Efficient derivation of stable sheep embryonic stem cells opens a new avenue for agricultural and biomedical application.

Miaohan Jin: Conceptualization, Investigation, Writing-Original Draft. Shuhong Huang: Formal analysis, Data Curation, Writing-Original Draft. Shiwei Zhou: Resources. Wenwen Shen: Investigation. Jing Cao: Investigation. Shengjiao Song: Investigation. Yinghui Wei: Writing-Review & Editing. Peter Kalds: Writing-Review & Editing. Jinlian Hua: Supervision. Baohua Ma: Supervision. Pablo Juan Ross: Supervision. Xiaolong Wang: Writing-Review & Editing, Supervision, Funding acquisition. Yulin Chen: Supervision, Project administration, Funding acquisition.

In order to derive sheep ESCs, different culture conditions were tested. In vivo derived blastocysts were used to establish sheep ESC lines under four different culture conditions (NBFR, bEPSCM, 3i/LAF, and TePR) [ 10 , 11 , 23 ] ( Fig. 1 A and Fig. S1 A, Supporting Information ). Total 28 whole blastocysts were used to derive sheep ESCs (each culture conditions 7 whole blastocysts) ( Fig. S1 B, Supporting Information ). ICM outgrowths were passaged once per week until passage three (P3) when colonies were evident ( Fig. S1 A, Supporting Information ). Under these four different culture conditions, the cells were able to form colonies with a round, dome-shaped morphology, the derivation efficiency was reached to 57 % in TePR culture condition, and the derivation efficiency of other three culture conditions were about 29 % ( Fig. S1 B, Supporting Information ). Notably, cells cultured in TePR maintained undifferentiated. In contrast, in other conditions, cells started to differentiate and did not form colonies until P10 ( Fig. S1 A, Supporting Information ). Finally, we obtained 4 sheep ESCs lines were established under the TePR condition, termed TePR-sESCs, were used for further validation. Fig. 1 Generation and characterization of TePR-sESCs . (A) Schematic diagram of the derivation of TePR-sESCs at day 6 in vivo fertilization embryos. (B) Bright field images and AP staining of TePR-sESCs, Scale bars, 40 µm. (C) Karyotype analysis of TePR-sESCs. (D) Gene expression analysis using RT-qPCR of core pluripotency genes ( POU5F1 , SOX2 , NANOG , LIN28 , and SALL4 ) of TePR-sESCs. The relative expression was normalized to SFF and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments). (E) Single-cell cloning efficiency of TePR-sESCs (P10) and TePR-sESCs (P100). (F) In vitro EB differentiation assay. RT-PCR for the ectodermal ( OTX2 , PAX3 , KRT8 , and NESTIN ), mesodermal ( PDGFRA and ACTA2 ) and endoderm ( FOXA2 and GATA6 ) specific genes. The relative expression was normalized to TePR-sESCs and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments).(G) In vivo teratoma formation assay. H&E staining of teratoma derived from TePR-sESCs, Scale bars, 100 µm. (H) Immunostaining POU5F1, SOX2, NANOG, SSEA4, TRA-1–81, and TRA-1–60, Scale bars, 100 µm. (I) Immunostaining AFP, GATA6, SMA, Nestin, and Tublin, Scale bars, 400 µm. For (B), P10 and P47 represent the TePR-sESCs at passage 10 and 47, respectively. For (C), P15 and P100 represent the TePR-sESCs at passage 15 and 100, respectively. For (D), the error bar indicates ± SD (n = 3, independent experiments); cell line 1 at passage 10 (TePR-P10) and cell line 2 at passage 50 (TePR-P50) were used. For (E), the error bar indicates ± SD (n = 3, independent experiments); cell line 4 at passage 10 and cell line 1 at passage 100 were used. For (F), the error bar indicates ± SD (n = 3, independent experiments); cell line 1 at passage 47 were used. For (G), (H), (I) similar results were obtained in three independent experiments. Generation and characterization of TePR-sESCs . (A) Schematic diagram of the derivation of TePR-sESCs at day 6 in vivo fertilization embryos. (B) Bright field images and AP staining of TePR-sESCs, Scale bars, 40 µm. (C) Karyotype analysis of TePR-sESCs. (D) Gene expression analysis using RT-qPCR of core pluripotency genes ( POU5F1 , SOX2 , NANOG , LIN28 , and SALL4 ) of TePR-sESCs. The relative expression was normalized to SFF and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments). (E) Single-cell cloning efficiency of TePR-sESCs (P10) and TePR-sESCs (P100). (F) In vitro EB differentiation assay. RT-PCR for the ectodermal ( OTX2 , PAX3 , KRT8 , and NESTIN ), mesodermal ( PDGFRA and ACTA2 ) and endoderm ( FOXA2 and GATA6 ) specific genes. The relative expression was normalized to TePR-sESCs and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments).(G) In vivo teratoma formation assay. H&E staining of teratoma derived from TePR-sESCs, Scale bars, 100 µm. (H) Immunostaining POU5F1, SOX2, NANOG, SSEA4, TRA-1–81, and TRA-1–60, Scale bars, 100 µm. (I) Immunostaining AFP, GATA6, SMA, Nestin, and Tublin, Scale bars, 400 µm. For (B), P10 and P47 represent the TePR-sESCs at passage 10 and 47, respectively. For (C), P15 and P100 represent the TePR-sESCs at passage 15 and 100, respectively. For (D), the error bar indicates ± SD (n = 3, independent experiments); cell line 1 at passage 10 (TePR-P10) and cell line 2 at passage 50 (TePR-P50) were used. For (E), the error bar indicates ± SD (n = 3, independent experiments); cell line 4 at passage 10 and cell line 1 at passage 100 were used. For (F), the error bar indicates ± SD (n = 3, independent experiments); cell line 1 at passage 47 were used. For (G), (H), (I) similar results were obtained in three independent experiments. TePR-sESCs displayed compact dome-shaped colonies with smooth colony edges could be passaged at the single-cell level by enzymatic dissociation every 2–3 days (1:4 passaging ratio; Fig. 1 B). TePR-sESCs could also maintain robust growth after long-term cultures (>100 passages; Fig. S1 C, Supporting Information ) and the colony formation efficiency was 24 % and 30.7 % at P10 and P100, respectively ( Fig. 1 E and Fig. S1 D, Supporting Information ). Alkaline phosphatase staining was positive at P100 ( Fig. 1 B and Fig. S1 E, Supporting Information ). Karyotype analysis at P15 (2n = 54, 23/30 76.7 %) and P100 (25/30 83.3 %) showed that TePR-sESCs were genetically stable with a normal chromosome number ( Fig. 1 C). Cell cycle analysis showed a higher proportion of cells in the S phase compared to sheep fetal fibroblasts (SFFs; Fig. S1 F, Supporting Information ). TePR-sESCs expressed pluripotent stem cell marker genes including POU5F1 , SOX2 , NANOG , LIN28 , and SALL4 ( Fig. 1 D and H). In addition, the embryoid body (EB) differentiation assay showed that TePR-sESCs have the potential to differentiate into three primordial germ lineages ( Fig. 1 F). We could also generate neurons from TePR-sESCs via directed differentiation ( Fig. S1 G, Supporting Information ). Moreover, TePR-sESCs showed the potential to form teratoma, which contained tissues from all three germ layers ( Fig. 1 G and I). These results indicate the successfully establishment and the potential of the derived sheep ESCs. To investigate the transcriptomic characteristics of TePR-sESCs, we performed a transcriptomic analysis of sheep preimplantation embryos at different developmental stages (MII oocytes, zygote, 2-cell, 8-cell, morula, early blastula, and late blastula; Fig. 2 A and Fig. S2 A, Supporting Information ). Each sample group contained at least 3 biological replicates, with the mapping rate around 85 %, and could be used for the further analysis ( Fig. 2 B and Table S2 , Supporting Information ). In consistency with a previously reported study [ 30 ], we noticed that the gene expression levels of sheep embryos were dramatically changed at 8-cell stage to morula stage, indicating that the sheep embryo genome activation (EGA) may occur during between these stages ( Fig. 2 C). The principal-component analysis (PCA) showed that TePR-sESCs have similar transcriptional features of sheep 8-cell stage to morula stage embryos ( Fig. 2 D). Fig. 2 Transcription characterization of TePR-sESCs. (A) Schematic diagram of transcriptome sequencing. (B) Heatmap of Sample correlation (MII oocytes, zygote, 2-cell, 8-cell, morula, early blastula, and late blastula and TePR-sESCs). (C) Gene expression levels (counts) across different embryo stages (MII oocytes, zygote, 2-cell, 8-cell, morula, early blastula, and late blastula as well as TePR-sESCs). (D) Principal-component analysis of TePR-sESCs and different embryo stages.(E) Boxplot of marker genes expression in 8-cell stage embryo, morula, and TePR-sESCs. (F) Volcano plot of DEGs between 8-cell stage embryo and TePR-sESCs. (G) GO enrichment analysis (Biological processes) about the DEGs between 8-cell stage embryo and TePR-sESCs. (H) Volcano plot of DEGs between morula and TePR-sESCs. (I) GO enrichment analysis (Biological processes) about the DEGs between morula and TePR-sESCs. Transcription characterization of TePR-sESCs. (A) Schematic diagram of transcriptome sequencing. (B) Heatmap of Sample correlation (MII oocytes, zygote, 2-cell, 8-cell, morula, early blastula, and late blastula and TePR-sESCs). (C) Gene expression levels (counts) across different embryo stages (MII oocytes, zygote, 2-cell, 8-cell, morula, early blastula, and late blastula as well as TePR-sESCs). (D) Principal-component analysis of TePR-sESCs and different embryo stages.(E) Boxplot of marker genes expression in 8-cell stage embryo, morula, and TePR-sESCs. (F) Volcano plot of DEGs between 8-cell stage embryo and TePR-sESCs. (G) GO enrichment analysis (Biological processes) about the DEGs between 8-cell stage embryo and TePR-sESCs. (H) Volcano plot of DEGs between morula and TePR-sESCs. (I) GO enrichment analysis (Biological processes) about the DEGs between morula and TePR-sESCs. In order to figure out the difference among sheep 8-cell stage embryos, morula stage embryos, and TePR-sESCs. Based on the transcriptomic data, we firstly analyzed the expression levels of several key pluripotent genes ( POU5F1 , SOX2 , NANOG , LIN28A , and SALL4 ). Interestingly, we found that POU5F1 was highly expressed in morula than 8-cell stage embryos and TePR-sESCs ( Fig. 2 E). On the other hand, the expression of DPPA4 was in a similar level in 8-cell stage embryos and TePR-sESCs ( Fig. 2 E). These results show the dramatic changes in the gene expression levels at different embryonic stages and the potential of the transcriptomic analysis in understanding the gene regulation network. The transcriptional differences between sheep 8-cell stage embryo and TePR-sESCs (termed 8-cell vs. TePR-sESCs) were also analyzed. Between these two developmental stages, there were 705 upregulated genes and 1,405 downregulated genes identified ( Fig. 2 F, Fig. S3 A, B and Table S1 , Supporting Information ). Functional enrichment analyses showed that upregulated differentially expressed genes (DEGs) were enriched in the WNT and BMP signaling pathways and the embryonic organ morphogenesis, while the downregulated DEGs were enriched in nervous system and epithelial cell development ( Fig. 2 G, Table S3 , 4, Supporting Information ). This is consistent with the recently reported results that the WNT activation has a negative effect on the generation of naïve ESCs [ 31 ]. The Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis showed that the DEGs were enriched in HIF-1 signaling, tight junction, and AMPK signaling pathways ( Fig. S2 C and Table S3 , 4, Supporting Information ). We also found that there were 5,263 genes upregulated and 4,188 genes downregulated between morula and TePR-sESCs (termed morula vs. TePR-sESCs; Fig. 2 H, Fig. S3 C, D and Table S1 , Supporting Information ). The Gene Ontology (GO) enrichment analysis showed that these DEGs were enriched in translation, oxidative phosphorylation, Pi-body, and histone H3K4 demethylase activity ( Fig. 2 I, and Table S5 , 6, Supporting Information ). The KEGG pathway enrichment analysis indicated that these DEGs were enriched in signaling pathways regulating pluripotency of stem cells ( Fig. S2 D and Table S5 , 6, Supporting Information ). These results showed that TePR-sESCs might be amenable to modulation through metabolic regulators and histone demethylase activity, consistent with the previously reported results in human and mouse ESCs [ 25 , 32 ]. Interestingly, TFAP2C (naïve gene), ZSCAN4 (classical totipotency gene), DUSP10 , and MYC were upregulated, while LIN28A was downregulated in both the 8-cell vs. TePR-sESCs and morula vs. TePR-sESCs groups ( Fig. 2 F, H). We also noticed that POU5F1 and SALL4 were downregulated in the morula vs. TePR-sESCs group ( Fig. 2 H). In TePR-sESCs, the pluripotency genes ( POU5F1, SOX2, NANOG and SALL4 ) showed high H3K4me3 peaks and low H3K27me3 peaks at the promoter regions ( Fig. S2 E, Supporting Information ), similar to human and porcine EPSCs [ 6 ]. These results reveal the dynamic expressional changes of pluripotency related genes in sheep embryos and provide more evidence to investigate the conversion processes of sheep ESCs. In order to measure the global epigenetic features of TePR-sESCs, we sampled TePR-sESCs and SFFs by an assay for transposase-accessible chromatin with high sequencing (ATAC-seq). Based on the ATAC-seq, we could assess the chromatin accessibility signature between TePR-sESCs and SFFs. Based on the ATAC-seq results, we noticed that the average differentially accessible regions (DARs) of TePR-sESCs were higher than SFFs ( Fig. 3 A). In addition, the PCA analysis showed that TePR-sESCs clustered tightly as a group separate from SFFs ( Fig. S4 A, Supporting Information ). The DARs were predominantly located in the promoter region and were dramatically changed between TePR-sESCs and SFFs ( Fig. 3 B, C). We observed that the DARs of TePR-sESCs were enriched in motifs corresponding to pluripotent transcription factors (TFs), such as POU5F1 , SOX2 , and NANOG ( Fig. 3 D). In addition, the CCCTC-binding factor (CTCF) was also enriched. It showed that CTCF involved in mediating the pluripotency changes of TePR-sESCs ( Fig. 3 D). The chromatin accessibility analysis of these genes also indicated that these loci were more accessible in SFFs than in TePR-sESCs ( Fig. S4 C-F, Supporting Information ). Specifically, upregulated and downregulated DARs were identified between TePR-sESCs and SFFs ( Fig. 3 E, Fig. S4 B and Table S1 , Supporting Information ). Based on the DARs, we further annotated the DEGs. KEGG enrichment analysis showed that the upregulated DEGs were enriched in cAMP, cGMP-PKG, and WNT signaling pathways ( Fig. 3 E and Table S7 , 8, Supporting Information ), while the downregulated DEGs were enriched in AGE-RAGE, TNF, PI3K-Akt, and MAPK signaling pathways ( Fig. 3 E and Table S7 , 8, Supporting Information ). Fig. 3 Epigenetic features of TePR-sESCs. (A) Total ATAC-seq levels of TePR-sESCs and SFFs. (B) Distribution of DARs within different genomic regions (promoter, intron, coding exon, and distal intergenic regions) between TePR-sESCs and SFFs. (C) The violin plots illustrate the abundance of DARs in each genomic region in terms of their log2(Fold Change) ratio. (D) Motif enrichment analysis of ATAC-seq peaks in TePR-sESCs and SFFs. (E) Pathways enriched in genes with distinct chromatin accessibility in TePR-sESCs compared to SFFs. (F) DNA methylation levels of TePR-sESCs and SFFs.(G) Distribution of DMRs within different genomic regions (promoter, exon, intergenic, and intron regions) between TePR-sESCs and SFFs. (H) Averaged DNA methylation levels and DAR distribution of the transcription start sites (TSS ± 3 Kb). Epigenetic features of TePR-sESCs. (A) Total ATAC-seq levels of TePR-sESCs and SFFs. (B) Distribution of DARs within different genomic regions (promoter, intron, coding exon, and distal intergenic regions) between TePR-sESCs and SFFs. (C) The violin plots illustrate the abundance of DARs in each genomic region in terms of their log2(Fold Change) ratio. (D) Motif enrichment analysis of ATAC-seq peaks in TePR-sESCs and SFFs. (E) Pathways enriched in genes with distinct chromatin accessibility in TePR-sESCs compared to SFFs. (F) DNA methylation levels of TePR-sESCs and SFFs.(G) Distribution of DMRs within different genomic regions (promoter, exon, intergenic, and intron regions) between TePR-sESCs and SFFs. (H) Averaged DNA methylation levels and DAR distribution of the transcription start sites (TSS ± 3 Kb). DNA methylation is an important regulator of gene expression and undergoes dynamic changes in early embryo development [ 33 ]. In order to study DNA methylation changes between TePR-sESCs and SFFs, we performed whole genome bisulfite sequencing (WGBS) for TePR-sESCs and SFFs. The sample correlation was 89 % in each group ( Fig. S4 G, Supporting Information ). The WGBS results indicated that TePR-sESCs have high levels of DNA methylation around 82 % ( Fig. 3 F), similar to human, porcine, and bovine EPSCs [6a, 7b]. The differentially methylated regions (DMRs; Table S9 , 10, Supporting Information ) predominantly located in the intergenic region and intron region ( Fig. 3 G). Of note, the DARs and DMRs were near transcription start sites (TSSs), similar to bovine ESCs [ 33 , 34 ] ( Fig. 3 H). We compared the levels of DNA methylation between TePR-sESCs and Sheep blastocysts [ 29 ]. The results showed that TePR-sESCs exhibit higher methylation than blastocysts ( Fig. S4 H, Supporting Information ), likely reflecting their developmental stage prior to EGA. This is consistent with their transcriptomic similarity to 8-cell embryos, where parental methylation patterns are still predominant. The inclusion of preimplantation embryo epigenomic data provides a more biologically relevant benchmark for defining naive-like pluripotency in sheep. In order to understand the gene regulation network of TePR-sESCs and SFFs, we analyzed the further transcriptomic analyses. We found that there are 5,761 DEGs between TePR-sESCs and SFFs. Of these DEGs, 2,905 were upregulated and 2,856 were downregulated ( Fig. S2 B, Fig. S5 A, B and Table S1 , Supporting Information ). The GO enrichment analysis showed that these DEGs were enriched in embryonic organ morphogenesis, cell–cell junction, cell–cell contact zone, DNA-methyltransferase activity, and WNT receptor activity ( Fig. S5 C and Table S1 1, 12, Supporting Information ). On the other hand, the KEGG enrichment analysis indicated that the upregulated DEGs were enriched in Notch, cAMP, Rap1, and MAPK signaling pathways ( Fig. S5 D and Table S1 1, 12, Supporting Information ), while downregulated DEGs were enriched in AGE-RAGE, MAPK, Rap1, TNF, and TGF-beta signaling pathways ( Fig. S5 D and Table S1 1, 12, Supporting Information ). We also analyzed the expression of several imprinted genes in TePR-sESCs and SFFs. Compared to SFFs, the expression of IGF2 , IGF2R , MEST , PEG10 , PEG3 , and PLAGL1 were lower in TePR-sESCs ( Fig. 4 D). We also found that the histone genes ( H1-1 , H1-4 , H1-5 , H3F3A , and H2AZ1 ) were highly expressed in TePR-sESCs compared to SFFs ( Fig. 4 E). Fig. 4 Specific gene network features of TePR-sESCs (A) PCA analysis comparing TePR-sESCs, pEPSC, bESC, hNaive ESC, hEPS, and hPrimed ESC. (B) Venn of TePR-sESCs, pEPSC, bESC, hNaive ESC, hEPS, and hPrimed ESC. (C) GO enrichment analysis (Biology process) of the TePR-sESCs unique expressed genes. (D) Expression of imprinted genes in TePR-sESCs and SFFs. (E) Expression levels of histone genes in TePR-sESCs and SFFs. (F) DMR, DAR and DEG overlap of TePR-sESCs and SFFs. Specific gene network features of TePR-sESCs (A) PCA analysis comparing TePR-sESCs, pEPSC, bESC, hNaive ESC, hEPS, and hPrimed ESC. (B) Venn of TePR-sESCs, pEPSC, bESC, hNaive ESC, hEPS, and hPrimed ESC. (C) GO enrichment analysis (Biology process) of the TePR-sESCs unique expressed genes. (D) Expression of imprinted genes in TePR-sESCs and SFFs. (E) Expression levels of histone genes in TePR-sESCs and SFFs. (F) DMR, DAR and DEG overlap of TePR-sESCs and SFFs. In order to explore the specific gene regulation network of sheep ESCs, the RNA-seq data of pEPSC [ 6 ], bESC [ 7 ], human primed ESCs (hPrimed) [ 27 ], human EPSCs (hEPS) [ 28 ], and human naïve ESCs (hNaive) [ 27 ] were used to conjointly analyze the expression patterns of TePR-sESCs. The PCA analysis showed that the expression patterns of TePR-sESCs were closed to porcine EPSCs and human naïve ESCs ( Fig. 4 A). Interestingly, there are 356 genes co-expressed in these six types of cells and only 292 genes were expressed in TePR-sESCs ( Fig. 4 B and Table S1 , Supporting Information ). Functional enrichment analyses showed that the TePR-sESCs specifically expressed genes were enriched in ribosome biogenesis and protein neddylation ( Fig. 4 C and Table S1 3, Supporting Information ), while the co-expressed genes were enriched in translation, gene expression, and spliceosome ( Fig. S6 A and Table S1 4, Supporting Information ). Based on these results, we performed joint transcriptomic analyses with 8-cell embryos and morula, 13 (TPT1, CLDN6, CK8, MDK, KRT18, LIN28A, MAZ, GD11, CNN2, ARL2, MTA2, NPC2, and CD9) and 43 co-expressed genes were identified, respectively ( Fig. S6 E, F, and Table S1 5 Supporting Information ). We also analyzed the co-expressed genes in bESCs and TePR-sESCs by GO enrichment analysis. The results showed that these genes were enriched in fibroblast growth factor receptor signaling pathway, histone ubiquitination, DNA methylation, and oxidative phosphorylation ( Fig. S6 B and Table S1 6, Supporting Information ). To determine the relationship between chromatin accessibility, gene expression, and DNA methylation in TePR-sESCs, we integrated multi-omics investigations by co-jointly analyzing the DEGs, differentially methylated genes (DMGs), and differentially ATAC-enriched genes DAGs. A total of 13 genes ( PDE4D , SIX6 , TBX18 , MECOM , NNAT , LOC101122274 , GPM6A , HOXA4 , KIAA1217 , BDKRB1 , LOC101111217 , GLT8D2 , and GAB2 ) were identified ( Fig. 4 F and Fig. S6 C, Supporting Information ). The functional enrichment analyses showed these genes were enriched in cAMP, MAPK, Cell-cell junction, and Ras signaling pathways ( Fig. S6 D, Supporting Information ). Taken together, these results provide new potential genes for studying the sheep embryo development and imply that these genes and pathways may be important for maintaining the pluripotency of TePR-sESCs. Based on the previous study, WNT signaling pathway-related inhibitors play important role in maintaining the pluripotency of ESCs [ 6 ]. In this study, we also tested the requirements of IWR-1 for long term in vitro culturing TePR-sESCs by removing IWR-1 from the culture medium. We noticed that the removal of IWR-1 would disrupt the morphology of TePR-sESCs, the AP staining results were weakened and RT-PCR results showed that the expression of POU5F1, SOX2 and SALL4 were significantly decreased than TePR-sESCs ( Fig. 5 A, B). All of these results indicated IWR-1 is indispensable requirement to sustain pluripotency of TePR-sESCs. Fig. 5 WNT inhibitors maintaining the pluripotency of TePR-sESCs. Comparison of morphology and AP staining in TePR-sESCs cultured without IWR-1, and with XAV939. Scale bar, 40 μm. (B) Quantification of mRNA expression of proliferation-associated genes by qRT-PCR. The relative expression was normalized to IWR-1 and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments). (C) Immunostaining of POU5F1, SOX2, and NANOG, Scale bars, 200 µm. WNT inhibitors maintaining the pluripotency of TePR-sESCs. Comparison of morphology and AP staining in TePR-sESCs cultured without IWR-1, and with XAV939. Scale bar, 40 μm. (B) Quantification of mRNA expression of proliferation-associated genes by qRT-PCR. The relative expression was normalized to IWR-1 and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments). (C) Immunostaining of POU5F1, SOX2, and NANOG, Scale bars, 200 µm. Then, we tested whether IWR-1 could be replaced with 2.5 μM XAV939. TePR-sESCs were adapted to the culture medium containing XVA939, after 8 passages, cells were used for further analyses. The results showed that cells still exhibited dome-shaped morphology ( Fig. 5 A), positive AP staining ( Fig. 5 A) and no detriment of pluripotency factor expression ( Fig. 5 A-C, Fig. S7 , Supporting Information ). All of these results indicated that IWR-1 could be replaced by XAV939. In order to figure out whether TePR-sESCs can contribute to form chimeras, we injected mCherry-labeled TePR-sESCs into the sheep early blastulae ( Fig. S8 A, Supporting Information ). After injection, we transferred the injected blastocysts into the recipient sheep. A total 37 new born lambs were obtained for testing. Unfortunately, no mCherry-positive cell signals were observed ( Fig. S8 B, C, Supporting Information ). We hypothesize that insufficient coordination between the cell cycle phase of TePR-sESCs and the developmental stage of the host embryo disrupts cellular proliferation and integration. The negative results from current sheep chimera assays highlight technical challenges in sheep ESC research, which still need further research. Gene editing in large animals would facilitate agricultural development and biological research. A previous study of pig pgEpiSCs has reported that pgEpiSCs could tolerate several rounds of gene editing [ 11 ]; however, whether TePR-sESCs could also tolerate gene editing warranted further investigation. In order to figure out the editability of TePR-sESCs, we preformed two rounds of genomic manipulation ( Fig. 6 A). First, we generated mCherry-labeled TePR-sESCs by PiggyBac transposition. The mCherry-positive cells were sorted by flow cytometry ( Fig. 6 B). After that, we performed CRISPR/Cas9-mediated knock-out of the myostatin ( MSTN ) gene, a significant gene known to promote muscle differentiation and growth in animals [ 22 ]. MSTN knocked-out cell colonies were selected by infinite dilution method and were subjected to the Sanger sequencing and targeted deep sequencing. Sanger sequencing based genotyping of 15 colonies indicated that 13.33 % (2/15) of the cell colonies were homozygous. One of the homozygously edited cell colonies showed a single base pair insertion [termed MSTN -1(+1)] and the other homozygously edited cell colony showed a 4-base pair deletion [termed MSTN -5(−4); Fig. 6 C, D]. Targeted deep sequencing results showed that the editing efficiency of these two homozygous colonies were up to 93.9 % and 97.4 % ( Fig. 6 E), respectively. Western blotting results also indicated that these two single cell colonies did not express the MSTN protein ( Fig. 6 F). At the same time, we also tested the pluripotency of MSTN -1(+1) and MSTN -5(−4). MSTN -1(+1) and MSTN -5(−4) displayed dome-shaped, and the RT-PCR results showed that the pluripotency genes were expressed ( Fig. 6 G). These results indicated that the TePR-sESCs could be used for genome modification, including traditional transgenic insertion and modern genome editing, such as CRISRP/Cas9-based gene knockout. Fig. 6 Gene editing expression of TePR-sESCs . (A) Gene editing Pipeline. (B) Fluorescence images of mCherry labeled TePR-sESCs, Scale bars, 100 µm. (C) Image of MSTN -/- TePR-sESCs, Scale bars, 40 µm and 400 µm. Sanger sequencing result. (D) Editing efficiency of MSTN -/- TePR-sESCs. (E)Western blotting results of MSTN -/- TePR-sESCs. (F) qRT-PCR of pluripotent genes in MSTN -/- TePR-sESCs. The relative expression was normalized to SFFs and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments). For (E), (G), the error bar indicates ± SD (n = 3, independent experiments). The cell line 1 at passage 47 were used for gene editing and positive-colony selection. Gene editing expression of TePR-sESCs . (A) Gene editing Pipeline. (B) Fluorescence images of mCherry labeled TePR-sESCs, Scale bars, 100 µm. (C) Image of MSTN -/- TePR-sESCs, Scale bars, 40 µm and 400 µm. Sanger sequencing result. (D) Editing efficiency of MSTN -/- TePR-sESCs. (E)Western blotting results of MSTN -/- TePR-sESCs. (F) qRT-PCR of pluripotent genes in MSTN -/- TePR-sESCs. The relative expression was normalized to SFFs and housekeeping gene GAPDH. Data are represented as ± SD (n = 3, independent experiments). For (E), (G), the error bar indicates ± SD (n = 3, independent experiments). The cell line 1 at passage 47 were used for gene editing and positive-colony selection.

All procedures related to animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) at Northwest A&F university, China (Approval no. 2024–1216). In vivo produced embryos were collected from ewes according to a previously reported protocol [ 22 ]. In brief, EAZI-BREED TM CIDR® Sheep Devices were inserted into the vagina of donor ewes for 14 days. The superovulation was performed 60 h prior to the removal of the CIDR devices. A total of 260 mg follicle-stimulating hormone (FSH) was injected in 7 dosages. The interval between each dosage was 12 h, and the first dose was 70 mg and others were decreased progressively to 25 mg. A total of 0.1 mg cloprostenol was injected 60 h later. After the withdrawal of CIDR for 12 h, the estrous conditions of donor ewes were detected, and matings were performed every 8 h. Sheep zygotes were surgically collected after 48 h of mating. Zygotes were immediately transferred to embryo in vitro culture medium (BO-IVC, IVF Bioscience) and cultured at 38 °C, 5 % CO 2 . Zygotes developed into 2-cell, 8-cell, morula, early blastula, and late blastula were collected for further sequencing. Sheep ovaries were collected from slaughterhouse and transported to the lab in 37 °C, 0.9 % physiological saline containing 1 × penicillin–streptomycin (BI, 03–031-1B). Ovaries were washed and kept in warm 0.9 % physiological saline. Sheep cumulus oocyte complexes (COCs) were squeezed from the follicles using 1 mL syringes and sorted by mouth pipettes. Oocytes with several layers of cumulus cells were selected and transferred to in vitro maturation (IVM) medium. In vivo produced blastocysts were surgically collected at day 7, as we previously described [ 22 ]. Zone pellucida was removed by Tyrode’s solution, Acidic (Sigma, T1788-100 mL), and washed three times. Blastocysts were plated in 4-well plates on Mitomycin C (Selleck, S8146) treated mouse embryonic fibroblast (MEF) cells in four different culture conditions: NBFR [ 23 ], bEPSCM [ 10 ], 3i/LAF [ 11 ], and TePR, supplemented with 10 μM Y-27632 (Selleck, S6390). After 48 h, the medium was changed daily without Y-27632 for one week until the outgrowths were visible. Then, outgrowths were digested and passaged by TrypLE TM Select (Gibco, 2481905) to new MEF feeders supplemented with 10 μM Y-27632 at a 1:1 ratio. Cell colonies were observed 2–3 days after passage. The medium was changed daily, and further passages were performed every 3–4 days at a 1:4 ratios. TePR-sESCs were cultured in TePR medium [mTeSR PLUS-based medium (Stem Cell Technology, 100–0276), supplemented with 2.5 μM IWR-1 (Sigma, I0161)] on MEF feeder plates pre-coated with 0.1 % gelatin (Sigma, G1890-100G). TePR-sESCs were cultured at 37 °C with 5 % CO 2 , and the medium was changed daily. For passaging TePR-sESCs, cells were digested into single cells by TrypLE TM Select (Gibco, 12563011) and reseeded to a new MEF feeder plate at a 1:4–5 ratio every 3–4 days. Cells were fixed in 4 % paraformaldehyde at room temperature for 10 min. AP staining was based on the Alkaline Phosphatase Staining Kit (Beyotime Biotechnology, C3250S), following the manufacturer’s protocol. A total of 1 % KaryoMAX Solution (Gibco, 15212012) was added to the TePR-sESC culture medium, and TePR-sESCs were incubated for 2 h. TePR-sESCs were harvested by TrypLE TM Select (Gibco, 2481905) and resuspended in 5 mL of a 37 °C pre-warmed 0.075 M potassium chloride (KCL) solution. After incubation at 37 °C for 30 min, 3 mL precooling fixative solution (3:1, methanol:acetic acid) were added and centrifuged at 1,000  rpm for 5 min. The supernatant was discarded, and the pellet was resuspended in a 5 mL fixative solution and incubated for 20 min on ice. Then the mixture was centrifuged at 1,000  rpm for 10 min. The fixing process was repeated three times. After that, cells were resuspended by a 100 μL fixative solution and dropped onto a cold slide. Slides were dried at room temperature and stained with Giemsa (Sangon, E607314-0001). TePR-sESCs were dissociated into single cells by TrypLE TM Select (Gibco, 2481905), and feeder cells were separated by differential attachment. Fetal fibroblasts were dissociated by 0.5 % Trypsin-EDTA (Gibco, 15400054). Cells were fixed by precooling 70 % ethanol and stained with propidium iodide (PI)/RNase solution (eBioscience, 00–6990-50) for 15 min, avoided from light. After that, cells were analyzed by a fluorescence-activated cell sorting (FACS) flow cytometer (BD, FACSAria III). TePR-sESCs were seeded at a 12-well plate. The density of 500 cells per well were plated in triplicate. The colonies were counted 7 days later using AP staining. The colony formation efficiency was evaluated as a percentage of colony number per number of cell seeded. TePR-sESCs were seeded at a density of 2 × 10 5 cells per well. The cell numbers were counted every 24 h. Cells were digested and counted using a Countess™ 3 Automated Cell Counter (Invitrogen™, AMQAX2000). Each sample have three repeats. Total RNA was extracted by TRIzol Plus RNA Purification Kit (Invitrogrn, 12183555CN) following the manufacturer’s instructions. cDNA was synthesized with the PrimeScript TM RT reagent kit with DNA Eraser (RR047A, TaKaRa). RT-PCR was performed by TB Green Premix EX Taq TM II (RR820Q, TaKaRa). GAPDH was used as an internal normalization control. Primers used in this study are listed in the Table S17 ( Supporting Information ). Cells were fixed with 4 % paraformaldehyde at room temperature for 10 min and washed with DPBS (Gibco, 10010023). Then the cells were permeabilized and blocked in 0.3 % Trition-X100 and 3 % Normal Donkey serum for 30 min at room temperature. The cells were incubated with primary antibodies for 1 h at room temperature and diluted with 0.3 % Trition-X100 and 1 % normal donkey serum. Secondary antibodies were incubated for 1 h at room temperature. The nuclei were stained with DAPI (Beyotime, C1005) for 3 min at room temperature. Antibodies used in this study are listed in Table S18 ( Supporting Information ). TePR-sESCs were dissociated by TrypLE TM Select (Gibco, 2481905), and feeder cells were separated by differential attachment. Then, TePR-sESCs were seeded into AggreWell TM 400 24-well (Stem Cell Technology, 34411). After 3 days, the generated embryoid bodies were transferred to gelatin-coated dishes and cultured for an additional 2 weeks in DMEM medium supplemented with 10 % FBS (Gibco, 2075800) and FGF2 (Peprotech, 100-18B). TePR-sESCs were digested into single cells by TrypLE TM Select (Gibco, 2481905), and 1 × 10 6 cells were suspended in 100 μL cold DMEM/F12 medium (Gibco, 11320–033) and 100 μL cold Matrigel (Corning, 354277). Then the complexes were injected into 6–8 week-old immunodeficient NOD-SCID-IL2Rg mice. After 10 weeks, teratomas were dissected and fixed with 4 % paraformaldehyde. Fixed teratomas were sent to Servicebio (Wuhan, China) for hematoxylin and eosin (H&E) staining and immunofluorescence analysis. Antibodies used in this study are listed in Table S17 ( Supporting Information ). In order to differentiate TePR-sESCs into neurons, we followed the previous published protocol [ 24 ]. Briefly, TePR-sESCs were plated in 0.1 % gelatin-coated 6 well plate at a density of cells in TePR-Activin A (TePR + Activin A) medium. When the cells reach a density at which they can be passaged, the cells were passaged at 1:4–6 ratio into 0.1 % gelatin-coated 6 well plate in MEF-contained sheep ESCs medium contains 10 μM Y-27632. On the next day, replace the medium with fresh MEF-contains sheep ESCs medium, Incubate under these conditions for 4–6 days. Then the medium was changed to neural induction medium (Neural maintenance medium is supplemented with 500 ng/ml noggin and 10 μM SB431542 or with 1 μM Dorsomorphin and 10 μM SB431542. Store the medium at 4 °C and use it within 5 days) for additional 5–7 days. After induction, cells were passaged at a 1:4–6 ratio into 0.1 % gelatin-coated plate in neural maintenance medium (This is a 1:1 mixture of N-2 and B-27containing medium. N-2 medium consists of DMEM GlutaMAX, 1 × N-2, 5 μg/mL insulin, 1 mM L-glutamine, 100 μm nonessential amino acids, 100 μM 2-mercaptoethanol, 50 U/mL penicillin and 50 mg/mL streptomycin. B-27 medium consists of Neurobasal, 1 × B-27, 200 mM L-glutamine, 50 U/mL penicillin and 50 mg/mL streptomycin. Store the medium at 4 °C and use it within 3 weeks) for additional 7–12 days. Then, cells were collected for analysis. Total RNA of TePR-sESCs and SFFs (three biologically replications, respectively) were extracted by TRIzol Plus RNA Purification Kit (Invitrogrn, 12183555CN) following the manufacturer’s instructions. RNA were analyzed by NanoDrop 2000C Spectrophotometer (ThermoFisher Scientific), the total RNA with good quality were shipped to Beijing Genomics Institute for library construction and sequencing. All raw reads underwent quality control using fastp (version 0.22.0), followed by alignment to reference genomes (ARS-UI_Ramb_v2.0, Sscrofa11.1, ARS-UCD1.3, GRCh38.p14) using HISAT2 (version 2.2.1). Uniquely mapped reads were retained for gene expression quantification using featureCounts (version 2.0.3), ComBat_seq algorithm implemented in SVA (R package version 3.4.2) was used to remove batch effect, gene counts were transformed into FPKM (Fragments Per Kilobase Million) values to identify highly expressed genes. A 2-fold variance in expression levels and an adjusted P value less than 0.05 were used as cutoffs to define differentially expressed genes (DEGs). Differential gene expression analysis was conducted using DESeq2, and KEGG pathway enrichment for DEGs was performed with KOBAS ( https://bioinfo.org/kobas/ ). GO terms, encompassing Biological Processes, Cellular Components, and Molecular Functions for DEGs, were identified using enrichR (R package version 3.2). Data visualization was achieved using the pheatmap package (R package version 1.0.12) and ggplot2 (R package version 3.4.2) to generate informative figures. Smart-seq2 was performed by E-GENE (Shenzhen, China), Shenzhen based on the previous study [ 25 ]. In brief, embryos at different stages were harvested and loaded into a 96-well plate, then cells were subjected to cell lysis, reverse transcription and cDNA amplification. Then, the cDNA libraries were labelled with specific barcode and sequenced by a BGISEQ-500 sequencer. Each sample have at least three biologically replications. Approximately 8,000–10,000 cells of each sample were collected (three biologically replications). Cells were washed with cold D-PBS and re-suspended in lysis buffer based on a previous study [ 26 ]. Library preparation was performed following the manufacturer’s protocol (TD501-02, Vazyme and TD202 96rxn, Vazyme) and sequenced by a BGISEQ-500 sequencer at E-GENE, Shenzhen. The ATAC-seq analysis was conducted followed the established ENCODE analysis pipeline. Initial quality control for raw sequence data was carried out using fastp (version 0.22.0). Subsequently, clean reads were aligned to the ARS-UI_Ramb_v2.0 reference genome using Bowtie2 (version 2.5.1). To ensure data integrity, PCR duplicates within the alignments were removed, and only unique alignments within each sample were retained for further analyses. ATAC-seq peaks were independently called for each sample using MACS2 (version 2.2.7.1) with specific parameters: −f BAMPE −g 2.6e9 −B −q 0.05 −-nomodel −-shift −75 −-extsize 150 −-keep-dup all −-call-summits. Genomic feature annotation packages were constructed using GenomicFeatures (R package version 1.50.4) and the NCBI ARS-UI_Ramb_v2.0 gtf annotation file. For in-depth analysis, transcription factor motif enrichment within these peaks was evaluated using the findMotifsGenome.pl script from HOMER ( https://homer.ucsd.edu/homer/motif/ ). Differential ATAC-seq enrichment peaks between TePR-sESCs and SFFs groups were identified using DiffBind (R package version 3.8.4), focusing on differential peaks located within 2000 bp of the TSS and exhibiting a fold change > 1. These differentially ATAC-enriched genes (DAGs) were subjected to GO enrichment analysis using enrichR (R package version 3.2) and KEGG pathway enrichment analysis using KOBAS ( https://bioinfo.org/kobas/ ). Data visualization was expertly achieved using the pheatmap package (R package version 1.0.12) and ggplot2 (R package version 3.4.2) to generate informative figures. WGBS libraries construction was following the previous study [ 18 ]. Shortly, genomic DNA of TePR-sESCs and SFFs (three biologically replications, respectively) were extracted by using Universal Genomic DNA Kit (CWBIO, CW2298). Labraries construction were following the manufacturer’s protocol (Zymo Research) and sequenced by a BGISEQ-500 sequencer at E-GENE, Shenzhen. The BS conversion was applied to the ARS-UI_Ramb_v2.0 reference genome using bismark's (version 0.24.0) bismark_genome_preparation function. Quality control for raw sequence data was executed with fastp (version 0.22.0). To ensure data quality, PCR-duplicated reads were removed using deduplicate_bismark. To mitigate sequencing bias, only reads with a minimum 10x coverage were utilized in subsequent analyses. Methylation levels of individual CpG sites were calculated, and the methylation status of each sample was determined by averaging consecutive genomic windows of 100 bp tiles with a 50 bp step size for methylation. The differentially methylated regions (DMRs) between two compared groups were defined using methylkit (R package version 1.24.0). DMRs were characterized as having methylation levels ≥ 75 % in one group and ≤ 25 % in the other, with statistical significance determined by Fisher's exact test (P-value ≤ 0.05, FDR ≤ 0.05). Hyper- and hypo-methylated tiles were those with DNA methylation levels ≥ 75 % and ≤ 25 %, respectively. Only DMRs located within 2000 base pairs of the TSSs and containing more than 10 DMR were annotated as differentially methylated genes (DMGs). GO enrichment analysis of DMGs was performed using enrichR (R package version 3.2), and KEGG pathway enrichment analysis was conducted using KOBAS ( https://bioinfo.org/kobas/ ). Data visualization was expertly achieved using ggplot2 (R package version 3.4.2) to generate informative figures. TePR-sESCs were digested into single cells by TrypLE TM Select (Gibco, 2481905). 5 × 10 5 cells were electroporated at 175 V, 7.5 ms by NEPA21 instrument (NEPAGENE, Japan). 1 μg PBase plasmid and 5 μg Pb-mCherry plasmid were performed to get mCherry-labeled TePR-sESCs. mCherry-labeled cells were sorted by FACS flow cytometer (BD, FACSAria III). For MSTN gene knockout in TePR-sESCs, 104 pmol CRISPR/Cas9 nuclease (IDT, 10008100) and 120 pmol sgRNA (Transgene, China) were transferred to TePR-sESCs. As we previously described [ 22 ]. In vivo produced blastocysts were surgically collected at day 7. The mCherry-labeled cells were digested into single cells by TrypLE TM Select (Gibco, 12563011). Then, the cells were centrifuged at 1000 rpm at room temperature for 3 min. Removing the supernatant and resuspended the cell with fresh culture medium. Each blastocyst injected 10 cells, after injection, blastocysts were cultured in TePR and embryo in vitro culture medium (BO-IVC, IVF Bioscience) (1:1) for 24 h for analysis to check if the TePR-sESCs incorporated the Inner cell mass (ICM). For chimeric embryo transfer, the blastocysts were transplanted to the recipient ewe after injection. Homozygously edited cell colonies were lysed in RIPA Lysis and Extraction Buffer (Thermo Scientific™, #89900) supplemented with Halt™ Protease Inhibitor Cocktail (EDTA-free, 100X; Thermo Scientific™, #78425) and phosphatase inhibitors on ice for 10 min. After centrifugation at 14,000 × g for 15 min at 4 °C, the supernatant was collected. Protein concentration was determined using the BCA Assay Kit (Vazyme, #ZJ101). Equal amounts of protein (10–30 μg) were mixed with 5 × SDS Loading Buffer (Vazyme, #LT103), denatured at 95 °C for 5 min, and cooled on ice. Samples were separated on a 10 % ACE future PAGE™ Precast Gel (ACE Biotechnology, #F11010Gel). Electrophoresis was initiated at 80 V until the dye front entered the separating gel, then increased to 120 V until the dye front reached the gel bottom. Proteins were wet-transferred onto PVDF membrane (100 V, 90 min). The membrane was blocked with Protein-Free Rapid Blocking Buffer (Epizyme, #PS108p) for 1 h at room temperature. Primary antibodies (1:1000–1:5000 dilution in blocking buffer) were incubated overnight at 4 °C. Membranes were washed three times (10 min each) with 1 × TBST on a rocking shaker at RT. HRP-conjugated secondary antibodies (1:2000–1:10,000 dilution) were applied for 1 h at room temperature, followed by identical TBST washes. Protein bands were detected using Omni-ECL Ultra-Sensitive Chemiluminescent Substrate (Epizyme, #SQ201) and imaged with a chemiluminescence detection system. GAPDH served as the loading control. All quantitative data were presented as the mean ± SD. Experiments were repeated at least three times. For RNA-seq, ATAC-Seq and WGBS, Benjamini-Hochberg (BH) was used to perform statistical analysis. For qRT-PCR results and other quantitative data, a multiple t test analysis with Prism (GraphPad Software) was used. P  < 0.05 were considered to be statistically significant. RNA-seq, WGBS and ATAC-seq sequencing data of sheep embryos at different stage, TePR-sESCs, and SFFs reported in this study were submitted to NCBI under accession number: PRJNA1185625. The RNA-seq data used in this study from pEPSC [6a] were downloaded from Genome Sequence Archive (GSA) ( https://ngdc.cncb.ac.cn/gsa/ ) of China National Center for Bioinformation-National Genomics Data Center (CNCB-NGDC) with accession code: CRA003960. The data about bESC [ 7 ] were downloaded from GEO (accession no. GSE110040 ). The accession number of human primed ESCs (hPrimed) and human naïve ESCs (hNaive) data were GSE87452 [ 27 ] and the accession number of human EPSCs (hEPS) were GSE89303 [ 28 ]. The data about sheep blastocysts WGBS [ 29 ] were downloaded from GEO (accession no. GSE190746 ).

All Institutional and National Guidelines for the care and use of animals (fisheries) were followed.

In summary, we provide a new approach to establish stable sheep ESCs in vitro. Sheep ESCs display distinct molecular properties and developmental potential. The transcriptional and epigenetic characterizations of sheep ESCs provide new evidence to study the development of sheep embryo and the pluripotency changes of sheep ESCs. Meanwhile, sheep ESCs could tolerate two rounds of gene editing (gene knock-in and knockout), providing great potential for multiple genetic modification. This study highlights novel insights on mammalian embryonic development and pluripotency changes. Finally, the results of the current ESC-based research might have important implications for not only therapeutic cloning by somatic cell nuclear transfer [ [54] , [55] , [56] ], but also clinical medicine targeted at specimen-specific regenerative, reconstructive and anti-senescence treatments focused on transgenic or non-transgenic stem cell-mediated tissue and organ engineering in sheep and other livestock species [ [57] , [58] , [59] ].

Establishment of bovine and porcine pluripotent stem cells has made significant progress to enhance the research of the livestock ESCs [ 10 , 11 ]. However, deriving stable sheep ESCs is still a challenge [ 21 ]. The study of sheep ESCs will provide a platform for biomedical and agriculture researches. For biomedical applications, their robust self-renewal, differentiation capacity, and physiological relevance to humans (due to comparable organ size, longer gestation, and singleton pregnancies) position them uniquely [ 35 ]. Sheep ESCs will provide a critical foundation for generating complex in vitro embryo models, such as ovine gastruloids, offering a more translatable large-mammal system than rodents for studying early patterning, morphogenesis, and conceptus-maternal interactions [ 36 , 37 ]. Furthermore, their genetic tractability is pivotal for xenotransplantation research. Stable ovine ESCs enable the generation of sheep-human chimeric embryos in vitro , facilitating studies on overcoming molecular barriers to interspecies chimerism and the potential generation of humanized tissues/organs within sheep hosts [ 38 , 39 ]. Genetic modification of sheep ESCs could introduce key compatibility genes (e.g., for immune modulation or vascularization), directly enhancing xenotransplantation feasibility [ 40 ]. Beyond biomedical applications, sheep ESCs also offer transformative prospects for precision livestock breeding. Their capacity for targeted trait knock-in (e.g., introducing disease-resistance genes) or knock-out (e.g., suppressing allergenic proteins) could accelerate the development of elite livestock varieties with enhanced productivity, climate resilience, and welfare traits [ 41 ]. Furthermore, ovine ESCs present a scalable, ethically streamlined platform for agricultural genetic engineering [ 16 ]. sheep ESCs bridge the gap between cutting-edge biotechnology and practical breeding programs, paving the way for sustainable, genome-edited livestock tailored to meet global agricultural challenges [ 9 ]. In this study, we explored a new culture condition to derive sheep ESC lines from whole blastocysts. This could be achieved by adding a tankyrase/WNT inhibitor (IWR-1) into the mTeSR PLUS basal medium, which was named TePR. This culture condition is much simpler than other previously reported culture conditions, in which there is no need to prepare a custom-made base medium and only IWR-1 should be added. Under TePR culture condition, we successfully derived sheep ESCs, termed TePR-sESCs. TePR-sESCs display a dome-shaped colony, express pluripotency markers, have the potential to differentiate into three germ layers after long-term passages, and maintain normal karyotypes, indicating their pluripotency. Recently, there are several researches have reported the effect of WNT/β-catenin signaling pathway in embryo development. By using human embryonic stem cells to generate human gastruloids-three-dimensional multicellular aggregates, the researchers revealed that WNT/β-catenin signaling involved in gastruloids establishing and maintaining patterning [ 42 ]. Another study has shown that adding Cardamonin (a β-catenin inhibitor) to human embryos culture medium won’t affect the expression of POU5F1, SOX2, NANOG and SALL4 , but the trophectoderm marker CDX2 was downregulated [ 43 ]. The exposure of bovine embryos to Wnt-C59 (non-canonical WNT inhibitor) promotes the proliferation of ICM cells, which indicates the ICM proliferation were regulated by WNT signaling pathway [ 44 ]. There also several reports have shown canonical WNT signaling pathway is associated with lineage fate determination in bovine preimplantation embryos [ 45 , 46 ]. Taken together, work in embryos could provide more evidences to support our results and explain the effect of WNT inhibitor for derivation and maintenance of TePR-sESCs. Consistent with previous findings, the removal of IWR-1 induced differentiation and loss of pluripotency in TePR-sESCs [ 21 ]. Recent studies have reported that IWR-1 and XAV939 are essential for deriving and maintaining the pluripotency of porcine [ 11 ], bovine [ 10 ], and ovine ESCs [ 21 ]. These results strongly suggest that the canonical WNT/β-catenin signaling pathway participates in regulating ESCs pluripotency. We replaced IWR-1 with XAV939 in the culture of TePR-sESCs and found that XAV939 could substitute for IWR-1 without compromising cell expansion or pluripotency factor expression across multiple passages. The β-catenin destruction complex comprises key components, including adenomatous polyposis coli (APC), axin, glycogen synthase kinase (GSK3β), and casein kinase (CK1α) [ 47 ]. Tankyrase 1 and 2 (TNKS1/TNKS2) regulate β-catenin stability, with IWR-1 shown to modulate β-catenin stability by inhibiting TNKS1/2 and stabilizing axin [ 48 ]. In contrast, XAV939 targets distinct domains of the tankyrase enzyme and demonstrates broader inhibitory activity against poly (ADP-ribose) polymerases [ 49 , 50 ]. Notably, the specific contributions of IWR-1 and XAV939 to tankyrase inhibition, axin stabilization, β-catenin degradation, and their combined effects require further investigation. In order to further understand the transcriptional characterizations of TePR-sESCs, we performed bulk RNA-seq of TePR-sESCs and SFFs to reveal the molecular mechanism of sheep ESCs differentiation. We noticed that the DEGs were enriched in multiple functional categories such as Notch signaling pathway, cAMP, Rap1, MAPK, AGE-RAGE, TNF, TGF-β signaling pathways, DNA-methyltransferase activity, Wnt receptor activity. Recent works showing that Notch and WNT signaling pathways involved in the generation of naïve pluripotent stem cells [ 25 ] and the Notch signaling pathways also enriched in bEPSCs and bovine fetal fibroblasts [ 10 ]. Cross-species comparison is advancing our understanding to study the similarity and difference between each specie. In this study, we analyzed the RNA-seq data of TePR-sESCs, porcine EPSCs, bovine primed ESCs, human primed ESCs, human EPSCs, and human naïve ESCs to further study the specific gene network of sheep ESCs. The results indicated that TePR-sESCs more resembled human naïve ESCs than other ESCs. This provides new evidence that sheep could be used as a powerful tool in basic and regenerative biomedicine research. Pluripotency describes cells that have the ability to generate cells from all three embryonic germ layers. The pluripotent cells can be recapitulated from different stages of embryonic development in vitro . The studies concerning mice and humans have provided the gold standards for naïve, formative, and primed ESCs [ 51 ]. Recently, the studies of porcine [ 11 ] and bovine [ 10 ] ESCs have provided new evidences to research the pluripotency about livestock; however, the investigation of the sheep pluripotency changes still lacking. In this study, we firstly combined the RNA-seq results of TePR-sESCs and sheep preimplantation embryos to feature the transcriptional characterizations of TePR-sESCs. The results have shown that TePR-sESCs were close to 8-cell stage embryo and morula. We also noticed that gene expression levels of sheep embryos were dramatically changed at 8-cell to morula stage, indicating that the sheep ZGA may occur during 8-cell to morula stage. Based on these results, TePR-sESCs could be used as a promising tool for studying sheep embryonic development and pluripotency changes. Genomic methylation and chromatin accessibility changes play an important role in embryo development [ 33 ], ESC pluripotency maintainance and differentiation [ 25 ]. In the current study, we compared the genomic methylation levels and chromatin accessibility changes of the TePR-sESCs and SFFs in order to figure out the epigenetic characterization changes of sheep ESCs differentiation. The classical pluripotency loci ( POU5F1, SOX2 , and NANOG ) were observed. These findings support a distinctive mode of transcriptional regulation for sheep ESCs and SFFs. Studying the changes of epigenetic characterizations could provide a new approach to understand sheep ESC differentiation. Based on the multi-omics technologies, we firstly combined the DEGs, DAGs, and DMGs of TePR-sESCs to find the specific regulators of sheep ESCs. There were 13 genes identified ( PDE4D, SIX6 , TBX18 , MECOM , NNAT , LOC101122274 , GPM6A , HOXA4 , KIAA1217 , BDKRB1 , LOC101111217 , GLT8D2 , and GAB2 ). SIX6 [ 52 ] and MECOM [ 53 ] were reported to be associated with stem cell proliferation and differentiation. However, how these genes affect sheep ESC proliferation and differentiation still unknown. Thus, further investigations are needed to reveal the potential functions of these genes. Sheep is one of the most important farm animals in agriculture. It is also an excellent model for biological research and biomedicine. Thus, stable sheep ESCs, combined with powerful gene editing tools, have great potential to facilitate animal breeding and biological research. We here report the first successful gene-edited sheep ESCs by simultaneously applying piggyBac transposition and CRISPR/Cas9-based gene knockout. The success of efficiently editing the ovine ESCs provides a promising tool for advancing further biological research, agricultural and regenerative biomedicine.

Embryonic stem cells (ESCs) have the potential to differentiate into all cell types of an adult individual and are optimal research materials for studying embryogenesis in vitro [ 1 ]. Under specific culture conditions, ESCs of mice [ 2 ], rats [ 3 ], humans [ 4 ], non-human primate species [ 5 ], pigs [ 6 ], and cattle [ 7 ] were established from the inner cell mass (ICM) of blastocyst-stage embryos. Stable livestock ESCs are expected to accelerate animal molecular breeding [ 8 ] and provide promising applications for biomedicine and understanding of cell fate decisions during early development [ 9 ]. In cattle, the bovine expanded potential stem cells (bEPSCs) [ 10 ] and primed bovine ESCs (CTFR-bESCs) were previously developed [ 7 ]. bEPSCs have the ability to differentiate into cells of the three germ layers in vitro , as well as embryonic and extraembryonic cell lineages in chimeras. bEPSCs are cultured in mTeSR1 base medium supplemented with leukemia inhibitory factor (LIF), activin-A, CHIR99021, WH-4–023 and WNT/TNK inhibitor (IWR-1 or XAV939). On the other hand, CTFR-bESCs are cultured in the CTFR condition, which consists of the custom mTeSR1 base medium, FGF2 and IWR-1 [ 7 ]. CTFR-bESCs display typical characteristics of mouse and human-primed ESCs; however, without the ability to generate chimeras. Porcine E10 pregastrulation epiblasts (pgEpiSCs) were established by using 3i/LAF culture medium containing CHIR99021, IWR-1-endo, WH-4–023, LIF, activin-A, and FGF2 [ 11 ]. pgEpiSCs also exhibited several pluripotency features, tolerated several rounds of genome editing and enabled the generation of gene-edited cloned live piglets. All of these studies have showed encouraging results and a promising future for livestock ESCs research. Sheep are a significant large livestock species that play an important role in food production systems. Sheep also offer a uniquely powerful platform for modeling human diseases due to their physiological and anatomical convergence with humans—a critical advantage over rodent models [ 12 ]. Their human-scale cardiovascular system enables faithful recapitulation of cardiac pathologies (e.g., myocardial infarction models for testing regenerative therapies) [ 13 , 14 ], while shared metabolic features like prolonged gestation permit robust study of developmental disorders and transgenerational epigenetic effects [ 15 ]. Crucially, ovine models excel in complex conditions where small-animal systems fall short: spontaneous osteoarthritis progression mirrors human joint degeneration timelines [ 16 ], their gynecological anatomy supports high-fidelity endometriosis modeling [ 17 ], and their sizable airways allow clinically relevant studies of cystic fibrosis therapeutics [ 18 ]. This translational fidelity, combined with ethical practicality for longitudinal interventions, positions sheep as indispensable for bridging preclinical discoveries and human clinical applications—particularly in organ systems where physiological scale directly impacts disease mechanisms [ 19 ]. Recently, owing to the rapid development in the multi-omics technologies, the investigation of early embryonic development, lineage trajectories of embryos, and embryo-to-stem-cell transition have been well established in mice and humans [ 20 ]. However, the understanding of sheep embryogenesis and the efficient ESC establishment from major large animal species, such as sheep, are still lacking. Vilarino et al . reported the first establishment of sheep primed ESCs (CTFR-sESCs) from in vitro produced ovine blastocysts [ 21 ]. Even though the CTFR culture system consistently supported the growth of CTFR-sESCs in an undifferentiated state after more than 40 passages. Nevertheless, the epigenetic features of CTFR-sESCs were uncharacterized, and it is worth to mention that the custom mTeSR1 base medium is cumbersome and not readily available from commercial sources. All of these challenges highlight the need for further investigation of sheep embryogenesis and sheep embryonic stem cells (sheep ESCs). To solve these issues, in this study, we applied mTeSR PLUS base medium supplemented with a WNT/TNK inhibitor (IWR-1), abbreviated as TePR, leading to the successful in vitro establishment of sheep ESC lines (termed TePR-sESCs). TePR-sESCs exhibited distinct features to differentiate into all three germ lineages and supported the application of genome editing for desired genetic alterations. We also collected sheep preimplantation embryos from different stages [embryonic day (E)_0 to E6] and preformed Smart-seq2 to reveal the molecular basis of sheep embryogenesis. Multi-omics analysis and cross-species analysis revealed the transcriptomic landscapes of TePR-sESCs and identified specifically expressed genes in sheep ESCs. These results open a new avenue for studying sheep preimplantation embryo development, improvement of animal breeding programs, and the advancement of large animal models for biomedical research.

The author declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.