SETDB1 enables development beyond cleavage stages by extinguishing the MERVL-driven two-cell totipotency transcriptional program in the mouse embryo

Tie-Bo Zeng; Zhen Fu; Mary F Majewski; Ji Liao; Marie Adams; Piroska E Szabó

doi:10.7554/eLife.109248

. 2026 Feb 16;15:RP109248. doi: 10.7554/eLife.109248

SETDB1 enables development beyond cleavage stages by extinguishing the MERVL-driven two-cell totipotency transcriptional program in the mouse embryo

Tie-Bo Zeng ^1,^†,^‡, Zhen Fu ^2,^†, Mary F Majewski ³, Ji Liao ¹, Marie Adams ³, Piroska E Szabó ^1,^✉

Editors: Yan Song⁴, Lori Sussel⁵

PMCID: PMC12908936 PMID: 41697236

Abstract

Loss of maternal SETDB1, a histone H3K9 methyltransferase, leads to developmental arrest prior to implantation, with very few mouse embryos advancing beyond the eight-cell stage, which is currently unexplained. We genetically investigate SETDB1’s role in the epigenetic control of the transition from totipotency to pluripotency—a process demanding precise timing and forward directionality. Through single-embryo total RNA sequencing of two-cell and eight-cell embryos, we find that Setdb1^mat-/+ embryos fail to extinguish one-cell and two-cell transient genes—alongside persistent expression of MERVL retroelements and MERVL-driven chimeric transcripts that define the totipotent state in mouse two-cell embryos. Comparative bioinformatics reveals that SETDB1 acts at MT2 LTRs and MERVL-driven chimeric transcripts, which normally acquire H3K9me3 during early development. The dysregulated targets substantially overlap with DUXBL-responsive genes, indicating a shared regulatory pathway for silencing the two-cell transcriptional program. We establish maternal SETDB1 as a critical chromatin regulator required to extinguish retroelement-driven totipotency networks and ensure successful preimplantation development.

Research organism: Mouse

Introduction

SETDB1 is a histone 3 lysine 9 (H3K9) di- and tri-methyl transferase (Yang et al., 2002). In the mouse embryo, it localizes to the inner cell mass of the blastocyst (Cho et al., 2012) and is essential for postimplantation development (Dodge et al., 2004). Homozygous knockout of Setdb1 results in embryonic lethality between 3.5 and 5.5 days post coitum (dpc) and prevents the derivation of embryonic stem cells (ESCs). SETDB1 is required for the silencing of developmental gene promoters (Bilodeau et al., 2009). Although Setdb1^-/- zygotic mutants retain maternally deposited SETDB1 protein from the oocyte, conditional knockout of Setdb1 during oogenesis produces an earlier and more severe maternal-effect phenotype: most oocytes arrest during meiosis, and the few that are fertilized typically fail to develop beyond the two-cell (2c) or eight-cell (8c) stages and never reach the blastocyst (Eymery et al., 2016; Kim et al., 2016). The basis of this cleavage-stage arrest in the Setdb1^mat-/+ embryos remains unknown.

This developmental block coincides with the transition from totipotency to pluripotency, prompting us to investigate whether SETDB1 regulates this critical process. Following fertilization, parental epigenomes undergo extensive reprogramming to enable zygotic genome activation (ZGA). In mice, minor ZGA occurs at the late one-cell (1c) stage and major ZGA at the 2c stage—both essential for development (Abe et al., 2018). The 2c stage marks the onset of totipotency, while from the four-cell (4c) stage onward, blastomeres begin committing to either the pluripotent inner cell mass or the trophectoderm lineage (Piotrowska-Nitsche et al., 2005). In ESCs, dedifferentiation to two-cell-like cells (2CLCs) (Macfarlan et al., 2012) depends on DUX, a pioneer transcription factor that activates ZGA genes (De Iaco et al., 2017). Additional ZGA drivers include OBOX, NFYA, NR5A2, and SRF (De Iaco et al., 2017; Iaco et al., 2020; Gassler et al., 2022; Hermant et al., 2025; Ji et al., 2023; Lu et al., 2016), which open chromatin and activate transcription in a combinatorial or compensatory fashion (Kojima et al., 2025). Although DUX and individual OBOX proteins are dispensable for ZGA in embryos (Chen and Zhang, 2019; Iaco et al., 2020; Guo et al., 2019; Ji et al., 2023), their combined loss (Ji et al., 2023), or loss of both DUX and OBOX4 (Guo et al., 2024), impairs development, suggesting functional redundancy. While recent studies shed light on ZGA initiation, the mechanisms responsible for extinguishing the 2c transcriptional program remain poorly understood.

Endogenous retroviruses (ERVs) are key regulators of early development. Murine endogenous retrovirus-L (MERVL) elements are transcriptionally active in 2c embryos (Svoboda et al., 2004) and drive chimeric transcripts that define the totipotent state (Macfarlan et al., 2012; Peaston et al., 2004). MERVL activation is required for blastocyst formation (Sakashita et al., 2023) and is mediated by DUX/OBOX and SRF binding at MT2 long terminal repeats (LTRs) (Hermant et al., 2025; Yang et al., 2024). However, sustained MERVL expression—such as that induced by forced DUX expression—disrupts totipotency exit and embryo development (Guo et al., 2019; Percharde et al., 2018). Although timely repression of MERVLs and the 2c program requires DUXBL (Vega-Sendino et al., 2024), the mechanisms by which DUXBL silences totipotency-related genes remain unknown.

Because genome-wide DNA demethylation occurs after fertilization, preimplantation embryos rely heavily on histone modifications to regulate transcription. During cleavage-stage development, H3K9me3 is deposited at retrotransposons; MERVL elements begin to acquire this repressive mark by the 2c stage, as DNA methylation levels decline (Wang et al., 2018). SETDB1, a key H3K9 methyltransferase, is thus a strong candidate for regulating MERVL silencing and early embryonic transcriptional control. In ESCs, SETDB1 represses transposable elements (TEs) via TRIM28 (KAP1)-dependent H3K9me3 deposition and by maintaining DNA methylation at LTRs and imprinted DMRs (Karimi et al., 2011; Leung et al., 2014; Matsui et al., 2010). However, its role at MERVLs in ESCs is debated: some studies report minimal repression (Maksakova et al., 2013), while others show that SETDB1 silences MERVLs to prevent dedifferentiation into 2CLCs (Wu et al., 2020). In oocytes, Setdb1 loss leads to derepression of ERVK and ERVL-MaLR retrotransposons (Eymery et al., 2016; Kim et al., 2016). Whether SETDB1 also regulates MERVLs and totipotency exit in cleavage-stage embryos remains unresolved.

To address this question and understand why Setdb1^mat-/+ embryos fail to exit the cleavage stages, we employed a conditional knockout strategy. We previously showed that maternal SETDB1 protects the maternal pronucleus from TET3-mediated DNA demethylation in zygotes (Zeng et al., 2019). Here, we investigate its role in preimplantation development by analyzing the transcriptomes of 2c and rare 8c Setdb1^mat-/+ embryos. We find that maternally deposited SETDB1 is required to silence MERVL expression and 2c-specific totipotency transcripts, enabling successful progression beyond the cleavage stages.

Results

Setdb1 maternal knockout embryos fail preimplantation development

A complete understanding of early embryonic development—and its intrinsic directionality and tempo—requires in vivo approaches. To address the in vivo role of SETDB1 in controlling the earliest stages of embryo development, we applied a genetic approach. We crossed Setdb1^f/f; Zp3-cre females with wild-type JF1/Ms males to obtain maternal knockout (Setdb1^mat-/+) (KO) embryos. Control Setdb1^f/f (WT) embryos were derived from Setdb1^f/f females lacking the Zp3-cre transgene. We previously reported that H3K9me2 and H3K9me3 histone marks are globally reduced in Setdb1^mat-/+ zygotes following Zp3-cre-driven excision of the SET domain from growing oocytes (Zeng et al., 2019).

The maternal Setdb1 transcript is present in oocytes and persists throughout preimplantation development, while zygotic expression from the paternal allele begins only at the blastocyst stage (Dodge et al., 2004). Consequently, Setdb1^mat-/+ embryos lack SETDB1 protein function from the oocyte through the 8c stage.

At 1.5 dpc, over 70% of control eggs had reached the 2c stage, with a few progressing to the 4c stage (Figure 1A). In contrast, fewer than 30% of Setdb1^mat-/+ eggs reached the 2c stage, with many arrested at GV, MI, or MII stages. By 2.5 dpc, ~80% of control embryos reached the 8c stage, while only ~4% of mutant embryos did so. These findings agree with earlier studies (Eymery et al., 2016; Kim et al., 2016) and show that preimplantation development is severely impacted in Setdb1^mat-/+ embryos.

Figure 1—figure supplement 1. — (A) Quantification of *Setdb1*^mat-/+ (KO) and *Setdb1*^fl/+ (WT) embryo stages from the following number of total recovered embryos: KO (n=638), WT (n=484) at 1.5 dpc, and KO (n=310) and WT (n=80) at 2.5 dpc. (B) Principal component analysis of single-embryo total RNA-seq data from 2cWT (n=6), 2cKO (n=15), 8cWT (n=8), and 8cKO (n=8) embryos (Figure 1—source data 1). (C) Schematic of four pairwise comparisons defining requirements for normalcy and development. (D) Volcano plots highlighting differentially expressed genes (DEGs) using |log₂FC|>1 and adjusted p<0.05. (**E, F**) Four-way DEG comparisons (Figure 1—source data 2) visualized by Venn diagrams: (E) downregulated; (F) upregulated. (G–J) Heatmaps of DEGs from Venn compartments, showing stage- and genotype-specific patterns.

Figure 1—source data 1. Sample information.
Details of all single embryos analyzed in the study, including genotype, developmental stage, and sequencing sample ID.

elife-109248-fig1-data1.xlsx^{(10.5KB, xlsx)}

Figure 1—source data 2. Differential gene expression (DGE) analysis.
DGE analysis was conducted using the four pairwise comparisons shown in Figure 1C and D, with thresholds of |log₂FC|>1 and adjusted p<0.05. The ‘expression’ columns indicate whether each transcript is up- or downregulated in any given comparison. The ‘FOUR_WAY_UP_COMP’ and ‘FOUR_WAY_DOWN_COMP’ columns assign DEGs to compartments corresponding to the Venn diagrams in Figure 1E and F. Sample columns (TZSC118–TZSC116) contain log₂-transformed counts per million (CPM) values, generated using the cpm function in edgeR (v4.4.1).

elife-109248-fig1-data2.xlsx^{(6.4MB, xlsx)}

Single-embryo total RNA sequencing of Setdb1^mat-/+ embryos

As no Setdb1^mat-/+ embryos develop into blastocysts, we performed single-embryo RNA sequencing at the 2c and 8c stages (1.5 and 2.5 dpc, respectively) to determine how maternal SETDB1 supports development beyond the cleavage stage. The KO embryos that did develop to the 2c or 8c stage were morphologically indistinguishable from WT under light microscopy (Figure 1—figure supplement 1A). To optimally assess both regular and chimeric transcription, we used total RNA sequencing. We profiled both KO and WT embryos with >6 biological replicates (Figure 1). Examples of normally expressing transcripts are displayed in Figure 1—figure supplement 1B.

Principal component analysis (PCA) revealed that samples clustered primarily by developmental stage (PC1), and secondarily by genotype (PC2) (Figure 1B), indicating that stage-specific transcriptional programs remain largely intact—more distinctly so in the rare 8c-stage embryos. We defined differentially expressed genes (DEGs) using thresholds of |log₂FC|>1 and adjusted p<0.05 across four pairwise comparisons (Figure 1C).

Volcano plots (Figure 1D) visualize the four key comparisons, and we provide the matching gene set enrichment analysis (GSEA) to each pairwise contrast (Figure 1—figure supplement 2): ‘To be normal’ at the 2c (2cWT vs. 2cKO) and 8c (8cWT vs. 8cKO) stages; and ‘To develop’ from 2c to 8c in WT (8cWT vs. 2cWT) and KO (8cKO vs. 2cKO) embryos. The volcano plots and an intersection analysis (Figure 1E and F) using a four-way Venn diagram approach revealed that most DEGs fell into the ‘To develop’ category, consistent with the PCA results.

Transcriptional programs of development do not collapse in the absence of maternal SETDB1

We first examined DEGs associated with the ‘To develop’ category. We identified 1722 genes downregulated during development in both WT (8cWT vs. 2cWT) and KO (8cKO vs. 2cKO) embryos (Figure 1E), suggesting their normal downregulation is not dependent on SETDB1. This trend was confirmed by heatmap clustering (Figure 1—figure supplement 3A).

Gene ontology (GO) terms associated with these downregulated DEGs included DNA binding, transcriptional activation, ubiquitination, and phospholipid binding (Figure 1—figure supplement 3B). Similarly, 828 genes were consistently upregulated during development in both WT and KO embryos (Figure 1F) and were enriched in RNA production and translation (Figure 1—figure supplement 3D). Together, these results indicate that maternal SETDB1 is not required for the global execution of 2c-to-8c transcriptional transitions, and that the developmental arrest observed is not due to a complete transcriptional collapse.

The two-cell program is misregulated in the absence of maternal SETDB1

To understand which developmental change occurs specifically in Setdb1^mat-/+ embryos, we focused on genes differentially expressed only in KO embryos during the 2c to 8c transition. We identified 2379 downregulated and 1166 upregulated DEGs in the 8cKO vs. 2cKO contrast, but not in the WT developmental comparison (Figure 1E and F).

Heatmaps revealed that many genes exhibited aberrant expression in 2cKO embryos, with levels reverting to near-normal by the 8c stage (Figure 1—figure supplement 4A and C). Overrepresentation analysis (ORA) indicated that 2cKO embryos underexpressed genes related to ribosomal and mitochondrial function (Figure 1—figure supplement 4B), while overexpressing genes associated with later developmental processes such as pattern specification and organ morphogenesis (Figure 1—figure supplement 4D). These observations suggest that mis-timed activation or repression occurs in the absence of maternal SETDB1.

In a related heat map comparison (Figure 1—figure supplement 4E), we identified genes uniquely overexpressed in 2cKO embryos that are enriched in ribosomal and mitochondrial pathways (Figure 1—figure supplement 4F). Thus, maternal SETDB1 is required to properly regulate two-cell-stage gene expression dynamics.

Maternal SETDB1 has a broad effect on transposable elements in 2c and 8c embryos

We next examined whether maternal SETDB1 regulates TE expression during preimplantation. PCA of TE expression in our RNA-seq data revealed clustering first by stage and then by genotype, with increased variability among 8cKO samples (Figure 1—figure supplement 5A). Multimapped (Figure 1—figure supplement 5B) and unimapped (Figure 1—figure supplement 5—source data 2) TE analysis showed that maternal SETDB1 is required to suppress TEs in WT vs. KO embryos and to control TEs during 2c to 8c development.

The tallies of ‘To be normal’ DE TEs by TE families show that several TE families—including ERVK, ERVL, ERVL-MaLR, ERV1, and L1 elements—require suppression by SETDB1 in WT embryos (Figure 1—figure supplement 5C). The ERVK, ERVL-MaLR, and ERV1 families were most affected in both 2cWT vs. 2cKO and 8cWT vs. 8cKO contrasts, while LINEs required SETDB1 for repression mainly at 2c. The tallies of ‘To develop’ suggest that SETDB also affects developmental TE changes during 2c-to-8c transitions. ERVK and LINE-1 elements appear to be developmentally activated in WT but less so in KO embryos (Figure 1—figure supplement 5D), which we investigate further below.

Heatmaps of DE TEs (Figure 1—figure supplement 5E) revealed strong derepression of nearly all repeat families in KO embryos at both the 2c and 8c stages. In WT embryos, L1 LINEs and many ERVKs were upregulated between 2c and 8c, while MERVLs and ERVL-MaLRs were appropriately silenced (Figure 1—figure supplement 5F). This developmental regulation was disrupted in KO embryos: 8cKO vs. 2cKO showed partial MERVL/ERV1 repression and inappropriate activation of some ERVKs and LINEs—due to the highly derepressed state in 2cKO embryos.

These results collectively show that maternal SETDB1 is a major regulator of TE expression, with broad effects on multiple TE families including the 2c hallmark MERVL during early development.

Maternal SETDB1 is required for timely suppression of cleavage-stage transcripts

To further explore SETDB1’s role in establishing a normal 2c transcriptome, we looked at 2cWT vs. 2cKO DEGs and found 171 downregulated and 238 upregulated transcripts (Figure 1E and F). Among the 69 downregulated genes exclusive to 2cWT vs. 2cKO, many are prematurely activated in 2cKO embryos, including Klf10, Lig1, and H1f3 (Figure 1G). Another 59 DEGs downregulated in 2cWT vs. 2cKO overlapped with 8cKO vs. 2cKO DEGs and were misexpressed only at the 2c stage (Figure 1H), suggesting these transcripts are normally repressed by SETDB1 at 2c but not at 8c. A notable example is Zscan4c, a key regulator of the 2c program in vitro (Eckersley-Maslin et al., 2019).

To understand SETDB1’s role at the 8c stage, we inspected the 636 downregulated and 302 upregulated DEGs in 8cWT vs. 8cKO embryos (Figure 1). Of these, 270 overlapped with 8cWT vs. 2cWT DEGs (Figure 1E), indicating that maternal SETDB1 normally suppresses these transcripts during the 2c to 8c transition (Figure 1I). This group included Usp17ld, Tdpoz1, and Tdpoz3, transcripts expressed transiently at the 1c, 2c, or 4c stages (Park et al., 2015).

An additional 121 DEGs overlapped with both developmental and genotype comparisons, suggesting that their silencing occurs more slowly in Setdb1^mat-/+ embryos than in controls (Figure 1J). These included the 2c transient genes (Park et al., 2015) Obox3 and Fzd7, further implicating maternal SETDB1 in regulating 2c-specific transcriptional shutdown. Taken together, this four-way comparison approach highlights maternal SETDB1 as a key factor for the timely suppression of specific transcripts at both the 2c and 8c stages.

Maternal SETDB1 controls transient gene expression during ZGA

We next asked whether maternal SETDB1 regulates specific ZGA gene classes as defined in the ‘Database of Transcriptome in Mouse Early Embryos’ (DBTMEE) (Park et al., 2015). ORA (Figure 2A) using DBTMEE gene clusters from Sakashita et al., 2023 revealed distinct patterns across four two-way DEG comparisons.

Figure 2. — (A) Bubble plot showing overrepresentation analysis of Database of Transcriptome in Mouse Early Embryos (DBTMEE) (Park et al., 2015)-defined transcript sets among differentially expressed genes (DEGs) identified in the four pairwise comparisons (Figure 2—source data 1). (B) IGV browser snapshots of representative transcripts from the DBTMEE-defined minor ZGA to MGA, two-cell transient, and four-cell transient gene sets across five biological replicates. Venn diagram compartments are indicated above each track. Units in brackets represent normalized counts per million (CPM). Bigwig tracks are shown in the transcriptional direction matching the depicted gene. (C) Boxplots of selected DEGs *(|log₂FC|>1, adjusted p<0.05) from each Venn compartment, based on data from 2cWT (n=6), 2cKO (n=15), 8cWT (n=8), and 8cKO (n=8) embryos.

Figure 2—source data 1. Overrepresentation analysis (ORA) with Database of Transcriptome in Mouse Early Embryos (DBTMEE) categories.
ORA was performed using DBTMEE-defined (Park et al., 2015) gene sets as in Sakashita et al., 2023 and DEGs up-/downregulated from the pairwise comparisons. The analysis was run using the function enricher from the R package ClusterProfiler (v4.14.4). Results are shown only for sets with significant enrichment.

elife-109248-fig2-data1.xlsx^{(817.1KB, xlsx)}

Maternal RNA and minor ZGA genes were appropriately downregulated in both WT and KO embryos from the 2c to 8c stage, as expected during normal development. However, 2c-specific transient transcripts were significantly downregulated in 8cWT vs. 2cWT embryos but not in 8cKO vs. 2cKO embryos. Moreover, 8cWT vs. 8cKO downregulated DEGs were enriched in this gene set, indicating a requirement for maternal SETDB1 to suppress 2c-transient transcripts by the 8c stage.

Similarly, 1c transient genes at the 2c stage and minor ZGA genes at the 8c stage were also insufficiently silenced in KO embryos. By contrast, mid-preimplantation gene activation (MGA) genes and their upstream regulators were upregulated normally in both genotypes. Notably, major ZGA and MGA transition-related genes were also underexpressed in 2cKO embryos, suggesting that SETDB1 may also support their proper induction.

Representative IGV browser tracks illustrate misregulation patterns in five biological replicates per condition (Figure 2B), including Zscan4c (minor ZGA), Usp17ld (1c transient) Dbf4 (2c transient), and Tdpoz3 (4c transient). Differential expression status is mapped across Venn diagram segments (Figure 1E), and selected transcripts are shown in boxplots (Figure 2C).

These results demonstrate that maternal SETDB1 is required to suppress 1c transient genes by the 2c stage, and minor ZGA and 2c-specific genes by the 8c stage, ensuring transcriptional fidelity across cleavage stages.

Maternal SETDB1 suppresses two-cell-specific MERVL-chimeric transcripts

MERVL retroelements act as alternative promoters for chimeric transcripts during the 2c stage and serve as molecular hallmarks of totipotency (Macfarlan et al., 2012). Although SETDB1 represses ERVK and ERVL-MaLR retrotransposons in oocytes, it does not affect MERVLs at that stage (Kim et al., 2016).

We found significantly higher transcript levels of MERVLint:ERVL:LTR and MT2_Mm:ERVL:LTR in 2cKO than in 2cWT embryos (Figure 3A). These transcripts were efficiently extinguished in WT embryos by the 8c stage but remained partially expressed in KO embryos. The difference was even more pronounced at 8c: MT2 expression was downregulated in 8cWT vs. 8cKO with LFC = –2.65 and p adj=1.31E–16, compared to LFC = –1.86 and p adj=1.41E–08 in 2cWT vs. 2cKO.

Similarly, many MERVL chimeric transcripts defined by Macfarlan et al., 2012 were upregulated in both 2cKO and 8cKO embryos (Figure 3B), with most clustering on the left side of volcano plots (downregulated in WT vs. KO) (Figure 3C and D). Two shared DEGs reached significance at 2c, and nine at 8c (p adj <0.05, |log₂FC|>1). IGV views show misregulation of representative transcripts Slc19a3 and Trak2 (Figure 3E).

Using a combined list of previously identified (Modzelewski et al., 2021) and novel chimeric transcripts, we performed a four-way comparison to assess SETDB1 dependency (Figure 3F). SETDB1 had a strong suppressive effect on TE-driven, including MERVL-driven transcripts at both 2c and 8c stages, with no clear activating role. Thus, maternal SETDB1 is essential to silence 2c-specific MERVL chimeric transcripts during preimplantation development.

Maternal SETDB1 silences MERVL MT2 LTR-activated transcripts

MT2s represent major MERVL subfamilies. Their LTRs can act as alternative promoters or distal enhancers for endogenous genes, depending on their genomic position and orientation. A recent study by Yang et al., 2024 used a CRISPR-based epigenetic MT2 LTR inactivation system (MT2i) to identify MT2-controlled genes in mouse embryos. Using ChIP-seq data from Wang et al., 2018 we found that H3K9me3 deposition at MT2-dependent transcripts is first reduced between E2c and L2c but later increases from 4c to the 8c stage (Figure 4A). MT2s that drive those transcripts accumulate H3K9me3 peaks progressively from the E2c to 8c stages (Figure 4B), and these peaks are stronger and more spatially defined. Peaks localized at the MT2 edges are consistent with low internal mappability in repeat elements. IGV browser views show that the expression of MT2-controlled DEGs such as Arsk and Usp17ld was derepressed in 2cKO and 8cKO embryos, while their flanking MERVLs and MT2s displayed strong H3K9me3 peaks in WT embryos (Figure 4C–E).

Heatmaps reveal that the expression of MT2-controlled DEGs in general was derepressed in 2cKO and 8cKO embryos (Figure 4F). These genes clustered on the left of volcano plots (Figure 4G and H), indicating strong SETDB1-mediated repression. Yang et al., 2024 identified 341 genes requiring MT2 for activation by early 2c (E2c) and 1111 by late 2c (L2c). We observed that many of these genes were derepressed in SETDB1 KO embryos across time (Figure 5A–E). These results provide genetic evidence that maternal SETDB1 is essential for silencing MT2-activated genes.

Figure 5. — (A) Heatmaps comparing MT2i-responsive (Yang et al., 2024) early two-cell (E2c) differentially expressed genes (DEGs) with the *Setdb1* KO transcriptomes. (**B, C**) Volcano plots highlighting MT2-regulated E2c DEGs in 2cWT vs. 2cKO and 8cWT vs. 8cKO pairwise comparisons. (D) Heatmaps of MT2i late two-cell (L2c) DEGs. (E) Volcano plot highlighting MT2-regulated L2c DEGs.

Maternal SETDB1 is required to extinguish DUXBL-responsive transcripts

DUXBL is a known repressor of ZGA and MERVL genes, acting in the footsteps of DUX during the exit from totipotency (Vega-Sendino et al., 2024). We tested whether maternal SETDB1 is required for DUXBL-mediated suppression. We compared our RNA-seq data to the top 50 DEGs from Duxbl^-/- embryos (Vega-Sendino et al., 2024) and found >2-fold derepression of 7 and 12 DUXBL-suppressed transcripts in 2cKO and 8cKO embryos, respectively (Figure 6A), suggesting that maternal SETDB1 acts specifically in their silencing. Volcano plots (Figure 6B) showed DUXBL-upregulated genes clustering on the left in both 2cWT vs. 2cKO and 8cWT vs. 8cKO comparisons, indicating again SETDB1-dependent suppression.

IGV browser views confirmed that Dennd4c, Tmem132c (Figure 6D and E), Antxr1, and Aqr (Figure 6—figure supplement 1A and B) were derepressed in KO embryos and associated with H3K9me3 peaks flanking MERVL elements in WT embryos. Although H3K9me3 signal within MERVLs is low due to mappability limits, adjacent flanks exhibited strong enrichment beginning at the 2c stage and increasing through 4c to morula (Wang et al., 2018). Two additional genes, Nelfa and Zscan4d, appeared derepressed in KO embryos (Figure 6—figure supplement 1C and D), despite not crossing significance thresholds. MERVLs serve as alternative promoters or distal enhancers for these transcripts, with H3K9me3 deposited near these elements in WT but not KO embryos.

Together, these data provide genetic evidence that DUXBL requires maternal SETDB1 to silence a subset of its targets via H3K9me3 deposition at MERVL LTRs.

Overrepresentation analysis confirms shared targets with MT2 and DUXBL pathways

We performed ORA to test for overlap between SETDB1-sensitive genes and those misregulated by MT2i (Yang et al., 2024) or DUXBL KO (Vega-Sendino et al., 2024; Figure 7A). We found highly significant overlap between proximal MT2-controlled DEGs and DEGs downregulated in 2cWT vs. 2cKO (p adj=3.19E–04); proximal MT2-controlled DEGs and DEGs in 8cWT vs. 8cKO (p adj=7.17E–04); E2c MT2i-downregulated genes and SETDB1-repressed DEGs at 2c (p adj=8.18E–05); and E2c MT2i-downregulated genes and SETDB1-repressed DEGs at 8c (p adj=5.77E–11). This suggests that genes activated by MT2s during ZGA also require maternal SETDB1 for their timely silencing later.

Figure 7. — (A) Bubble plot of overrepresentation analysis showing enrichment of previously identified MERVL-chimeric, MT2i-responsive, and *Duxbl* KO-responsive gene sets (Macfarlan et al., 2012; Vega-Sendino et al., 2024; Yang et al., 2024) in *Setdb1* KO pairwise differentially expressed genes (DEGs) (Figure 7—source data 1). (B) Summary model: Maternal SETDB1 deposits H3K9me3 at MT2 elements to silence MERVL-driven two-cell transcripts in coordination with DUXBL, enabling exit from totipotency.

Figure 7—source data 1. ORA with MERVL-chimeric, MT2i, and *Duxbl* KO gene sets.
Overrepresentation analysis of up/down pairwise DEGs using gene sets from Macfarlan et al., 2012 (MERVL-chimeric), Yang et al., 2024 (MT2i), and Vega-Sendino et al., 2024 (*Duxbl* KO). ORA was performed if there were more than 50 genes using function enricher from R package ClusterProfiler (v4.14.4). Results are shown only for sets with significant enrichment.

elife-109248-fig7-data1.xlsx^{(28.6KB, xlsx)}

We also observed strong enrichment between DUXBL-suppressed genes and SETDB1-repressed DEGs in 8cWT vs. 8cKO embryos (p adj = 0.0071), providing genetic evidence for a collaboration between DUXBL and SETDB1 in extinguishing the 2c program.

In summary, these findings establish SETDB1, a H3K9 methyltransferase, as a broad-acting chromatin regulator in the preimplantation-stage embryo required for: silencing MERVL retroviral elements; extinguishing minor ZGA and transient totipotency-associated genes; and terminating transcription of MT2-activated and DUXBL-regulated transcripts at key regulatory elements in cleavage-stage embryos.

Discussion

SETDB1 is essential for mammalian development due to its role in establishing transcriptionally repressive H3K9me3 marks. Our findings indicate that the maternal-effect lethality observed in Setdb1^mat-/+ preimplantation embryos is not due to a global transcriptional collapse, but rather the failure to silence specific transcriptional programs that must be extinguished for normal progression through, and beyond, the cleavage stages. Exit from the totipotency state is essential for embryonic development (Guo et al., 2019; Percharde et al., 2018). While SETDB1 affects a broad range of genes and retroelements during preimplantation development, its specific role in silencing the totipotency program alone underscores its critical role as a chromatin regulator in early embryogenesis.

Maternal SETDB1 functions in a highly specific and dynamic manner. At the 2c stage, it suppresses 1c transient and minor ZGA transcripts; by the 8c stage, it silences 2c transient and MERVL-associated chimeric transcripts. In normal development, these silencing events coincide with dynamic H3K9me3 deposition at MERVLs and MT2 LTRs (Figure 4). In ESCs, SETDB1 prevents dedifferentiation into 2c-like cells by depositing H3K9me3 at or near totipotency genes (Wu et al., 2020), suggesting that a similar mechanism may operate during preimplantation embryo development in vivo. However, the rarity of Setdb1^mat-/+ 8c embryos limits our ability to directly confirm H3K9me3 depletion at these loci—a limitation of this study.

SETDB1’s mechanism likely involves recruitment to specific retroelement loci, such as MT2 LTRs by transcription factors, such as DUXBL. Our genetic data show that SETDB1 is required for silencing both MT2 LTRs and DUXBL-responsive transcripts. Detecting overlap of SETDB1-regulated genes with those affected by MT2 inactivation (Yang et al., 2024) and DUXBL loss (Vega-Sendino et al., 2024), along with the accumulation of H3K9me3 at these loci during normal preimplantation development, suggests a coordinated silencing mechanism in the embryo. Phenotypic parallels between Setdb1^mat-/+ and Duxbl^-/- embryos also support this model. While Duxbl^-/- embryos arrest at the 4c stage (Vega-Sendino et al., 2024), a few Setdb1^mat-/+ embryos progress to the 8c stage, suggesting that SETDB1 may act redundantly with, or downstream of, DUXBL.

In conclusion, we identify maternal SETDB1 as a critical in vivo epigenetic regulator that extinguishes totipotency-associated transcriptional programs, particularly those driven by retroelements and responsive to DUXBL. Its role at MT2 elements supports a model in which SETDB1 provides an epigenetic memory, silencing retroelements shortly after their activation by DUX and OBOX. This mechanism rapidly restricts the 2c transcriptional program, enabling the transition to pluripotency. By repressing MT2 LTRs, SETDB1 ensures successful progression beyond the cleavage stages and supports the earliest cell fate transitions in the mammalian embryo (Figure 7B).

Materials and methods

Key resources table.

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Gene (Mus musculus)	Setdb1	NCBI	Gene: 84505
Genetic reagent (M. musculus)	Setdb1 ^{tm1a(EUCOMM)Wtsi}	European Mouse Mutant Archive	EMMA ID EM:04814	Skarnes et al., 2011
Genetic reagent (M. musculus)	B6.Cg-Tg^{(Pgk1-flpo)10Sykr/J}	The Jackson Laboratory	RRID:IMSR_JAX:011065	Wu et al., 2009
Genetic reagent (M. musculus)	C57BL/6-Tg^{(Zp3-cre)93Knw/J}	The Jackson Laboratory	RRID:IMSR_JAX:003651	de Vries et al., 2000
Software, algorithm	StringTie2	https://github.com/skovaka/stringtie2	RRID:SCR_016338	Kovaka et al., 2019
Software, algorithm	Trim Galore (v0.60)	https://github.com/FelixKrueger/TrimGalore	RRID:SCR_011847	Krueger, 2025 Martin, 2011
Software, algorithm	edgeR (v4.4.1)	https://bioconductor.org/packages/edgeR/	RRID:SCR_012802	Chen et al., 2016; McCarthy et al., 2012; Robinson et al., 2010
Software, algorithm	ClusterProfiler (v4.14.4)	https://bioconductor.org/packages/clusterProfiler/	RRID:SCR_016884	Wu et al., 2021
Software, algorithm	STAR (v2.7.8)	https://github.com/alexdobin/STAR	RRID:SCR_004463	Dobin et al., 2013
Software, algorithm	Samtools (v1.17)	https://github.com/samtools/samtools/	RRID:SCR_002105	Li et al., 2009
Software, algorithm	FeatureCounts (Subread v2.0.0)	https://subread.sourceforge.net/featureCounts.html	RRID:SCR_012919	Liao et al., 2014
Software, algorithm	DESeq2 (v1.46.0)	https://github.com/thelovelab/DESeq2	RRID:SCR_000154	Love et al., 2014
Software, algorithm	BWA (v0.7.1)	https://github.com/lh3/bwa	RRID:SCR_010910	Li and Durbin, 2009
Software, algorithm	deepTools2 v3.5.2	https://github.com/deeptools/deepTools	RRID:SCR_016366	Ramírez et al., 2016
Software, algorithm	ggplot2	https://ggplot2.tidyverse.org	RRID:SCR_014601	Wickham, 2016
Software, algorithm	ComplexHeatmap (R)	https://jokergoo.github.io/ComplexHeatmap-reference/book/	RRID:SCR_017270	Gu et al., 2016
Software, algorithm	ggvenn (v0.1.10)	https://github.com/yanlinlin82/ggvenn	RRID:SCR_025300	Yan, 2023
Software, algorithm	TEtranscripts (v2.2.3)	https://github.com/mhammell-laboratory/TEtranscripts	RRID:SCR_015687	Jin et al., 2015
Chemical compound, drug	CARD HyperOva	Cosmo Bio	Cat. No. KYD-010-EX
Commercial assay or kit	SMART-Seq Stranded Kit	Takara Biosciences USA, Mountain View CA	Cat. No. 634444
Commercial assay or kit	RNA-Bee	Amsbio	Cat. No. CS-105B
Peptide, recombinant protein	RNasin ribonuclease inhibitor	Promega	Cat. No. N2515
Commercial assay or kit	DNA-free Kit	Ambion	Cat. No. AM1906
Commercial assay or kit	Agilent High Sensitivity DNA Kit	Agilent Technologies, Inc.	Part Number:5067-4626
Commercial assay or kit	QuantiFluor dsDNA System	Promega Corp., Madison, WI, USA	Cat. No. E2671
Commercial assay or kit	Kapa Illumina Library Quantification qPCR assays	Kapa Biosystems	Cat. No. KR0405

Open in a new tab

Mice

All animal experiments were conducted in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals, under Institutional Animal Care and Use Committee-approved protocol (AUP ref.no. 23-12-032) at the Van Andel Institute (VAI). The Setdb1 conditional knockout mouse line (Setdb1^f/f) was derived from Setdb1^{tm1a(EUCOMM)Wtsi} (Skarnes et al., 2011) (European Mouse Mutant Archive) as previously described (Zeng et al., 2019). The Pgk-neo cassette was removed via FLPE-mediated recombination using B6.Cg-Tg^{(Pgk1-flpo)10Sykr/J} (Wu et al., 2009). Zp3-cre transgenic mice C57BL/6-Tg^{(Zp3-cre)93Knw/J} (de Vries et al., 2000) were used to excise SET-domain–encoding exons in growing oocytes. Setdb1^f/f; Zp3-cre females were experimental animals; Setdb1^f/f females without the transgene served as controls.

Collection of preimplantation mouse embryos

Embryos were collected at late 2c (L2c) and 8c stages. For 2c embryos, 6–8-week-old females were superovulated with 5 IU PMSG followed by 5 IU hCG after 46–48 h, then mated with WT males. For 8c embryos, CARD HyperOva was used in 26–30-day-old females to increase efficiency. Embryos were harvested in M2 medium at 44 h (1.5 dpc, 2c) and 68 h (2.5 dpc, 8c) post-hCG. Embryos were washed in PBS, placed into 1.5 mL tubes, lysed in RNA-Bee, flash frozen, and stored at –80°C.

RNA isolation

RNA was isolated from single embryos using RNA-Bee. Linear polyacrylamide (LPA) was used for carrier-mediated isopropanol precipitation. RNasin (Promega) was added during resuspension. DNA contamination was removed using rDNase I (Ambion). RNA was recovered after DNase inactivation and ethanol precipitation and eluted in RNasin-containing Tris buffer.

Library preparation and sequencing

Libraries were prepared with the SMART-Seq Stranded Kit (Takara). Samples were randomized and processed with fragmentation to 250 bp (85°C, 10  min embryo RNA or 6  min control RNA). Unique dual indexes were added during PCR1 (10 cycles), and eight samples were pooled for PCR2 (12 cycles). Libraries were QC’d using Agilent chips, QuantiFluor, and KAPA qPCR, and sequenced on a NovaSeq S2 (paired-end 50  bp, ~25M reads/library).

RNA-seq analysis

Reads were trimmed using Trim Galore (v0.60) (Krueger, 2025; Martin, 2011) and aligned to mm10 with STAR (v2.7.8) (Dobin et al., 2013) using --quantMode GeneCounts. DGE analysis was performed with edgeR (v4.4.1) (Chen et al., 2016; McCarthy et al., 2012; Robinson et al., 2010). Briefly, gene counts were filtered to remove lowly expressed genes and normalized using the trimmed mean of M-values (TMM) method. Gene-wise dispersion was calculated and differential expression was assessed using a negative binomial generalized linear model in edgeR. P-values were adjusted for multiple testing using Benjamini–Hochberg adjustment (adjusted p<0.05). Normalized log₂ counts were used in figures. Bigwig tracks were created using bamCoverage (deepTools v3.5.2, CPM normalization) (Ramírez et al., 2016). GSEA and ORA against GO and custom gene sets were performed with ClusterProfiler (v4.14.4) (Wu et al., 2021).

Transposable element (TE) analysis

Two sets of data were analyzed for TE: multimapped reads and uni-mapped reads. For multimapped reads, STAR (v2.7.8) (Dobin et al., 2013) was run with --winAnchorMultimapNmax 200, --outFilterMultimapNmax 100, and --outSAMmultNmax 1. TEtranscripts (v2.2.3) (Jin et al., 2015) was used for multi-mapped TE counts. Uni-mapped reads were filtered with -q 255 using Samtools (v1.17) (Li et al., 2009). FeatureCounts (Subread v2.0.0) (Liao et al., 2014) was used with mm10_rmsk_TE.gtf to generate count matrices. Differential expression was assessed using DESeq2 (v1.46.0) (Love et al., 2014).

Chimeric transcript analysis

Chimeric transcripts were quantified with Retrotransposon (v37), which uses STAR (Liao et al., 2014) and StringTie2 (Kovaka et al., 2019). Count tables were processed using DESeq2 (Love et al., 2014) for DGE (adjusted p<0.05).

Public datasets

ChIP-seq fastq files (Wang et al., 2018) (GSE98149) were trimmed using Trim Galore (v0.60) (Krueger, 2025; Martin, 2011), aligned to mm10 with BWA (v0.7.1) (Li and Durbin, 2009), and filtered with Samtools (v1.17) (Li et al., 2009). BAMs were converted to BigWig and visualized with deepTools (Ramírez et al., 2016). Supplementary data was used from other studies (Eymery et al., 2016; Macfarlan et al., 2012; Park et al., 2015; Vega-Sendino et al., 2024; Yang et al., 2024), for comparative analyses.

Data visualization

Heatmaps were generated with ComplexHeatmap (R) (Gu et al., 2016), bubble plots were generated using ggplot2 (Wickham, 2016), and Venn diagrams using ggvenn (v0.1.10) (Yan, 2023).

Acknowledgements

We gratefully acknowledge the support from the Van Andel Institute (VAI) Genomics Core (RRID:SCR_022913), the VAI Bioinformatics and Biostatistics Core (RRID:SCR_024762), and the VAI Vivarium (RRID:SCR_023211). This work was supported by the Van Andel Institute.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Piroska E Szabó, Email: piroska.szabo@vai.org.

Yan Song, Peking University, Beijing, China.

Lori Sussel, University of Colorado Anschutz Medical Campus, Aurora, United States.

Funding Information

This paper was supported by the following grant:

Van Andel Institute to Piroska E Szabó.

Additional information

Competing interests

No competing interests declared.

Author contributions

Investigation, Writing – review and editing.

Formal analysis, Visualization, Writing – original draft, Writing – review and editing.

Investigation, Methodology, Writing – review and editing.

Investigation, Writing – review and editing.

Investigation, Methodology, Writing – review and editing.

Conceptualization, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing – original draft, Project administration, Writing – review and editing.

Ethics

All animal experiments were conducted in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals, under Institutional Animal Care and Use Committee-approved protocol (AUP ref.no. 23-12-032) at the Van Andel Institute (VAI).

Additional files

MDAR checklist

elife-109248-mdarchecklist1.docx^{(88.4KB, docx)}

Data availability

RNA sequence data were deposited in GEO (accession number GSE269417). All data generated or analyzed during this study are included in the manuscript and supporting files; source data files have been provided for all figures.

The following previously published datasets were used:

Wang C, Liu X, Gao Y, Yang L, Li C, Liu W, Chen C, Kou X, Zhao Y, Wang H, Zhang Y, Gao S. 2018. Reprogramming of H3K9me3-dependent heterochromatin during mammalian early embryo development [ChIP-seq] NCBI Gene Expression Omnibus. GSE98149

Szabó PE, Zeng T, Fu Z. 2025. Single embryo total RNA sequencing to map the effect of maternal SETDB1 on the 2-celll and 8-cell transcriptome. NCBI Gene Expression Omnibus. GSE269417

References

Abe KI, Funaya S, Tsukioka D, Kawamura M, Suzuki Y, Suzuki MG, Schultz RM, Aoki F. Minor zygotic gene activation is essential for mouse preimplantation development. PNAS. 2018;115:E6780–E6788. doi: 10.1073/pnas.1804309115. [DOI] [PMC free article] [PubMed] [Google Scholar]
Bilodeau S, Kagey MH, Frampton GM, Rahl PB, Young RA. SetDB1 contributes to repression of genes encoding developmental regulators and maintenance of ES cell state. Genes & Development. 2009;23:2484–2489. doi: 10.1101/gad.1837309. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen Y, Lun ATL, Smyth GK. From reads to genes to pathways: differential expression analysis of RNA-Seq experiments using Rsubread and the edgeR quasi-likelihood pipeline. F1000Research. 2016;5:1438. doi: 10.12688/f1000research.8987.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen Z, Zhang Y. Loss of DUX causes minor defects in zygotic genome activation and is compatible with mouse development. Nature Genetics. 2019;51:947–951. doi: 10.1038/s41588-019-0418-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Cho S, Park JS, Kwon S, Kang YK. Dynamics of Setdb1 expression in early mouse development. Gene Expression Patterns. 2012;12:213–218. doi: 10.1016/j.gep.2012.03.005. [DOI] [PubMed] [Google Scholar]
De Iaco A, Planet E, Coluccio A, Verp S, Duc J, Trono D. DUX-family transcription factors regulate zygotic genome activation in placental mammals. Nature Genetics. 2017;49:941–945. doi: 10.1038/ng.3858. [DOI] [PMC free article] [PubMed] [Google Scholar]
Deng Q, Ramsköld D, Reinius B, Sandberg R. Single-cell RNA-seq reveals dynamic, random monoallelic gene expression in mammalian cells. Science. 2014;343:193–196. doi: 10.1126/science.1245316. [DOI] [PubMed] [Google Scholar]
de Vries WN, Binns LT, Fancher KS, Dean J, Moore R, Kemler R, Knowles BB. Expression of Cre recombinase in mouse oocytes: a means to study maternal effect genes. Genesis. 2000;26:110–112. [PubMed] [Google Scholar]
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21. doi: 10.1093/bioinformatics/bts635. [DOI] [PMC free article] [PubMed] [Google Scholar]
Dodge JE, Kang YK, Beppu H, Lei H, Li E. Histone H3-K9 methyltransferase ESET is essential for early development. Molecular and Cellular Biology. 2004;24:2478–2486. doi: 10.1128/MCB.24.6.2478-2486.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
Eckersley-Maslin M, Alda-Catalinas C, Blotenburg M, Kreibich E, Krueger C, Reik W. Dppa2 and Dppa4 directly regulate the Dux-driven zygotic transcriptional program. Genes & Development. 2019;33:194–208. doi: 10.1101/gad.321174.118. [DOI] [PMC free article] [PubMed] [Google Scholar]
Eymery A, Liu Z, Ozonov EA, Stadler MB, Peters AHFM. The methyltransferase Setdb1 is essential for meiosis and mitosis in mouse oocytes and early embryos. Development. 2016;143:2767–2779. doi: 10.1242/dev.132746. [DOI] [PubMed] [Google Scholar]
Gassler J, Kobayashi W, Gáspár I, Ruangroengkulrith S, Mohanan A, Gómez Hernández L, Kravchenko P, Kümmecke M, Lalic A, Rifel N, Ashburn RJ, Zaczek M, Vallot A, Cuenca Rico L, Ladstätter S, Tachibana K. Zygotic genome activation by the totipotency pioneer factor Nr5a2. Science. 2022;378:1305–1315. doi: 10.1126/science.abn7478. [DOI] [PubMed] [Google Scholar]
Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32:2847–2849. doi: 10.1093/bioinformatics/btw313. [DOI] [PubMed] [Google Scholar]
Guo M, Zhang Y, Zhou J, Bi Y, Xu J, Xu C, Kou X, Zhao Y, Li Y, Tu Z, Liu K, Lin J, Yang P, Gao S, Wang Y. Precise temporal regulation of Dux is important for embryo development. Cell Research. 2019;29:956–959. doi: 10.1038/s41422-019-0238-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Guo Y, Kitano T, Inoue K, Murano K, Hirose M, Li TD, Sakashita A, Ishizu H, Ogonuki N, Matoba S, Sato M, Ogura A, Siomi H. Obox4 promotes zygotic genome activation upon loss of Dux. eLife. 2024;1:95856. doi: 10.7554/eLife.95856. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hermant C, Mourra-Díaz CM, Oomen ME, Altamirano-Pacheco L, Pal M, Nakatani T, Torres-Padilla M-E. The transcription factor SRF regulates MERVL retrotransposons and gene expression during zygotic genome activation. Genes & Development. 2025;39:490–509. doi: 10.1101/gad.352270.124. [DOI] [PMC free article] [PubMed] [Google Scholar]
Iaco A, Verp S, Offner S, Grun D, Trono D. DUX is a non-essential synchronizer of zygotic genome activation. Development. 2020;147:177725. doi: 10.1242/dev.177725. [DOI] [PMC free article] [PubMed] [Google Scholar]
Ji S, Chen F, Stein P, Wang J, Zhou Z, Wang L, Zhao Q, Lin Z, Liu B, Xu K, Lai F, Xiong Z, Hu X, Kong T, Kong F, Huang B, Wang Q, Xu Q, Fan Q, Liu L, Williams CJ, Schultz RM, Xie W. OBOX regulates mouse zygotic genome activation and early development. Nature. 2023;620:1047–1053. doi: 10.1038/s41586-023-06428-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
Jin Y, Tam OH, Paniagua E, Hammell M. TEtranscripts: a package for including transposable elements in differential expression analysis of RNA-seq datasets. Bioinformatics. 2015;31:3593–3599. doi: 10.1093/bioinformatics/btv422. [DOI] [PMC free article] [PubMed] [Google Scholar]
Karimi MM, Goyal P, Maksakova IA, Bilenky M, Leung D, Tang JX, Shinkai Y, Mager DL, Jones S, Hirst M, Lorincz MC. DNA methylation and SETDB1/H3K9me3 regulate predominantly distinct sets of genes, retroelements, and chimeric transcripts in mESCs. Cell Stem Cell. 2011;8:676–687. doi: 10.1016/j.stem.2011.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kim J, Zhao H, Dan J, Kim S, Hardikar S, Hollowell D, Lin K, Lu Y, Takata Y, Shen J, Chen T. Maternal Setdb1 is required for meiotic progression and preimplantation development in mouse. PLOS Genetics. 2016;12:e1005970. doi: 10.1371/journal.pgen.1005970. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kojima ML, Hoppe C, Giraldez AJ. The maternal-to-zygotic transition: reprogramming of the cytoplasm and nucleus. Nature Reviews. Genetics. 2025;26:245–267. doi: 10.1038/s41576-024-00792-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kovaka S, Zimin AV, Pertea GM, Razaghi R, Salzberg SL, Pertea M. Transcriptome assembly from long-read RNA-seq alignments with StringTie2. Genome Biology. 2019;20:19101. doi: 10.1186/s13059-019-1910-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
Krueger F. Trim Galore: A wrapper around Cutadapt and FastQC to consistently apply adapter and quality trimming to FastQ files, with extra functionality for RRBS data. a46184bGithub. 2025 https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/
Leung D, Du T, Wagner U, Xie W, Lee AY, Goyal P, Li Y, Szulwach KE, Jin P, Lorincz MC, Ren B. Regulation of DNA methylation turnover at LTR retrotransposons and imprinted loci by the histone methyltransferase Setdb1. PNAS. 2014;111:6690–6695. doi: 10.1073/pnas.1322273111. [DOI] [PMC free article] [PubMed] [Google Scholar]
Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25:1754–1760. doi: 10.1093/bioinformatics/btp324. [DOI] [PMC free article] [PubMed] [Google Scholar]
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, Genome Project Data Processing S. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25:2078–2079. doi: 10.1093/bioinformatics/btp352. [DOI] [PMC free article] [PubMed] [Google Scholar]
Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30:923–930. doi: 10.1093/bioinformatics/btt656. [DOI] [PubMed] [Google Scholar]
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology. 2014;15:5508. doi: 10.1186/s13059-014-0550-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Lu F, Liu Y, Inoue A, Suzuki T, Zhao K, Zhang Y. Establishing chromatin regulatory landscape during mouse preimplantation development. Cell. 2016;165:1375–1388. doi: 10.1016/j.cell.2016.05.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
Macfarlan TS, Gifford WD, Driscoll S, Lettieri K, Rowe HM, Bonanomi D, Firth A, Singer O, Trono D, Pfaff SL. Embryonic stem cell potency fluctuates with endogenous retrovirus activity. Nature. 2012;487:57–63. doi: 10.1038/nature11244. [DOI] [PMC free article] [PubMed] [Google Scholar]
Maksakova IA, Thompson PJ, Goyal P, Jones SJ, Singh PB, Karimi MM, Lorincz MC. Distinct roles of KAP1, HP1 and G9a/GLP in silencing of the two-cell-specific retrotransposon MERVL in mouse ES cells. Epigenetics & Chromatin. 2013;6:8935615. doi: 10.1186/1756-8935-6-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.Journal. 2011;17:171200. doi: 10.14806/ej.17.1.200. [DOI] [Google Scholar]
Matsui T, Leung D, Miyashita H, Maksakova IA, Miyachi H, Kimura H, Tachibana M, Lorincz MC, Shinkai Y. Proviral silencing in embryonic stem cells requires the histone methyltransferase ESET. Nature. 2010;464:927–931. doi: 10.1038/nature08858. [DOI] [PubMed] [Google Scholar]
McCarthy DJ, Chen Y, Smyth GK. Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Research. 2012;40:4288–4297. doi: 10.1093/nar/gks042. [DOI] [PMC free article] [PubMed] [Google Scholar]
Modzelewski AJ, Shao W, Chen J, Lee A, Qi X, Noon M, Tjokro K, Sales G, Biton A, Anand A, Speed TP, Xuan Z, Wang T, Risso D, He L. A mouse-specific retrotransposon drives a conserved Cdk2ap1 isoform essential for development. Cell. 2021;184:5541–5558. doi: 10.1016/j.cell.2021.09.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
Park SJ, Shirahige K, Ohsugi M, Nakai K. DBTMEE: a database of transcriptome in mouse early embryos. Nucleic Acids Research. 2015;43:D771–D776. doi: 10.1093/nar/gku1001. [DOI] [PMC free article] [PubMed] [Google Scholar]
Peaston AE, Evsikov AV, Graber JH, de Vries WN, Holbrook AE, Solter D, Knowles BB. Retrotransposons regulate host genes in mouse oocytes and preimplantation embryos. Developmental Cell. 2004;7:597–606. doi: 10.1016/j.devcel.2004.09.004. [DOI] [PubMed] [Google Scholar]
Percharde M, Lin CJ, Yin Y, Guan J, Peixoto GA, Bulut-Karslioglu A, Biechele S, Huang B, Shen X, Ramalho-Santos M. A LINE1-Nucleolin partnership regulates early development and ESC identity. Cell. 2018;174:391–405. doi: 10.1016/j.cell.2018.05.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
Piotrowska-Nitsche K, Perea-Gomez A, Haraguchi S, Zernicka-Goetz M. Four-cell stage mouse blastomeres have different developmental properties. Development. 2005;132:479–490. doi: 10.1242/dev.01602. [DOI] [PubMed] [Google Scholar]
Ramírez F, Ryan DP, Grüning B, Bhardwaj V, Kilpert F, Richter AS, Heyne S, Dündar F, Manke T. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Research. 2016;44:W160–W165. doi: 10.1093/nar/gkw257. [DOI] [PMC free article] [PubMed] [Google Scholar]
Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–140. doi: 10.1093/bioinformatics/btp616. [DOI] [PMC free article] [PubMed] [Google Scholar]
Sakashita A, Kitano T, Ishizu H, Guo Y, Masuda H, Ariura M, Murano K, Siomi H. Transcription of MERVL retrotransposons is required for preimplantation embryo development. Nature Genetics. 2023;55:484–495. doi: 10.1038/s41588-023-01324-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
Skarnes WC, Rosen B, West AP, Koutsourakis M, Bushell W, Iyer V, Mujica AO, Thomas M, Harrow J, Cox T, Jackson D, Severin J, Biggs P, Fu J, Nefedov M, de Jong PJ, Stewart AF, Bradley A. A conditional knockout resource for the genome-wide study of mouse gene function. Nature. 2011;474:337–342. doi: 10.1038/nature10163. [DOI] [PMC free article] [PubMed] [Google Scholar]
Svoboda P, Stein P, Anger M, Bernstein E, Hannon GJ, Schultz RM. RNAi and expression of retrotransposons MuERV-L and IAP in preimplantation mouse embryos. Developmental Biology. 2004;269:276–285. doi: 10.1016/j.ydbio.2004.01.028. [DOI] [PubMed] [Google Scholar]
Vega-Sendino M, Lüttmann FF, Olbrich T, Chen Y, Kuenne C, Stein P, Tillo D, Carey GI, Zhong J, Savy V, Radonova L, Lu T, Saykali B, Kim K-P, Domingo CN, Schüler L, Günther S, Bentsen M, Bosnakovski D, Schöler H, Kyba M, Maity TK, Jenkins LM, Looso M, Williams CJ, Kim J, Ruiz S. The homeobox transcription factor DUXBL controls exit from totipotency. Nature Genetics. 2024;56:697–709. doi: 10.1038/s41588-024-01692-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
Wang C, Liu X, Gao Y, Yang L, Li C, Liu W, Chen C, Kou X, Zhao Y, Chen J, Wang Y, Le R, Wang H, Duan T, Zhang Y, Gao S. Reprogramming of H3K9me3-dependent heterochromatin during mammalian embryo development. Nature Cell Biology. 2018;20:620–631. doi: 10.1038/s41556-018-0093-4. [DOI] [PubMed] [Google Scholar]
Wickham H. Ggplot2: Elegant Graphics for Data Analysis. Springer; 2016. [Google Scholar]
Wu Y, Wang C, Sun H, LeRoith D, Yakar S. High-efficient FLPo deleter mice in C57BL/6J background. PLOS ONE. 2009;4:e8054. doi: 10.1371/journal.pone.0008054. [DOI] [PMC free article] [PubMed] [Google Scholar]
Wu K, Liu H, Wang Y, He J, Xu S, Chen Y, Kuang J, Liu J, Guo L, Li D, Shi R, Shen L, Wang Y, Zhang X, Wang J, Pei D, Chen J. SETDB1-mediated cell fate transition between 2C-like and pluripotent states. Cell Reports. 2020;30:25–36. doi: 10.1016/j.celrep.2019.12.010. [DOI] [PubMed] [Google Scholar]
Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, Fu X, Liu S, Bo X, Yu G. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation. 2021;2:100141. doi: 10.1016/j.xinn.2021.100141. [DOI] [PMC free article] [PubMed] [Google Scholar]
Xue Z, Huang K, Cai C, Cai L, Jiang C, Feng Y, Liu Z, Zeng Q, Cheng L, Sun YE, Liu J, Horvath S, Fan G. Genetic programs in human and mouse early embryos revealed by single-cell RNA sequencing. Nature. 2013;500:593–597. doi: 10.1038/nature12364. [DOI] [PMC free article] [PubMed] [Google Scholar]
Yan L. Draw venn diagram by “ggplot2”. 0.1.19cran r. 2023 https://cran.r-project.org/web/packages/ggvenn/ggvenn.pdf
Yang L, Xia L, Wu DY, Wang H, Chansky HA, Schubach WH, Hickstein DD, Zhang Y. Molecular cloning of ESET, a novel histone H3-specific methyltransferase that interacts with ERG transcription factor. Oncogene. 2002;21:148–152. doi: 10.1038/sj.onc.1204998. [DOI] [PubMed] [Google Scholar]
Yang J, Cook L, Chen Z. Systematic evaluation of retroviral LTRs as cis-regulatory elements in mouse embryos. Cell Reports. 2024;43:113775. doi: 10.1016/j.celrep.2024.113775. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zeng TB, Han L, Pierce N, Pfeifer GP, Szabó PE. EHMT2 and SETDB1 protect the maternal pronucleus from 5mC oxidation. PNAS. 2019;116:10834–10841. doi: 10.1073/pnas.1819946116. [DOI] [PMC free article] [PubMed] [Google Scholar]

eLife. doi: 10.7554/eLife.109248.2.sa0

eLife Assessment

Yan Song ¹

This study presents a valuable finding on maternal SETDB1 as a key chromatin repressor that shuts down the 2C gene program and enables normal mouse embryonic development. The evidence supporting the claims of the authors is solid, although the inclusion of a causality test, a mechanistic understanding of SETDB1 targeting, and phenotypic quantification would have greatly strengthened the study. The work will be of broad interest to biologists working on embryonic development, stem cells and gene regulation.

eLife. doi: 10.7554/eLife.109248.2.sa1

Reviewer #1 (Public review):

Anonymous

Summary:

During the earliest stages of mouse development, the zygote and 2-cell (2C) embryo are totipotent, capable of generating all embryonic and extra-embryonic lineages, and they transiently express a distinctive set of "2C-stage" genes, many driven by MERVL long terminal repeat (LTR) promoters. Although activation of these transcripts is a normal feature of totipotency, they must be rapidly silenced as development proceeds to the 4-cell and 8-cell stages; failure to shut down the 2C program results in developmental arrest. This study examines the role of maternal SETDB1, a histone H3K9 methyltransferase, in suppressing the 2C transcriptional network. Using an oocyte-specific conditional knockout that removes maternal Setdb1 while leaving the paternal allele intact, the authors demonstrate that embryos lacking maternal SETDB1 arrest during cleavage, with very few progressing beyond the 8-cell stage and no morphologically normal blastocysts forming. Transcriptomic analyses reveal persistent expression of MERVL-LTR-driven transcripts and other totipotency markers, indicating a failure to terminate the totipotent state. Together, the data demonstrate that maternally deposited SETDB1 is required to silence the MERVL-driven 2C program and enable the transition from totipotency to pluripotency. More broadly, the work identifies maternal SETDB1 as a key chromatin repressor that deposits repressive H3K9 methylation to shut down the transient 2C gene network and to permit normal preimplantation development.

Strengths:

(1) Closes a key knowledge gap.

The study tackles a central open question - how embryos exit the totipotent 2-cell (2C) state - and provides direct in vivo evidence that epigenetic repression is required to terminate the 2C program for development to proceed. By identifying maternal SETDB1 as the responsible factor, the work substantially advances our understanding of the maternal-to-zygotic transition and early lineage specification.

(2) Clean genetics paired with rigorous genomics.

An oocyte-specific Setdb1 knockout cleanly isolates a maternal-effect requirement, ensuring that early phenotypes arise from loss of maternal protein. The resulting cleavage-stage arrest is unambiguous (most embryos stall before or around the 8-cell stage). State-of-the-art single-embryo RNA-seq across stages - well-matched to low-cell-number constraints - captures genome-wide mis-expression, including persistent 2C transcripts in mutants, strongly supporting the conclusions.

(3) Compelling molecular linkage to phenotype.

Transcriptome data show that without maternal SETDB1, embryos fail to repress a suite of 1-cell/2C-specific genes by the 8-cell stage. The tight correlation between continued activation of the MERVL-driven totipotency network and developmental arrest provides a specific molecular explanation for the observed failure to progress.

(4) Mechanistic insight grounded in chromatin biology.

SETDB1, a H3K9 methyltransferase classically linked to heterochromatin and transposon repression, targets MERVL LTRs and MERVL-driven chimeric transcripts in early embryos. Bioinformatic evidence indicates that these loci normally acquire H3K9me3 during the 2C→4C transition. The data articulate a coherent mechanism: maternal SETDB1 deposits repressive H3K9me3 at 2C gene loci to shut down the totipotency network, extending observations from ESC systems to bona fide embryos.

(5) Broad implications for development and stem-cell biology.

By pinpointing a maternal gatekeeper of the totipotent-to-pluripotent transition, the work suggests that some cases of cleavage-stage arrest (e.g., in IVF) may reflect faulty epigenetic silencing of transposon-driven genes. It also informs stem-cell efforts to control totipotent-like states in vitro (e.g., 2C-like cells), linking epigenetic reprogramming, transposable-element regulation, and developmental potency.

Weaknesses:

(1) Causality not directly demonstrated.

The link among loss of SETDB1, persistence of 2C transcripts, and developmental arrest is compelling but remains correlative. No rescue experiments test whether dampening the 2C/MERVL program restores development. Targeted interventions-e.g., knocking down key 2C drivers (such as Dux) or pharmacologically curbing MERVL-linked transcription in maternal Setdb1 mutants-would strengthen the claim that unchecked 2C activity is causal rather than a by-product of other SETDB1 functions.

(2) Limited mechanistic resolution of SETDB1 targeting.

The study establishes a requirement for maternal SETDB1 but does not define how it is recruited to MERVL loci. Given SETDB1's canonical cooperation with TRIM28/KAP1 and KRAB-ZNFs, upstream sequence-specific factors and/or pre-existing chromatin features likely guide targeting. Direct occupancy and mark-placement evidence (e.g., SETDB1/TRIM28 CUT&RUN or ChIP, and H3K9me3 profiling at MERVL LTRs during the 2C→4C window) would convert inferred mechanisms into demonstrated ones.

(3) Narrow scope on MERVL; broader epigenomic consequences underexplored.

Maternal SETDB1 may restrain additional repeat classes or genes beyond the 2C network. A systematic repeatome analysis (LINEs/SINEs/ERV subfamilies) would clarify specificity versus a general loss of heterochromatin control. Moreover, potential effects on imprinting or DNA methylation balance are not examined; perturbations there could also contribute to arrest. Bisulfite-based DNA methylation maps at imprinted loci and allele-specific expression analyses would help rule in/out these mechanisms.

(4) Phenotype quantitation and transcriptomic breadth could be clearer.

The developmental phenotype is described qualitatively ("very few beyond 8-cell") without precise stage-wise arrest rates or representative morphology. Tabulated counts (2C/4C/8C/blastocyst), images, and statistics would increase clarity. On the RNA-seq side, the narrative emphasizes known 2C markers; reporting novel/unannotated misregulated transcripts, as well as downregulated pathways (e.g., failure to activate normal 8-cell programs, metabolism, or early lineage markers), would present a fuller portrait of the mutant state.

eLife. doi: 10.7554/eLife.109248.2.sa2

Reviewer #2 (Public review):

Anonymous

Zeng et al. report that Setdb1-/- embryos fail to extinguish the 1- and 2-cell embryo transcriptional program and have permanent expression of MERVL transposable elements. The manuscript is technically sound and well performed, but, in my opinion, the results lack conceptual novelty.

(1) The manuscript builds on previous observations that: 1, Setbd1 is necessary for early mouse development, with knockout embryos rarely reaching the 8-cell stage; 2, SETB1 mediates H3K9me3 deposition at transposable elements in mouse ESCs; 3, SETB1silences MERVLs to prevent 2CLC-state acquisition in mouse ESCs. The strength of the current work is the demonstration that this is not due to a general transcriptional collapse; but otherwise, the findings are not surprising. The well-known (several Nature papers of years ago) crosstalk between m6A RNA modification and H3K9me3 in preventing 2CLC generation also partly compromises the novelty of this work.

(2) The conclusions regarding H3K9me3 deposition are inferred based on previously reported datasets, but there is no direct demonstration.

(3) The detection of chimeric transcripts is somewhat unreliable using short-read sequencing.

eLife. 2026 Feb 16;15:RP109248. doi: 10.7554/eLife.109248.2.sa3

Author response

Tie-Bo Zeng ¹, Zhen Fu ², Mary F Majewski ³, Ji Liao ⁴, Marie Adams ⁵, Piroska E Szabó ⁶

eLife Assessment

This study presents a valuable finding on maternal SETDB1 as a key chromatin repressor that shuts down the 2C gene program and enables normal mouse embryonic development. The evidence supporting the claims of the authors is solid, although the inclusion of a causality test, a mechanistic understanding of SETDB1 targeting, and phenotypic quantification would have greatly strengthened the study. The work will be of broad interest to biologists working on embryonic development, stem cells and gene regulation.

Thank you for this positive evaluation of our work. Please find the point-by point responses to the Reviewer’s comments below.

Public Reviews:

Reviewer #1 (Public review):

Summary:

During the earliest stages of mouse development, the zygote and 2-cell (2C) embryo are totipotent, capable of generating all embryonic and extra-embryonic lineages, and they transiently express a distinctive set of "2C-stage" genes, many driven by MERVL long terminal repeat (LTR) promoters. Although activation of these transcripts is a normal feature of totipotency, they must be rapidly silenced as development proceeds to the 4-cell and 8-cell stages; failure to shut down the 2C program results in developmental arrest. This study examines the role of maternal SETDB1, a histone H3K9 methyltransferase, in suppressing the 2C transcriptional network. Using an oocyte-specific conditional knockout that removes maternal Setdb1 while leaving the paternal allele intact, the authors demonstrate that embryos lacking maternal SETDB1 arrest during cleavage, with very few progressing beyond the 8-cell stage and no morphologically normal blastocysts forming. Transcriptomic analyses reveal persistent expression of MERVL-LTR-driven transcripts and other totipotency markers, indicating a failure to terminate the totipotent state. Together, the data demonstrate that maternally deposited SETDB1 is required to silence the MERVL-driven 2C program and enable the transition from totipotency to pluripotency. More broadly, the work identifies maternal SETDB1 as a key chromatin repressor that deposits repressive H3K9 methylation to shut down the transient 2C gene network and to permit normal preimplantation development.

Strengths:

(1) Closes a key knowledge gap.

The study tackles a central open question - how embryos exit the totipotent 2-cell (2C) state - and provides direct in vivo evidence that epigenetic repression is required to terminate the 2C program for development to proceed. By identifying maternal SETDB1 as the responsible factor, the work substantially advances our understanding of the maternal-to-zygotic transition and early lineage specification.

(2) Clean genetics paired with rigorous genomics.

An oocyte-specific Setdb1 knockout cleanly isolates a maternal-effect requirement, ensuring that early phenotypes arise from loss of maternal protein. The resulting cleavage-stage arrest is unambiguous (most embryos stall before or around the 8-cell stage). State-of-the-art single-embryo RNA-seq across stages - well-matched to low-cell-number constraints - captures genome-wide mis-expression, including persistent 2C transcripts in mutants, strongly supporting the conclusions.

(3) Compelling molecular linkage to phenotype.

Transcriptome data show that without maternal SETDB1, embryos fail to repress a suite of 1-cell/2C-specific genes by the 8-cell stage. The tight correlation between continued activation of the MERVL-driven totipotency network and developmental arrest provides a specific molecular explanation for the observed failure to progress.

(4) Mechanistic insight grounded in chromatin biology.

SETDB1, a H3K9 methyltransferase classically linked to heterochromatin and transposon repression, targets MERVL LTRs and MERVL-driven chimeric transcripts in early embryos. Bioinformatic evidence indicates that these loci normally acquire H3K9me3 during the 2C→4C transition. The data articulate a coherent mechanism: maternal SETDB1 deposits repressive H3K9me3 at 2C gene loci to shut down the totipotency network, extending observations from ESC systems to bona fide embryos.

(5) Broad implications for development and stem-cell biology.

By pinpointing a maternal gatekeeper of the totipotent-to-pluripotent transition, the work suggests that some cases of cleavage-stage arrest (e.g., in IVF) may reflect faulty epigenetic silencing of transposon-driven genes. It also informs stem-cell efforts to control totipotent-like states in vitro (e.g., 2C-like cells), linking epigenetic reprogramming, transposable-element regulation, and developmental potency.

We thank Reviewer 1 for recognizing the strengths in our work and for the suggestions below.

Weaknesses:

(1) Causality not directly demonstrated.

The link among loss of SETDB1, persistence of 2C transcripts, and developmental arrest is compelling but remains correlative. No rescue experiments test whether dampening the 2C/MERVL program restores development. Targeted interventions-e.g., knocking down key 2C drivers (such as Dux) or pharmacologically curbing MERVL-linked transcription in maternal Setdb1 mutants-would strengthen the claim that unchecked 2C activity is causal rather than a by-product of other SETDB1 functions.

We agree that rescue experiments might strengthen causality. Those experiments, however, would be extremely challenging technically because the knockdowns would need to be precisely timed to follow (and not prevent) the wave of 2c-specific activation. Knocking down 2c drivers in the zygote, for example, may prevent switching on the totipotency program. In addition, while sustained MERVL expression—such as that induced by forced DUX expression—disrupts totipotency exit and embryo development (1, 2), derepression of transcription is very broad in Setdb1^mat-/+ embryos and knocking down individual 2C drivers may not be sufficient to rescue development or restore the exit from totipotency.

(2) Limited mechanistic resolution of SETDB1 targeting.

The study establishes a requirement for maternal SETDB1 but does not define how it is recruited to MERVL loci. Given SETDB1's canonical cooperation with TRIM28/KAP1 and KRAB-ZNFs, upstream sequence-specific factors and/or pre-existing chromatin features likely guide targeting. Direct occupancy and mark-placement evidence (e.g., SETDB1/TRIM28 CUT&RUN or ChIP, and H3K9me3 profiling at MERVL LTRs during the 2C→4C window) would convert inferred mechanisms into demonstrated ones.

We do show H3K9me3 patterns at MERVL LTRs during the early2c-late2c-2c-4c-8c-morula window from a published dataset. Please see the genome browser images in Figures 4C, 4D, 4E, 6D, 6E and Figure S6. We agree that mapping of SETDB1/TRIM28 to those locations would strengthen the mechanistic insight. However, ChIPseq or CUT&RUN of those proteins in preimplantation embryos are not technically feasible. We do provide genetic evidence for the collaboration between SETDB1 and DUXBL, a DNA-binding factor, by showing that DUXBL cannot switch off its top targets without SETDB1 (Figure 6). Future studies will characterize the molecular mechanisms underlying this (likely indirect) collaboration. We do not think that DUXBL and SETDB1 directly interact, because such interaction was not detected by DUXBL IP-MS (3).

(3) Narrow scope on MERVL; broader epigenomic consequences underexplored.

Maternal SETDB1 may restrain additional repeat classes or genes beyond the 2C network. A systematic repeatome analysis (LINEs/SINEs/ERV subfamilies) would clarify specificity versus a general loss of heterochromatin control. Moreover, potential effects on imprinting or DNA methylation balance are not examined; perturbations there could also contribute to arrest. Bisulfite-based DNA methylation maps at imprinted loci and allele-specific expression analyses would help rule in/out these mechanisms.

We did examine genes and repeat elements beyond the 2c network. We evaluated gene and TE expression changes using four-way comparisons. Please find the results regarding gene expression in Figure 1C-J, Figure S2, Figure S3, Figure S4., Table S2, Table S3, and Table S4. Please find results on TE expression in Figure S5. Table S6, Table S7, and Table S8 and in the text. We agree that DNA methylation may be altered in Setdb1^mat-/+ embryos. In our hands, evaluating this possibility using bisulfite sequencing requires a larger number of embryos than what we can feasibly obtain (the number of obtained mutant embryos is very small). Regarding imprinted gene expression, one cannot fully assess and interpret imprinted gene expression in preimplantation stage embryos before the maternally deposited transcripts are gone. We reported earlier that clear somatic parental-specific patterns of imprinted gene expression may only start later in development, around 8.5 dpc (4).

(4) Phenotype quantitation and transcriptomic breadth could be clearer.

The developmental phenotype is described qualitatively ("very few beyond 8-cell") without precise stage-wise arrest rates or representative morphology. Tabulated counts (2C/4C/8C/blastocyst), images, and statistics would increase clarity. On the RNA-seq side, the narrative emphasizes known 2C markers; reporting novel/unannotated misregulated transcripts, as well as downregulated pathways (e.g., failure to activate normal 8-cell programs, metabolism, or early lineage markers), would present a fuller portrait of the mutant state.

Tabulated counts are displayed in Figure 1A, and morphology is shown in Figure S1A. We do say that 4% Setdb1^mat-/+ embryos reached the 8-cel stage by 2.5 dpc. We recovered zero Setdb1^mat-/+ blastocysts at 4.5 dpc (not shown). On the RNA-seq side we do report a more global assessment of transcription of genes and TEs (please see above at point 3), including novel chimeric transcripts (Table S6). Developmental pathways are shown in Figure S3 and Figure S4. Metabolic pathways are displayed in Figure S2.

Reviewer #2 (Public review):

Zeng et al. report that Setdb1-/- embryos fail to extinguish the 1- and 2-cell embryo transcriptional program and have permanent expression of MERVL transposable elements. The manuscript is technically sound and well performed, but, in my opinion, the results lack conceptual novelty.

(1) The manuscript builds on previous observations that: 1, Setbd1 is necessary for early mouse development, with knockout embryos rarely reaching the 8-cell stage; 2, SETB1 mediates H3K9me3 deposition at transposable elements in mouse ESCs; 3, SETB1silences MERVLs to prevent 2CLC-state acquisition in mouse ESCs. The strength of the current work is the demonstration that this is not due to a general transcriptional collapse; but otherwise, the findings are not surprising. The well-known (several Nature papers of years ago) crosstalk between m6A RNA modification and H3K9me3 in preventing 2CLC generation also partly compromises the novelty of this work.

We thank the Reviewer for appreciating the technical quality of our work. Regarding novelty, please consider that prior work in ES cells included contradictory findings (please see our Introduction). Prior embryology work (please see our Introduction) did not explain the preimplantation-stage phenotype. We highly appreciate those earlier works. Our work here answers the expectations drawn from prior studies and unequivocally shows that SETDB1 carries out the developmentally essential function of suppressing MERVLs and the 2-cell program in the mouse embryo.

(2) The conclusions regarding H3K9me3 deposition are inferred based on previously reported datasets, but there is no direct demonstration.

Dynamic H3K9me3 deposition is displayed at MERVL LTRs during the early2c-late2c-2c-4c-8c-morula window (Figures 4C, 4D, 4E, 6D, 6E and Figure S6) from a published work that has very high-quality data. We agree that demonstrating loss off H3K9me3 in Setdb1^mat-/+ embryos would confirm that the H3K9me3 histone methyltransferase function of SETDB1 (as opposed to any, yet unidentified, non-HMT specific activity of SETDB1) is responsible for shutting down MERVL LTRs. However, ChIP-seq, CUT&RUN, or similar assays are not feasible due to the rarity of Setdb1^mat-/+ embryos.

(3) The detection of chimeric transcripts is somewhat unreliable using short-read sequencing.

We used single embryo total RNA-seq and we report detecting chimeric transcripts (Table S6), which is considered more reliable than mRNA-seq for detecting chimeric transcripts, because many are not polyadenylated. We acknowledge, however, that long-read sequencing, which recently is becoming available, but which is still very expensive, is currently the most powerful method for detecting chimeric transcripts. This, however, does not affect the major conclusions or the significance of our work.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

Wang C, Liu X, Gao Y, Yang L, Li C, Liu W, Chen C, Kou X, Zhao Y, Wang H, Zhang Y, Gao S. 2018. Reprogramming of H3K9me3-dependent heterochromatin during mammalian early embryo development [ChIP-seq] NCBI Gene Expression Omnibus. GSE98149 [DOI] [PubMed]
Szabó PE, Zeng T, Fu Z. 2025. Single embryo total RNA sequencing to map the effect of maternal SETDB1 on the 2-celll and 8-cell transcriptome. NCBI Gene Expression Omnibus. GSE269417

Supplementary Materials

Figure 1—source data 1. Sample information.

Details of all single embryos analyzed in the study, including genotype, developmental stage, and sequencing sample ID.

elife-109248-fig1-data1.xlsx^{(10.5KB, xlsx)}

Figure 1—source data 2. Differential gene expression (DGE) analysis.

DGE analysis was conducted using the four pairwise comparisons shown in Figure 1C and D, with thresholds of |log₂FC|>1 and adjusted p<0.05. The ‘expression’ columns indicate whether each transcript is up- or downregulated in any given comparison. The ‘FOUR_WAY_UP_COMP’ and ‘FOUR_WAY_DOWN_COMP’ columns assign DEGs to compartments corresponding to the Venn diagrams in Figure 1E and F. Sample columns (TZSC118–TZSC116) contain log₂-transformed counts per million (CPM) values, generated using the cpm function in edgeR (v4.4.1).

elife-109248-fig1-data2.xlsx^{(6.4MB, xlsx)}

Figure 1—figure supplement 2—source data 1. Gene set enrichment analysis (GSEA) on gene ontology (GO).

GO term enrichment for DEGs identified in the four pairwise comparisons by gseGO in R package ClusterProfiler (v4.14.4).

elife-109248-fig1-figsupp2-data1.xlsx^{(164.4KB, xlsx)}

Figure 1—figure supplement 3—source data 1. Overrepresentation analysis (ORA) against gene ontology (GO) analysis by differentially expressed gene (DEG) Venn compartment.

GO enrichment for DEG subsets defined by intersecting Venn diagram compartments (Figure 1E, F). The analysis was run using the function enrichGO from the R package ClusterProfiler (v4.14.4). Results are shown only for sets with significant enrichment.

elife-109248-fig1-figsupp3-data1.xlsx^{(75.5KB, xlsx)}

Figure 1—figure supplement 5—source data 1. Differential TE expression (DTE): multimapping reads.

TE expression analysis by DESeq2 (v146.0) using multimapped reads across four comparisons.

elife-109248-fig1-figsupp5-data1.xlsx^{(363.1KB, xlsx)}

Figure 1—figure supplement 5—source data 2. Differential TE expression (DTE): unimapping reads.

TE expression analysis DESeq2 (v146.0) using uniquely mapped reads across four comparisons.

elife-109248-fig1-figsupp5-data2.xlsx^{(10.3MB, xlsx)}

Figure 2—source data 1. Overrepresentation analysis (ORA) with Database of Transcriptome in Mouse Early Embryos (DBTMEE) categories.

ORA was performed using DBTMEE-defined (Park et al., 2015) gene sets as in Sakashita et al., 2023 and DEGs up-/downregulated from the pairwise comparisons. The analysis was run using the function enricher from the R package ClusterProfiler (v4.14.4). Results are shown only for sets with significant enrichment.

elife-109248-fig2-data1.xlsx^{(817.1KB, xlsx)}

Figure 3—source data 1. Chimeric transcript analysis.

Chimeric transcripts previously identified from the datasets of Xue et al., 2013, and Deng et al., 2014 by Modzelewski et al., 2021, and de novo identified ones according to Modzelewski et al., 2021 were analyzed across four pairwise comparisons and annotated with their corresponding Venn diagram compartments from the four-way analysis. Values in the sample columns for the four pairwise DE transcript comparisons (last four sheets) are normalized counts generated using DESeq2 (v1.46.0).

elife-109248-fig3-data1.xlsx^{(4.3MB, xlsx)}

Figure 7—source data 1. ORA with MERVL-chimeric, MT2i, and Duxbl KO gene sets.

Overrepresentation analysis of up/down pairwise DEGs using gene sets from Macfarlan et al., 2012 (MERVL-chimeric), Yang et al., 2024 (MT2i), and Vega-Sendino et al., 2024 (Duxbl KO). ORA was performed if there were more than 50 genes using function enricher from R package ClusterProfiler (v4.14.4). Results are shown only for sets with significant enrichment.

elife-109248-fig7-data1.xlsx^{(28.6KB, xlsx)}

MDAR checklist

elife-109248-mdarchecklist1.docx^{(88.4KB, docx)}

Data Availability Statement

The following previously published datasets were used:

Szabó PE, Zeng T, Fu Z. 2025. Single embryo total RNA sequencing to map the effect of maternal SETDB1 on the 2-celll and 8-cell transcriptome. NCBI Gene Expression Omnibus. GSE269417

[bib1] Abe KI, Funaya S, Tsukioka D, Kawamura M, Suzuki Y, Suzuki MG, Schultz RM, Aoki F. Minor zygotic gene activation is essential for mouse preimplantation development. PNAS. 2018;115:E6780–E6788. doi: 10.1073/pnas.1804309115. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib2] Bilodeau S, Kagey MH, Frampton GM, Rahl PB, Young RA. SetDB1 contributes to repression of genes encoding developmental regulators and maintenance of ES cell state. Genes & Development. 2009;23:2484–2489. doi: 10.1101/gad.1837309. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib3] Chen Y, Lun ATL, Smyth GK. From reads to genes to pathways: differential expression analysis of RNA-Seq experiments using Rsubread and the edgeR quasi-likelihood pipeline. F1000Research. 2016;5:1438. doi: 10.12688/f1000research.8987.2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib4] Chen Z, Zhang Y. Loss of DUX causes minor defects in zygotic genome activation and is compatible with mouse development. Nature Genetics. 2019;51:947–951. doi: 10.1038/s41588-019-0418-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] Cho S, Park JS, Kwon S, Kang YK. Dynamics of Setdb1 expression in early mouse development. Gene Expression Patterns. 2012;12:213–218. doi: 10.1016/j.gep.2012.03.005. [DOI] [PubMed] [Google Scholar]

[bib6] De Iaco A, Planet E, Coluccio A, Verp S, Duc J, Trono D. DUX-family transcription factors regulate zygotic genome activation in placental mammals. Nature Genetics. 2017;49:941–945. doi: 10.1038/ng.3858. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] Deng Q, Ramsköld D, Reinius B, Sandberg R. Single-cell RNA-seq reveals dynamic, random monoallelic gene expression in mammalian cells. Science. 2014;343:193–196. doi: 10.1126/science.1245316. [DOI] [PubMed] [Google Scholar]

[bib8] de Vries WN, Binns LT, Fancher KS, Dean J, Moore R, Kemler R, Knowles BB. Expression of Cre recombinase in mouse oocytes: a means to study maternal effect genes. Genesis. 2000;26:110–112. [PubMed] [Google Scholar]

[bib9] Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21. doi: 10.1093/bioinformatics/bts635. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] Dodge JE, Kang YK, Beppu H, Lei H, Li E. Histone H3-K9 methyltransferase ESET is essential for early development. Molecular and Cellular Biology. 2004;24:2478–2486. doi: 10.1128/MCB.24.6.2478-2486.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] Eckersley-Maslin M, Alda-Catalinas C, Blotenburg M, Kreibich E, Krueger C, Reik W. Dppa2 and Dppa4 directly regulate the Dux-driven zygotic transcriptional program. Genes & Development. 2019;33:194–208. doi: 10.1101/gad.321174.118. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib12] Eymery A, Liu Z, Ozonov EA, Stadler MB, Peters AHFM. The methyltransferase Setdb1 is essential for meiosis and mitosis in mouse oocytes and early embryos. Development. 2016;143:2767–2779. doi: 10.1242/dev.132746. [DOI] [PubMed] [Google Scholar]

[bib13] Gassler J, Kobayashi W, Gáspár I, Ruangroengkulrith S, Mohanan A, Gómez Hernández L, Kravchenko P, Kümmecke M, Lalic A, Rifel N, Ashburn RJ, Zaczek M, Vallot A, Cuenca Rico L, Ladstätter S, Tachibana K. Zygotic genome activation by the totipotency pioneer factor Nr5a2. Science. 2022;378:1305–1315. doi: 10.1126/science.abn7478. [DOI] [PubMed] [Google Scholar]

[bib14] Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32:2847–2849. doi: 10.1093/bioinformatics/btw313. [DOI] [PubMed] [Google Scholar]

[bib15] Guo M, Zhang Y, Zhou J, Bi Y, Xu J, Xu C, Kou X, Zhao Y, Li Y, Tu Z, Liu K, Lin J, Yang P, Gao S, Wang Y. Precise temporal regulation of Dux is important for embryo development. Cell Research. 2019;29:956–959. doi: 10.1038/s41422-019-0238-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib16] Guo Y, Kitano T, Inoue K, Murano K, Hirose M, Li TD, Sakashita A, Ishizu H, Ogonuki N, Matoba S, Sato M, Ogura A, Siomi H. Obox4 promotes zygotic genome activation upon loss of Dux. eLife. 2024;1:95856. doi: 10.7554/eLife.95856. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] Hermant C, Mourra-Díaz CM, Oomen ME, Altamirano-Pacheco L, Pal M, Nakatani T, Torres-Padilla M-E. The transcription factor SRF regulates MERVL retrotransposons and gene expression during zygotic genome activation. Genes & Development. 2025;39:490–509. doi: 10.1101/gad.352270.124. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] Iaco A, Verp S, Offner S, Grun D, Trono D. DUX is a non-essential synchronizer of zygotic genome activation. Development. 2020;147:177725. doi: 10.1242/dev.177725. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] Ji S, Chen F, Stein P, Wang J, Zhou Z, Wang L, Zhao Q, Lin Z, Liu B, Xu K, Lai F, Xiong Z, Hu X, Kong T, Kong F, Huang B, Wang Q, Xu Q, Fan Q, Liu L, Williams CJ, Schultz RM, Xie W. OBOX regulates mouse zygotic genome activation and early development. Nature. 2023;620:1047–1053. doi: 10.1038/s41586-023-06428-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] Jin Y, Tam OH, Paniagua E, Hammell M. TEtranscripts: a package for including transposable elements in differential expression analysis of RNA-seq datasets. Bioinformatics. 2015;31:3593–3599. doi: 10.1093/bioinformatics/btv422. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib21] Karimi MM, Goyal P, Maksakova IA, Bilenky M, Leung D, Tang JX, Shinkai Y, Mager DL, Jones S, Hirst M, Lorincz MC. DNA methylation and SETDB1/H3K9me3 regulate predominantly distinct sets of genes, retroelements, and chimeric transcripts in mESCs. Cell Stem Cell. 2011;8:676–687. doi: 10.1016/j.stem.2011.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib22] Kim J, Zhao H, Dan J, Kim S, Hardikar S, Hollowell D, Lin K, Lu Y, Takata Y, Shen J, Chen T. Maternal Setdb1 is required for meiotic progression and preimplantation development in mouse. PLOS Genetics. 2016;12:e1005970. doi: 10.1371/journal.pgen.1005970. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib23] Kojima ML, Hoppe C, Giraldez AJ. The maternal-to-zygotic transition: reprogramming of the cytoplasm and nucleus. Nature Reviews. Genetics. 2025;26:245–267. doi: 10.1038/s41576-024-00792-0. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] Kovaka S, Zimin AV, Pertea GM, Razaghi R, Salzberg SL, Pertea M. Transcriptome assembly from long-read RNA-seq alignments with StringTie2. Genome Biology. 2019;20:19101. doi: 10.1186/s13059-019-1910-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib25] Krueger F. Trim Galore: A wrapper around Cutadapt and FastQC to consistently apply adapter and quality trimming to FastQ files, with extra functionality for RRBS data. a46184bGithub. 2025 https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/

[bib26] Leung D, Du T, Wagner U, Xie W, Lee AY, Goyal P, Li Y, Szulwach KE, Jin P, Lorincz MC, Ren B. Regulation of DNA methylation turnover at LTR retrotransposons and imprinted loci by the histone methyltransferase Setdb1. PNAS. 2014;111:6690–6695. doi: 10.1073/pnas.1322273111. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25:1754–1760. doi: 10.1093/bioinformatics/btp324. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib28] Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, Genome Project Data Processing S. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25:2078–2079. doi: 10.1093/bioinformatics/btp352. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib29] Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30:923–930. doi: 10.1093/bioinformatics/btt656. [DOI] [PubMed] [Google Scholar]

[bib30] Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology. 2014;15:5508. doi: 10.1186/s13059-014-0550-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib31] Lu F, Liu Y, Inoue A, Suzuki T, Zhao K, Zhang Y. Establishing chromatin regulatory landscape during mouse preimplantation development. Cell. 2016;165:1375–1388. doi: 10.1016/j.cell.2016.05.050. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib32] Macfarlan TS, Gifford WD, Driscoll S, Lettieri K, Rowe HM, Bonanomi D, Firth A, Singer O, Trono D, Pfaff SL. Embryonic stem cell potency fluctuates with endogenous retrovirus activity. Nature. 2012;487:57–63. doi: 10.1038/nature11244. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib33] Maksakova IA, Thompson PJ, Goyal P, Jones SJ, Singh PB, Karimi MM, Lorincz MC. Distinct roles of KAP1, HP1 and G9a/GLP in silencing of the two-cell-specific retrotransposon MERVL in mouse ES cells. Epigenetics & Chromatin. 2013;6:8935615. doi: 10.1186/1756-8935-6-15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib34] Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.Journal. 2011;17:171200. doi: 10.14806/ej.17.1.200. [DOI] [Google Scholar]

[bib35] Matsui T, Leung D, Miyashita H, Maksakova IA, Miyachi H, Kimura H, Tachibana M, Lorincz MC, Shinkai Y. Proviral silencing in embryonic stem cells requires the histone methyltransferase ESET. Nature. 2010;464:927–931. doi: 10.1038/nature08858. [DOI] [PubMed] [Google Scholar]

[bib36] McCarthy DJ, Chen Y, Smyth GK. Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Research. 2012;40:4288–4297. doi: 10.1093/nar/gks042. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib37] Modzelewski AJ, Shao W, Chen J, Lee A, Qi X, Noon M, Tjokro K, Sales G, Biton A, Anand A, Speed TP, Xuan Z, Wang T, Risso D, He L. A mouse-specific retrotransposon drives a conserved Cdk2ap1 isoform essential for development. Cell. 2021;184:5541–5558. doi: 10.1016/j.cell.2021.09.021. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib38] Park SJ, Shirahige K, Ohsugi M, Nakai K. DBTMEE: a database of transcriptome in mouse early embryos. Nucleic Acids Research. 2015;43:D771–D776. doi: 10.1093/nar/gku1001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib39] Peaston AE, Evsikov AV, Graber JH, de Vries WN, Holbrook AE, Solter D, Knowles BB. Retrotransposons regulate host genes in mouse oocytes and preimplantation embryos. Developmental Cell. 2004;7:597–606. doi: 10.1016/j.devcel.2004.09.004. [DOI] [PubMed] [Google Scholar]

[bib40] Percharde M, Lin CJ, Yin Y, Guan J, Peixoto GA, Bulut-Karslioglu A, Biechele S, Huang B, Shen X, Ramalho-Santos M. A LINE1-Nucleolin partnership regulates early development and ESC identity. Cell. 2018;174:391–405. doi: 10.1016/j.cell.2018.05.043. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib41] Piotrowska-Nitsche K, Perea-Gomez A, Haraguchi S, Zernicka-Goetz M. Four-cell stage mouse blastomeres have different developmental properties. Development. 2005;132:479–490. doi: 10.1242/dev.01602. [DOI] [PubMed] [Google Scholar]

[bib42] Ramírez F, Ryan DP, Grüning B, Bhardwaj V, Kilpert F, Richter AS, Heyne S, Dündar F, Manke T. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Research. 2016;44:W160–W165. doi: 10.1093/nar/gkw257. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib43] Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–140. doi: 10.1093/bioinformatics/btp616. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] Sakashita A, Kitano T, Ishizu H, Guo Y, Masuda H, Ariura M, Murano K, Siomi H. Transcription of MERVL retrotransposons is required for preimplantation embryo development. Nature Genetics. 2023;55:484–495. doi: 10.1038/s41588-023-01324-y. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib45] Skarnes WC, Rosen B, West AP, Koutsourakis M, Bushell W, Iyer V, Mujica AO, Thomas M, Harrow J, Cox T, Jackson D, Severin J, Biggs P, Fu J, Nefedov M, de Jong PJ, Stewart AF, Bradley A. A conditional knockout resource for the genome-wide study of mouse gene function. Nature. 2011;474:337–342. doi: 10.1038/nature10163. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib46] Svoboda P, Stein P, Anger M, Bernstein E, Hannon GJ, Schultz RM. RNAi and expression of retrotransposons MuERV-L and IAP in preimplantation mouse embryos. Developmental Biology. 2004;269:276–285. doi: 10.1016/j.ydbio.2004.01.028. [DOI] [PubMed] [Google Scholar]

[bib47] Vega-Sendino M, Lüttmann FF, Olbrich T, Chen Y, Kuenne C, Stein P, Tillo D, Carey GI, Zhong J, Savy V, Radonova L, Lu T, Saykali B, Kim K-P, Domingo CN, Schüler L, Günther S, Bentsen M, Bosnakovski D, Schöler H, Kyba M, Maity TK, Jenkins LM, Looso M, Williams CJ, Kim J, Ruiz S. The homeobox transcription factor DUXBL controls exit from totipotency. Nature Genetics. 2024;56:697–709. doi: 10.1038/s41588-024-01692-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib48] Wang C, Liu X, Gao Y, Yang L, Li C, Liu W, Chen C, Kou X, Zhao Y, Chen J, Wang Y, Le R, Wang H, Duan T, Zhang Y, Gao S. Reprogramming of H3K9me3-dependent heterochromatin during mammalian embryo development. Nature Cell Biology. 2018;20:620–631. doi: 10.1038/s41556-018-0093-4. [DOI] [PubMed] [Google Scholar]

[bib49] Wickham H. Ggplot2: Elegant Graphics for Data Analysis. Springer; 2016. [Google Scholar]

[bib50] Wu Y, Wang C, Sun H, LeRoith D, Yakar S. High-efficient FLPo deleter mice in C57BL/6J background. PLOS ONE. 2009;4:e8054. doi: 10.1371/journal.pone.0008054. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib51] Wu K, Liu H, Wang Y, He J, Xu S, Chen Y, Kuang J, Liu J, Guo L, Li D, Shi R, Shen L, Wang Y, Zhang X, Wang J, Pei D, Chen J. SETDB1-mediated cell fate transition between 2C-like and pluripotent states. Cell Reports. 2020;30:25–36. doi: 10.1016/j.celrep.2019.12.010. [DOI] [PubMed] [Google Scholar]

[bib52] Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, Fu X, Liu S, Bo X, Yu G. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation. 2021;2:100141. doi: 10.1016/j.xinn.2021.100141. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib53] Xue Z, Huang K, Cai C, Cai L, Jiang C, Feng Y, Liu Z, Zeng Q, Cheng L, Sun YE, Liu J, Horvath S, Fan G. Genetic programs in human and mouse early embryos revealed by single-cell RNA sequencing. Nature. 2013;500:593–597. doi: 10.1038/nature12364. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib54] Yan L. Draw venn diagram by “ggplot2”. 0.1.19cran r. 2023 https://cran.r-project.org/web/packages/ggvenn/ggvenn.pdf

[bib55] Yang L, Xia L, Wu DY, Wang H, Chansky HA, Schubach WH, Hickstein DD, Zhang Y. Molecular cloning of ESET, a novel histone H3-specific methyltransferase that interacts with ERG transcription factor. Oncogene. 2002;21:148–152. doi: 10.1038/sj.onc.1204998. [DOI] [PubMed] [Google Scholar]

[bib56] Yang J, Cook L, Chen Z. Systematic evaluation of retroviral LTRs as cis-regulatory elements in mouse embryos. Cell Reports. 2024;43:113775. doi: 10.1016/j.celrep.2024.113775. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib57] Zeng TB, Han L, Pierce N, Pfeifer GP, Szabó PE. EHMT2 and SETDB1 protect the maternal pronucleus from 5mC oxidation. PNAS. 2019;116:10834–10841. doi: 10.1073/pnas.1819946116. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

SETDB1 enables development beyond cleavage stages by extinguishing the MERVL-driven two-cell totipotency transcriptional program in the mouse embryo

Tie-Bo Zeng

Zhen Fu

Mary F Majewski

Ji Liao

Marie Adams

Piroska E Szabó

Roles

Abstract

Introduction

Results

Setdb1 maternal knockout embryos fail preimplantation development

Figure 1. Maternal SETDB1 is essential for development beyond the eight-cell stage.

Figure 1—figure supplement 1. Normal features of Setdb1 KO embryos.

Figure 1—figure supplement 2. Gene set enrichment analysis (GSEA) of Setdb1 KO differentially expressed genes (DEGs).

Figure 1—figure supplement 3. Developmental transcriptional changes independent of maternal SETDB1.

Figure 1—figure supplement 4. Developmental misregulation in Setdb1 KO embryos.

Figure 1—figure supplement 5. SETDB1 suppresses transposable elements at cleavage stages.

Single-embryo total RNA sequencing of Setdb1mat-/+ embryos

Transcriptional programs of development do not collapse in the absence of maternal SETDB1

The two-cell program is misregulated in the absence of maternal SETDB1

Maternal SETDB1 has a broad effect on transposable elements in 2c and 8c embryos

Maternal SETDB1 is required for timely suppression of cleavage-stage transcripts

Maternal SETDB1 controls transient gene expression during ZGA

Figure 2. Maternal SETDB1 extinguishes two-cell transient gene expression.

Maternal SETDB1 suppresses two-cell-specific MERVL-chimeric transcripts

Figure 3. Maternal SETDB1 regulates MERVL-driven chimeric transcripts.

Maternal SETDB1 silences MERVL MT2 LTR-activated transcripts

Figure 4. Maternal SETDB1 suppresses MT2 LTR-regulated genes.

Figure 5. SETDB1 regulates MT2-activated genes across time.

Maternal SETDB1 is required to extinguish DUXBL-responsive transcripts

Figure 6. SETDB1 represses DUXBL-responsive transcripts.

Figure 6—figure supplement 1. Examples of DUXBL targets derepressed in Setdb1 KO embryos.

Overrepresentation analysis confirms shared targets with MT2 and DUXBL pathways

Figure 7. SETDB1 collaborates with DUXBL to repress totipotency programs.

Discussion

Materials and methods

Key resources table.

Mice

Collection of preimplantation mouse embryos

RNA isolation

Library preparation and sequencing

RNA-seq analysis

Transposable element (TE) analysis

Chimeric transcript analysis

Public datasets

Data visualization

Acknowledgements

Funding Statement

Contributor Information

Funding Information

Additional information

Competing interests

Author contributions

Ethics

Additional files

Data availability

References

eLife Assessment

Yan Song

Roles

Reviewer #1 (Public review):

Anonymous

Roles

Reviewer #2 (Public review):

Anonymous

Roles

Author response

Tie-Bo Zeng

Zhen Fu

Mary F Majewski

Ji Liao

Marie Adams

Piroska E Szabó

Roles

Associated Data

Data Citations

Supplementary Materials

Data Availability Statement

Single-embryo total RNA sequencing of Setdb1^mat-/+ embryos