420
RNA encoding the germ cell-specific protein dazl, which is a component of Xenopus
germ plasm (Houston et al. 1998 ), is not similarly localized in axolotl oocytes. On
this basis they concluded that urodele oocytes do not contain germ plasm, implying
a fundamental difference in the development of anuran and urodele embryos.
However, in contrast to previous work, these authors considered the disparate mech-
anisms for PGC specification within a modern phylogenetic context in which extant
amphibians are known to have evolved from a common tetrapod ancestor. Because
urodeles retain basal amphibian traits, relative to anurans, they proposed that PGC
specification by induction, also known as epigenesis, is conserved in vertebrates.
On the other hand, they postulated that germ plasm was an evolutionary innovation
of frogs, and other organisms (Johnson et al. 2001 ). This hypothesis conflicted
directly with the prevailing assumption that germ plasm, in some form or other, is
conserved across the animal kingdom. In later work, Johnson and colleagues postu-
lated that germ plasm has evolved by convergence throughout the animal kingdom
in response to selective pressures (Johnson et al. 2003a, b, 2011 ). This controversial
hypothesis enhances the concept that germ cells are not fundamentally different
than somatic cells. Furthermore, it suggests a heretofore unappreciated ability of
organisms to evolve novel mechanisms for the control of early development.
Using axolotl embryos as a model, Chatfield et al. ( 2014 ) defined the basal
mechanism for PGC specification in vertebrates. By modulating gene expression
in vivo, or in vitro (using the animal cap system), they showed that PGCs are induced
by a combination of fibroblast growth factor (FGF) and bone morphogenetic protein
(BMP) signaling. Moreover, it was possible to expand or eliminate the PGC pool by
modulating FGF signaling levels, suggesting that presumptive somatic cells could be
recruited to germ line development, or vice versa, in response to an appropriate signal-
ing regime. This supports the hypothesis that PGCs are derived from pluripotent pre-
cursors. Accordingly, the cells that make up axolotl animal caps express conserved
orthologs of the pluripotency genes oct4 and nanog (Bachvarova et al. 2004 ; Dixon
et al. 2010 ), which are not found in Xenopus (Frankenberg and Renfree 2013 ; Hellsten
et al. 2010 ). These transcription factors are required to establish the ground state of
pluripotency in mammals, suggesting that the mechanisms which govern early verte-
brate development are conserved between urodeles and mammals.
Chatfield et al. ( 2014 ) also showed that downstream of FGF, PGCs could be
induced, along with related somatic cell types, by the mesodermal determinant
Brachyury. In addition, they showed that pluripotent cells in the animal cap were
directed to either a germ line or somatic fate by the competing effects of extracellular
signaling. These data suggest that PGC specification in axolotls is a stochastic event,
which does not involve the early acting germ line determinants found in other animal
models. Further, these authors used a variety of methods to determine when cells
undergo restriction to the germ lineage in axolotls. They showed that irreversible germ
line commitment occurs during tailbud stages, which is days after the completion of
gastrulation. Prior to this, PGCs can readily be converted to somatic cells. Up until
lineage restriction, germ cell potential is maintained within a multipotent mesodermal
domain by MAP kinase signaling. The mechanisms acting downstream of MAPK
signaling are presently unknown; however, their abrogation eliminates the PGCs
T. Aguero et al.