The research team, led by Chengcheng Su and corresponding author Xiujuan Shan of the State Key Laboratory of Mariculture Biobreeding and Sustainable Goods at the Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, carried out time-series RNA sequencing across ten developmental stages. The sampling began at the zygote stage, the very first moment after fertilization when the egg contains only maternal gene products, and extended all the way to the pre-hatching period, when the embryo is nearly ready to break free of its chorion. By profiling gene expression at each of these milestones, the researchers built a continuous molecular narrative of how a fish embryo comes to be.

What emerged from the data was anything but a smooth, linear progression. Instead, the analysis revealed dynamic, non-linear transcriptomic transitions, meaning that the embryo’s gene activity does not simply ramp up or down gradually but reorganizes itself in bursts at critical junctures. These discontinuities correspond to major developmental events, and they highlight how embryogenesis is punctuated by sharp molecular turning points rather than a steady march. For developmental biologists, such stage-resolved resolution is essential, because averaging across broad developmental windows can obscure precisely the transitions that matter most.

One of the study’s central findings concerns the maternal-to-zygotic transition, often abbreviated as MZT, one of the most fundamental events in animal development. In the earliest hours of life, an embryo is transcriptionally silent: everything that happens is directed by messenger RNAs and proteins deposited in the egg by the mother. At some point, the embryo’s own genome switches on and begins producing its own transcripts, while the maternal stockpile is actively degraded. This handover of control is known as zygotic genome activation, or ZGA, and its timing and structure vary across species. The new data suggest that in marine medaka, ZGA follows a dual-wave architecture, with two distinct surges of embryonic gene expression rather than a single activation event.

The functional signatures of the two waves are strikingly different. The early wave of zygotic activation was associated mainly with chromatin-related and transcriptional regulatory functions, consistent with the idea that the first genes switched on in the embryo are those that remodel the genome itself and set up the regulatory machinery for everything that follows. The later wave, by contrast, was enriched for ribosome biogenesis and RNA processing, reflecting the embryo’s growing need to build its protein-making infrastructure as cell division accelerates and differentiation begins. This two-phase pattern echoes findings from other model organisms and suggests a broadly conserved logic governing how vertebrate embryos take command of their own development.

Beyond the global architecture of genome activation, the team used network-based analyses to identify candidate regulatory modules, groups of genes whose coordinated expression suggests shared control and shared function. Among these modules were networks involving pluripotency-associated factors, the molecular custodians of the embryo’s undifferentiated state in its earliest stages. Other modules captured components of maternal transcript clearance, the machinery responsible for sweeping away the maternal mRNAs as the zygotic genome assumes control. Still others corresponded to stage-specific developmental gene sets, providing a framework for connecting individual gene networks to particular morphological milestones.

Perhaps the most visually evocative finding is a transient ciliogenesis-associated expression program that appears during a narrow developmental window corresponding to the formation of Kupffer’s vesicle. Kupffer’s vesicle is a transient organ unique to fish and other teleost embryos, and it plays an outsized role: the cilia inside it generate a directional fluid flow that establishes the left-right asymmetry of the body plan, determining which side the heart and other organs will occupy. The appearance of a coordinated cilia-building gene program precisely during this window ties the transcriptomic data directly to a morphological structure with clear functional importance, and it offers researchers a molecular handle for studying how organ asymmetry is established in fish.

To place marine medaka in a broader comparative context, the authors summarized their findings against other teleost models, comparing the timing of zygotic genome activation, the developmental timing of left-right asymmetry establishment, and the activation of key genes. Such cross-species comparisons are valuable because they reveal which features of embryonic development are conserved across fish lineages and which have diverged. Marine medaka is particularly attractive for such comparisons because, unlike its freshwater relative the Japanese medaka, it tolerates a wide range of salinities, making it an ideal subject for studies of how environmental conditions, including ocean pollution and climate-related stressors, affect early development.

Indeed, the practical significance of this resource extends well beyond basic developmental biology. Marine medaka is widely used in ecotoxicology, where embryos are exposed to contaminants, endocrine disruptors, microplastics, and other environmental hazards to assess their effects. Transcriptomic responses in such experiments are typically interpreted against a baseline of normal development, and until now that baseline has been underdeveloped for this species. By providing a stage-resolved reference of normal embryonic gene expression, the study gives toxicologists a far more accurate yardstick. A gene that appears dysregulated after chemical exposure can now be evaluated against its expected expression trajectory at the exact developmental stage being studied, reducing false positives and sharpening the detection of genuine developmental toxicity.

The study, which was funded by the State Key Laboratory of Mariculture Biobreeding and Sustainable Goods, the National Key Research and Development Program of China, and the Taishan Scholar Project, also carries implications for aquaculture. Understanding the molecular events that govern normal embryogenesis in a marine fish supports breeding programs, embryo quality assessment, and the development of new farmed species. The authors describe their dataset as a framework for future functional, comparative, and exposure-related studies, and with the full data openly accessible, laboratories around the world can now interrogate the earliest chapters of a marine fish’s life with a precision that was previously unavailable. As genomic resources for non-traditional model organisms continue to expand, studies like this one are steadily closing the gap between a handful of classic laboratory species and the vast diversity of life in the ocean.

Subject of Research: Stage-resolved transcriptomic dynamics of embryonic development in the marine medaka, Oryzias melastigma

Article Title: Transcriptomic analysis of marine medaka embryonic development

Article References: Su, C., Li, S., Jin, X., Shao, C., & Shan, X. (2026). Transcriptomic analysis of marine medaka embryonic development. BMC Genomics. https://doi.org/10.1186/s12864-026-13342-1

Image Credits: AI Generated

DOI: 10.1186/s12864-026-13342-1

Keywords: marine medaka, Oryzias melastigma, transcriptomics, embryonic development, maternal-to-zygotic transition, zygotic genome activation, Kupffer's vesicle, ciliogenesis, RNA sequencing, developmental biology, marine ecotoxicology, BMC Genomics

The review, authored by Jiaoqi Mei, Bianling Xu, Zhi Chen, Yuxin Dong, Xueping Liu, Xiaodong Li and colleagues at the First Hospital of Hebei Medical University in Shijiazhuang, China, frames its argument around a distinctive cellular structure known as the mitochondria-associated ribonucleoprotein domain, or MARDO. During the growth phase of the oocyte, mitochondria cluster together with RNA-protein complexes into these specialized domains, which serve as physical warehouses for dormant maternal mRNAs. The authors synthesize evidence that ZAR1 is not a passive bystander in this architecture. Through its RNA-binding capacity, ZAR1 helps tether specific maternal transcripts to MARDO, effectively deciding which messages are stored where, and positioning them for the moment when translational silencing must give way to activity.

The timing could hardly be more critical. Oocyte maturation and the earliest rounds of embryonic development occur during a period of transcriptional silence. The oocyte progressing through meiosis, and the one-cell embryo immediately after fertilization, cannot transcribe new genes in any meaningful way. Every protein needed to drive chromosome segregation, spindle assembly, and the first cleavage divisions must be manufactured from pre-existing maternal mRNAs. The regulation is therefore extraordinarily dependent on three linked operations: the storage of maternal mRNAs in a translationally inactive state, their timely activation for protein production, and their programmed clearance once their work is done. ZAR1 and ZAR2, the review argues, sit at the junction of all three.

At the molecular level, ZAR1 and its homolog ZAR2, also called ZAR1-like, belong to a small family of maternal-effect proteins whose importance first became apparent when mouse embryos lacking ZAR1 arrested at the very first stages after fertilization, the phenotype that gave the protein its name. Since then, a growing toolbox of techniques, including linear amplification of complementary DNA ends and sequencing, known as LACE-seq, has allowed researchers to map the transcripts that ZAR1 and ZAR2 physically bind. The review emphasizes that these proteins associate with a shared cast of RNA-handling factors, including the Y-box binding protein 2 (YBX2), the DEAD-box helicase 6 (DDX6), and the LSM family member 14B (LSM14B), all of which are implicated in stabilizing and silencing stored transcripts. Together these factors form a ribonucleoprotein network that keeps the maternal message archive intact during oocyte growth.

What happens when this network falters? Evidence from efficient mouse models lacking both Zar1 and Zar2 suggests the consequences cascade through every stage of the maternal mRNA lifecycle. The review highlights that combined loss of these proteins may destabilize maternal mRNAs, disrupt the dynamics of their poly(A) tails, the stretches of adenosine residues whose length acts as a molecular throttle on translation, and impair the activation of protein synthesis at fertilization. Polyadenylation in the maturing egg is a tightly choreographed event: selected dormant transcripts receive extended poly(A) tails that recruit the translation machinery, while others are deadenylated and marked for decay. If ZAR1 and ZAR2 help determine which transcripts receive which treatment, their absence scrambles the schedule, producing messages that are translated too early, too late, or not at all.

One of the most striking threads in the review concerns the clearance arm of the system. Maternal mRNAs cannot simply linger forever. Their degradation is essential for the maternal-to-zygotic transition (MZT), the handover of developmental control from the maternal message archive to the embryo’s own zygotic genome. A key executioner of maternal mRNA destruction is the protein B-cell translocation gene 4 (BTG4), which recruits the deadenylase CNOT6L, a subunit of the CCR4-NOT transcription complex, to strip protective poly(A) tails from maternal transcripts and condemn them to decay. The review presents evidence that combined ZAR1/ZAR2 loss may compromise this BTG4-mediated clearance pathway, leaving embryo-killing maternal messages to persist beyond their expiry date. The result is an embryo that fails to complete the MZT and cannot activate its own genome, a process known as zygotic genome activation, or ZGA.

The developmental fallout of these molecular failures is predictable and severe. The review catalogues meiotic abnormalities in oocytes lacking ZAR1-family function, including defective spindle assembly, the structural apparatus that must segregate chromosomes with near-perfect fidelity during the divisions that halve the egg’s genome and then drive the first embryonic cleavages. Errors in spindle formation produce aneuploid eggs and embryos, a leading cause of miscarriage and failed in vitro fertilization cycles in humans. Downstream, impaired maternal mRNA regulation culminates in arrest at the zygote stage, precisely the phenotype observed in the mouse knockouts that first identified ZAR1 as a maternal-effect gene. The thread running from RNA storage in MARDO to chromosome segregation and embryonic genome activation illustrates how a single protein family can coordinate events that span multiple cellular compartments and developmental stages.

For reproductive medicine, the implications are tantalizing but the review is careful to draw boundaries. The authors stress that the link between reduced ZAR1/ZAR2 expression and the abnormal epigenetic modifications seen in aged oocytes remains to be proven through additional functional experiments. Similarly, the connection between human ZAR1 sequence variants and clinical syndromes of oocyte maturation failure or preimplantation embryonic arrest requires firmer genetic evidence from patient cohorts. The reviewers explicitly caution against directly extrapolating findings from animal models to human clinical conclusions, noting that species differences, experimental model limitations, and varying levels of evidence all impose limits on what can currently be claimed. This restraint matters for a field in which assisted reproductive technologies, from in vitro fertilization and intracytoplasmic sperm injection to in vitro maturation of oocytes, are constantly seeking molecular markers that could predict oocyte quality and embryo viability.

Even within those limits, the review makes a compelling case that the ZAR1/ZAR2-MARDO-maternal mRNA regulatory axis deserves a central place in the biology of reproduction. It recasts the oocyte not as a passive vessel but as an information-dense package whose cargo management determines whether development launches at all. It connects seemingly disparate observations, from mitochondrial clustering and RNA granule formation to poly(A) tail dynamics and BTG4-dependent decay, into a single mechanistic narrative. And it charts a research agenda: defining the full inventory of ZAR1- and ZAR2-bound transcripts, resolving how MARDO architecture changes as the oocyte matures, testing whether epigenetic drift in aging eggs disrupts this axis, and screening infertile patients for variants in ZAR1 that could explain otherwise mysterious failures of oocyte maturation. If those efforts succeed, the proteins that guard a mother’s molecular legacy may one day point the way to new diagnostics and therapies for infertility, turning a once-obscure maternal-effect gene into a cornerstone of reproductive medicine.

Subject of Research: The role of ZAR1 and ZAR2 proteins in regulating maternal mRNA storage, translation, and clearance during oocyte maturation and early embryonic development.

Article Title: ZAR1/ZAR2-mediated maternal mRNA fate control in oocyte maturation and early embryonic development

Article References: ZAR1/ZAR2-mediated maternal mRNA fate control in oocyte maturation and early embryonic development. (n.d.). https://doi.org/10.1186/s13048-026-02262-z

Image Credits: AI Generated

DOI: 10.1186/s13048-026-02262-z

Keywords: ZAR1, ZAR2, maternal mRNA, MARDO, oocyte maturation, maternal-to-zygotic transition, zygotic genome activation, female infertility, RNA binding proteins, BTG4, early embryonic development, reproductive biology

PRC2, or Polycomb repressive complex 2, is best known for placing the histone modification H3K27me3 on chromatin. Histones are proteins around which DNA is wrapped, and chemical marks on these proteins help regulate whether genes remain accessible for transcription. H3K27me3 is generally associated with gene repression, although its effects depend on developmental context and the presence of other regulatory signals. During embryogenesis, the genome is extensively reorganized after fertilization. Maternal messenger RNAs and proteins initially control development, but the embryo must soon activate its own genome. This handover, called the maternal-to-zygotic transition, requires precise changes in chromatin structure and gene activity. The new study shows that PRC2 is an important part of this transition rather than a passive regulator acting later in development.

The researchers used the dTAG system to remove PRC2 components rapidly and selectively at defined developmental stages. This approach differs from conventional genetic knockout experiments, which eliminate a gene throughout development and can make it difficult to determine when a protein is actually required. In the dTAG system, a small chemical molecule targets a tagged protein for destruction, allowing investigators to examine the consequences of acute depletion at particular time points. By applying this strategy during preimplantation development, the team could distinguish PRC2 functions in the fertilized egg, early cleavage-stage embryos and the blastocyst from functions that emerge later. The results revealed that the complex has distinct requirements across these stages, indicating that its role changes as embryonic cell states become progressively specialized.

During preimplantation development, the embryo passes through a series of divisions before forming the blastocyst, a structure containing the future embryonic cells, the trophectoderm and the primitive endoderm. The study found that PRC2 contributes to the maternal-to-zygotic transition and to the formation of the epiblast, the cell population that will generate the embryo proper. When PRC2 activity was disrupted at inappropriate times, embryos showed defects in developmental progression and in the establishment of the epiblast. These observations suggest that PRC2 helps prevent premature or misplaced gene-expression programs while allowing the correct developmental network to emerge. Its activity therefore appears to provide temporal control: genes must be silenced at the right moment, but the chromatin landscape must also remain capable of being remodeled as cells acquire new identities.

A major focus of the work was the relationship between H3K27me3 and H3K4me3. H3K4me3 is commonly associated with gene activation and is enriched near promoters of actively transcribed genes. When H3K27me3 and H3K4me3 occur together at the same regulatory region, the chromatin is described as bivalent. Bivalent domains have traditionally been viewed as “poised” regulatory regions, carrying both repressive and activating signals so that developmental genes can respond rapidly when a cell commits to a new fate. However, the precise timing of bivalency formation in early embryos has remained uncertain, partly because embryonic chromatin changes rapidly and because different studies have used different definitions of a bivalent domain.

By systematically tracking both histone modifications, the researchers propose a revised model for when bivalency is established. Rather than appearing as a fully formed state at a single developmental moment, bivalency emerges through a stepwise process. One modification can be detected before the other, and the two marks become combined as development proceeds and cellular identities sharpen. This sequence challenges the idea that bivalent chromatin is installed in one coordinated event. It instead suggests that early embryonic cells build regulatory competence gradually, integrating repressive and active information over time. PRC2 is central to this process because it supplies the H3K27me3 component, while the distribution of H3K4me3 reflects the changing transcriptional and chromatin environment around developmental genes.

The study also identifies an unexpected connection between PRC2 activity outside the embryo proper and the production of primordial germ cells, or PGCs. PGCs are the embryonic precursors of sperm and eggs, and their numbers must be tightly controlled: too few can compromise reproductive potential, while abnormal expansion may disrupt tissue organization and developmental balance. The researchers found that PRC2 regulates appropriate PGC numbers in the epiblast through its control of Esrrb expression in the extraembryonic ectoderm. The extraembryonic ectoderm does not become part of the fetus, but it produces signals that influence neighboring embryonic cells. This places PRC2 within a non-cell-autonomous pathway, in which an epigenetic event in one tissue changes the behavior of cells in another.

The role of Esrrb helps explain how this intercellular effect is transmitted. Esrrb encodes a transcription factor, a protein that binds DNA and regulates suites of genes involved in cell identity and development. According to the study, changes in PRC2-dependent regulation of Esrrb in the extraembryonic ectoderm alter the developmental environment experienced by epiblast cells and consequently affect PGC formation. The finding illustrates that epigenetic regulation does not operate solely within the boundaries of a single cell. A chromatin complex can modify gene expression in one embryonic compartment, which then changes signals or conditions that guide fate decisions elsewhere. This epigenetic-to-transcription-factor-to-cell-communication pathway adds another layer to the already complex regulation of germline specification.

Taken together, the findings expand the biological definition of PRC2 during early embryogenesis. The complex acts cell-autonomously to organize chromatin and support preimplantation development, yet it also has broader developmental consequences through interactions between embryonic and extraembryonic tissues. The researchers’ results emphasize that the same epigenetic regulator can have different effects depending on developmental stage, cellular location and the surrounding transcriptional network. They also show why acute protein-depletion tools such as dTAG can be valuable for developmental biology: timing is often as important as molecular identity. By connecting histone-mark dynamics, transcription-factor control and intercellular signaling, the study offers a framework for understanding how embryos convert rapidly changing epigenetic information into coordinated developmental outcomes.

Subject of Research: Stage-specific functions of PRC2 in preimplantation development, epigenetic bivalency and primordial germ cell formation.

Article Title: Decoding stage-specific functions of PRC2 in early embryogenesis uncovers roles in preimplantation development and primordial germ cell fate.

Article References: Zhou, C., Wang, M., Chen, Z. et al. “Decoding stage-specific functions of PRC2 in early embryogenesis uncovers roles in preimplantation development and primordial germ cell fate.” Nature Cell Biology 28, 1700–1714 (2026). https://doi.org/10.1038/s41556-026-02002-x

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41556-026-02002-x; publication date: August 2026

Keywords: PRC2, embryogenesis, epigenetics, H3K27me3, H3K4me3, bivalency, maternal-to-zygotic transition, epiblast, primordial germ cells, Esrrb, preimplantation development, dTAG system

Right after fertilization, the highly organized chromatin architecture characteristic of differentiated cells is disrupted. Among the most important organizational features of chromatin are topologically associating domains (TADs), which are contiguous regions within chromosomes where DNA sequences preferentially interact with each other. During the one-cell zygote stage, these canonical TADs dissolve, erasing much of the conventional chromosomal folding landscape. This dissolution coincides with a remarkable reprogramming event where the embryo’s genome is essentially resetting, stripping away prior epigenetic patterns to allow a clean slate for new developmental programs.

The resurgence of structured chromatin architecture happens gradually from the two-cell stage onward, culminating around the eight-cell stage with the reestablishment of TADs and finer chromatin domains. Central to this process are the DNA-binding protein CTCF and the cohesin complex. CTCF has long been recognized as a master architectural protein that helps demarcate domain boundaries within chromosomes. In this study, the authors found that CTCF already occupies chromatin continuously throughout early embryonic development, even when TADs dissolve in the zygote. On the other hand, cohesin, which is essential for loop extrusion and chromatin domain formation, exhibits minimal chromatin binding at the one-cell stage. Cohesin binding then increases progressively from the two-cell to the eight-cell stages, paralleling the slow reestablishment of three-dimensional chromatin organization.

Perhaps the most unexpected discovery lies in the emergence of what the researchers term “genic cohesin islands” or GCIs during this developmental window. These GCIs are concentrated cohesin enrichments specifically localized across gene bodies of highly active genes. Strikingly, these genes are not random but are enriched for cell identity and key regulators critical for embryonic development. The chromatin environment surrounding GCI genes is marked by broad domains of histone H3 lysine 4 trimethylation (H3K4me3) at promoters, a modification known to reflect active transcription initiation states. Moreover, enhancers in the proximity of GCI genes robustly recruit the cohesin loader NIPBL and various transcription factors, highlighting a hotspot of regulatory activity.

Crucial to these findings is the realization that hypertranscription—an unusually elevated level of transcriptional activity—characterizes the early embryo precisely during the period when GCIs appear and TADs are being rebuilt. Using pharmacological inhibition and genetic manipulation, the authors demonstrate that active transcription is not simply a consequence of chromatin architecture but is necessary for GCI formation. This causal link indicates that transcription itself shapes the chromatin landscape, possibly by enhancing cohesin loading or stabilization over gene bodies.

Conversely, when transcription is experimentally induced, GCIs form in response, further supporting the idea that gene activity can directly influence chromatin topology. This two-way relationship establishes a feedback loop wherein hypertranscription promotes specialized cohesin binding patterns, and these cohesin islands in turn contribute to gene regulation by creating local insulation boundaries. These boundaries physically separate GCI genes from neighboring chromatin regions, allowing the formation of discrete contact domains with nearby CTCF-bound sites.

Functionally, this chromatin structure provides a stabilizing effect on gene expression. Genes bearing GCIs exhibit enhanced transcriptional output and reduced variability in expression levels—the hallmarks of robust, stable gene regulation essential for early developmental decisions. Therefore, the physical organization of the genome into three-dimensional architectures is not merely a structural phenomenon but a dynamic participant in the control of developmental gene expression programs.

The study’s insights extend our understanding of how embryonic cells transition from a totipotent, relatively unstructured state to a more hierarchically organized genome poised for lineage commitment. It also emphasizes the plasticity and interdependence of transcriptional activity and chromatin topology in shaping cell fate trajectories. By uncovering these genic cohesin islands and defining their functional role, the research offers a compelling model of early embryonic genome organization that reconciles chromatin folding principles with the unique transcriptional demands of rapid developmental progression.

One striking implication of this work is the reconsideration of established chromatin architectural features—such as TADs—not as static entities but as dynamic structures whose establishment depends on and reinforces transcriptional states. Early embryos employ a distinct mode of genome folding that co-opts transcriptional hyperactivity to direct cohesin loading, establishing chromatin domains that then bolster gene expression fidelity. This paradigm shift opens new avenues for exploring how dysregulation of chromatin-transcription interplay might lead to developmental disorders or contribute to reprogramming inefficiencies in regenerative medicine.

Moreover, the discovery of GCIs may have broader significance beyond early embryogenesis. Similar mechanisms could operate in other contexts of intense transcriptional demand such as in stem cells, activated immune cells, or cancers. Understanding how cohesin and transcription factors collaborate to sculpt the genome might reveal targets for therapeutic intervention or synthetic biology applications aiming to modulate gene expression landscapes.

In summary, Yu and colleagues’ work unravels the intricate choreography of genome architecture construction in the earliest stages of life, demonstrating an intimate crosstalk between hypertranscription and chromatin organization. This dual relationship enables embryos to orchestrate the complex activation of their genomes, setting the stage for all subsequent development. As we push deeper into the mysteries of the three-dimensional genome, studies like this remind us that form and function in biology are inseparable, coevolving facets of the living cell.

Subject of Research:

Molecular mechanisms underlying the establishment of three-dimensional chromatin architecture and the relationship between chromatin folding and transcriptional activity during early mammalian embryonic development.

Article Title:

Establishment of chromatin architecture interplays with embryo hypertranscription.

Article References:

Yu, G., Xu, K., Xia, W. et al. Establishment of chromatin architecture interplays with embryo hypertranscription. Nature (2025). https://doi.org/10.1038/s41586-025-09400-5

Image Credits:

AI Generated