The cellular and molecular origins of beak morphology

Cell-autonomous and nonautonomous actions of endothelin-A receptor signaling in craniofacial and cardiovascular development

Craniofacial and cardiac development relies on the proper patterning of the neural crest-derived ectomesenchyme of the pharyngeal arches, from which many craniofacial and great vessel structures arise. One of the intercellular signaling molecules that is involved in this process, endothelin-1 (ET-1), is expressed in the arch epithelium and influences arch development by binding to its cognate receptor, the endothelin A (ET(A)) receptor, found on ectomesenchymal cells. We have previously shown that absence of ET(A) signaling in ET(A)(-/-) mouse embryos disrupts neural crest cell development, resulting in craniofacial and cardiovascular defects similar in many aspects to those in mouse models of DiGeorge syndrome. These changes may reflect a cell-autonomous requirement for ET(A) signaling during crest cell development because the ET(A) receptor is an intracellular signaling molecule. However, it is also possible that some of the observed defects in ET(A)(-/-) embryos could arise from the absence of downstream signaling that act in a non-cell-autonomous manner. To address this question, we performed chimera analysis using ET(A)(-/-) embryonic stem cells. We observe that, in almost all early ET(A)(-/-) --> (+/+) chimeric embryos, ET(A)(-/-) cells are excluded from the caudoventral aspects of the pharyngeal arches, suggesting a cell-autonomous role for ET(A) signaling in crest cell migration and/or colonization. Interestingly, in the few embryos in which mutant cells do reach the ventral arch, structures derived from this area are either composed solely of wild type cells or are missing, suggesting a second cell-autonomous role for ET(A) signaling in postmigratory crest cell differentiation. In the cardiac outflow tract and great vessels, ET(A)(-/-) cells are excluded from the walls of the developing pharyngeal arch arteries, indicating that ET(A) signaling also acts cell-autonomously during cardiac neural crest cell development.

Patterning of the hyoid cartilage depends upon signals arising from the ventral foregut endoderm

Hyoid bone is a part of the visceral skeleton which arises from both Hox-expressing (Hox+) and Hox-nonexpressing (Hox-) cephalic neural crest cells. In a previous work, we have demonstrated that the Hox- neural crest domain behaves as a naïve entity to which the ventral foregut endoderm confers patterning cues to specify the shape and orientation of the nasal and mandibular skeleton. By using ablation and grafting approaches, we have extended our study to the formation of the hyoid bone and tested the patterning ability of more caudal levels of the lateroventral foregut endoderm in the chick embryo at the early neurula stage. In this study, endodermal stripes have first been delineated according to the projection of mid- and posterior rhombencephalic structures. The extirpation of endodermal transverse stripes along the anteroposterior axis selectively hampers the formation of the ceratobranchials and epibranchials. Thus defined, the patterning ability of the endodermal stripes was further explored in their medial and lateral parts. When homotopically engrafted on the migration pathway of cephalic neural crest cells, ventromedial zones of endoderm lead to the formation of supernumerary basihyal and basibranchial, while lateral zones generate additional cartilaginous pieces recognizable as ceratobranchial and epibranchial. Taken together, our data demonstrate that, early in development, the ventral foregut endoderm exerts a regionalized patterning activity on the cephalic neural crest to build up the primary facial and visceral skeleton in jaws and neck and enable a map of the endodermal skeletogenic areas to be drawn. This map reveals that a cryptic metamerization of the anterior foregut endoderm precedes the formation of the branchial arches.

Unraveling the genetic and developmental mysteries of 22q11 deletion syndrome

Birth defects occur in nearly 5% of all live births and are the major cause of infant mortality and morbidity. Despite the recent progress in molecular and developmental biology, the underlying genetic etiology of most congenital anomalies remains unknown. Heterozygous deletion of the 22q11.2 locus results in the most common human genetic deletion syndrome, known as DiGeorge syndrome, and has served as an entry to understanding the basis for numerous congenital heart and craniofacial anomalies, among many other defects. Extensive human genetic analyses, mouse modeling and studies of developmental molecular cascades involved in 22q11 deletion syndrome are revealing complex networks of signaling and transcriptional events that are essential for normal embryonic development. Armed with this knowledge, we can now begin to consider the multiple genetic "hits" that might contribute to developmental anomalies, some of which could provide targets for in utero prevention of birth defects.

Mesenchymal patterning by Hoxa2 requires blocking Fgf-dependent activation of Ptx1

Hox genes are known key regulators of embryonic segmental identity, but little is known about the mechanisms of their action. To address this issue, we have analyzed how Hoxa2 specifies segmental identity in the second branchial arch. Using a subtraction approach, we found that Ptx1 was upregulated in the second arch mesenchyme of Hoxa2 mutants. This upregulation has functional significance because, in Hoxa2(-/-);Ptx1(-/-) embryos, the Hoxa2(-/-) phenotype is partially reversed. Hoxa2 interferes with the Ptx1 activating process, which is dependent on Fgf signals from the epithelium. Consistently, Lhx6, another target of Fgf8 signaling, is also upregulated in the Hoxa2(-/-) second arch mesenchyme. Our findings have important implications for the understanding of developmental processes in the branchial area and suggest a novel mechanism for mesenchymal patterning by Hox genes that acts to define the competence of mesenchymal cells to respond to skeletogenic signals.

About face: signals and genes controlling jaw patterning and identity in vertebrates

The embryonic vertebrate face is composed of similarly sized buds of neural crest-derived mesenchyme encased in epithelium. These buds or facial prominences grow and fuse together to give the postnatal morphology characteristic of each species. Here we review the role of neural crest cells and foregut endoderm in differentiating facial features. We relate the developing facial prominences to the skeletal structure of the face and review the signals and genes that have been shown to play an important role in facial morphogenesis. We also examine two experiments one at the genetic level and one at the signal level in which transformation of facial prominences and subsequent change of jaw identity was induced. We propose that signals such as retinoids and BMPs and downstream transcription factors such as Distal-less related genes specify jaw identity.

Cranial skeletal biology

To artists, the face is a mirror of the soul. To biologists, the face reflects remarkable structural diversity--think of bulldogs and wolfhounds or galapagos finches. How do such variations in skeletal form arise? Do the same mechanisms control skeletogenesis elsewhere in the body? The answers lie in the molecular machinery that generates neural crest cells, controls their migration, and guides their differentiation to cartilage and bone.

A zone of frontonasal ectoderm regulates patterning and growth in the face

A fundamental set of patterning genes may define the global organization of the craniofacial region. One of our goals has been to identify these basic patterning genes and understand how they regulate outgrowth of the frontonasal process, which gives rise to the mid and upper face. We identified a molecular boundary in the frontonasal process ectoderm, defined by the juxtaposed domains of Fibroblast growth factor 8 and Sonic hedgehog, which presaged the initial site of frontonasal process outgrowth. Fate maps confirmed that this boundary region later demarcated the dorsoventral axis of the upper beak. Ectopic transplantation of the ectodermal boundary region activated a cascade of molecular events that reprogrammed the developmental fate of neural crest-derived mesenchyme, which resulted in duplications of upper and lower beak structures. We discuss these data in the context of boundary/morphogen models of patterning, and in view of the recent controversy regarding neural crest pre-patterning versus neural crest plasticity.

A celebration of the new head and an evaluation of the new mouth

Twenty years ago now, Carl Gans and Glen Northcutt proposed that the main invention of vertebrates was a new head, with its full array of sensory organs involved in an active predatory lifestyle. Tracing back the embryological origin of these structures, they showed how all are primarily derived from the neural crest and the placodes, two transient ectodermal cell populations in the embryo. These cell types were then used for further innovations, such as a new mouth in jawed vertebrates. The interplay between patterning and plasticity of the neural crest is largely responsible for the endless variation of vertebrate craniofacial features in evolution.