Development of the Central Nervous System

Abstract

The human fertilized egg divides iteratively generating blastomeres. From the stage 16 blastomeres, the peripheral cells form the trophectoderm, whereas the central cells regroup under the name of internal cell mass. The latter evolve to form two cavities separated by an embryonic disk composed of two layers (epiblast and primitive endoderm). From the epiblast, three layers are set up during the gastrulation process. Thus, the superficial ectoderm, the intermediate mesoderm, and the deep endoderm are produced. Neural induction takes place at the level of the ectoderm and allows the divergence of the neurectoderm and the surface ectoderm. This induction is a complex phenomenon involving molecular sequences still imperfectly described and known.

Neurulation leads to the formation of a neural tube. In amniotes, several modes of neurulation are described making difficult the comprehension of the complete process. Primary neurulation drives the conversion of the neural plate to the neural tube. It occurs in successive stages. First, the neural plate extends in the rostro-caudal direction and converges toward the midline. Then, cellular hinges are formed which lead to a deformation of the neural plate which gives rise to a groove then to the neural tube after fusion of the folds. Animal models show differences in their primary neurulation, which should lead to caution before extrapolating the experimental data to humans. Secondary neurulation is performed by cavitation of the medullary cord. This mode of neurulation forms the caudal part of the spinal cord. The mapping of morphogenetic movements taking place during this phase reveals similarities with those of primary neurulation. This continuity between the two neurulations is now established. It is important to note that the tissue of the caudal neural plate produces the secondary neural tube. Thus, a defect of internalization of this tissue will generate and open neural tube defect. Ventro-dorsal polarization of the spinal cord involves secreted molecules. Ventralization is essentially provided by sonic hedgehog. Dorsalization involves several systems: BMP and especially Wnt.

The hindbrain is segmented into metameric units called rhombomeres. This segmentation has been described since the nineteenth century, but the understanding of the mechanisms involved in its control has only recently emerged. A rhombomere is an anatomical metameric unit separated from others by constrictions. At the beginning of development, these rhombomeres behave like autonomous segments whose cells do not mix. Some genes define a positional identity of rhombomeres. Their mutation causes malformations, including in human beings. Genetic regulation controlling the positional identity of rhombomeres is a very fertile field of investigation. This problem is very complex and involves many molecular players.

The cerebellum derives from rhombomers 0 and 1 of the neural tube. The cerebellar primordia are initially paired. They undergo morphogenetic movements which lead to a rotation and a fusion on the midline. Two germinative layers are described. The ventricular layer is responsible for the formation of GABAergic neurons, while the rhombic lip generates glutamatergic neurons. The meninges play a major role in the control of cerebellar development.

The roof of the fourth ventricle has been scarcely studied in human beings. This leads to a confusing natural history we have at our disposal. Therefore, it is advisable to be very careful before proposing a physiopathology for the cystic lesions of the posterior fossa.

The rostral region of the neural tube is segmented like the rhombomeres. This notion is at the heart of the neuromeric model, which is a work in progress.

The development of the cerebral cortex has undergone profound changes in concepts. Several anatomical zones produce neurons of the future cortex. The migration of these cells is complex and not only radial. Inhibitory interneurons come mainly from the ventral telencephalon. The germinative zone of the dorsal telencephalon is highly complex in man with a ventricular, internal subventricular, and external ventricular layers.

Somites give rise to sclerotome which is the precursor of the vertebrae. Sclerotome resegments so each vertebra is deriving from a caudal hemi-sclerotome associated with the rostral hemi-sclerotome from the next adjacent somite. Each sclerotome gives rise to four compartments that differentiate into different parts of the vertebra. The molecular control of these different anatomical parts is different.

Embryogenesis of the base of the skull is not well known. This region, which is anatomically complex, differs according to the experimental models used. Moreover, even in mice, a model currently used by developmental biology as the closest to humans, the base of the skull is not similar to that of humans. Thus, human interpolations must remain cautious. Schematically, we distinguish a prechordal base whose origin is essentially the neural crest (excluding hypochiasmatic cartilages). This region depends on the prechordal plate and is affected by severe forms of holoprosencephaly. The second region is the chordal base that develops in contact with the notochord. Its origin is the mesoderm (whether cephalic or somitic). The occipital bone is an even more complex structure whose embryological origin is not perfectly known. Only experiments in birds have been made. All other mappings are not based on solid data, which requires readers to be very cautious before using the published results. A real job remains to be done in mammals.

The vault of the skull is derived from the cells of the neural crest and mesoderm. The bones that form it undergo a membranous ossification. They receive many influences from the surrounding tissues (neural tissue, dura mater, and surface ectoderm). The primordia of the bones of the vault are separated from each other by the sutures. These non-ossified regions are still poorly known, and their natural history deserves to be reviewed. Thanks to human genetics, many genes involved in craniosynostosis have been highlighted. Their role in the ossification and physiopathology of craniosynostosis is beginning to be unveiled.