127
Evolution of the Vertebrate Neural Crest
8.3. INSIGHTS INTO NEURAL CREST
DEVELOPMENT FROM XENOPUS
IN THE ERA OF MOLECULAR
BIOLOGY AND “-OMICS”
Up until the 1970s and 1980s, nearly all of our understanding of the mechanisms of embryonic development in
amphibians relied on either observation or relatively simple
methods of experimentation and biochemical characterization (Nieuwkoop 1973; Gurdon and Hopwood 2003). While
these approaches, led in large part by the pioneering work of
Peter Nieuwkoop and David Newth, fundamentally advanced
our understanding of vertebrate development, they could not
provide a mechanistic, molecular-genetic understanding of
these processes. The development of widely used molecular
techniques such as in situ hybridization and immunohistochemistry (Clarke et al. 1984; Hemmati-Brivanlou et al.
1990) positioned Xenopus as a system that could rapidly
provide key insights into the roles of molecules and genes
during vertebrate development (Gurdon and Hopwood 2003).
Within the last two decades, an expanded arsenal of powerful
new technologies, including morpholinos and antisense RNA
(Heasman et al. 2000), CRISPR/Cas9-mediated genome
editing (Nakayama et al. 2013; Aslan et al. 2017; DeLay et al.
2018), unbiased genome-wide approaches (RNA-Seq, ChIPSeq, ATAC-Seq) (Hellsten et al. 2010; Bright and Veenstra
2019; Gentsch and Smith 2019; Hontelez et al. 2019), transgenic lines (Alkobtawi et al. 2018; Li et al. 2019), proteomics
(Wühr et al. 2014), and single-cell sequencing (Briggs et al.
2018) became available, greatly expanding the Xenopus
developmental biologists’ toolkit and providing unparalleled
insights into embryo development.
Use of these tools in Xenopus has signif cantly enhanced
our understanding of the complex, interconnected web of
molecular-genetic interactions that drive neural crest development (Prasad et al. 2012). These gene-regulatory interactions progressively drive the development of the neural crest
from an early stem cell state through lineage diversif cation
and differentiation into derivatives such as cartilage, neurons, and pigment cells (Meulemans and Bronner-Fraser
2004; Prasad et al. 2012). Although the neural crest gene
regulatory network (GRN) is spatially and temporally continuous and hence cannot be reduced into fully separable
components, we can nonetheless identify distinct GRN
“subcircuits” or “modules” that serve dedicated functions at
certain times and places during development. For example,
some of these subcircuits direct cell migration, whereas
others govern pluripotency and lineage restriction. In the
following, we describe our current understanding of these
processes in Xenopus, focusing on the functional roles of
evolutionarily conserved GRN subcircuits.
8.3.1. NEURAL CREST PROGENITORS AND ESTABLISHMENT
OF THE NEURAL PLATE BORDER
Defnitive neural crest cells can frst be identif ed using
molecular markers during gastrulation (~stage 11/12) in
Xenopus (Mayor et al. 1995; O’Donnell et al. 2006). The
importance of tissue interactions in establishing the neural
crest was demonstrated in Xenopus as early as the 1940s
using in vivo grafting studies that juxtaposed lateral archenteron roof (paraxial and lateral plate mesoderm precursors)
to non-neural ectoderm (Raven and Kloos 1945). This was
later confrmed using molecular markers: combining nonneural ectoderm and mesoderm explants was suff cient to
induce expression of snai2 (one of the f rst identif ed neural crest markers, formerly known as slug) and promote
melanocyte formation (Bonstein et al. 1998). Such studies led to a search for the signals released by these tissues.
As in other vertebrates, establishment of def nitive neural
crest involves the combined activity of several evolutionarily conserved signaling pathways (Mayor et al. 1995;
LaBonne and Bronner-Fraser 1998; Klymkowsky et al.
2010; Prasad et al. 2012). This includes intermediate levels
of BMP activity, in contrast to the high and low levels that
characterize the epidermal and neural ectoderm, respectively (Wilson and Hemmati-Brivanlou 1995; Weinstein
and Hemmati-Brivanlou 1997; LaBonne and Bronner-Fraser
1998; Marchant et al. 1998). Decreased levels of BMP signaling are achieved via the activity of secreted BMP inhibitors, such as noggin (Lamb et al. 1993; Zimmerman et al.
1996 ), chordin (Piccolo et al. 1996 ), follistatin (Fainsod et
al. 1997; Iemura et al. 1998), and gremlin 1 (Hsu et al. 1998),
as well as other genes and pathways that modulate BMP
activity directly or indirectly (e.g. snai2, tsku, dll1/notch1,
traf4, snw1) (Glavic et al. 2004; Kalkan et al. 2009; Wu and
Hill 2009; Shi et al. 2011; Wu et al. 2011). Intermediate levels of BMP activity are not suffcient to establish a neural
crest state, however (LaBonne and Bronner-Fraser 1998).
There is evidence that both canonical and non-canonical
Wnt signaling and Wnt-mediated FGF signaling are key
players in this step (Mayor et al. 1997; Chang and HemmatiBrivanlou 1998; Villanueva et al. 2002; Monsoro-Burq et al.
2003; Wu et al. 2005; Abu-Elmagd et al. 2006; Hong et al.
2008; Steventon et al. 2009; Borday et al. 2018). Inhibition
of either signaling pathway abrogates expression of neural
crest markers and results in loss of neural crest derivatives,
whereas enhanced signaling can expand the pool of neural
crest progenitors. Work in Xenopus has made major contributions to identifying the novel proteins and mechanisms
that amplify or attenuate these signals, including Kctd15,
Apoc1, Hes3, Dkk2, Adam33, Adam19, Szl, and Daam1
(Table 8.1). There is also evidence for roles for Dll1/Notch1
(Glavic et al. 2004; Kuriyama et al. 2006), and retinoic acid
pathways (Li et al. 2018) in establishing neural crest, as well
as from other genes typically studied at slightly later stages
of neural crest ontogeny, such as tfap2a (Luo et al. 2003) and
adam33 (Wei et al. 2010). In addition to signals regulating
nuclear gene expression, some studies have identif ed pathways and proteins linked to mechanical processes including control of the cytoskeleton and tight junction assembly
as important for early neural crest formation. These include
Rhov (Guémar et al. 2007) and Marveld3-mediated attenuation of Jnk/Mapk8 signaling (Vacca et al. 2018). Similar
Evolution of the Vertebrate Neural Crest
8.3. INSIGHTS INTO NEURAL CREST
DEVELOPMENT FROM XENOPUS
IN THE ERA OF MOLECULAR
BIOLOGY AND “-OMICS”
Up until the 1970s and 1980s, nearly all of our understanding of the mechanisms of embryonic development in
amphibians relied on either observation or relatively simple
methods of experimentation and biochemical characterization (Nieuwkoop 1973; Gurdon and Hopwood 2003). While
these approaches, led in large part by the pioneering work of
Peter Nieuwkoop and David Newth, fundamentally advanced
our understanding of vertebrate development, they could not
provide a mechanistic, molecular-genetic understanding of
these processes. The development of widely used molecular
techniques such as in situ hybridization and immunohistochemistry (Clarke et al. 1984; Hemmati-Brivanlou et al.
1990) positioned Xenopus as a system that could rapidly
provide key insights into the roles of molecules and genes
during vertebrate development (Gurdon and Hopwood 2003).
Within the last two decades, an expanded arsenal of powerful
new technologies, including morpholinos and antisense RNA
(Heasman et al. 2000), CRISPR/Cas9-mediated genome
editing (Nakayama et al. 2013; Aslan et al. 2017; DeLay et al.
2018), unbiased genome-wide approaches (RNA-Seq, ChIPSeq, ATAC-Seq) (Hellsten et al. 2010; Bright and Veenstra
2019; Gentsch and Smith 2019; Hontelez et al. 2019), transgenic lines (Alkobtawi et al. 2018; Li et al. 2019), proteomics
(Wühr et al. 2014), and single-cell sequencing (Briggs et al.
2018) became available, greatly expanding the Xenopus
developmental biologists’ toolkit and providing unparalleled
insights into embryo development.
Use of these tools in Xenopus has signif cantly enhanced
our understanding of the complex, interconnected web of
molecular-genetic interactions that drive neural crest development (Prasad et al. 2012). These gene-regulatory interactions progressively drive the development of the neural crest
from an early stem cell state through lineage diversif cation
and differentiation into derivatives such as cartilage, neurons, and pigment cells (Meulemans and Bronner-Fraser
2004; Prasad et al. 2012). Although the neural crest gene
regulatory network (GRN) is spatially and temporally continuous and hence cannot be reduced into fully separable
components, we can nonetheless identify distinct GRN
“subcircuits” or “modules” that serve dedicated functions at
certain times and places during development. For example,
some of these subcircuits direct cell migration, whereas
others govern pluripotency and lineage restriction. In the
following, we describe our current understanding of these
processes in Xenopus, focusing on the functional roles of
evolutionarily conserved GRN subcircuits.
8.3.1. NEURAL CREST PROGENITORS AND ESTABLISHMENT
OF THE NEURAL PLATE BORDER
Defnitive neural crest cells can frst be identif ed using
molecular markers during gastrulation (~stage 11/12) in
Xenopus (Mayor et al. 1995; O’Donnell et al. 2006). The
importance of tissue interactions in establishing the neural
crest was demonstrated in Xenopus as early as the 1940s
using in vivo grafting studies that juxtaposed lateral archenteron roof (paraxial and lateral plate mesoderm precursors)
to non-neural ectoderm (Raven and Kloos 1945). This was
later confrmed using molecular markers: combining nonneural ectoderm and mesoderm explants was suff cient to
induce expression of snai2 (one of the f rst identif ed neural crest markers, formerly known as slug) and promote
melanocyte formation (Bonstein et al. 1998). Such studies led to a search for the signals released by these tissues.
As in other vertebrates, establishment of def nitive neural
crest involves the combined activity of several evolutionarily conserved signaling pathways (Mayor et al. 1995;
LaBonne and Bronner-Fraser 1998; Klymkowsky et al.
2010; Prasad et al. 2012). This includes intermediate levels
of BMP activity, in contrast to the high and low levels that
characterize the epidermal and neural ectoderm, respectively (Wilson and Hemmati-Brivanlou 1995; Weinstein
and Hemmati-Brivanlou 1997; LaBonne and Bronner-Fraser
1998; Marchant et al. 1998). Decreased levels of BMP signaling are achieved via the activity of secreted BMP inhibitors, such as noggin (Lamb et al. 1993; Zimmerman et al.
1996 ), chordin (Piccolo et al. 1996 ), follistatin (Fainsod et
al. 1997; Iemura et al. 1998), and gremlin 1 (Hsu et al. 1998),
as well as other genes and pathways that modulate BMP
activity directly or indirectly (e.g. snai2, tsku, dll1/notch1,
traf4, snw1) (Glavic et al. 2004; Kalkan et al. 2009; Wu and
Hill 2009; Shi et al. 2011; Wu et al. 2011). Intermediate levels of BMP activity are not suffcient to establish a neural
crest state, however (LaBonne and Bronner-Fraser 1998).
There is evidence that both canonical and non-canonical
Wnt signaling and Wnt-mediated FGF signaling are key
players in this step (Mayor et al. 1997; Chang and HemmatiBrivanlou 1998; Villanueva et al. 2002; Monsoro-Burq et al.
2003; Wu et al. 2005; Abu-Elmagd et al. 2006; Hong et al.
2008; Steventon et al. 2009; Borday et al. 2018). Inhibition
of either signaling pathway abrogates expression of neural
crest markers and results in loss of neural crest derivatives,
whereas enhanced signaling can expand the pool of neural
crest progenitors. Work in Xenopus has made major contributions to identifying the novel proteins and mechanisms
that amplify or attenuate these signals, including Kctd15,
Apoc1, Hes3, Dkk2, Adam33, Adam19, Szl, and Daam1
(Table 8.1). There is also evidence for roles for Dll1/Notch1
(Glavic et al. 2004; Kuriyama et al. 2006), and retinoic acid
pathways (Li et al. 2018) in establishing neural crest, as well
as from other genes typically studied at slightly later stages
of neural crest ontogeny, such as tfap2a (Luo et al. 2003) and
adam33 (Wei et al. 2010). In addition to signals regulating
nuclear gene expression, some studies have identif ed pathways and proteins linked to mechanical processes including control of the cytoskeleton and tight junction assembly
as important for early neural crest formation. These include
Rhov (Guémar et al. 2007) and Marveld3-mediated attenuation of Jnk/Mapk8 signaling (Vacca et al. 2018). Similar