The terminal nerve, or nervus terminalis, is the rostralmost cranial nerve (“nerve 0”). This nerve or complex of nerves is a diffusely organized system of neurons in the nasal cavity and rostral forebrain (Demski, 1993; Wirsig-Wiechmann et al., 2002; von Bartheld, 2004). The functions of the terminal nerve relate primarily to sensory (but not chemosensory) modalities, as well as neuromodulation and reproductive behavior. Although the terminal nerve associates with the olfactory nerve, it is functionally distinct from the classical olfactory system (Demski and Schwanzel-Fukuda, 1987; Demski, 1993; Wirsig-Wiechmann et al., 2002). For most of the last century, it was thought that the terminal nerve derives embryonically from the olfactory placode (Schwanzel-Fukuda and Pfaff, 1989; Wray et al., 1989). The olfactory placodes are bilateral patches of thickened anterior head ectoderm that give rise to the entire olfactory epithelium and the migratory glial cells that ensheath the olfactory nerve (reviewed in Baker and Bronner-Fraser, 2001). A recent fate-mapping study in zebrafish, however, challenged the dogma of an olfactory placode origin for nervus terminalis cells, and suggested that at least some of these cells derive from the neural crest (Whitlock et al., 2003).
Although both placode and neural crest cells originate from ectoderm at the border between the prospective neural plate and epidermis, placodes and neural crest cells are different cell populations with generally distinct fates (Le Douarin and Kalcheim, 1999; Baker and Bronner-Fraser, 2001). Since original reports regarding the embryonic origin of the terminal nerve in shark embryos also favored, at least in part, a neural crest origin (Locy, 1905; Johnston, 1913), these new data revive a century-old controversy (Larsell, 1918). A fresh look into the origin of terminalis cells may solve this dispute and clarify the roles of placodes and the neural crest in vertebrate forebrain development (Le Douarin and Kalcheim, 1999; Baker and Bronner-Fraser, 2001).
A neuroendocrine subpopulation of the terminal nerve cells expresses gonadotropin-releasing hormone (GnRH). GnRH is an important neurotransmitter that links the brain to the reproductive system. It also has neuromodulatory roles in the brain, and is a convenient neurochemical marker for a subpopulation of terminalis cells (Demski and Schwanzel-Fukuda, 1987; Dubois et al., 2002). A series of immunocytochemical and birthdating studies in 1989 showed that during mouse embryonic development, the GnRH-expressing cells of the terminalis system migrate from the region of the olfactory placode along the olfactory nerve pathway and into the forebrain (Schwanzel-Fukuda and Pfaff, 1989; Wray et al., 1989), where they are required for normal development of the gonads (Schwanzel-Fukuda and Pfaff, 1989). These data, together with ablation and grafting studies in chick, amphibian, and lungfish embryos, appeared to confirm that the terminalis system originates from the olfactory placode (e.g., Wang et al., 2001). Indeed, in both mouse and chick embryos, GnRH-immunoreactive cells have been seen within the epithelium of the olfactory placode itself (Wray et al., 1989; Mulrenin et al., 1999). In the chick embryo, GnRH mRNA expression has also been reported at very early stages in the anterior neural folds, where the anlagen of the olfactory placode and nonsensory nasal epithelium are located (Witkin et al., 2003). Although an ablation study in chicks suggested that the GnRH neurons may arise from nonsensory nasal epithelium, rather than olfactory sensory epithelium (El Amraoui and Dubois, 1993), the prevailing view was that GnRH neurons, and presumably other components of the terminal nerve complex, originated from the olfactory placode (Schwanzel-Fukuda and Pfaff, 1989; Wray et al., 1989; Demski, 1993).
In contrast, the recent fate-mapping study in zebrafish (Whitlock et al., 2003) proposed that “rostral” GnRH-expressing neuroendocrine cells, and thus sensory or neuromodulatory cells of the terminalis system, originate from the neural crest. In that study, fluorescent dextrans were injected into two or three individual cells at the border between the cranial neural crest domain and the olfactory placode domain. The location and phenotype of the clones derived from these cells were scored at a later stage. The authors concluded that cranial neural crest cells give rise to terminal nerve GnRH neurons.
How can a neural crest origin for rostral GnRH neurons in zebrafish be reconciled with the GnRH expression seen in both prospective olfactory placode ectoderm and cells within the olfactory epithelium itself, in the mouse and chick (Wray et al., 1989; Mulrenin et al., 1999; Witkin et al., 2003)? The anterior neural plate border, where the prospective olfactory placode ectoderm is located, does not appear to form neural crest cells in the chick (Couly and Le Douarin, 1988). It is possible that neural crest cells from the midbrain and caudal diencephalon, which give rise to the frontonasal mesenchyme surrounding the olfactory placode, infiltrate the olfactory placode epithelium at a later stage and give rise to the GnRH-immunoreactive cells detected there. Although this has not been reported in neural crest fate-mapping studies to date, this possibility should be explored in representatives of all vertebrate classes, given the new zebrafish data. It is also formally possible that teleost neural crest cells have independently acquired the capacity to generate GnRH neurons.
For the developmental neurobiologist, both the neural crest and the placodes present plausible alternatives as the embryonic source of the highly diverse cells of the terminalis system. Numerous cells of the terminalis system (not only the GnRH-expressing neurons) migrate from the olfactory epithelium toward and into the brain, and display an astonishing array of transmitter and other peptide phenotypes (Valverde et al., 1993; von Bartheld, 2004). Both neural crest- and placode-derived cells migrate significantly during development and give rise to a multitude of different cellular phenotypes (Le Douarin and Kalcheim, 1999; Baker and Bronner-Fraser, 2001). The neural crest generates several types of endocrine/neuroendocrine cells. However, the endocrine cells of the adenohypophysis (anterior pituitary) are derived from the hypophyseal placode, which arises from the rostralmost ectoderm of the neural plate border (Baker and Bronner-Fraser, 2001). The hypophyseal placode was identified as the source of hypothalamic GnRH neurons in the same zebrafish fate-mapping study that proposed a neural crest origin for terminal nerve GnRH neurons (Whitlock et al., 2003). Hence, there is no a priori reason to consider neural crest cells a more likely source of terminal nerve GnRH neurons than placodal cells.
For the developmental neurobiologist, both the neural crest and the placodes present plausible alternatives as the embryonic source of the highly diverse cells of the terminalis system.
By challenging the prevailing dogma of a placodal origin for terminal nerve GnRH neurons, the new data from Whitlock et al. (2003) focus attention on one of the remaining frontiers of embryonic fate mapping. To resolve this century-old controversy, additional neural crest and olfactory placode fate-mapping data from multiple vertebrate species, in combination with cell type-specific markers, will be required. Such a resolution will ultimately lead to a definitive framework for understanding the remarkable origin, fates, and functions of the cells that form the nervus terminalis complex.