The germ line is established in animal embryos with the formation of primordial germ cells (PGCs), which give rise to gametes. Therefore, the need to form PGCs can act as a developmental constraint by inhibiting the evolution of embryonic patterning mechanisms that compromise their development. Conversely, events that stabilize the PGCs may liberate these constraints. Two modes of germ cell determination exist in animal embryos: (a) either PGCs are predetermined by the inheritance of germ cell determinants (germ plasm) or (b) PGCs are formed by inducing signals secreted by embryonic tissues (i.e., regulative determination). Surprisingly, among the major extant amphibian lineages, one mechanism is found in urodeles and the other in anurans. In anuran amphibians PGCs are predetermined by germ plasm; in urodele amphibians PGCs are formed by inducing signals. To determine which mechanism is ancestral to the tetrapod lineage and to understand the pattern of inheritance in higher vertebrates, we used a phylogenetic approach to analyze basic morphological processes in both groups and correlated these with mechanisms of germ cell determination. Our results indicate that regulative germ cell determination is a property of embryos retaining ancestral embryological processes, whereas predetermined germ cells are found in embryos with derived morphological traits. These correlations suggest that regulative germ cell formation is an important developmental constraint in vertebrate embryos, acting before the highly conserved pharyngula stage. Moreover, our analysis suggests that germ plasm has evolved independently in several lineages of vertebrate embryos.

How germ cell specification occurs remains a fundamental question in embryogenesis. The embryos of several model organisms contain germ cell determinants (germ plasm) that segregate to germ cell precursors. In other animals, including mice, germ cells form in response to regulative mechanisms during development. To investigate germ cell determination in urodeles, where germ plasm has never been conclusively identified, we cloned a DAZ-like sequence from axolotls, Axdazl. Axdazl is homologous to Xdazl, a component of Xenopus germ plasm found in the vegetal pole of oocytes and eggs. Axdazl RNA is not localized in axolotl oocytes, and, furthermore, these oocytes do not contain the mitochondrial cloud that localizes Xdazl and other germ plasm components in Xenopus. Maternal Axdazl RNA is inherited in the animal cap and equatorial region of early embryos. At gastrula, neurula, and tailbud stages, Axdazl RNA is widely distributed. Axdazl first shows cell-specific expression in primordial germ cells (PGCs) approaching the gonad at stage 40, when nuage (germ plasm) appears in PGCs. These results suggest that, in axolotls, germ plasm components are insufficient to specify germ cells.

How germ cells are specified in the embryos of animals has been a mystery for decades. Unlike most developmental processes, which are highly conserved, embryos specify germ cells in very different ways. Curiously, in mouse embryos germ cells are specified by extracellular signals; they are not autonomously specified by maternal germ cell determinants (germ plasm), as are the germ cells in most animal model systems. We have developed the axolotl (Ambystoma mexicanum), a salamander, as an experimental system, because classic experiments have shown that the germ cells in this species are induced by extracellular signals in the absence of germ plasm. Here, we provide evidence that the germ cells in axolotls arise from naive mesoderm in response to simple inducing agents. In addition, by analysing the sequences of axolotl germ-cell-specific genes, we provide evidence that mice and urodele amphibians share a common mechanism of germ cell development that is ancestral to tetrapods. Our results imply that germ plasm, as found in species such as frogs and teleosts, is the result of convergent evolution. We discuss the evolutionary implications of our findings.

No abstract