From head to tail: regionalization of the neural crest

Abstract

The neural crest is regionalized along the anteroposterior axis, as demonstrated by foundational lineage-tracing experiments that showed the restricted developmental potential of neural crest cells originating in the head. Here, we explore how recent studies of experimental embryology, genetic circuits and stem cell differentiation have shaped our understanding of the mechanisms that establish axial-specific populations of neural crest cells. Additionally, we evaluate how comparative, anatomical and genomic approaches have informed our current understanding of the evolution of the neural crest and its contribution to the vertebrate body.

Keywords: Ectomesenchyme; Gene regulatory networks; Neural crest; Patterning; Stem cells.

Fig. 1.

Axial regionalization of the neural crest. Hox gene expression domains and neural crest derivatives are aligned to the body axis of a schematized amniote embryo. PG, paralog group. Fates of neural crest cells from all axial levels (yellow), cranial region only (blue, striped), trunk region only (green) and vagal/sacral regions (purple) are shown. The sacral neural crest occurs posterior to somite 28 in older embryos and is therefore shown alongside the unsegmented region of the pre-somitic mesoderm. The prosencephalon (P), metencephalon (M) and rhombencephalon (R) are labeled within the central nervous system. The otic vesicle (o), somites (s1-s5) and cranial neural crest streams migrating to the pharyngeal arches (PA) 1-3 are also indicated.

Fig. 2.

Transplantation approaches reveal that intrinsic factors and extrinsic signals underlie differences between cranial and trunk NC cells. Transplantation experiments reveal differences in the contributions of cranial and trunk neural crest. Derivatives of transplanted tissue are in red; quail embryos are shown in blue, chick embryos in orange. (A) Bilateral and heterotopic transplant of cranial (midbrain and anterior hindbrain) neural primordium from a quail donor (4-9 somite stage; ss) to the trunk of a chick host (24-26 ss) leads to the formation of skeletogenic derivatives, as well as chromaffin cells, at ectopic posterior positions. (B) The reciprocal transplant of trunk neural tube from a quail donor (11-29 ss) to the cranial (mid- and hindbrain) region of a chick host (4-10 ss) shows that trunk NC does not form skeletogenic derivatives. (C) A unilateral version of the transplant experiment shown in B demonstrates that host tissue can influence the migration and potential of transplanted cells. Donor cells form connective tissues alongside the host NC but cannot form skeletogenic derivatives. (D) When trunk NC cells cultured in bone-promoting media are transplanted into the mandibular and maxillary primordium of a chick host, the transplanted cells are able to form skeletogenic derivatives, demonstrating the importance of the NC signaling environment for cell fate decisions.

Fig. 3.

Differentiation of hSPCs into distinct axial subpopulations of NC. An undifferentiated human pluripotent stem cell (hPSC) passes through different intermediate states en route to a cranial or trunk neural crest cell fate. Each cell type and its characteristic gene expression is in red. The signals needed to promote each cell type are indicated next to the arrows. Derivatives formed by each cell type are color coded. ¹Hackland et al., 2017; ²Lee et al., 2007; ³Frith et al., 2018; ⁴Fattahi et al., 2016; ⁵Gomez et al., 2019b; ⁶Hackland et al., 2019.

Fig. 4.

Bipotent NMPs in the tailbud give rise to posterior tissues. (A) Gastrulation stage (E7.5) mouse embryo showing the primitive streak and the location of neuromesodermal progenitors (NMPs) within the node streak border (NSB) and caudal lateral epiblast (CLE). Gastrulation movements through and away from the primitive streak are shown using dashed arrows, while the anterior (A) to posterior (P) axis of the developing body is labeled with a blue arrow. (B) Early somite-stage mouse embryo (E8.5). Arrows show contributions of bipotent NMPs located in the NSB and CLE to both neuroectodermal tissues, such as the spinal cord (1), and mesodermal tissues, such as the somitic mesoderm (2) and notochord (3). (C) At later stages (E9.5-14.5), NMPs located in the chordoneural hinge (CNH) of the developing tailbud continue to contribute to both mesodermal and neuroectodermal tissues. A, anterior; FB, forebrain; HB, hindbrain; MB, midbrain; NC, notochord; P, posterior; PSM, pre-somitic mesoderm; SC, spinal cord.

Fig. 5.

Distribution of mineralized dermal tissues and neural crest derivatives across the vertebrate evolutionary tree. (A) Mineralized dermal tissues can be odontogenic (green) or osteogenic (gray), or both (striped). The extent of mineralized dermal tissues along the body axis of each species is indicated by the distribution of color. Contributions of the neural crest to cartilage in the pharyngeal arches and to odontogenic tissues of the teeth and scales are represented by symbols located above each species silhouette. Representative species are depicted for each branch of the tree. Extant groups (left to right): hagfish, lampreys, cartilaginous fishes, teleosts, non-teleost ray-finned fishes (e.g. Polypterus), lungfishes, anamniotes (e.g. Xenopus) and amniotes. Extinct groups (left to right): ostracoderms, placoderms, fossil ray-finned fishes (e.g. Cheirolepis), fossil lungfishes (e.g. Dipterus) and fossil tetrapods (e.g. Osteolepis). Asterisks indicate that ostracoderms and placoderms are both paraphyletic. (B) Representative cross-section of dermal skeletal elements with both odontogenic and osteogenic layers (left), odontogenic layer only (middle) and osteogenic layer only (right).

Fig. 6.

Evolutionary models for axial regionalization of the neural crest. (A) Paleontological data on the location of mineralized dermal tissues along the body axis suggest that ectomesenchymal (EM) potential originated either: (1) at the base of the vertebrate tree (orange); or (2) in the last common ancestor of jawed vertebrates (blue). Circles indicate acquisition of a trait, while lines indicate a loss. In both cases, EM potential would have been lost in an ancestor of modern teleosts and modern tetrapods. (B) Genetic data from a variety of species, including lamprey, shark, zebrafish and chick, suggest that multiple heterochronic shifts in transcription factor expression may have led to the restriction of EM potential to the cranial neural crest. In this scenario, the ancestral vertebrate neural crest would have resembled that found in the modern amniote trunk (pink).