*For correspondence:
angelike@caltech.edu
Competing interests:
The
authors declare that no
competing interests exist.
Funding:
See page 24
Received:
26 November 2019
Accepted:
22 July 2020
Published:
23 July 2020
Reviewing editor:
Oliver
Hobert, Howard Hughes Medical
Institute, Columbia University,
United States
Copyright Koromila et al. This
article is distributed under the
terms of the
Creative Commons
Attribution License,
which
permits unrestricted use and
redistribution provided that the
original author and source are
credited.
Odd-paired is a pioneer-like factor that
coordinates with Zelda to control gene
expression in embryos
Theodora Koromila
1
, Fan Gao
1
, Yasuno Iwasaki
2
, Peng He
1
, Lior Pachter
1
,
J Peter Gergen
2
, Angelike Stathopoulos
1
*
1
California Institute of Technology, Division of Biology and Biological Engineering,
Pasadena, United States;
2
Stony Brook University, Department of Biochemistry and
Cell Biology and Center for Developmental Genetics, Stony Brook, United States
Abstract
Pioneer factors such as Zelda (Zld) help initiate zygotic transcription in
Drosophila
early
embryos, but whether other factors support this dynamic process is unclear. Odd-paired (Opa), a
zinc-finger transcription factor expressed at cellularization, controls the transition of genes from
pair-rule to segmental patterns along the anterior-posterior axis. Finding that Opa also regulates
expression through enhancer
sog_Distal
along the dorso-ventral axis, we hypothesized Opa’s role
is more general. Chromatin-immunoprecipitation (ChIP-seq) confirmed its in vivo binding to
sog_Distal
but also identified widespread binding throughout the genome, comparable to Zld.
Furthermore, chromatin assays (ATAC-seq) demonstrate that Opa, like Zld, influences chromatin
accessibility genome-wide at cellularization, suggesting both are pioneer factors with common as
well as distinct targets. Lastly, embryos lacking
opa
exhibit widespread, late patterning defects
spanning both axes. Collectively, these data suggest Opa is a general timing factor and likely late-
acting pioneer factor that drives a secondary wave of zygotic gene expression.
Introduction
The transition from dependence on maternal transcripts deposited into the egg to newly transcribed
zygotic transcripts is carefully regulated to ensure proper development of early embryos. During the
maternal-to-zygotic transition (MZT), maternal products are cleared and zygotic genome activation
occurs (rev. in
Vastenhouw et al., 2019
;
Hamm and Harrison, 2018
). In
Drosophila
embryos, the
first 13 mitotic divisions involve rapid nuclear cycles (nc), that include only a short DNA replication S
phase and no G2 phase, and the nuclei are not enclosed in separate membrane compartments but
instead present in a shared cytoplasm (
Foe and Alberts, 1983
). This streamlined division cycle likely
relates to the fast development of
Drosophila
embryos, permitting rapid increase in cell number
before gastrulation in a matter of a few hours. Gene expression is initiated during the early syncytial
stage, as early as nc7, and continues to the cellularized blastoderm stage (
Ali-Murthy and Korn-
berg, 2016
;
Lott et al., 2011
;
Kwasnieski et al., 2019
). Gene expression patterns may be transient
or continuous, lasting through gastrulation or beyond (
Kvon et al., 2014
). This process is controlled
by a specific class of transcription factors called pioneer factors, which bind to closed chromatin cis-
regulatory regions to create accessible binding sites for additional transcription factors during devel-
opment (
Iwafuchi-Doi and Zaret, 2014
). The pioneer Zelda (Zld) is a ubiquitous, maternal factor
that binds to promoters of the earliest zygotically expressed genes and primes them for activation
(
Liang et al., 2008
;
Harrison et al., 2010
;
Harrison et al., 2011
). It was unknown whether a similar
regulation exists in other animals until the zebrafish Pou5f1, homolog of mammalian Oct4, was
shown to act in an analogous manner to Zld in that it controls zygotic gene activation in vertebrates
(
Leichsenring et al., 2013
).
Koromila
et al
. eLife 2020;9:e59610.
DOI: https://doi.org/10.7554/eLife.59610
1 of 30
RESEARCH ARTICLE
A complete understanding of how widespread activation of zygotic gene expression is achieved
is lacking, but several regulatory mechanisms have been proposed. One model suggests that a
decrease in histone levels over time due to dilution during nuclear division provides an opportunity
for the pioneer factors that drive zygotic gene expression to successfully compete for DNA access
and activate transcription (
Shindo and Amodeo, 2019
;
Hamm and Harrison, 2018
). Chromatin
accessibility can also be more specifically modulated by targeted action of transcriptional factors at
regulatory loci. For example, Zld is pivotal for the MZT as it increases accessibility of chromatin at
enhancers thereby allowing binding of other transcriptional activators at these DNA regions which
facilitates initiation of zygotic gene expression (
Xu et al., 2014
;
Harrison et al., 2011
;
Liang et al.,
2008
;
Nien et al., 2011
;
Ya
́
n
̃
ez-Cuna et al., 2012
). Zld binds nucleosomes, another characteristic of
pioneer factors (
McDaniel et al., 2019
), and therefore loss of Zld leads to a global decrease in
zygotic gene expression as many enhancer regions remain inaccessible (
Schulz et al., 2015
;
Sun et al., 2015
). Through its effects on chromatin accessibility, Zld has been shown to influence the
ability of morphogen transcription factors, Bicoid and Dorsal, to support embryonic patterning
(
Xu et al., 2014
;
Foo et al., 2014
). While Zld is clearly pivotal for supporting MZT, some genes con-
tinue to be expressed even in its absence (
Nien et al., 2011
). As chromatin accessibility in the early
embryo has recently been shown to be a dynamic process (
Blythe and Wieschaus, 2016b
;
Bozek et al., 2019
), it is possible that Zld contributes in a stage-specific manner and that other as
yet unidentified pioneer factors contribute to the extended process of zygotic genome activation.
The embryo undergoes a widespread state change after the 14th nuclear division, termed the
midblastula transition (MBT) (
Foe and Alberts, 1983
;
Shermoen et al., 2010
). This developmental
milestone is marked by dramatic slowing of the division cycle and cellularization of nuclei before the
onset of embryonic programs of morphogenesis and differentiation. Cell membranes encapsulate
nuclei to form a single-layered epithelium. In addition, at nc14, developmental changes relating to
DNA replication occur; namely a lengthened S-phase and the introduction of G2 phase into the cell
cycle. MBT is also associated with clearance of a subset of maternally provided mRNAs, large-scale
transcriptional activation of the zygotic genome, and an increase in cell cycle length (
Yuan et al.,
2016
;
Tadros and Lipshitz, 2009
). We hypothesized that other late-acting pioneer factors manage
the MBT in addition to or in place of Zld.
The
Drosophila
gene
odd-paired
(
opa
) encodes the founding member of the Zinc finger in the
cerebellum (Zic) protein family (
Aruga et al., 1996
;
Hursh and Stultz, 2018
). The important regula-
tory role of Zic (ZIC human ortholog) in early developmental processes has been established across
major animal models and also implicated in human pathology (rev. in
Aruga and Millen, 2018
;
Houtmeyers et al., 2013
).
opa
is a broadly expressed gene of relatively long transcript length (
~
17
kB) that is activated during mid-nc14 and serves a number of important functions throughout devel-
opment (
Cimbora and Sakonju, 1995
;
Benedyk et al., 1994
). Opa protein has a DNA-binding
domain containing five Cys2His2-type zinc fingers, and shares homology with mammalian Zic1, 2,
and three transcription factors. While mutants exhibit a pair-rule phenotype (
Ju
̈
rgens et al., 1984
),
the broad expression pattern of
opa
contrasts with the typical 7-stripe pattern of other pair-rule
genes. Rather than providing spatial information as do most other pair-rule transcription factors,
Opa instead acts as a timing factor to broadly regulate the expression of segment polarity genes
including the transition of pair-rule genes to segmental expression patterns (i.e. from 7- to 14-
stripes) (
Clark and Akam, 2016
;
Benedyk et al., 1994
).
opa
mutant embryos die before hatching
and in addition to aberrant segmentation, they also exhibit defects in larval midgut formation
(
Cimbora and Sakonju, 1995
). During midgut formation, Opa regulates expression of a pivotal
receptor tyrosine kinase required for proper morphogenesis of the visceral mesoderm (
Mendoza-
Garcı ́a et al., 2017
). In addition, at later stages, Opa supports temporal patterning of intermediate
neural progenitors of the
Drosophila
larval brain (
Abdusselamoglu et al., 2019
).
Previous studies suggested that Opa can influence the activity of other transcription factors to
promote gene expression. A well-characterized target of Opa in the early embryo is
sloppy-paired 1
(
slp1
), a gene exhibiting a segment polarity expression pattern and for which two distinct enhancers
have been identified that are capable of responding to regulation by Opa and other pair-rule tran-
scription factors including Runt (Run;
Cadigan et al., 1994
;
Prazak et al., 2010
). One of these, the
slp1
DESE enhancer, mediates both Run-dependent repression and activation and Opa plays a cen-
tral role by supporting Run’s activating input (
Hang and Gergen, 2017
). Additionally, our recent
study showed that Run regulates the spatiotemporal response of another enhancer,
sog_Distal
Koromila
et al
. eLife 2020;9:e59610.
DOI: https://doi.org/10.7554/eLife.59610
2 of 30
Research article
Developmental Biology
Genetics and Genomics
(
Ozdemir et al., 2011
; also known as
sog_Shadow
;
Hong et al., 2008
) to support its expression in a
broad stripe across the dorsal-ventral (DV) axis on both sides of the embryo (
Koromila and Statho-
poulos, 2019
). Using a combination of fixed and live imaging approaches, our analysis suggested
that Run’s role changes from repressor to activator over time in the context of
sog_Distal
; late
expression requires Run activating input. These analyses of
slp1
DESE and
sog_Distal
regulation sup-
port the view that Opa might provide temporal input at enhancers.
The current study was initiated to investigate whether Opa supports late expression through the
sog_Distal
enhancer. Previous studies had not linked Opa to the regulation of DV patterning. Never-
theless, through mutagenesis experiments coupled with in vivo imaging, we provide evidence that
Opa does regulate expression of the
sog_Distal
enhancer. Further, we show that Opa’s role is
indeed late-acting, occurring in embryos at mid-nc14 onwards, whereas the enhancer initiates
expression at nc10. Given its ability to regulate key embryonic enhancers in a temporal manner, we
hypothesized that Opa may play a general role in activating zygotic gene expression during late
MZT, much as Zld does earlier. To assay Opa’s genome-wide effects on gene expression and chro-
matin accessibility in the embryo, we used a combination of sequencing approaches: RNA-seq tran-
scriptome profiling, chromatin immunoprecipitation (ChIP-seq) and single-embryo Assay for
Transposase-Accessible Chromatin (ATAC-seq). Our whole-genome data demonstrate that Opa con-
tributes to patterning the embryo by serving as a general timing factor, and possibly as a pioneer, to
broadly influence zygotic transcription in nc14, as the embryo undergoes cellularization, during late
phase of the maternal-to-zygotic transition.
Results
Opa regulates the
sog_Distal
enhancer demonstrating a role for this
gene in DV axis patterning
In a previous study, we created a reporter in which the 650 bp
sog_Distal
enhancer sequence was
placed upstream of a heterologous promoter from the
even skipped
gene (
eve.p
), driving expression
of a compound reporter gene containing both a tandem array of MS2 sites and the gene
yellow
,
including its introns (
Koromila and Stathopoulos, 2017
). We used this reporter to assay gene
expression supported by the
sog_Distal
enhancer in the early embryo. While this enhancer becomes
active at nc10 and continues into gastrulation, in this study we focused on late expression through
sog_Distal
during nc13 and nc14. Due to its length (i.e.
~
45 min compared to
~
15 min for nc13 at
23 ̊C) nc14 was assayed in four, roughly 12 min intervals: nc14A, nc14B, nc14C, and nc14D. Live
movies were analyzed using a previously defined computational approach tailored to spatiotemporal
dynamics (
Koromila and Stathopoulos, 2019
).
In our previous study, mutation of the single Run binding site in the
sog_Distal
enhancer led to
expansion of reporter expression early (i.e. nc13 and early nc14) but loss of expression late (i.e.
nc14C and nc14D) (
Koromila and Stathopoulos, 2019
). These results suggested that Run’s role
switches from that of repressor to activator in the context of
sog_Distal
enhancer during this time.
Other studies also suggested that Run can function as either repressor or activator depending on
context, as the response of a given enhancer to Run is influenced by the presence or absence of
other transcription factors (
Hang and Gergen, 2017
;
Prazak et al., 2010
;
Swantek and Gergen,
2004
). This is the case for
slp1
, where Opa is required for Run-dependent activation of expression
(
Swantek and Gergen, 2004
). We therefore hypothesized that Opa might also influence Run’s activ-
ity with respect to the
sog_Distal
enhancer; specifically, that Opa functions to support late expres-
sion of
sog_Distal
, when Run switches to providing activating input (
Koromila and Stathopoulos,
2019
).
In concordance with this hypothesis, the
sog Distal
650 bp enhancer sequence contains five puta-
tive 12 bp Opa binding sites, based on comparison with the vertebrate Zic3 consensus motif (JAS-
PAR;
Figure 1I
). We introduced 2–4 bp mutations at these five sites (i.e.
sogD_
D
Opa
) and assayed
MS2-MCP reporter expression by in vivo imaging of nascent transcription (
Garcia et al., 2013
;
Lucas et al., 2013
). We found that expression was relatively normal up to stage nc14B but then
exhibited a visually apparent decrease at nc14C (
Figure 1C
compare to
Figure 1A
;
Video 1
). Quan-
titative analysis of MS2-MCP signal in embryos containing either the wildtype
sog_Distal
or
sogD_
D
Opa
reporters using a previously described analysis pipeline (
Koromila and Stathopoulos,
Koromila
et al
. eLife 2020;9:e59610.
DOI: https://doi.org/10.7554/eLife.59610
3 of 30
Research article
Developmental Biology
Genetics and Genomics
Figure 1.
Opa is required to support activation of reporter expression at late nc14, just preceding gastrulation. In this and all other subsequent figures
lateral views of embryos are shown with anterior to the left and dorsal up, unless otherwise noted. (
A,C
) Stills from movies (n = 3 for each) of the two
indicated
sog_Distal
MS2-yellow reporter variants
sog_Distal
(
A
) or
sogD_
D
Opa
(
C
) in which five predicted Opa-binding sites were mutated as shown
(
H
) and transcription detected in vivo via MS2-MCP-GFP imaging (
Koromila and Stathopoulos, 2019
) at three representative timepoints: nc13, nc14B,
and nc14C. Blue dots indicate presence of GFP+ signal, representing nascent transcripts labeled by the MS2-MCP system; thresholding was applied
and remaining signals identified by the Imaris Bitplane software, for visualization purposes only. Nuclei were labeled by Nup-RFP (
Lucas et al.,
2013
). Scale bar represents 50
m
m. (
B
) Plots of number of active nuclei, defined by counting dots (x-axis) versus relative DV axis embryo-width (EW)
position (y-axis), analyzed from representative stills from movies of three embryos at nc14C. (
D, E
) Anti-Opa (
D
) and anti-Zld (
E
) antibody staining of
early wild-type embryos at the indicated stages. (
F
) Integrative Genomics Viewer (IGV) genome browser track of the
sog
locus showing Zld and Opa
ChIP-seq data for embryos at two timepoints: nc13-14 and nc14 late for Zld (GSM763061 and GSM763061, respectively;
Harrison et al., 2011
) and 3 hr
and 4 hr for Opa. Zld nc13-14, Zld nc14 late and Opa 3 hr ChIP-seq samples are of overlapping timepoints, whereas Opa 4 hr ChIP-seq sample is later.
Gray shading marks the region of
sog_Distal
enhancer location. (
G
) JASPAR consensus binding site for Opa based on mammalian Zic proteins
identified by bacterial one-hybrid (
Sen et al., 2010
;
Noyes et al., 2008
). (
H
) Location of 5 sequences within the 650 bp
sog_Distal
enhancer region that
match the Jaspar Opa consensus binding site allowing 1 bp mismatch. Mutated Opa sites introduced to eliminate binding are shown in blue, creating
sogD_
D
Opa
(C; see Materials and methods). Bases in bold (7 bp) indicate matches to the Opa de novo motifs identified by ChIP-seq analysis (see J).
For sake of comparison to consensus sequence, reverse complement sequence is shown for a subset. (
I
) Consensus binding site for
Mus musculus
Zic3/
Opa homolog identified using ChIP-seq (
Lim et al., 2010
). (
J,K
) Sequence logo representations of the most significant and abundant motifs, likely
consensus binding sites, identified by HOMER de novo motif analysis in the Opa 3 hr and Opa 4 hr (
J
), or Zld nc13-14 and Zld nc14 late (
K
) ChIP-seq
Figure 1 continued on next page
Koromila
et al
. eLife 2020;9:e59610.
DOI: https://doi.org/10.7554/eLife.59610
4 of 30
Research article
Developmental Biology
Genetics and Genomics