What can vertebrates tell us about segmentation?

Abstract

Segmentation is a feature of the body plans of a number of diverse animal groupings, including the annelids, arthropods and chordates. However, it has been unclear whether or not these different manifestations of segmentation are independently derived or have a common origin. Central to this issue is whether or not there are common developmental mechanisms that establish segmentation and the evolutionary origins of these processes. A fruitful way to address this issue is to consider how segmentation in vertebrates is directed. During vertebrate development three different segmental systems are established: the somites, the rhombomeres and the pharyngeal arches. In each an iteration of parts along the long axis is established. However, it is clear that the formation of the somites, rhombomeres or pharyngeal arches have little in common, and as such there is no single segmentation process. These different segmental systems also have distinct evolutionary histories, thus highlighting the fact that segmentation can and does evolve independently at multiple points. We conclude that the term segmentation indicates nothing more than a morphological description and that it implies no mechanistic similarity. Thus it is probable that segmentation has arisen repeatedly during animal evolution.

Keywords: Evolution; Metamerism; Pharyngeal arches; Rhombomeres; Segmentation; Somites; Vertebrates.

Figure 1

Segmentation in vertebrates is governed by divergent molecular mechanisms: somites, rhombomeres and pharyngeal arches. (A) A whole mount phalloidin staining of a stage 11 chick embryo viewed dorsally. The somites are readily visible as iterated blocks on either side of the neural tube. (B) An image of a dissected hindbrain region of chick embryo at 3 days of development. The roofplate has been removed to show the rhombomeres. (C) A longitudinal section through the pharyngeal region of a dogfish embryo. The embryo has been stained with an anti-laminin antibody. The pharyngeal segments are clearly shown. (D) Schematic diagram of somitogenesis. As the embryonic axis elongates, the presomitic mesoderm is patterned by opposing gradients of FGF8/Wnt activity (high posterior-low anterior) and retinoic acid (RA; low posterior-high anterior). Cycling expression of Notch pathway components (NICD = Notch Intracellular Domain) and of the transcription factor Mesp2 acts as a pacemaker in this process. (E) Diagram of rhombomere (r) formation. R identity is regulated by a posteriorly increasing gradient of RA and by FGF8 signalling from the midbrain-hindbrain boundary. R boundary formation involves Notch activation (NICD) and they are maintained through cell sorting driven by alternating expression of EphA4 (in r3/5) and ephrin ligands (in r2, 4, 6). (F) The formation of all pharyngeal arches depends on FGF signalling, that of the posterior arches also on Wnt and RA signalling and on the transcription factor Tbx1. In all panels anterior is to the top.

Figure 2

The evolutionary history of segmentation in the vertebrate lineage. Three instances of segmentation are found in extant vertebrates that are conserved with different invertebrate groups. While the ancestor of all Bilateria probably developed with a posterior growth zone expressing Wnts, and perhaps oscillating Notch signalling along the A/P axis, overt morphological segmentation appears later in the phylogeny. Pharyngeal segmentation can be dated to the deuterostome ancestor, while somitogenesis dates to the chordate ancestor and rhombomeric organisation of the hindbrain to the vertebrate stem.