Identification of developmentally important genes that are down-regulated in SCNT embryos

We have previously shown that development of SCNT embryos is greatly enhanced by the combinational treatment of TSA and VC (Miyamoto et al, 2017). We hypothesized that these epigenetic modifiers allowed activation of key genes for embryonic development, which were otherwise repressed in SCNT embryos. We therefore compared down-regulated genes in control SCNT embryos at the late two-cell stage (control SCNT versus SCNT with TSA and VC) (Miyamoto et al, 2017) to those found between SCNT embryos and in vitro fertilized (IVF) embryos (a group of 301 genes that failed to be activated in SCNT embryos) (Matoba et al, 2014). Among the down-regulated genes, 16 genes were shared (Fig 1A and B). These genes are normally repressed in SCNT embryos, but are up-regulated when the development of cloned embryos is greatly enhanced by the TSA + VC treatment. We then searched for developmentally important genes among the list by knocking down their expression in IVF embryos with siRNA injection and specific siRNA sets for 15 genes were designed. The siRNA screening identified that knockdown of Alyref and Gabpb1 significantly impaired the development of IVF embryos to the blastocyst stage (Fig 1C). Successful knockdown of both genes was shown by qRT-PCR (Fig S1A). We also observed a significant decrease in preimplantation development after injecting the mixture of siRNAs that target genes, which have been shown to function as transcriptional activators including Gabpb1, Gtf2f2, and Taf9 (Rosmarin et al, 2004) (Fig 1C; activators). Inhibition of Alyref or Gabpb1 by siRNA injection resulted in developmental arrest at the morula stage (Figs 1D and S1B); especially, almost all embryos did not develop to the blastocyst stage in Alyref siRNA-injected embryos (Fig S1B). These results suggest that Alyref or Gabpb1 genes, repressed in SCNT embryos, are important for the progression to the blastocyst stage.

Abnormal expression of Alyref and Gabpb1 in SCNT embryos

We next examined if the abnormal expression of Alyref or Gabpb1 is indeed observed in SCNT embryos. Firstly, the protein localization of Alyref and Gabpb1 was examined during the preimplantation development. Alyref and Gabpb1 were localized in pronuclei of mouse IVF zygotes (Fig 2A and B), suggesting maternal accumulation of Alyref and Gabpb1 protein. Nuclear localization was seen throughout the preimplantation development. To note, dot-like localization of Alyref was observed in the nuclei of IVF embryos from the two-cell stage onwards (Figs 2A and S1C). Gabpb1 protein was accumulated around nucleoli from the four-cell stage onwards (Fig 2B). Secondly, the transcriptional expression of Alyref or Gabpb1 in SCNT embryos was examined using RNA-seq dataset in Matoba et al (2014) and Miyamoto et al (2017). Both Alyref and Gabpb1 were repressed in SCNT embryos compared with IVF embryos at the two-cell stage and this repression was rescued by the overexpression of Kdm4d (Fig S1D), which removes excess H3K9me3 observed in SCNT embryos (Matoba et al, 2014). Up-regulation of Alyref and Gabpb1 was also observed in SCNT embryos at the late two-cell stage treated with TSA and VC (Fig S1D), which also lowers the level of H3K9me3 (Miyamoto et al, 2017). Repression of Alyref and Gabpb1 in SCNT embryos at the late two-cell stage, as compared with TSA-VC–treated SCNT embryos, was confirmed by qRT-PCR analyses (Fig 2C). These results suggest that Alyref and Gabpb1 are repressed at the transcript level in SCNT embryos at the time of ZGA. Lastly, the protein expression of Alyref and Gabpb1 in SCNT embryos was examined at 28, 48, and 72 h post-activation (hpa). Both Alyref and Gabpb1 were significantly down-regulated in SCNT embryos at all stages tested when compared with IVF embryos (Fig 2D and E). These results suggest that abnormal down-regulation of Alyref and Gabpb1 is observed in SCNT embryos both at the transcript and protein levels.

Alyref gene is essential for preimplantation development

To clarify the requirement of Alyref for embryonic development, we generated knockout mice by the CRISPR/Cas9 gene-editing method. The coding region of Alyref gene was removed by introducing gRNAs and Cas9 protein to zygotes as reported before (Mashiko et al, 2013) (Fig 3A). The resulting heterozygous mutant mice (Alyref^+/−) were identified by PCR-based genotyping (Fig S2A). When Alyref^+/− mice were mated with WT mice, viable offspring was obtained (Fig 3B). However, the number of offspring was significantly reduced (Fig S2B) and no knockout mice (Alyref^−/−) were obtained after crossing Alyref ^+/− × Alyref^+/− (Fig 3B), suggesting the embryonic lethal phenotype of Alyref knockout mice. We then performed IVF using sperm and oocytes obtained from Alyref ^+/− mice to reveal when development is arrested. The significant decrease in percentages of embryos that reach the blastocyst stage was observed (Fig 3C and D). Immunofluorescent analyses revealed that all blastocyst embryos obtained by Alyref ^+/− × Alyref ^+/− showed Alyref signals, whereas some of the morula embryos lacked Alyref expression (Fig S2C). These results demonstrate that Alyref is required for mouse preimplantation development and knockout of Alyref causes developmental arrest at the morula stage.

Gabpb1 gene is essential for embryonic development

To test the requirement of Gabpb1 for embryonic development, we generated knockout mice by the CRISPR/Cas9 method. gRNA that targets exon 3 of Gabpb1 was designed and introduced to zygotes with Cas9 protein (Fig 4A). The resulting heterozygous mutant mice (Gabpb1^+/−) were identified by DNA sequencing (Fig S2D) and the frame shift mutation was confirmed in the knockout allele (Fig 4B), which resulted in a truncated protein. When Gabpb1^+/− mice were mated with WT mice, viable offspring was obtained (Fig 4C and D). However, when Gabpb1^+/− mice were mated with Gabpb1^+/−, the number of offspring was significantly reduced (Fig 4D) and no knockout mice (Gabpb1^−/−) were obtained (Fig 4C). We then carried out IVF using sperm and oocytes obtained from Gabpb1^+/− mice. The significant decrease in percentages of embryos that reach the morula and blastocyst stages was observed (Fig 4E). Nevertheless, a certain number of embryos reached blastocysts after crossing Gabpb1^+/− × Gabpb1^+/−, and we have therefore performed genotyping of the generated embryos. Many of Gabpb1^−/− embryos (75%) were arrested at the morula stage or degenerated, whereas 61% of Gabpb1^+/− embryos developed to the blastocyst state (Fig S2E). The lack of Gabpb1^−/− expression in the arrested morula embryos was also confirmed by immunofluorescent analyses (Fig 4F). These results demonstrate that Gabpb1 is required for mouse embryonic development.

Developmentally important pathways are disturbed after knocking out Alyref or Gabpb1

To gain mechanistic insight into the role of Alyref and Gabpb1, we performed RNA sequencing (RNA-seq) analysis. Based on the IVF experiments using sperm and oocytes of heterozygous mutant mice, most of knockout embryos were arrested at the morula stage at 96 h post-insemination (hpi) (Figs 3 and 4). Therefore, embryos remained as morulae at 96 hpi were collected and subjected to single-embryo RNA-seq together with control blastocyst embryos (Fig 5A). The large deletion of Alyref was confirmed in two morula embryos (Fig 5B; KOA-M1 and KOA-M2) because no sequence reads were detected at the deleted region. Similarly, the frameshift mutation of Gabpb1 was also confirmed in three morula embryos (Fig 5C; KOG-M1-M3). Off-target deletion by CRISPR/Cas9 was not observed in Alyref2 and Gabpb2, family genes of Alyref and Gabpb1, respectively (Fig S3A and B). Principal component analysis and hierarchical clustering analysis indicated that knockout morulae showed distinct transcriptomes from WT morulae and were well separated from those of blastocyst embryos (Figs 5D and E and S3C and D). These results suggest that knockout of Alyref and Gabpb1 causes abnormal gene expression in morulae, which may prevent the progression of knockout embryos to the blastocyst stage.

We next examined molecular pathways that were disturbed by the depletion of Alyref. Differentially expressed genes (DEGs) were identified between knockout and WT (and/or heterozygous) morulae (KOA-M1-2 versus A-M1-3) and 705 genes were misregulated (P < 0.05, twofold changes). Ingenuity pathway analysis (IPA) using this DEG list showed that genes related to mammalian pluripotency were misregulated (Fig 6A and Appendix Table S1). We then searched for upstream regulators responsible for the abnormal gene expression in Alyref^−/− morulae using IPA. Interestingly, Pou5f1, a master regulator for mammalian pluripotency, was identified (Fig 6B). We examined Pou5f1-positive cells in Alyref KO embryos. Expansion of strongly Pou5f1-positive cells and the proper formation of inner cell mass (ICM) were not observed in Alyref-negative embryos (Fig 6C). However, the abolishment of Oct4 signals was not observed in Alyref KO embryos (Fig 6C, bottom panels). These results suggest that Alyref may not be a direct upstream factor to regulate Pou5f1 expression in early embryos, but it is still possible that Alyref might play a role in ICM formation by cooperating with Pou5f1 and/or other factors. To test this idea, we next focused on Alyref’s function to regulate Nanog expression as Nanog-related pathways were disturbed in Alyref KO embryos (Fig 6A). Previous studies demonstrated that Alyref is a component of the transcription/export complex, and Thoc2 and Thoc5, other components of the transcription/export complex, are required for the self-renewal of embryonic stem (ES) cells (Wang et al, 2013). Interestingly, Thoc2 and Thoc5 regulate export and expression of Nanog mRNA, but not Pou5f1 in ES cells (Wang et al, 2013). We therefore tested if Alyref regulates the expression of Nanog for ICM formation in blastocyst embryos. Expression of Nanog protein was significantly reduced in Alyref knockdown and knockout embryos at 72 hpi (Fig 6D). Furthermore, Alyref inhibition by siRNA injection resulted in down-regulation of Nanog transcripts (Fig 6E). Interestingly, expression of Nanog intron transcripts in Alyref knockdown embryos was not significantly reduced, compared with that in control embryos (Fig 6E), suggesting that Nanog RNA processing for its translation, but not Nanog gene activation, may be defective after Alyref knockdown. Because Gapdh expression was also reduced in Alyref knockdown embryos, developmental progression of Alyref knockdown embryos might be delayed, which affected global transcript levels. Indeed, cell numbers were significantly reduced after Alyref knockdown (siAlyref embryos at 72 hpi: 11 ± 0.83 cells versus siControl embryos at 72 hpi: 18 ± 1.6 cells, P < 0.01, F- and t test). Finally, Alyref overexpression in one of the two-cell blastomeres increased Nanog-positive cells at the blastocyst stage at 96 hpi (Fig 6F). Thus, Alyref ensures Nanog expression and is required for the developmental progression to blastocyst embryos.

To gain mechanistic insight into the developmental arrest caused by Gabpb1 depletion, we performed hierarchical clustering of genes based on their expression levels using Gabpb1 knockout and WT embryos and 20 clusters were generated (Fig S4A). Genes that belong to cluster 1 showed very low expression in Gabpb1^−/− morulae, but were expressed in other samples with the strongest expression in WT blastocysts (Fig S4A). Cluster 1 was enriched with genes related to blastocyst formation, and thus, knockout of Gapbp1 leads to the incomplete activation of such developmentally important genes. Next, DEGs were identified between Gabpb1 knockout and WT morulae (KOG-M1-3 versus G-M1-2) and 1,321 genes were misregulated (P < 0.05, twofold changes). IPA revealed that anti-oxidation and metabolic pathways were abnormally regulated in Gabpb1^−/− embryos (Fig S4B, highlighted in green). Candidate upstream regulators related to this misexpression were identified (Fig S4C). Moreover, genes related to cell viability and apoptotic pathways showed abnormal expression (Fig 6G). We therefore examined the appearance of apoptotic cells after Gabpb1 knockdown. At 96 hpi, 39% ± 2.6% of cells underwent apoptosis when Gabpb1 siRNA was injected (Fig 6H). Many apoptotic cells were also found in arrested morulae embryos after crossing Gabpb1^+/− × Gabpb1^+/− (Fig S4D). Furthermore, abnormal mitochondrial membrane potential was detected in Gabpb1 knockdown embryos at 96 hpi (Fig 6I). These results indicate that suppression of Gabpb1 leads to abnormal gene expression related to anti-oxidation, metabolic pathways, and blastocyst formation; disturbed mitochondrial functions; and eventually the induction of apoptosis, resulting in the failure to form proper blastocyst embryos. In summary, the roles of Alyref and Gabpb1 in mouse preimplantation development are different, but both genes are required for the establishment of proper gene expression programs for blastocyst formation.

The relationship between H3K9me3-mediated gene repression and expression of Alyref and Gabpb1 expression

The above data have indicated that Alyref and Gabpb1 are required for preimplantation development, but repressed in SCNT embryos. Expression of these genes is repressed by H3K9me3-related mechanisms because overexpression of Kdm4d and treatment with VC, both of which can lower H3K9me3 levels in SCNT embryos (Matoba et al, 2014; Miyamoto et al, 2017), can up-regulate Alyref and Gabpb1 (Figs 2C and S1D). We therefore asked if the expression of Alyref and Gabpb1 is directly or indirectly regulated by H3K9me3. We reanalyzed published ChIP-seq dataset in which H3K9me3 enrichment was examined in fertilized embryos, SCNT embryos at the two-cell stage, and donor cumulus cells (Liu et al, 2016). H3K9me3 was not much accumulated around Alyref and Gabpb1 in donor cumulus cells, fertilized embryos, and SCNT embryos at the two-cell stage, compared with the Dux cluster region that is known to be marked with abundant H3K9me3 in cumulus cells and SCNT embryos (Yang et al, 2021) (Fig S5A). In addition, both Alyref and Gabpb1 were not located around aberrantly low acetylated regions, which correlate with H3K9me3-enriched reprogramming-resistant regions (Yang et al, 2021). Nevertheless, very weak signals of H3K9me3 were detected around transcription start sites of Alyref and Gabpb1 (Fig S5A, red highlighted regions), and we then tried to lower H3K9me3 levels in cumulus cells to see the effect on expression of Alyref and Gabpb1. siRNA treatment against Suv39h1/2, main regulators for catalyzing H3K9me3 (Matoba et al, 2014), successfully inhibited their expression and reduced H3K9me3 levels in cumulus cells (Fig S5B–D). However, transcriptional up-regulation of Alyref and Gabpb1 was not observed in cumulus cells even after lowing H3K9me3 (Fig S5C). It is noteworthy that both Alyref and Gabpb1 were expressed in donor cumulus cells even without interfering with H3K9me3 (Fig S5C). Taken together, the repressed chromatin state by H3K9me3 in donor cumulus cells cannot explain silencing of Alyref and Gabpb1 in SCNT embryos. Therefore, it is most likely that down-regulated expression of Alyref and Gabpb1 in SCNT embryos is indirectly regulated by H3K9me3.

Preimplantation development of SCNT embryos is rescued by exogenous supplementation of Alyref and Gabpb1

We asked if the insufficient expression of Alyref and Gabpb1 is a reason for the low developmental ability of SCNT embryos. EGFP-Alyref and EGFP-Gabpb1 mRNA was injected to SCNT embryos reconstructed with cumulus cells and subsequence preimplantation development was examined (Fig 7A, without TSA). Because the developmental rescue is often dose-dependent, different doses of mRNA mixtures of Alyref and Gabpb1 were injected (100, 400, 500, and 600 ng/μl). Alyref and Gabpb1 mRNA injection enhanced preimplantation development of SCNT embryos (Fig 7B) and the most prominent effect was observed when 400 ng/μl of Alyref and Gabpb1 mRNA (each 200 ng/μl) were injected (Fig 7B and C). As shown in Fig S1D, Alyref and Gabpb1 are activated by removing H3K9me3 with Kdm4d or lowering it with VC. We therefore hypothesized that the positive effect of VC on reprogramming in SCNT embryos can be substituted by the supplementation of Alyref and Gabpb1 mRNA. VC’s effect is most prominent when combined with TSA (Miyamoto et al, 2017), and therefore, we injected 400 ng/μl of Alyref and Gabpb1 mRNA to SCNT embryos treated with TSA (Fig 7A, with TSA). Strikingly, 87% of SCNT embryos developed to the blastocyst stage after Alyref and Gabpb1 mRNA injection (Fig 7D). Finally, we tested if Alyref and Gabpb1 mRNA injection can support efficient development of SCNT embryos to term. SCNT embryos were treated with TSA, together with or without Alyref and Gabpb1 mRNA injection, but significant improvement of the live offspring rate was not observed by supplementation of Alyref and Gabpb1 mRNA (Fig 7E). These results suggest that incomplete activation of Alyref and Gabpb1 is closely linked to preimplantation arrest of SCNT embryos, but restoration of their expression is not sufficient for supporting full-term development of SCNT embryos.

Klf16, repressed in SCNT embryos, can activate Alyref and Gabpb1 at the two-cell stage

Expression of Alyref and Gabpb1 is not repressed by H3K9me3 in donor cumulus cells (Fig S5C). However, their expression is up-regulated in SCNT embryos in which H3K9me3 levels were lowered at the two-cell stage (Figs 2C and S1D) (Matoba et al, 2014; Miyamoto et al, 2017). One possible mechanism of this could be that reprogramming-resistant H3K9me3-repressed genes in SCNT embryos contain upstream transcriptional activators for Alyref and Gabpb1. We therefore explored transcriptional regulators that are up-regulated after removing the H3K9me3 mark (Matoba et al, 2014; Yang et al, 2021), and that can activate Alyref and Gabpb1 in SCNT embryos. Firstly, gene regulatory regions of Alyref and Gabpb1 were predicted by taking advantage of ENCODE registry of candidate cis-regulatory elements (The ENCODE Project Consortium et al, 2020) and candidate transcription factors that can bind to the predicted gene regulatory regions were listed from JASPAR transcription factor binding site database (Castro-Mondragon et al, 2022). Among the candidate transcription factors, we searched for those that were up-regulated after overexpression of H3K9me3 demethylases in SCNT embryos (Matoba et al, 2014; Yang et al, 2021), leading to the identification of Klf16 and Tfap2c (Fig 8A and B). Down-regulation of Klf16 transcripts was observed in SCNT embryos at the two-cell stage, as compared with IVF embryos (Fig 8C). Supplementation of 20 ng/μl Klf16 mRNA in SCNT embryos significantly up-regulated Alyref and Gabpb1 transcripts at the late two-cell stage (Fig 8D, Alyref: 1.65-fold and Gabpb1: 1.82-fold increases). The magnitude of transcriptional up-regulation in Klf16-supplemented SCNT embryos is not as much as that in TSA-VC–treated SCNT embryos (Fig 2C, Alyref: 3.71-fold and Gabpb1: 2.23-fold increases), suggesting that an additional factor for fully activating Alyref and Gabpb1 likely exists. We then asked whether Tfap2c can boost Alyref and Gabpb1 expression when combined with the Klf16 treatment. However, the combinational treatment of Klf16 and Tfap2c mRNA overexpression diminished the positive effect of Klf16 on Alyref and Gabpb1 activation (Fig 8E). Therefore, we focused on Klf16 to examine its effect on development of SCNT embryos. Klf16 overexpression (20 ng/μl) in SCNT embryos alone did not significantly increase the development of SCNT embryos (Fig 8F). This may be because of the partial rescue of Alyref and Gabpb1 expression after supplementation of Klf16 (Fig 8D). As the positive effect of Alyref and Gabpb1 up-regulation on preimplantation development is enhanced by the treatment with TSA (Fig 7D), Klf16-supplemented SCNT embryos were incubated in TSA-containing medium. Klf16 overexpression together with TSA significantly enhanced the development of SCNT embryos to the blastocyst stage, as compared with the TSA-treated control SCNT embryos (Fig 8G; 49.7% versus 26.7%). However, Klf16 overexpression with TSA did not support the efficient full-term development (Fig 8H). In summary, Alyref and Gabpb1 are repressed in SCNT embryos, at least partially, because of the H3K9me3-mediated incomplete activation of Klf16.

Animals

Mice (C57BL/6, DBA/2, MCH [ICR] strains and B6D2F1 [C57BL/6J × DBA/2N]) at 8–10 wk of age were purchased from CLEA Japan or Japan SLC and maintained in light-controlled, air-conditioned rooms. B6D2F1 mice were used for IVF. This study was carried out in strict accordance with the recommendations in the Guidelines of Kindai University for the Care and Use of Laboratory Animals. Experimental protocols were approved by the Committee on the Ethics of Animal experiments of Kindai University (Permit Number: KABT-31-003). All mice were euthanized by cervical dislocation and all efforts were made to minimize suffering and to reduce the number of animals used in the present study.

KO mice were generated on the B6D2F1 background with CRISPR/Cas9-mediated genome editing as reported before (Mashiko et al, 2013; Oji et al, 2016). Genotyping of KO mice was performed as described in Figs 3 and 4. PCR products were directly used for Sanger sequencing. Sequences of gRNA and primers are listed in Appendix Table S2. Animal experiments for generating KO mice were approved by the Animal Care and Use Committee of the Research Institute for Microbial Diseases, Osaka University (Permit Number: H30-01-1 and R03-01-0).

Cell culture and transfection

Cumulus cells were collected from cumulus–oocyte complexes obtained from female B6D2F1 mice by treating with hyaluronidase (H3884; Merck). The collected cumulus cells were seeded in a 24 well and cultured in the medium containing DMEM (11054001; Thermo Fisher Scientific), sodium pyruvate, 10% FBS, and penicillin/streptomycin at 37°C under 5% CO₂ in air before transfection. siRNA transfection for knockdown of Suv39h1/2 was performed as reported previously (Matoba et al, 2014). Briefly, 5 pM siRNAs (Suv39h1 siRNA [s74607; Thermo Fisher Scientific], Suv39h2 siRNA [s82300; Thermo Fisher Scientific]) were transfected into cumulus cells at the 80% confluency using Lipofectamine RNAiMAX Transfection Reagent (13778030; Thermo Fisher Scientific) (day 1). 24 h after the first transfection, the culture medium was changed to a fresh one. On day 4, transfection was repeated as described above. Forty-eight h after the second transfection (day 6), the cumulus cells were subjected to Western blot and qRT-PCR analyses.

IVF and embryo culture

Collection of spermatozoa, oocytes, and zygotes were performed as described in previous studies (Okuno et al, 2020). Briefly, spermatozoa were collected from the cauda epididymis of B6D2F1 fertile male mice (>8 wk of age). The sperm suspension was incubated in human tubal fluid (HTF) medium (ARK Resource) for 1.5 h to allow for capacitation at 37°C under 5% CO₂ in air. Oocytes were collected from the excised oviducts of B6D2F1 female mice (>8 wk of age) that were superovulated with pregnant mare serum gonadotropin (PMSG; Serotropin, ASKA Pharmaceutical Co.) and 48 h later, human chorionic gonadotropin (hCG; ASKA Pharmaceutical Co.). Cumulus–oocyte complexes were recovered into pre-equilibrated HTF medium. The sperm suspension was added to the oocyte cultures and morphologically normal zygotes were collected 2 h post-insemination (hpi). The zygotes were cultured in potassium simplex-optimized medium KSOMaa (ARK Resource) at 37°C under 5% CO₂ in air.

SCNT

SCNT was carried out as described previously (Miyamoto et al, 2017). Briefly, enucleation of denuded MII oocytes was performed in drops of HCZB containing 5 μg/ml cytochalasin B (Sigma-Aldrich). After enucleation, a donor cell in HCZB with 6% dBSA was placed in the perivitelline space of an enucleated oocyte together with HVJ-E (GenomeONE-CF; Ishihara Sangyo) by tightly attaching the donor cell to the enucleated oocyte; the oocyte was then cultured in KSOMaa for 1 h at 37°C in air containing 5% CO₂, during which time, it fused with the donor cell. The reconstructed oocytes were activated by the incubation for 6 h in 5 mM SrCl₂ and 2 mM EGTA-containing KSOMaa supplemented with 5 μg/ml cytochalasin B, referred to as the activation medium (Kishigami & Wakayama, 2007), at 37°C in air containing 5% CO₂. For the treatment with TSA (T8552; Sigma-Aldrich), NT embryos were treated with 50 nM TSA for 8 h from the commencement of activation. The activated NT embryos were cultured at 37°C in air containing 5% CO₂ in KSOMaa.

Embryo transfer

Pseudopregnant ICR females were prepared as recipients for embryo transfer by mating with sterile ICR males. SCNT embryos at the two-cell stage were transferred to the oviducts of the pseudopregnant females at 0.5 d post-coitus (dpc). Cesarean section and uterine analysis of implantation sites were performed in all recipients at 19.5 dpc. If available, lactating foster mothers were used to raise live pups.

Plasmids

Constructs were produced either using a gateway system (Thermo Fisher Scientific) or in-fusion cloning (Clontech). Myc-tag-GFP was subcloned into the pCS2 vector (Miyamoto et al, 2011) for mRNA production as described previously (Baarlink et al, 2017). pCS2-myc-GFP and pcDNA3.1-Histone H2B-EGFP were reported previously (Baarlink et al, 2017; Hatano et al, 2022). To obtain the Gabpb1, Klf16, and Tfap2c mRNA-producing vectors for injection experiments, we first generated the pENTR/D-TOPO-Gabpb1 entry vector. After clonase-triggered recombination reaction, pCS2-EGFP-Gabpb1, pCS2-Klf16-HA, and pCS2-Tfap2c-HA were generated. Gabpb1 insert was amplified from cDNA of mouse blastocyst embryos by using oligos GW_Gabpb1_N_F and GW_Gabpb1_N_R (Appendix Table S2). Klf16 and Tfap2c inserts were amplified from cDNA of two-cell embryos by using oligos GW_mKlf16_C_F, GW_mKlf16_C_R, GW_mTfap2c_C_F, and GW_mTfap2c_C_R (Appendix Table S2). pCS2-EGFP-Alyref was produced by in-fusion cloning. For this subcloning, Alyref insert was amplified from the pCMV6-Entry-Alyref-Myc-DDK-tagged plasmid (MR220236; OliGene Technologies) using IF-Alyref-N_Fw and IF-Alyref-N_Rv3 primers. In-fusion reaction was performed after amplification of the linearized pCS2 vector, which was amplified by PCR using IF-Alyref-vectorF and IF-Alyref-vectorR primers. All primers used for cloning are listed in Appendix Table S2.

mRNA production

mRNAs were prepared from pCS2 vectors using mMESSAGE mMACHINE SP6 Transcription Kit (AM1340 or AM1344; Thermo Fisher Scientific). Briefly, to produce linearized vectors, ∼5-μg plasmids were digested overnight, with appropriate restriction enzymes. In the case of pCS2 vectors, mRNAs produced from the SP6 promoter were subjected to the addition of polyA tails (AM1350; Thermo Fisher Scientific). Produced mRNAs were purified using an Rneasy Mini Kit (74106; QIAGEN).

mRNA injection

After activation, NT embryos were collected at 6–7 hours post-activation (hpa) for mRNA injection. NT embryos were then washed with KSOMaa and kept in drops of CZB-HEPES medium for injection. mRNAs were injected using a piezo manipulator (Prime Tech). The final concentrations of the injected mRNA are as follows; 50–300 ng/μl EGFP-Alyref mixed with 50–300 ng/μl EGFP-Gabpb1, 100–600 ng Myc-tag-GFP for Fig 7B, 200 ng/μl EGFP-Alyref mixed with 200 ng/μl EGFP-Gabpb1 for Fig 7C–E, 20 ng Klf16 for Fig 8D–H, 20 or 50 ng Tfap2c for Fig 8E, 20–400 ng Myc-tag-GFP for Figs 7C–E and 8D–H. After injection, embryos were cultured in KSOMaa medium at 37°C in air containing 5% CO₂. For the TSA treatment, 50 nM TSA was supplemented in the KSOMaa medium and NT embryos were cultured in the TSA-containing medium. After injecting mRNA, embryos were immediately transferred to the inhibitor-containing medium.

For Fig 6F, IVF embryos were collected at 24–27 hpi for mRNA injection. Injection to two-cell embryos was performed in drops of CZB-HEPES medium. mRNAs were injected into one blastomere of a two-cell embryo using a piezo manipulator. The final concentrations of injected mRNA are as follows: 50 ng/μl of EGFP-Alyref or 50 ng of Histone H2B-EGFP in Fig 6F. After injection, embryos were cultured in the KSOMaa medium at 37°C in air containing 5% CO₂.

TdT-mediated dUTP Nick End Labeling (TUNEL) assay

For counting apoptotic cells after Gabpb1 knockdown or knockout, In Situ Cell Death Detection Kit (11684795910; Roche) was used according to the vendor’s instruction. Briefly, embryos at 96 hpi were fixed with 4% PFA/PBS at RT for 20 min, and were washed by 3 mg/ml PVP/PBS for three times. Zona pellucida was removed in Tyrode’s solution (T1788; Sigma-Aldrich) and the zona-free embryos were washed by PVP/PBS for three times. Then, the embryos were treated with 0.5% Triton X-100 in PBS at RT for 20 min. The TUNEL reaction mixture (5-μl enzyme solution and the 45-μl label solution) was prepared immediately before use, and the embryos were incubated in 50 μl of the TUNEL reaction mixture at 37°C. The samples were washed with PVP/PBS three times and then mounted on slides using VECTASHIELD Mounting Medium containing DAPI. The fluorescence signals were observed using an LSM800 microscope (Carl Zeiss), equipped with a laser module (405/488/561/640 nm) and GaAsP detector, using the same contrast, brightness, and exposure settings within the same experiments. Z-slice thickness was determined by using the optimal interval function in the ZEN software.

Mitochondrial membrane potential analysis

The membrane-sensitive dye JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolyl carbocyanine iodide) was used to detect mitochondrial membrane potential (ΔΨ m) according to the vendor’s instruction with some modifications (10009172; Cayman Chemical Company). Briefly, the JC-1 stock solution was diluted (x40) in the KSOMaa medium (KSOM + JC1). Embryos injected with siRNAs against Gabpb1 or control siRNAs were incubated in KSOM + JC1 for 30 min at 37°C in air containing 5% CO₂. After the incubation, embryos were washed three times in phenol red-free KSOMaa medium containing 0.00025% PVA (KSOM + PVA). Then, embryos were transferred to drops of KSOM + PVA on a glass-bottom dish (Matsunami) and placed in an incubation chamber stage (Tokai Hit) at 37°C under 5% CO₂ in air for live-cell imaging. The fluorescence signals were observed using an LSM800 confocal microscope, equipped with a laser module (405/488/561/640 nm) and GaAsP detector, using the same contrast, brightness, and exposure settings (3 μm interval). For detecting J-aggregates and monomers, the rhodamine and FITC filter settings were used, respectively. Membrane potential is expressed as a ratio of the average rhodamine fluorescence (J-aggregates) relative to the average FITC fluorescence (monomers) in embryo areas.

Immunofluorescence staining

Embryos were fixed in 4% PFA/PBS at RT for 15 min, and were washed by 3 mg/ml PVP/PBS for three times. Zona pellucida was removed in Tyrode’s solution and the zona-free embryos were washed by 3 mg/ml PVP/PBS for three times. Samples were next treated with 0.25% Triton X-100 in 3 mg/ml PVP/PBS at RT for 30 min. Blocking was performed in 3% BSA/PBS with 0.01% Tween 20 for 1 h at RT, then embryos were incubated with primary antibodies diluted in 3% BSA/PBS with 0.01% Tween 20 (1:50; anti-ALY antibody [sc-32311; Santa Cruz Biotechnology], 1:200; GABPB1 antibody [GTX103464; Gene Tex], 1:200; Oct4 [Pou5f1] antibody [09-0023; REPROCELL], 1:100; Oct3/4 [Pou5f1] antibody [sc5279; Santa Cruz Biotechnology], 1:300; Nanog Antibody [RCAB002P-F; REPROCELL]) at 4°C overnight. After three times washes by 1% BSA/PBS with 0.01% Tween 20, samples were further incubated in the dark with Alexa Fluor 488-labeled goat anti-mouse IgG antibody (1:2,000; A11001; Thermo Fisher Scientific), Alexa Fluor 488-labeled goat anti-rabbit IgG antibody (1:2,000; A11008; Thermo Fisher Scientific), Alexa Fluor 594-labeled donkey anti-mouse IgG antibody (1:2,000; A21203; Thermo Fisher Scientific) or Alexa Fluor 594-labeled donkey anti-rabbit IgG antibody (1:2,000; A21207; Thermo Fisher Scientific) at RT for 1 h. The samples were washed with 1% BSA/PBS with 0.01% Tween 20 three times and then mounted on slides using VECTASHIELD Mounting Medium containing DAPI. The fluorescence signals were observed using an LSM800 microscope, equipped with a laser module (405/488/561/640 nm) and GaAsP detector, using the same contrast, brightness, and exposure settings within the same experiments. Z-slice thickness was determined by using the optimal interval function in the ZEN software.

Image analysis

Images were analyzed using the ZEN software. For quantification of Nanog signals in IVF embryos (Fig 6D), all focal planes were merged by maximum intensity projection. Nanog staining did not show nonspecific cytoplasmic signals and it was localized in nuclei. Therefore, the maximum intensity projection using all focal planes was used for quantifying Nanog signals. After maximum intensity projection, intensities of Nanog in nuclei were quantified. Furthermore, background signals were subtracted from the nuclear signals and the average signal intensities of the merged focal planes were calculated by using the “Measure” function in the ZEN software. To quantify nuclear signal intensities of Alyref and Gabpb1 (Fig 2D and E), all focal planes that cover the nuclear region were used for quantifying nuclear signal intensities. In each focal plane, background signals were subtracted from the nuclear signals and the average of all focal planes was calculated by using the “Measure” function in the ZEN software.

For Fig 6F, the injected embryos were cultured till 96 hpi and blastocyst embryos were collected. The collected blastocysts were fixed and stained with Nanog and DAPI. Then, EGFP-positive cells and Nanog-positive ICM cells, together with the total number of cells with DAPI, were counted by the confocal microscopy.

Western blot

Cumulus cells grown in a 24-well were suspended in the SDS sample buffer (196-11022; Wako), and the lysed samples were subjected to SDS–PAGE (15% poly-acrylamide gel). The gels were transferred onto polyvinylidene difluoride membranes, and the membranes were blocked with 3% BSA in PBS-Tween (PBS-T) solution for 1 h. The blocked membranes were incubated with anti-H3K9me3 (ab8898; Abcam) or anti-ACTB antibodies (A5441; Merck) overnight at 4°C. After washing, the membranes were incubated with HRP-conjugated donkey anti-rabbit IgG secondary antibody (AP182P; Millipore) or IRDye 680RD goat anti-rabbit IgG secondary antibody (925-68071; LI-COR), and visualized using ECL prime Western blotting detection reagents (RPN2232; Amersham) or Odyssey imaging system (LI-COR).

RNA-seq analysis

Timings for sampling are shown in each corresponding figure. A single-NT embryo at the morula or blastocyst stage was treated with acid Tyrode, followed by three washes with 0.1% BSA/PBS, and was moved into 1 × reaction buffer from SMART-seq v4 Ultra Low Input RNA Kit (Z4888N; Takara). SMART-seq library preparation was performed using SMART-seq v4 Ultra Low Input RNA Kit and Nextera DNA Sample Preparation Kit (FC-131-1024; illumine) according to the vendor’s instruction. We followed the previously published protocol for RNA-seq of mouse embryos (Tomikawa et al, 2021).

Briefly, paired-end sequencing (50 + 25 bp) was obtained by the NextSeq platform (Illumina). Raw reads were first subjected to filtering to remove low-quality reads using Trimmomatic (Bolger et al, 2014). Reads of less than 20 bases and unpaired reads were also removed. Furthermore, adaptor, polyA, polyT, and polyG sequences were removed using Trim Galore! (https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/). The sequencing reads were then mapped to the mouse genome (mm10) using STAR (Dobin et al, 2013). Reads on annotated genes were counted using featureCounts (Liao et al, 2014). RNA-seq reads were visualized using Integrative Genome Viewer (Robinson et al, 2011). Fragments per kilobase of exon per million mapped values were calculated from mapped reads by normalizing to the total count and transcript. DEGs, P < 0.05 were then identified using DESeq2 (Love et al, 2014). A hierarchical clustering of the read count values was performed using hclust in TCC (unweighted pair group method with arithmetic mean). Gene ontology terms showing overrepresentation of genes that were up- or down-regulated were detected using DAVID tools (P < 0.05). Principal-component analysis of the global gene expression profile was performed using scikit-learn and the plots were depicted using plotly. Each gene list was further subjected to an Ingenuity Pathway Analysis (IPA; QIAGEN). Using IPA, enriched canonical pathways, upstream transcriptional regulators, and diseases and biological functions were investigated.

For Fig S1D, Fragments per kilobase of exon per million mapped values were obtained from GSE59073 and GSE95053.

ChIP-sequencing (ChIP-seq) analysis

The published ChIP-seq dataset (GSE70606) was used for examining H3K9me3 enrichment around Alyref and Gabpb1 in Fig S5A. Sequencing reads were trimmed to remove adapters using trim galore (v0.6.7). The trimmed reads were mapped to mouse genome (UCSC mm10) using Bowtie2 (v2.4.5) with the parameters: -t–no-mixed–no-discordant–no-unal (Langmead & Salzberg, 2012; Yang et al, 2021). Duplicated reads were removed using samtools (v0.6.8) (Tarasov et al, 2015). H3K9me3 signals were calculated using a 50-bp window and normalized to the uniquely mapped fragments using “bamCoverage” from deepTools (v3.5.1) with the parameters: --normalizeUsing RPKM.

qRT-PCR

Total RNA was extracted from a pool of three four-cell embryos at 48 hpi for Fig S1A, three two-cell embryos at 28 hpa or 28 hpi for Figs 2C and 8C–E, and three morula embryos at 72 hpi for Fig 6E using PicoPure RNA Isolation Kit (KIT0204; Thermo Fisher Scientific), or from cumulus cells grown in a 24-well using RNeasyMini Kit (74106; QIAGEN) according to the manufacturer’s protocol. cDNA was synthesized from the extracted RNA using appropriate RT primers (oligo [dT] primers for Figs 2C, 8C–E, and S1A, and random hexamers for Fig 6E) and SuperScript III Reverse Transcriptase (18080044; Thermo Fisher Scientific). cDNA samples were analyzed in a 7300 Real-Time PCR System (Applied Biosystems) or StepOne Real-Time PCR System (Applied Biosystems). The primer sequences are listed in Appendix Table S2.

siRNA knockdown in embryos

For investigating a function of candidate genes responsible for acquiring totipotency, specific siRNA was injected to IVF embryos at 6 hpi, and the injected embryos were further cultured in the KSOMaa medium to observe preimplantation development. Predesigned negative control siRNA (RNAi Inc.) was used as a control. Sequences for siRNA and qRT-PCR primers are listed in Appendix Table S2.

Statistical analysis

All of the statistical methods for each experiment can be found in the figure legends and in the relevant Materials and Methods section. For Figs 1C, 2C–E, 4D, 6D–F, H, and I, 8C and D, S1A, S2B, and S5C, statistical significance was calculated by two-sided F- and t tests. For Figs 1D, 3C, 4E, 7C and D, 8F and G, and S1B, statistical significance was calculated by chi-square test. For Fig 8E, statistical significance was calculated by ANOVA. For Figs 7E and 8H, statistical significance was calculated by Fisher’s exact test. Sample size was estimated based on our preliminary experiments and no statistical method was used. Embryos were randomly selected for image and gene expression analyses. Blinding was applied for the analysis of image data. Embryos that showed abnormal appearance such as degenerated embryos were not subjected for further image and gene expression analyses.