the lateral, frontal, and axial projections. The data were presented as mean plots of the craniofacial region including the calvaria, cranial base, orbits, nasal bone, maxilla, mandible, cervical column, pharynx, and soft-tissue profile. The most pronounced deviations in the UCCLP group were observed in the maxillary complex and the mandible. The most striking findings were: markedly increased width of the maxilla, a short mandible, and bimaxillary retrognathia except for the premaxillary area, which was relatively protruding and asymmetric. The study did not support the hypothesis previously suggested in the literature that cleft lip and palate is a craniofacial anomaly as size and shape of the calvaria and cranial base were found to be normal. The etiology of cleft lip and palate is still incompletely understood. Based on the present study, we suggest that facial type may be a liability factor that could represent a developmental threshold increasing the probability of cleft lip and palate.     

The ontogeny of cranial base angulation in humans and chimpanzees and its implications for reconstructing pharyngeal dimensions

This paper examines differences in the processes by which the cranial base flexes in humans and extends in chimpanzees. In addition, we test the extent to which one can use comparisons of cranial base angles in humans and non-human primates to predict vocal tract dimensions. Four internal cranial base angles and one external cranial base angle were measured in a longitudinal sample of Homo sapiens and a cross-sectional sample of Pan troglodytes. These data show that the processes of cranial base angulation differ substantially in these species. While the human cranial base flexes postnatally in a rapid growth trajectory that is complete by two years, the cranial base in P. troglodytes extends postnatally in a more prolonged skeletal growth trajectory. These comparisons also demonstrate that the rate of cranial base angulation is comparable for different measures, but that angles which incorporate different anterior cranial base measurements correlate poorly. We also examined ontogenetic relationships between internal and external cranial base angles and vocal tract growth in humans to test the hypothesis that cranial base angulation influences pharyngeal dimensions and can, therefore, be used to estimate vocal tract proportions in fossil hominids. Our results indicate that internal and external cranial base angles are independent of the horizontal and vertical dimensions of the vocal tract. Instead, a combination of mandibular and palatal landmarks can be used to predict dimensions of the vocal tract in H. sapiens. The developmental contrasts in cranial base angulation between humans and non-human primates may have important implications for testing hypotheses about the relationship between cranial base flexion and other craniofacial dimensions in hominid evolution.     

Maxillary changes and occlusal traits in crania with artificial fronto-occipital deformation

Artificial fronto-occipital deformation of the cranial vault was typical of pre-Columbian cultures in the central Andean coastal regions. We have studied the influence of this deformation on maxillary and mandibular morphology. Measurements were performed on 86 adult Ancon skulls with anteroposterior deformation. Undeformed skulls from the area of Makatampu (n = 52) were used as the control group. To explore the influence of the deformity on occlusion, the skulls were categorized using the Angle classification and the alignment of the interincisor midline. In the group of deformed skulls, there was an increase in lateral growth of the vault and of the base of the skull (P < 0.001), giving rise to a greater interpterygoid width of the maxilla (P < 0.001), and an increase in the transverse diameter of the palatal vault. The mandible presented an increase in the length of the rami (P < 0.001) and in the intercondylar width, with no alteration of mandibular length. The deformed skulls had normal (class I) occlusion, with no displacement of the midline. The difference in the asymmetry index between the two groups was not statistically significant. Artificial fronto-occipital deformation of the cranial vault provoked compensatory lateral expansion of the base that was correlated with the transverse development of the maxilla and mandible. Occlusion and sagittal intermaxillary position were not affected by the cranial deformity. These results provide evidence of the integration between the neurocranium and the viscerocranium in craniofacial development, and support the hypothesis of a compensatory effect of function.     

Shape of the craniofacial complex in 45,X females: cephalometric study

The shape of the craniofacial complex was analysed cephalometrically in sixty-four adult 45,X females (Turner syndrome) using lateral skull radiographs, and the subjects were compared with first-degree female relatives and control females. The results showed that 45,X females have marked changes in relatively few craniofacial areas compared to the controls. Most of the changes are located in the cranial base, so that the face is retrognathic. The mandible is short, whereas the maxilla is of normal length. The results support the view that the morphology of the cranial base is markedly affected in 45,X females, whereas most other craniofacial changes could be considered secondary to the cranial base abnormality. It is suggested that retarded cartilage growth may be a factor leading to the present findings.     

Cranial base angulation and prognathism related to cranial and general skeletal maturation in human fetuses.

The purpose of the present study was to describe normal midsagittal craniofacial morphology in second trimester human fetuses. Measurements of the cranial base angle and the prognathism of the maxilla and the mandible were performed on radiographs of cranial midsagittal tissue blocks of 52 fetuses with a gestational age from 13 to 27 weeks. Special procedures were developed for the definitions of the nasion and sella reference points on the radiographs in the early stages of fetal development. Mean data were reported for stages of crown rump length (CRL) and maturation of the fetal cranial base (MSS), usable as reference in assessment of pathological fetal crania in reports and autopsy procedures. Regression equations were determined for the regression of the angular values on CRL, MSS, and general skeletal maturation (TNO). The cranial base angle was found to decrease significantly, and the angles of prognathism to increase significantly with increasing CRL, TNO, and MSS values. It was suggested that these simultaneous and similar changes in the three angles could be accounted for by the upwards movement of the sella point produced by a cranial displacement of the pituitary fossa caused by local cartilagenous growth and bony remodelling during the period of study. The study thus reflects the influence of cranial skeletal maturation on the early development in shape of the craniofacial complex.     

Studies on orthocephalization. 10. Behaviour of the visceral part of the rat head during the first 14 days after gestation

The present paper considers the significance of interosseous flexions of the palatal complex in the process of orthocephalization of the rat skull between birth and 7 d p.n. The study is based on a sample of 90 rats divided into 4 age groups, i.e. 0, 4, 7, and 14 d. These rats have been X-rayed, and their photographs subsequently analysed. During the studied period, the constituents of the bony palate, i.e. the horizontal part of the palatine bone, the palatal process of maxilla and the palatal part of premaxilla, increase markedly in length, but with individual differences in growth rate. There is, in the period, a marked decrease in angulation between the cranial base and the palatal plane. This means that the rat skull becomes more orthocranial. There is also a straightening (orthopalatalization) of the palate, as the angle between maxilla and premaxilla becomes more obtuse, and a marked decrease in angulation between the palatine bone and the cranial base. The patterns of angular changes suggest that the process of orthocephalization in the period between birth and 14 d p.n. primarily is a result of an upwards rotation of the palatine bone relative to the cranial base, while interosseous deflections in the palate only play a minor role.     

Palate development after fetal tongue removal in cortisone-treated mice

Morphological studies of cortisone-induced cleft palate have shown retardation in the rotation of palatine shelves from a sagittal to a transverse plane. Cortisone also reduces fetal muscular movements, which may explain why displacement of the tongue from between the palatine shelves is delayed. Previous work with extrauterine development of control fetuses demonstrated that fetal membranes and tongue were major obstacles to shelf rotation. Thus, removal of these obstacles might permit rotation and fusion of palatine shelves in cortisone-treated fetuses. In the present experiment, fetuses from cortisone-treated strain CD-1 mice were released from uterus and membranes and allowed to develop for eight hours in a fluid medium with the umbilical cord left intact. Compared to 4% fusion in utero, there was palatal fusion in 20% of fetuses released from membranes. When the fetal tongue was removed during extrauterine development, the frequency of fusions increased to 61%. Fusion appeared normal by the criteria applicable through light microscopy. Thus, cortisone induces cleft palate primarily through interference with shelf rotation. The palatine shelves of treated fetuses retain their ability to fuse when they can come in contact during the normal time for palate closure.     

A study of the os goniale in man.

The anterior process of the malleus of the middle ear develops irrespective of Meckel's cartilage through an intramembranous ossification center that appears in the human embryo of 26.5 mm crown-rump length at a caudomedial position in relation to Meckel's cartilage. The malleus has a double origin: the anterior process originates from the os goniale through intramembranous ossification, and the rest from Meckel's cartilage, through endochondral ossification.     

A craniofacial growth maturity gradient for males and females between 4 and 16 years of age

Differential growth of the craniofacial complex implies variation in ontogenetic patterns of development. This investigation quantifies the relative maturity—as defined by percent adult status—of nine cephalometric dimensions and stature. Analysis is based on 663 lateral cephalograms from a mixed longitudinal sample of 26 males and 25 females between 4 and 16 years of age. Graphic comparison of maturity status across the age range shows that variation is intergraded between the neural and somatic growth maturity patterns, as described by head height and stature, respectively. The maturity gradient moves from head height through anterior cranial base, posterior cranial base and maxillary length, upper facial height, corpus length, and ramus height to stature. After 9 years of age ramus height is less mature than stature. Anterior maxillary and mandibular heights diminish during transitional dentition and thereafter exhibit maturity patterns that compare to corpus length. Although females are consistently more mature than males, the gradient of variation between dimensions is sex independent.     

Cranial base and jaw relationship

The lateral skull radiographs of 124 boys aged approximately 10 years divided equally between the four angle classes were digitized in an effort to establish the relationship between cranial base size and shape and jaw relationship. Comparison of the means for occlusal groups showed a trend from class II to class III as cranial base dimensions and angle decreased. The condyle was also more distally positioned with respect to nasion, point A and the Pterygomaxillary vertical in the class II groups. Cranial base length correlated strongly with maxillary length but weakly with mandibular length. Nevertheless, the size of the maxilla did not influence its prognathism. The cranial base angle was strongly correlated (-0.7) with angle sella-nasion-point B. It is concluded that cranial base size and shape influence mandibular prognathism by determining the anteroposterior position of the condyle relative to the facial profile.     

Craniofacial distraction with a modular internal distraction system: evolution of design and surgical techniques.

The present report summarizes the evolution of design for a modular internal distraction system that is applicable throughout the craniofacial region. Eleven patients (5 boys, 6 girls), whose ages ranged from 4 months to 10 years at the time of distraction, constitute the basis for this study. The clinical indications for distraction were exorbitism with corneal exposure (n = 1), obstructive sleep apnea (n = 4), tracheostomy decannulation (n = 1), severe maxillary hypoplasia with class III malocclusion (n = 3), severe vertical and sagittal maxillary deficiency with anophthalmia (n = 1), and relapse following frontoorbital advancement in a case of rare craniofacial clefting (n = 1). Twenty-two distraction devices were used in these 11 patients. Two initial prototypes were tested (prototype 1 = 8 devices; prototype 2 = 2 devices) until the modular internal distraction system (MIDS, Howmedica-Leibinger, Inc.) was developed (n = 12 devices). The craniofacial osteotomies used were Le Fort III (n = 4), monobloc (n = 3), mandibular (n = 3), Le Fort I (n = 2), and cranial (n = 1). The distraction distances ranged from 11 to 28 mm. One patient undergoing mandibular distraction developed transient swelling in the left mandibular region, which responded to antibiotics. There were no other complications. Depending on the age of the patient and the length of distraction, the distraction gap was allowed to consolidate from 6 weeks to 3 months. The devices were then removed on either an outpatient or a 23-hour-stay basis. The modular internal distraction system permits widespread application of easily customizable, buried distraction devices throughout the craniofacial region.     

Compensatory defects associated with mutations in Hoxa1 restore normal palatogenesis to Hoxa2 mutants.

The rhombencephalic neural crest play several roles in craniofacial development. They give rise to the cranial sensory ganglia and much of the craniofacial skeleton, and are vital for patterning of the craniofacial muscles. The loss of Hoxa1 or Hoxa2 function affects the development of multiple neural crest-derived structures. To understand how these two genes function together in craniofacial development, an allele was generated that disrupts both of these linked genes. Some of the craniofacial defects observed in the double mutants were additive combinations of those that exist in each of the single mutants, indicating that each gene functions independently in the formation of these structures. Other defects were found only in the double mutants demonstrating overlapping or synergistic functions. We also uncovered multiple defects in the attachments and trajectories of the extrinsic tongue and hyoid muscles in Hoxa2 mutants. Interestingly, the abnormal trajectory of two of these muscles, the styloglossus and the stylohyoideus, blocked the attachment of the hyoglossus to the greater horn of the hyoid, which in turn correlated exactly with the presence of cleft palate in Hoxa2 mutants. We suggest that the hyoglossus, whose function is to depress the lateral edges of the tongue, when unable to make its proper attachment to the greater horn of the hyoid, forces the tongue to adopt an abnormal posture which blocks closure of the palatal shelves. Unexpectedly, in Hoxa1/Hoxa2 double mutants, the penetrance of cleft palate is dramatically reduced. We show that two compensatory defects, associated with the loss of Hoxa1 function, restore normal attachment of the hyoglossus to the greater horn thereby allowing the palatal shelves to lift and fuse above the flattened tongue.