192
Xenopus
a specifc biological process and provide a framework with
constraints for building GRNs. Xenopus tropicalis scRNAseq data were recently reported describing the emergence of
cell states across early embryogenesis (Briggs et al., 2018).
While these scRNA-seq data are useful in identifying the
lineages and the number of cell states appearing during
early Xenopus development, the current data are not of high
enough resolution to build GRNs due to the low sequencing
depth of scRNA-seq and frequent dropouts of transcripts.
Despite these challenges, the scRNA-seq approach will
become an integral part of GRN building as the integration of these data with other high-throughput data (e.g., bulk
RNA-seq, DNA-seq) becomes easier.
12.4. OUR CONTRIBUTION TO
THIS FIELD—INSIGHTS
Hans Spemann and Hilde Mangold discovered the organizer in 1924, a tissue above the dorsal blastopore lip in
the amphibian early gastrula that induced development of a
secondary body axis containing a central nervous system.
In 1991, we cloned the frst gene expressed specif cally in
the organizer, which encodes Goosecoid (Gsc), a homeodomain TF, and began investigating the molecular mechanism
regulating gsc expression using a reporter gene microinjection assay (Watabe et al., 1995; Laurent et al., 1997). The
importance of synergistic inputs directly into gsc CRMs,
mediated by TFs that are regulated by Tgf-β and Wnt growth
factors, was demonstrated for Spemann organizer formation.
Our frst attempt to build a GRN controlling mesendoderm
development was published in 2005 (Koide et al., 2005),
and in 2017, we updated the network with new information
(Charney et al., 2017a). While we had established the most
comprehensive vertebrate mesendodermal GRN at the time,
it provided only a limited view of the process due to the inherent limitations of the traditional one-gene-at-a-time approach
used to address biological questions. To initiate systems-level
analyses of GRN function, we adopted computational modeling and deep sequencing to infer a more comprehensive
architecture of the mesendoderm GRN (Chiu et al., 2014;
Charney et al., 2017b; Paraiso et al., 2019; Afouda et al.,
2020; Mukherjee et al., 2020; Jansen et al., 2022).
Over the past several years, we have spent considerable
energy to develop genomic resources to better understand
GRNs in early development. We have determined the absolute amount of all known transcripts and the kinetics of their
accumulation in X. tropicalis embryos from 1-cell to tadpole (organogenesis) stages ( Owens et al., 2016). We mapped
lncRNAs that are expressed during Xenopus development
(Forouzmand et al., 2017) and created a manually curated
catalog of all 1240 TFs encoded by the X. tropicalis genome
(Blitz et al., 2017). Through such analyses, we have investigated the roles of maternal TFs in mesendoderm gene activation during zygotic genome activation. Upon examining
the DNA binding behaviors of several critical maternal TFs
(Fox, Sox, T-box, homeodomain, Lef/Tcf families) important for germ layer specifcation, we and others discovered
that these maternal TFs bind to the genome at select sites
before any sign of gene transcription (Charney et al., 2017b;
Paraiso et al., 2019; Gentsch et al., 2019). Furthermore,
many of these TFs (Foxh1, Otx1, Vegt, Tcf, Sox3/7) co-bind
and form enhanceosomal complexes on endodermal CRMs
and are responsible for the activity of super-enhancers (SEs)
(Paraiso et al., 2019; Gentsch et al., 2019). These data support the notion that maternal TFs act as pioneer factors,
binding to select CRMs to pre-mark the genome in advance
of activation of the subsequent germ layer-specif c GRN.
12.5. FUTURE DIRECTIONS AND
IMPORTANT QUESTIONS
The insights into regulatory functions of GRNs will uncover
new network connections that will generate specif c hypotheses on the molecular components that are identif ed. GRN
building based on genomic data will provide an enormous
number of new links between TFs and CRMs, and it is
necessary to test the validity of the connections experimentally using high-throughput approaches. Development
of both single- and multi-locus gene perturbation assays
through the use of morpholinos, genome editing techniques
such as CRISPR/Cas9 (Blitz et al., 2013; Nakayama et al.,
2013; Guo et al., 2014) and RNA cleaving Cas13 derivatives
(Kushawah et al., 2020) will be extremely valuable in accelerating the rate of validation analysis to build GRNs based
on genomic data.
A comprehensive understanding of the logic of GRNs
should one day inform synthetic biology approaches, by
which we can create or manipulate regulatory genetic circuits
to modify cell functions. For example, by manipulating key
nodes in a GRN, we could prevent or reverse a diseased cell
state. Changing the identity of terminally differentiated cells
or progenitor cells toward other cell types for regenerative
medicine may be achieved by redesigning network circuits
regulating the specifcation of cell fates. This realization can
be also assisted by the use of CRISPR/Cas systems. Based on
the current state of the GRN feld, we do not have suff cient
information to understand how GRNs control the diverse
biological functions in various organisms, nor are we in the
position to predict the changes in GRN function underlying
human disease. However, provisional Xenopus GRNs regulating neural crest, ectodermal placodes, blood differentiation, eye, and podocyte development have also been reported
(Seal et al., 2020, Maharana and Schlosser, 2018; Ciau-Uitz
and Patient, 2019; Lee et al., 2014; Zuber et al., 2003; White
et al., 2010). In the future, our current early mesendodermal
GRN network should be expanded, refned, and linked to
later GRNs to expand our current knowledge on the architecture of GRNs that control biology.
Genome-wide association studies (GWAS) have been a
useful approach to link specifc genetic variations such as
single-nucleotide polymorphisms (SNPs) with particular
diseases. Many GWAS publications have identif ed diseaseand trait-associated SNPs in the human genome, and SNPs
mapped within transcriptional regulatory regions were
Xenopus
a specifc biological process and provide a framework with
constraints for building GRNs. Xenopus tropicalis scRNAseq data were recently reported describing the emergence of
cell states across early embryogenesis (Briggs et al., 2018).
While these scRNA-seq data are useful in identifying the
lineages and the number of cell states appearing during
early Xenopus development, the current data are not of high
enough resolution to build GRNs due to the low sequencing
depth of scRNA-seq and frequent dropouts of transcripts.
Despite these challenges, the scRNA-seq approach will
become an integral part of GRN building as the integration of these data with other high-throughput data (e.g., bulk
RNA-seq, DNA-seq) becomes easier.
12.4. OUR CONTRIBUTION TO
THIS FIELD—INSIGHTS
Hans Spemann and Hilde Mangold discovered the organizer in 1924, a tissue above the dorsal blastopore lip in
the amphibian early gastrula that induced development of a
secondary body axis containing a central nervous system.
In 1991, we cloned the frst gene expressed specif cally in
the organizer, which encodes Goosecoid (Gsc), a homeodomain TF, and began investigating the molecular mechanism
regulating gsc expression using a reporter gene microinjection assay (Watabe et al., 1995; Laurent et al., 1997). The
importance of synergistic inputs directly into gsc CRMs,
mediated by TFs that are regulated by Tgf-β and Wnt growth
factors, was demonstrated for Spemann organizer formation.
Our frst attempt to build a GRN controlling mesendoderm
development was published in 2005 (Koide et al., 2005),
and in 2017, we updated the network with new information
(Charney et al., 2017a). While we had established the most
comprehensive vertebrate mesendodermal GRN at the time,
it provided only a limited view of the process due to the inherent limitations of the traditional one-gene-at-a-time approach
used to address biological questions. To initiate systems-level
analyses of GRN function, we adopted computational modeling and deep sequencing to infer a more comprehensive
architecture of the mesendoderm GRN (Chiu et al., 2014;
Charney et al., 2017b; Paraiso et al., 2019; Afouda et al.,
2020; Mukherjee et al., 2020; Jansen et al., 2022).
Over the past several years, we have spent considerable
energy to develop genomic resources to better understand
GRNs in early development. We have determined the absolute amount of all known transcripts and the kinetics of their
accumulation in X. tropicalis embryos from 1-cell to tadpole (organogenesis) stages ( Owens et al., 2016). We mapped
lncRNAs that are expressed during Xenopus development
(Forouzmand et al., 2017) and created a manually curated
catalog of all 1240 TFs encoded by the X. tropicalis genome
(Blitz et al., 2017). Through such analyses, we have investigated the roles of maternal TFs in mesendoderm gene activation during zygotic genome activation. Upon examining
the DNA binding behaviors of several critical maternal TFs
(Fox, Sox, T-box, homeodomain, Lef/Tcf families) important for germ layer specifcation, we and others discovered
that these maternal TFs bind to the genome at select sites
before any sign of gene transcription (Charney et al., 2017b;
Paraiso et al., 2019; Gentsch et al., 2019). Furthermore,
many of these TFs (Foxh1, Otx1, Vegt, Tcf, Sox3/7) co-bind
and form enhanceosomal complexes on endodermal CRMs
and are responsible for the activity of super-enhancers (SEs)
(Paraiso et al., 2019; Gentsch et al., 2019). These data support the notion that maternal TFs act as pioneer factors,
binding to select CRMs to pre-mark the genome in advance
of activation of the subsequent germ layer-specif c GRN.
12.5. FUTURE DIRECTIONS AND
IMPORTANT QUESTIONS
The insights into regulatory functions of GRNs will uncover
new network connections that will generate specif c hypotheses on the molecular components that are identif ed. GRN
building based on genomic data will provide an enormous
number of new links between TFs and CRMs, and it is
necessary to test the validity of the connections experimentally using high-throughput approaches. Development
of both single- and multi-locus gene perturbation assays
through the use of morpholinos, genome editing techniques
such as CRISPR/Cas9 (Blitz et al., 2013; Nakayama et al.,
2013; Guo et al., 2014) and RNA cleaving Cas13 derivatives
(Kushawah et al., 2020) will be extremely valuable in accelerating the rate of validation analysis to build GRNs based
on genomic data.
A comprehensive understanding of the logic of GRNs
should one day inform synthetic biology approaches, by
which we can create or manipulate regulatory genetic circuits
to modify cell functions. For example, by manipulating key
nodes in a GRN, we could prevent or reverse a diseased cell
state. Changing the identity of terminally differentiated cells
or progenitor cells toward other cell types for regenerative
medicine may be achieved by redesigning network circuits
regulating the specifcation of cell fates. This realization can
be also assisted by the use of CRISPR/Cas systems. Based on
the current state of the GRN feld, we do not have suff cient
information to understand how GRNs control the diverse
biological functions in various organisms, nor are we in the
position to predict the changes in GRN function underlying
human disease. However, provisional Xenopus GRNs regulating neural crest, ectodermal placodes, blood differentiation, eye, and podocyte development have also been reported
(Seal et al., 2020, Maharana and Schlosser, 2018; Ciau-Uitz
and Patient, 2019; Lee et al., 2014; Zuber et al., 2003; White
et al., 2010). In the future, our current early mesendodermal
GRN network should be expanded, refned, and linked to
later GRNs to expand our current knowledge on the architecture of GRNs that control biology.
Genome-wide association studies (GWAS) have been a
useful approach to link specifc genetic variations such as
single-nucleotide polymorphisms (SNPs) with particular
diseases. Many GWAS publications have identif ed diseaseand trait-associated SNPs in the human genome, and SNPs
mapped within transcriptional regulatory regions were