Is There a Prechordal Region and an Acroterminal Domain in Amphioxus?

This essay re-examines the singular case of the supposedly unique rostrally elongated notochord described classically in amphioxus. We start from our previous observations in hpf 21 larvae [Albuixech-Crespo et al.: PLoS Biol. 2017;15(4):e2001573] indicating that the brain vesicle has rostrally a rather standard hypothalamic molecular configuration. This correlates with the notochord across a possible rostromedian acroterminal hypothalamic domain. The notochord shows some molecular differences that specifically characterize its pre-acroterminal extension beyond its normal rostral end under the mamillary region. We explored an alternative interpretation that the putative extension of this notochord actually represents a variant form of the prechordal plate in amphioxus, some of whose cells would adopt the notochordal typology, but would lack notochordal patterning properties, and might have some (but not all) prechordal ones instead. We survey in detail the classic and recent literature on gastrulation, prechordal plate, and notochord formation in amphioxus, compare the observed patterns with those of some other vertebrates of interest, and re-examine the literature on differential gene expression patterns in this rostralmost area of the head. We noted that previous literature failed to identify the amphioxus prechordal primordia at appropriate stages. Under this interpretation, a consistent picture can be drawn for cephalochordates, tunicates, and vertebrates. Moreover, there is little evidence for an intrinsic capacity of the early notochord to grow rostralwards (it normally elongates caudalwards). Altogether, we conclude that the hypothesis of a prechordal nature of the elongated amphioxus notochord is consistent with the evidence presented.