354
Other studies indicate that Nodal signaling is required to pattern the animal–veg-
etal (anteroposterior axis) instead of the dorsoventral axis. Fate mapping and gene
expression analysis in ndr1/sqt;ndr2/cyc double mutants revealed that marginal
cells adopt neural fates in the complete absence of Nodal signaling (Feldman et al.
2000 ). Similar results were observed in mutants in which the levels of the Nodal
co-receptor Oep are reduced or completely eliminated (Gritsman et al. 1999 , 2000 ).
Finally, fate mapping of sqt−/−;cyc+/+ and sqt−/−;cyc+/− embryos, in which Nodal
signaling is reduced to different extents, showed that marginal cells adopt more
animal (anterior) fates when Nodal levels are abnormally low (Dougan et al. 2003 ).
This analysis also showed that dorsal cells require higher levels of Nodal signaling
to adopt marginal fates than do cells in the ventrolateral margin.
7.8.2 The Nodal Reaction-Diffusion Mechanism
Nodal signaling fits all the criteria laid out by Turing’s Reaction-Diffusion mecha-
nism (Solnica-Krezel 2003 ). As predicted by Turing, Nodal proteins induce their
own expression as well as that of a Nodal antagonist, Antivin/Lefty (Fig. 7.11a)
(Meno et al. 1999 ; Thisse and Thisse 1999 ; Cheng et al. 2000 ). Consistent with the
Reaction-Diffusion Model, Nodal-related proteins diffuse farther than normal in the
absence of Antivin/Lefty, and the mesoderm is expanded (Meno et al. 2001 ;
Branford and Yost 2002 ; Chen and Schier 2002 ; Feldman et al. 2002 ; Sakuma et al.
2002 ). The prediction of the Reaction-Diffusion model that the antagonist should
diffuse faster than the inducer has been more difficult to test. Fluorescently tagged
versions of Nodal and Lefty2 showed that Lefty had a longer range than Nodal
when overexpressed in Xenopus animal caps (Sakuma et al. 2002 ). The different
ranges of the two proteins could be explained by differential degradation or by dif-
ferential rates of diffusion. To distinguish between these possibilities, Ndr1/Sqt,
Ndr2/Cyc, and Lefty were tagged with GFP or a photoconvertable fluorescent pro-
tein, Dendra2.5 (Muller et al. 2012 ). Cells overexpressing the tagged proteins were
transplanted into a wild type host embryo, resulting in a localized source of the
signaling molecule in the animal pole. The kinetics of diffusion and degradation
were determined by measuring the rates of recovery after photobleaching or photo-
conversion. These studies ruled out the idea that Nodal and Lefty proteins were
differentially degraded, and demonstrated that Antivin/Lefty proteins diffuse at a
faster rate than either tagged Ndr1/Sqt or Ndr2/Cyc in the zebrafish blastoderm.
Differential regulation of translation by microRNA-430, which delays translation of
Antivin/Lefty, contributes to the different ranges of Nodal and Antivin/Lefty in the
zebrafish (van Boxtel et al. 2015 ). This data supports a model in which diffusion of
a Nodal protein establishes a spatial gradient within the responding tissue (Fig.
7.11b). At short distances from the source of Nodal signals, Nodal activity predomi-
nates and Nodal signaling induces its own expression, the expression of Antivin/
Lefty and mesodermal and endodermal target genes. At long distances from the
source of Nodal signals, the concentration of Antivin/Lefty increases above the
threshold required to inhibit Nodal autoregulation and mesoderm induction. Cell
W. Tseng et al.