356
7.8.3 The Temporal Gradient Model
Some recent experiments indicate that cells interpret the length of time they are
exposed to Nodal signals (Gritsman et al. 2000 ; Hagos and Dougan 2007 ; Hagos
et al. 2007 ). Late expression of the Oep/Cripto co-receptor can rescue notochord
formation in null oep mutants, but does not rescue formation of the prechordal plate
(Gritsman et al. 2000 ). This suggests that transient Nodal signaling is sufficient to
specify notochord, but continuous Nodal signaling is necessary to induce the pre-
chordal plate. Experiments using the drug SB431542 to conditionally inactivate or
activate the Type I Nodal receptor, ActRIb/ACVR1b at different times showed that
transient Nodal signaling can induce notochord at any point in the blastoderm stage,
and that specification of the prechordal plate requires prolonged exposure to Nodal
signals (Hagos and Dougan 2007 ; Hagos et al. 2007 ). Similar results were obtained
with other mesodermal cells types, indicating that this phenomenon is not restricted
to the shield. In a separate set of experiments, the embryo was exposed to uniformly
high levels of Nodal signals and the Nodal receptor was inactivated at different
times. When the receptor was blocked at early stages, all cells adopted a low thresh-
old response and expressed the notochord marker, no-tail (ntl). When the Nodal
receptor was blocked at later stages, cells adopted a high threshold response and
expressed the prechordal plate marker, goosecoid (gsc) (Hagos and Dougan 2007 ).
Experiments using microfluidic chambers to culture embryonic stem cells revealed
that mouse cells measure the rate of increase in Nodal signals as well as measuring
the absolute dose (Sorre et al. 2014 ).
These results suggest a Temporal Gradient Model of Nodal signaling (Fig. 7.11c).
According to this model, the Reaction-Diffusion interaction between Nodal and
Antivin/Lefty may act to establish a domain in which cells are exposed to Nodal
signals. Nodal signals from extraembryonic tissues, such as the zebrafish YSL or the
mouse VE may act to maintain a stable domain of Nodal expression despite the cell
movements in the blastula and gastrula stages. The Temporal Gradient Model could
explain how Nodal signals pattern dynamically moving and proliferative tissues,
such as the zebrafish margin or the amniote primitive streak. Cells that are next to
each other at any given point in time may have different histories of exposure to
Nodal signals based on their movement paths in the blastoderm and adopt different
fates. This could explain the juxtaposition of mesoderm and endoderm precursors in
many blastoderm fate maps, which can be viewed as a snapshot of the distribution of
precursors at a single moment of time. If the cell movements in the blastoderm are
random, then the Temporal Gradient Model predicts that mesodermal and endoder-
mal cell identities are specified stochastically. Precedent for stochastic specification
of cell fates comes from the invertebrate retina, in which cone cell identities are
chosen at random (Wernet et al. 2006 ; Perry et al. 2016 ). The Temporal Gradient
Model also predicts that long-term exposure to a low dose of Nodal signals will have
the same effect on cells as a short-term exposure to a high dose, which has been veri-
fied experimentally (Hagos and Dougan 2007 ). More experiments are necessary to
determine if cell movements can actually influence cell fates, and a refined Reaction-
Diffusion Model needs to be developed to take into account these movements.
W. Tseng et al.