Neural tube closure in humans initiates at multiple sites: evidence from human embryos and implications for the pathogenesis of neural tube defects
- Cite this article as:
- Nakatsu, T., Uwabe, C. & Shiota, K. Anat Embryol (2000) 201: 455. doi:10.1007/s004290050332
The closure of the neural tube (NT) in the human embryo has generally been described as a continuous process that begins at the level of the future cervical region and proceeds both rostrally and caudally. On the other hand, multiple initiation sites of NT closure have been demonstrated in mice and other animals. In humans, based on the study of neural tube defects (NTD) in clinical cases, van Allen et al. (1993) proposed a multi-site NT closure model in which five closure sites exist in the NT of human embryos. In the present study, we examined human embryos in which the NT was closing (Congenital Anomaly Research Center, Kyoto University) grossly and histologically, and found that NT closure in human embyos initiates at multiple sites but that the mode of NT closure in humans is different from that in many other animal species. In addition to the future cervical region that is widely accepted as an initiation site of NT closure (Site A), the mesencephalic-rhombencephalic boundary was found to be another initiation site (Site B). The second closure initiating at Site B proceeds bidirectionally and its caudal extension meets the first closure from Site A over the rhombencephalon, and the rostral extension of the second closure meets another closure extending from the rostral end of the neural groove (Site C) over the prosencephalon, where the anterior neuropore closes. The caudal extension of the first closure initiating at Site A was found to proceed all the way down to the caudal end of the neural groove where the posterior neuropore is formed, indicating that in humans, NT closure does not initiate at the caudal end of the neural groove to proceed rostrally. Since there is a considerable species difference in the mode of NT closure, we should be careful when extrapolating the data from other animals to the human. It seems that the type of NTD affects the intrauterine survival of abnormal embryos. Almost all the embryos with total dysraphism appear to die by 5 weeks of gestation, those with an opening over the rhombencephalon by 6.5 weeks, and those with a defect at the frontal and parietal regions survive beyond 7 weeks.