Evolution of Development: The Details Are in the Entrails

Abstract

Historically, the position of the site of gastrulation has been used to understand the developmental basis for body plan diversity. A recent molecular study, however, challenges long-held views and shows that molecular patterning mechanisms can be used to understand body plan evolution despite variation in gastrulation movements.

Organismal and evolutionary biologists have long tried to use changes in developmental features to help explain major transitions in animal body form. The study of gastrulation, in particular the site of gastrulation and the fate of the blastopore (the region of the embryo that gives rise to the internal tissues such as the gut), has played a central role in our attempts to understand the evolution of body plan diversity. Indeed, the two major branches of metazoan animals, the Protostomia and Deuterostomia (Figure 1A), were named due to the relationship between the larval/adult mouth and the site of gastrulation [1]. In protostomes, the site of gastrulation is said to give rise to the mouth, while in deuterostomes it forms the anus with the mouth arising from a second opening distant to the site of gastrulation.

Figure 1. Implications of the site of gastrulation with respect to body plan formation in the Metazoa.

(A) Phylogenetic relationship of major metazoan groups. A recent paper [2] suggests the Protostomes are misnamed because the site of gastrulation does not form the mouth. Note that the relative branching order of sponges and ctenophores remains unresolved. (B) The position of the site of gastrulation and the mouth relative to the animal–vegetal axis in four major metazoan lineages. Note that the two sensory structures derived from the vegetal pole in cnidarians and ctenophores are completely different in structure and function.

However, a new paper reporting the development of priapulids (‘penis worms’) published recently in Current Biology from Martin-Durá n et al. [2] shows that this important group of ancient marine animals, firmly established as members of the Protostomia, actually gastrulates exactly like deuterostomes. Furthermore, the expression patterns of molecular markers for the blastopore, mouth, and anus in priapulids follow expression patterns found in deuterostomes, rather than protostomes. These findings reveal that the terms Protostomia and Deuterostomia as labels for taxonomic purposes are no longer instructive and may actually obscure our understanding of the phylogenetic relationships between metazoan groups [3]. As is now apparent, using gastrulation as a criterion for describing major metazoan radiations turns out to have been an unfortunate choice as many forms of gastrulation (e.g., ingression, epiboly, delamination) do not generate an opening that can be associated with any larval/adult structure. Gastrulation patterns in many ‘minor’ metazoan taxa have not yet been carefully described and variation in gastrulation patterns even within individual metazoan groups [4,5] have not been thoroughly explored (Figure 1A).

Although previous workers have tried to make broad generalizations from observations of development, fate-mapping experiments are the only way to demonstrate the relationship between the site of gastrulation and the larval and/or adult body plan (Figure 1B). Such experiments have shown that in both protostomes [2,6] and deuterostomes [7] endoderm (i.e., gut) arises from cells derived from the vegetal pole of the egg. Cells from the animal pole give rise to oral and anterior neural structures associated with the feeding apparatus. In all deuterostomes and some protostomes [2], the site of gastrulation persists through development and becomes the anus. However, in protostomes the site of gastrulation (i.e., position of endoderm formation) rarely ever becomes the mouth, although differential growth often displaces vegetal tissue towards the oral opening. These data indicate that endodermal tissue originates from the vegetal pole in both protostomes and deuterostomes and that the mouth bears no conserved relationship to the site of gastrulation. It is tempting to speculate that this may be due to the wide range of modes of gastrulation observed in the protostomes and the conserved persistence of the blastopore in deuterostomes as a constraint of epithelial invagination.

Despite the lack of a relationship between the site of gastrulation and mouth formation, there is a growing body of evidence supporting the homology of the metazoan mouth [2,3], with the possible exception of the chordate mouth [8]. Fate-mapping experiments in cnidarians (e.g., sea anemones, corals, and ‘jellyfish’)—the sister group to the bilaterian clade (Figure 1A) — and ctenophores (an even earlier branching metazoan taxon) (Figure 1A), show that their mouths form from a region derived from the animal pole, just like bilaterians [9,10] (Figure 1B), and express many of the genes expressed around the mouth in bilaterians [2]. The homology of the metazoan mouth makes sense from a functional perspective as the adaptive significance of an intermediate stage with no, or multiple, mouths is doubtful.

There are two important differences in gastrulation between cnidarians and bilaterians. First, adult cnidarians possess a bifunctional gastrodermal layer lining the gastric cavity and an outer epidermal layer, but do not have a separate mesodermal germ layer (e.g., muscle, parenchyma, nephridia) characteristic for bilaterians. The cnidarian gastrodermis functions in both digestion and contraction and expresses genes typically involved in both bilaterian endoderm and mesoderm development [11–13], suggesting it represents an evolutionary precursor of both bilaterian endodermal and mesodermal tissue layers. Understanding the molecular basis for the development of the gastrodermis in cnidarians might provide tremendous insight into the evolutionary origins of distinct endodermal and mesodermal gene regulatory networks in bilaterians.

The second important difference between gastrulation and the formation of the mouth in cnidarians (and ctenophores) and bilaterians is that gastrulation occurs at the animal pole, not the vegetal pole, making them, by definition, the only true extant ‘protostome’ clades. Unfortunately, we do not know what this relationship is in the two other ‘prebilaterian’ taxa (Figure 1A). No viable embryos have ever been recovered from placozoans (e.g., Trichoplax) and a fate-map incorporating the primary egg axis has never been generated in any sponge species (not to mention that there is no clear tissue that can be homologized to bilaterian endoderm). The change in the site of gastrulation from the animal pole in cnidarians and ctenophores to the vegetal pole in bilaterians has been argued to be the most profound developmental change [9] — not the mode of gastrulation, or the existence of bilateral symmetry — that facilitated the tremendous radiation of bilaterian body plans [14].

A detailed understanding of the components of the cnidarian gastrodermal gene regulatory network is essential to understand the evolution of mesoderm [9] in the Metazoa. For example, ctenophores (Figure 1A) possess several mesodermal cell types (e.g., muscle cell and mesenchymal cells) and branched off from the rest of the metazoan lineage before the cnidarian–bilaterian ancestor. This implies that mesoderm evolved early in the Metazoa and was lost in cnidarians, placozoans, and possibly sponges (depending on the true phylogenetic position of sponges). The loss of mesodermal cell types in these lineages is surprising; however, this scenario assumes that the mesoderm in ctenophores is homologous to the mesoderm found in bilaterian taxa. If mesoderm evolved independently in ctenophores and bilaterians, it provides us with an opportunity to study the molecular basis for the appearance of similar cell types in animal evolution. Functional genomic approaches are being successfully employed to understand the development of mesodermal lineages in different echinoderm clades [15]. An analysis of the gene regulatory network underlying ctenophore mesoderm in comparison to the cnidarian bifunctional gastrodermis will be particularly interesting with respect to the evolutionary origin of mesodermal cell types.

Gastrulation is obviously an emergent property of complex metazoan body plans that allowed not only the formation and specialization of the lining of a digestive surface but, perhaps more importantly, the formation of internal tissues such as endoderm and mesoderm. The appearance of internal tissues allowed the evolution of complex inductive interactions between tissue layers both during and after gastrulation, broadly utilized in all bilaterian development, that would not be possible in a hollow ball of cells. It is of some interest that the lip of the blastopore in the cnidarian Nematostella has organizing activity [16] that can induce a new oral–aboral axis, likely due to the expression of several signaling family molecules such as BMPs, FGFs, and Wnts [17–19]. Thus, when the site of gastrulation changed in bilaterians, not only did the gene regulatory network that activates gastrodermal tissue change its spatial position, but the inductive activity responsible for axial patterning also moved along with it. A better understanding of the molecular basis of gastrulation will keep our bellies full of new insight into the developmental basis for body plan evolution, regardless of where our mouths might form.