Chemical Screening Identifies Danofloxacin as a Self‐Renewal Agonist of Embryonic Stem Cells through Alleviation of HDAC1‐Mediated Deacetylation of Tert and Prdm10 Yan Zhang, Junxiang Ji, Baoxing Dong, Zhenhua Zhu, Xucheng Lv, Peng Chen, Wenqiang Liu, Min Zhou, Shou‐Dong Ye Abstract While significant advances have been made in maintaining stem cell pluripotency, the successful establishment of embryonic stem cells (ESCs) from various animal species in vitro remains challenging. This highlights the need for further optimization of culture conditions. Through compound library screening, we identified Danofloxacin, a small molecule, as a promoter of mouse ESC stemness. Danofloxacin enhances the formation of undifferentiated colonies and the generation of chimeric mice. Mechanistically, it suppresses HDAC1 expression and activity, resulting in increased acetylation of histone H3 at lysines 9 (H3K9ac) and 27 (H3K27ac), which in turn upregulates the expression of Tert and Prdm10. Functional assays revealed that overexpressing Tert or Prdm10 mimics Danofloxacin's effect on ESC self‐renewal, whereas knockdown of either gene disrupts this process. Moreover, the elevated levels of H3K9ac and H3K27ac activate the PI3K/AKT and HIF‐1α signaling pathways, both of which contribute to the transcriptional upregulation of Tert and Prdm10. These findings offer new insights into the pluripotency regulatory network and propose potential strategies for isolating ESCs from other species in vitro. This study identified Danofloxacin as a novel supporter of the maintenance of mouse embryonic stem cell (ESC) pluripotency. It functions partially by inhibiting HDAC1, which increases specific histone acetylation marks and activates key genes like Tert and Prdm10. This maintains the ESCs’ capacity for self‐renewal, genetic manipulation, and full differentiation, including the ability to contribute to chimeric embryos and germline transmission. Introduction The quest for optimal culture conditions for embryonic stem cells (ESCs) began in 1981 when Evans and Martin isolated mouse ESCs (mESCs) from 4.5‐day‐old embryos, demonstrating their diploid structure and differentiation potential [, ]. In 1984, Wobus made a significant breakthrough by introducing mitotically inactivated mouse embryonic fibroblasts (MEFs) for mESC culture, which enhanced stemness []. This method, which relies on inactivated MEFs and fetal bovine serum (FBS) [], supports the long‐term self‐renewal of mESCs in vitro. A key cytokine secreted by MEFs is leukemia inhibitory factor (LIF) [, ], which maintains mESC stemness primarily through activation of the JAK/STAT3 signaling pathway. Several STAT3 targets, including Tfcp2l1, Gbx2, and Klf4, have been identified as crucial to this process [, , ]. However, culturing mESCs with serum and feeder cells limits our ability to conduct in‐depth analyses of the mechanisms underlying stem cell maintenance. To overcome these limitations, researchers have increasingly turned to well‐defined small molecules to culture ESCs in vitro [, , ]. The use of these small molecules has enabled more efficient and cost‐effective research, advancing our understanding of stem cell biology. One notable discovery is the 2i combination—CHIR99021 (CHIR) and PD0325901 (PD). CHIR inhibits glycogen synthase kinase 3 beta (Gsk3β), while PD inhibits mitogen‐activated protein kinase kinase (MEK) []. CHIR activates the Wnt/β‐catenin pathway to reduce Tcf3 levels, a transcriptional repressor of pluripotency markers such as Nanog and Esrrb [, , ]. PD stabilizes the protein of Klf2 by inhibiting ERK, a direct substrate of MEK []. Other small molecules, such as GO6983 [], CID755673 [], and Spautin‐1 [], have also shown promise in promoting ESC maintenance. While ESCs have been successfully derived from various species, including mice, rats, pigs, chickens, humans, and monkeys, only mouse and rat ESCs have been shown to generate germline‐competent chimeric animals when injected into embryos [, , , , ], a capacity that remains largely unexplored in other species. Therefore, developing new culture conditions to maintain ESC self‐renewal and pluripotency in vitro is a critical challenge. In this study, using a Rex1‐GFP reporter mESC line, we identified Danofloxacin (DFA) from a small molecule library as a promoter of ESC pluripotency. DFA maintains green fluorescence in the Rex1‐GFP cell line and enhances the expression of pluripotency genes and the transcription of Tert and Prdm10. DFA treatment was accompanied by reduced HDAC1 expression and increased acetylation of histone H3 at lysines 9 (H3K9ac) and 27 (H3K27ac), suggesting a possible mechanistic link. This modification is enriched at the Tert and Prdm10 loci, promoting their transcription and thereby supporting ESC self‐renewal. Results Compound Library Screening Identifies Danofloxacin as a Self‐Renewal‐Promoting Factor To identify novel factors that promote self‐renewal in mESCs, we screened a chemical library of 1,383 compounds targeting various cellular components. Each compound was added individually to the medium of Rex1‐GFP mESCs in the absence of LIF. After 7 days, GFP‐positive cells were observed in many of the DFA‐treated mESCs (Figure ). To validate these results, 46C mESCs were cultured for 7 days in basal mESC medium supplemented with either DFA or LIF, with cells cultured in basal medium without LIF serving as the control. Alkaline phosphatase (AP) staining, a marker of mESC identity, was then performed. LIF‐ and DFA‐treated mESCs exhibited enhanced AP activity, whereas the control group showed significant differentiation (Figure ; Figure ). These findings indicate that DFA promotes mESC self‐renewal. To determine the optimal concentration, four concentrations (3 μm, 5 μm, 10 μm, and 15 μm) were tested. After 7 days, mESCs treated with 5 μm DFA exhibited the strongest AP activity compared to other concentrations (Figure ; Figure ), making 5 μm the concentration selected for subsequent experiments. However, the undifferentiated state of mESCs diminished after three passages, suggesting that DFA cannot sustain stemness in vitro over prolonged periods. To optimize DFA‐containing culture conditions, mESCs were cultured in basal medium supplemented with CHIR, PD, 2i, DFA, CHIR/DFA, PD/DFA, or 2i/DFA, with cells maintained in basal medium serving as the control. After two passages, we observed that the 2i‐containing conditions most effectively preserved the undifferentiated state of mESCs, as indicated by robust AP activity and sustained high expression of the self‐renewal marker Oct4 (Figure ; Figure ). Similar results were observed in mESCs treated with PD/DFA (PDF) or 2i/DFA (Figure ; Figure ). Western blot and RT‐qPCR analyses revealed increased expression of pluripotency genes, including Klf4, Nanog, Oct4, Sox2, Gbx2, Tfcp2l1, Klf2, and Klf5, under the PDF condition, compared to the PD‐only condition (Figure ; Figure ). To assess pluripotency under PDF conditions, cells were transferred to serum‐containing medium without PDF to form embryoid bodies (EBs). After 8 days, these cells were digested and plated on gelatin‐coated dishes for an additional 5 days. PDF‐treated mESCs retained their ability to differentiate into Tuj1+ neurons, Myosin+ cardiomyocytes, and Gata4+ primitive endoderm cells in vitro (Figure ). Cells cultured under PDF conditions could be stably maintained and passaged for 50 passages while retaining a normal karyotype (Figure ). Furthermore, immunostaining for phosphorylated histone H2AX revealed a reduced accumulation of DNA double‐strand breaks in both PD and PDF culture conditions (Figure ), indicating that PDF does not overtly increase genomic stress. Furthermore, an in vitro embryotoxicity assay was performed in which zygotes were cultured in G1‐Plus medium supplemented with DFA. At a concentration of 5 μm, DFA did not adversely affect blastocyst formation, with the efficiency of development to the blastocyst stage comparable to that of the control group cultured in G1‐Plus medium alone. In contrast, exposure to DFA at concentration of 15 μm resulted in varying degrees of developmental delay (Figure ). Blastocysts cultured under 5 μm DFA conditions were transferred into pseudopregnant ICR recipient females at 2.5 d.p.c. Embryos were harvested at E12.5 and subjected to hematoxylin and eosin (H&E) staining. Comparable to blastocysts cultured in G1‐Plus medium alone, blastocysts derived from DFA‐treated zygotes developed into morphologically normal E12.5 mouse embryos (Figure ). Furthermore, the PDF‐treated mESCs generated chimeric mice when injected into 3.5‐day post‐coitum blastocysts (Figure ; Figure ). These results demonstrate the potential of mESCs under PDF conditions for multi‐lineage differentiation. To explore the effect of DFA on ESCs from other species, we tested human induced pluripotent stem cells (hiPSCs). Different concentrations of DFA were added to the culture medium of hiPSCs in the absence of Activin A and bFGF. After 5 days, hiPSCs treated with 2.5 μm DFA showed stronger AP activity and higher expression levels of SOX2, OCT4, and NANOG compared to other groups (Figure ). However, after three passages, all cells had fully differentiated, indicating that DFA only promotes short‐term self‐renewal in hiPSCs under these conditions. Addition of Danofloxacin induces Tert and Prdm10 transcription To explore how DFA promotes self‐renewal, total RNA was extracted from mESCs treated with PD or PDF for various durations and subjected to high‐throughput sequencing. After 12 h of PDF treatment, we identified 161 upregulated and 207 downregulated genes (|log2FC| ≥ 1.000, p ≤ 0.05) relative to the PD control group. At 24 h, these numbers rose to 358 upregulated and 441 downregulated genes (Figure ). The Venn diagram illustrates the distribution of differentially expressed genes (DEGs) between the PD and PDF groups, revealing that 226 DEGs are shared between the two comparisons (Figure ). Among genes enriched in pathways related to mESC pluripotency, Tert, Prdm10, Sox15, and Trim43b were highlighted (Figure ). RT‐qPCR validation confirmed that Tert and Prdm10 expression was significantly increased following PDF treatment at both 12 and 24 h (Figure ). Meanwhile, we also examined the expression of TERT and PRDM10 in human iPSCs. The experimental results demonstrated that, following the addition of DFA, the expression levels of both TERT and PRDM10 were significantly upregulated, consistent with the patterns observed in mice (Figure ). Based on these findings, we focused further investigation on Tert and Prdm10. To assess the role of Tert and Prdm10 in mESC self‐renewal, we cultured cells without LIF and treated them with retinoic acid (RA) to induce monolayer differentiation. After four days, RT‐qPCR analysis revealed a decrease in the expression of pluripotency‐associated genes (Nanog, Oct4, Rex1, Sox2, and Klf2), while differentiation‐associated genes (Gata4, Gata6, and Cdx2) were upregulated in the differentiated cells compared to LIF‐treated mESCs (Figure ). Notably, the expression of Tert and Prdm10 was significantly reduced during differentiation (Figure ), suggesting their critical role in maintaining mESC stemness. Tert and Prdm10 Mediate the Effects of Danofloxacin on mESC Self‐Renewal To investigate whether Tert and Prdm10 can phenocopy the effects of DFA, we generated Flag‐tagged Tert‐ or Prdm10‐overexpressing mESCs using the PiggyBac vector system (F‐Tert and F‐Prdm10), which effectively enhanced the expression of Tert and Prdm10 (Figure ). An empty PiggyBac vector carrying Flag was used as a control (FLAG) (Figure ). These cell lines were seeded at low density in serum‐containing medium without LIF. After eight days, the Tert‐ and Prdm10‐overexpressing cells exhibited increased AP activity and higher levels of pluripotency markers (Sox2, Oct4, Klf4, Nanog, Klf2, Tfcp2l1, Tbx3, and Esrrb) compared to FLAG control cells, with lower expression of differentiation‐associated genes (Cdx2, Otx2, Gata4 and Gata6) (Figures ; Figure ). Furthermore, after culturing in PD‐containing medium for two passages, the Tert‐ and Prdm10‐overexpressing cells formed more AP‐positive colonies than FLAG control cells (Figure ). These results indicate that enforced expression of Tert and Prdm10 can promote the maintenance of the undifferentiated state of mESCs. To examine whether knockdown of Tert or Prdm10 impairs the function of DFA in mESCs, we designed two short hairpin RNAs (shRNAs) targeting Tert and Prdm10 transcripts (Tert sh#1, Tert sh#2, Prdm10 sh#1 and Prdm10 sh#2). A construct with a nonspecific shRNA sequence (Scrambled shRNA) was used as a control. After infecting 46C mESCs with shRNA lentiviral particles for one week, we confirmed the successful downregulation of Tert and Prdm10 by RT‐qPCR, and the cells were maintained in LIF/serum medium (Figure ). After two passages in DFA‐containing medium without LIF, Tert shRNA‐infected cells exhibited significantly reduced AP activity (Figure ; Figure ), while Prdm10 shRNA cells showed partial differentiation (Figure ; Figure ). Additionally, since Tert is a telomere‐associated gene influencing telomere lengthening [, ], we performed telomere Q‐FISH staining on cells cultured under PDF conditions and found that, compared with cells cultured only under PD conditions, cells under PDF conditions exhibited longer telomere lengths (Figure ). This is consistent with our observation of higher Tert expression under PDF conditions. Similarly, RT‐qPCR analysis revealed significantly longer telomeres (Tel) in Tert‐overexpressing cells compared to FLAG control cells (Figure ). Similar results were observed under DFA‐containing conditions (Figure ). Collectively, these findings suggest that the effect of DFA on mESC maintenance depends on Tert and Prdm10, with a close association to telomere maintenance. Danofloxacin Inhibits the Expression and Activity of Histone deacetylase 1 To investigate how DFA induces Tert and Prdm10 expression, we first predicted its molecular structure using SMILES notation (Figure ). The structure was input into the Swiss Target Prediction database, which identified 100 potential target proteins, 27% of which were enzymes (Figure ). Among these, histone deacetylase 1 (HDAC1) stood out. mESCs with stable HDAC1 knockdown exhibited partial self‐renewal after LIF withdrawal (Figure ; Figure ), suggesting HDAC1 could be a potential target of DFA. Next, molecular docking analysis was performed to assess the binding of HDAC1 and DFA, revealing multiple binding sites (Figure ). However, in vitro enzyme activity assays showed that, unlike trichostatin A (TSA), an HDAC inhibitor, DFA did not directly inhibit HDAC1 protein activity (Figure ). Despite this, the DFA‐treated group exhibited lower HDAC1 protein levels than the control group (Figure ; Figure ). RT‐qPCR analysis further confirmed that DFA suppressed HDAC1 transcription (Figure ). Additionally, increased H3K9ac and H3K27ac levels were detected following DFA treatment, resembling the pattern observed with TSA treatment (Figure ; Figure ). CUT&Tag experiments using specific antibodies for H3K9ac and H3K27ac demonstrated enhanced DNA fragment binding after DFA treatment (Figure ). To contextualize these findings, we employed TSA as a functional reference to observe the kinetic accumulation of these acetylated marks. Similar to the robust increase induced by the reference inhibitor over a time course of 8 to 48 h, DFA administration effectively mimicked this hyperacetylation effect (Figure ). This suggests that while DFA does not directly bind the HDAC1 catalytic site, it effectively establishes a repressive chromatin state at the HDAC1 gene itself, ultimately leading to the sustained global elevation of H3K9ac and H3K27ac levels required for mESC self‐renewal (Figure ). To further investigate the epigenetic regulatory mechanism of DFA, we performed Chromatin Immunoprecipitation sequencing (ChIP‐seq) to profile various histone modifications across the HDAC1 gene locus. Our results revealed that DFA treatment (PDF group) led to a marked increase in the enrichment of H3K9me3 and H3K27me3, two hallmark repressive histone marks specifically at the HDAC1 promoter region. In contrast, the enrichment levels of H3K4me3, a modification typically associated with active transcription, showed no significant changes upon DFA administration (Figure ). These findings suggest that DFA specifically modulates inhibitory histone methylation to establish a more compact, transcriptionally repressive chromatin state at the HDAC1 promoter. We performed a dual‐luciferase reporter assay (Figure ). Our results demonstrated that DFA treatment (PDF group) significantly inhibited the luciferase activity driven by the HDAC1 promoter compared to the control group (Figure ). Together, these results suggest that DFA exerts part of its effects in association with changes in HDAC1 activity and levels of H3K9ac and H3K27ac. Induction of Histone Acetylation by Danofloxacin Enables Transcription of Tert and Prdm10 HDAC1 protein expression declined in a time‐dependent manner (Figure ; Figure ). Notably, at the 12‐hour time point, increased expression of Tert and Prdm10 was observed in the presence of PDF (Figure ), pointing to a possible relationship between these changes. To validate this relationship, we performed HDAC1 knockdown experiments in mESCs using a lentiviral system. Knockdown of HDAC1 (HDAC1 sh#1 and HDAC1 sh#2) resulted in increased expression of Tert and Prdm10 compared to scrambled shRNA control cells (Figure ; Figure ), further supporting a possible association between HDAC1 levels and the expression of Tert and Prdm10. Next, we analyzed the distribution of H3K9ac and H3K27ac modifications at the transcription start sites (TSS) of Tert and Prdm10. Following DFA treatment, there was an increase in the presence of H3K9ac and H3K27ac on the gpromoters of both genes (Figure ). To further explore this, we conducted ChIP experiments with antibodies against H3K9ac and H3K27ac in mESCs. RT‐qPCR analysis revealed that DFA treatment significantly increases H3K9ac and H3K27ac enrichment at the Tert and Prdm10 gene loci (Figure ). Together, these findings suggest that the effect of DFA is linked to reduced HDAC1 activity, accompanied by increased histone acetylation and transcriptional upregulation of Tert and Prdm10. PI3K/AKT and HIF‐1α Signaling Pathways Contribute to Histone Acetylation‐Mediated Transcription of Tert and Prdm10 KEGG analysis of CUT&Tag sequencing data for H3K9ac and H3K27ac enrichment in mESCs cultured under DFA‐containing conditions revealed significant activation of the PI3K/AKT and hypoxia‐inducible factor‐1α (HIF‐1α) signaling pathways (Figure ). To assess the involvement of these pathways in the expression of Tert and Prdm10, we treated cells with LY294002, a specific inhibitor of PI3K, and observed a marked reduction in the expression of both genes following AKT pathway inhibition (Figure ). In agreement with these results, both DFA and TSA treatment increased AKT phosphorylation (Figure ; Figure ), suggesting that the enhancing effect of DFA on Tert and Prdm10 expression may, at least in part, be associated with activation of the PI3K/AKT signaling pathway. Furthermore, to evaluate the role of HIF‐1α, we downregulated its expression using shRNA lentiviruses (Figure ). This was accompanied by a significant reduction in Tert and Prdm10 expression (Figure ), suggesting that HIF‐1α may also be involved in the gene activation observed following DFA treatment. In conclusion, our data indicate that DFA‐induced changes in H3K9ac and H3K27ac were accompanied by alterations in the PI3K/AKT and HIF‐1α signaling pathways, along with increased expression of Tert and Prdm10. Discussion In this study, we established a novel culture condition by introducing a small molecule DFA to promote the maintenance of mouse and human ESC pluripotency. Although DFA is primarily recognized as a fluoroquinolone antibiotic [], its mechanism of action in ESCs remains poorly understood. To address this, we employed high‐throughput sequencing and molecular cell biology experiments, which revealed that DFA treatment was associated with the upregulation of two key genes involved in mESC fate decision, Tert and Prdm10 (Figure ). Telomerase, consisting of the catalytic subunit Tert and the RNA component Terc, is crucial for telomere repair and elongation [, , ]. Terc is widely expressed, while Tert expression is tightly regulated. In ESCs, Tert is highly expressed, whereas in somatic cells, both Tert expression and telomerase activity are often low or undetectable []. Telomeres are specialized structures at the ends of chromosomes, and the absence of functional telomeres significantly impairs pluripotency in iPSCs, preventing teratoma formation []. mESCs without Tert exhibit abnormal differentiation [], due to the global changes in chromatin accessibility and gene expression []. Similarly, Tert knockout in human ESCs reduces their proliferative capacity []. Notably, Tert expression in human ESCs is synergistically activated by transcription factors Sp1 and c‐Myc, which are closely linked to stem cell fate commitment []. These findings underscore the importance of Tert in regulating ESC fate decisions and maintaining pluripotency. In our study, we found that DFA treatment was accompanied by upregulation of Tert, a change that contributes to the maintenance of mESC stemness (Figure ), with Tert expression gradually decreasing as cells differentiate (Figure ). Unlike Tert, the molecular and functional properties of Prdm10—also known as tristanin and a member of the same phylogenetic subfamily as Prdm15—are not as well characterized []. Prdm10 contains an N‐terminal PR domain, ten C2H2 zinc fingers, and a C‐terminal glutamine‐rich transactivation domain. It is expressed in various embryonic and adult tissues []. Homozygous deletion of Prdm10 in mice results in embryonic lethality [], with developmental arrest occurring before blastocyst formation []. Similarly, acute deletion of Prdm10 in mESCs leads to reduced growth and increased apoptosis []. Notably, the effects of Prdm10 deficiency are partly mediated through EIF3B‐dependent global translation regulation []. Moreover, Prdm10 influences Bcl‐2 expression: depletion of Prdm10 reduces Bcl‐2 levels, while overexpression of Prdm10 promotes Bcl‐2 expression []. Luciferase reporter assays have shown that Prdm10 binds to the Bcl‐2 promoter in vivo []. In our study, we observed that Prdm10 is activated alongside Tert upon DFA treatment (Figure ), highlighting its critical role in maintaining the undifferentiated state of mESCs. HDAC1 appears to play a role in DFA‐mediated transcription of Tert and Prdm10. DFA suppresses the expression of HDAC1 (Figure ), and this regulatory effect is associated with the increased H3K9me3 and H3K27me3 levels at the HDAC1 promoter (Figure ). HDAC1 is a member of class I histone deacetylases, which also include HDAC2, HDAC3, and HDAC8 []. These enzymes negatively regulate histone acetylation by removing acetyl groups from histone tails []. In mESCs, HDAC1 binds to the regulatory regions of numerous genes, many of which are upregulated upon HDAC1 depletion, including key regulators of ESC self‐renewal and differentiation []. Notably, the effects of complete HDAC1 knockout differ from partial inhibition of its activity in mESCs. HDAC1 is essential for early embryonic development, as loss of it leads to early embryonic lethality, but ESCs can maintain pluripotency in the absence of HDAC1 or its family member HDAC2 [], although the expression of pluripotency genes such as Oct4, Nanog, Sox2, and Rex1 is reduced in these cells [, ]. Additionally, ESCs deficient in HDAC1 exhibit smaller embryoid bodies with abnormal differentiation patterns, including spontaneous rhythmic contraction and increased expression of cardiomyocyte and neuronal markers []. Interestingly, inhibition of HDAC activity using specific inhibitors such as TSA, valproic acid (VPA), or butyrate promotes ESC self‐renewal in various species and enhances the reprogramming efficiency of iPSCs [, , , , , ]. Consistent with these findings, our study showed that DFA treatment was accompanied by increased H3K9ac and H3K27ac levels, along with elevated transcription of Tert and Prdm10 (Figure ). Notably, HDAC1 suppression represents one of the primary mechanisms through which DFA exerts its effects, but it is not the sole biochemical property of this compound. Genome‐wide sequencing using CUT&Tag revealed that DFA‐induced histone acetylation can stimulate the PI3K/AKT and HIF‐1α signaling pathways (Figure ). This is in line with previous reports demonstrating that the PI3K/AKT pathway can upregulate Tert expression in mice []. DFA may activate the PI3K/AKT pathway by regulating upstream regulatory genes of this pathway, and this is further corroborated by the observation that HDAC1 silencing promotes AKT phosphorylation in mouse myotubes, human mesenchymal stem cells, and acute myocardial ischemia and reperfusion injury [, , ]. Additionally, activation of the PI3K/AKT signaling pathway has been shown to maintain self‐renewal and support multipotent differentiation in both human and mouse pluripotent stem cells [, ]. Moreover, HIF‐1α, a master transcription factor regulating gene expression in stem cell metabolism and behavior, has been implicated in Tert regulation [, ]. Overexpression of exogenous HIF‐1α cDNA in murine ESCs promotes cardiogenesis []. HIF‐1α also plays a critical role in the metabolic and functional transition of ESCs to the epiblast stem cell‐like stage []. In addition, upregulation of HIF‐1α inhibits mESC self‐renewal by negatively regulating the LIF/Stat3 pathway through direct binding to the reverse hypoxia‐responsive element located in the LIFR promoter []. In contrast, suppression of HIF‐1α facilitates the maintenance of ESCs in their undifferentiated state under hypoxia in vitro []. We found that DFA treatment was accompanied by increased H3K9ac and H3K27ac levels in mESCs, along with changes in the PI3K/AKT and HIF‐1α pathways and increased expression of Tert and Prdm10. Notably, several key aspects remain unresolved in this study. First, the upstream molecular events by which DFA represses HDAC1 transcription remain undefined; complementary approaches, including CRISPR‐based screening and chromatin conformation assays, will be valuable for identifying the direct effectors involved. Second, whether increased histone acetylation directly activates the PI3K/AKT and HIF‐1α pathways or represents a parallel downstream event remains unclear. Additionally, the mechanisms by which AKT and HIF‐1α activate Tert and Prdm10 remain uncertain. Finally, the potential for reciprocal regulation or synergistic interactions between Tert and Prdm10 in DFA‐induced pluripotency maintenance requires further investigation. Conclusion In summary, this study identifies DFA as a self‐renewal‐promoting factor and elucidates its underlying mechanism, thereby providing an alternative strategy for stem cell culture. DFA treatment was associated with reduced HDAC1 expression and increased acetylation of H3K9 and H3K27. These changes were accompanied by elevated expression of Tert and Prdm10, possibly in association with histone acetylation‐related recruitment of RNA polymerase II and changes in the PI3K/AKT and HIF‐1α signaling pathways (Figure ). Together, these findings expand the current understanding of the stem cell regulatory network and offer a useful framework for developing strategies to maintain ESC pluripotency. Future comprehensive genome‐wide analyses, such as whole‐genome sequencing, will further strengthen the safety evaluation for translational applications. Experimental Section Cell Culture 46C mESCs, kindly provided by Qi‐Long Ying (University of Southern California, USA), were cultured in 0.1% gelatin‐coated dishes at 37°C in 5% carbon dioxide. The medium used for routine maintenance was DMEM (C3113‐0500, Vivacell, China) supplemented with 10% fetal bovine serum (FND500, ExCell Bio, Australia), 1×MEM nonessential amino acids (11140‐050, Gibco, USA), 2 mm GlutaMax (21051024, Gibco, USA) and 0.1 mm β‐mercaptoethanol (M3148, Sigma). Additionally, 1000 U/mL LIF (LIF1010, Millipore, USA) was added to sustain mESC stemness. The basal medium used to culture hiPSCs was N2B27 supplemented with 10% KnockOut Serum Replacement (KSR) (10828, Life Technologies). N2B27 consisted of the following components: Neurobasal medium (21103‐049, Life Technologies) and DMEM/F12 medium (11330‐032, Life Technologies) mixed at a 1:1 ratio; 1×non‐essential amino acids solution, 2 mm GlutaMAX, 0.1 mm β‐mercaptoethanol, 1×N2 (17502‐048, Life Technologies); and 1× B27 (17504‐044, Life Technologies). All cell lines used in this study were free from mycoplasma contamination for all experiments. Plasmid Construction The coding regions of the Tert and Prdm10 genes were inserted into PiggyBac transposon vectors carrying a Flag tag. The primers for gene cloning are listed in Table . shRNA‐expressing plasmids were constructed according to the protocol for the pLKO.1‐TRC vector (Addgene plasmid #10878). Targeted specific shRNA sequences for Tert, Prdm10, HDAC1, and HIF‐1α are listed in Table . AP activity Assay Cells were fixed in 4% paraformaldehyde for 2 min at room temperature, followed by gentle washing with PBS three times. AP staining reagent (C3206, Beyotime Biotechnology, China) was prepared in the dark, and cells were incubated with the reagent at room temperature for 30 min. After incubation, cells were gently washed with PBS three times, and finally, the cells were visualized under a Leica DMI8 microscope. Reverse Transcription–Quantitative PCR (RT‐qPCR) Total RNA was extracted using the Cell/Tissue Total RNA Kit (19221ES50, YEASEN). 1 μg of total RNA was used to synthesize cDNA according to the instructions of HiScript III All‐in‐one RT SuperMix Perfect for qPCR (R333‐01, Vazyme). RT‐qPCR was performed using Hieff qPCR SYBR Green Master Mix (11201ES03, YEASEN) to prepare the PCR reaction mixture, and experiments were conducted on a PikoReal real‐time PCR machine (Thermo Scientific, USA). The relative expression levels were calculated using the 2‐ΔΔCt method, with β‐actin expression used as the internal reference for normalization. The primers used in the experiment are listed in Table . Western Blot Cells were lysed using ice‐cold RIPA lysis buffer (P0013B, Beyotime Biotechnology, China) containing a protease inhibitor cocktail. The proteins were separated by electrophoresis on 10% SDS‐PAGE gels and subsequently transferred onto PVDF membranes. For antibody detection, specific primary antibodies were incubated overnight at 4°C, followed by incubation with horseradish peroxidase‐conjugated secondary antibodies at 25°C for 1 h. The primary antibodies used included Flag (F9291, Sigma,1:1000), β‐tubulin (250007, ZEN BIO,1:1000), Oct4 (11263‐1‐AP, Proteintech,1:1000), Klf4 (381633, ZEN BIO,1:1000), Sox2 (66411‐1‐Ig, Proteintech,1:1000), Nanog (14295‐1‐AP, Proteintech,1:1000), HDAC1 (10197‐1‐AP, Proteintech,1:1000), H3K9ac (ab4441, Abcam,1:1000), H3K27ac (ab4729, Abcam,1:1000), AKT (10176‐2‐AP, Proteintech,1:1000) and Pho‐AKT (66444‐1‐Ig, Proteintech,1:1000). Immunofluorescence Staining Immunostaining was performed according to the standard protocol. First, cells were fixed in 4% paraformaldehyde for 20 min, gently washed three times with PBS, and then incubated in blocking buffer (PBS containing 5% BSA and 0.2% Triton X‐100) at 37°C for 1 h. Subsequently, the cells were incubated overnight at 4°C with primary antibodies against Oct4 (11263‐1‐AP, Proteintech, 1:100), Sox2 (66411‐1‐Ig, Proteintech, 1:100), Tuj1 (R&D Systems, MAB1195, 1:100), Gata4 (Cell Signaling Technology, 36966S, 1:100), Myosin (Abcam, ab50967, 1:100), H3K9ac (ab4441, Abcam,1:200) and H3K27ac (ab4729, Abcam,1:200). After being gently washed three times with PBS, the cells were incubated with secondary antibody. The nuclei were stained with Hoechst 33342 (H3570, Invitrogen; 1:10,000). CUT&Tag The CUT&Tag experiment was performed according to the protocol of the Hyperactive Universal CUT&Tag Assay Kit for Illumina (Vazyme). Fresh and viable cells, approximately 50,000, were first collected and centrifuged at 300 × g for 2 min. The supernatant was discarded, and the cells were washed with pre‐prepared Wash Buffer, followed by another centrifugation at 300 × g for 2 min. The cells were then resuspended in Wash Buffer, and 10 μL of pre‐activated ConA beads were added to each sample. The mixture was incubated at room temperature for 10 min to form a cell‐bead complex. After discarding the supernatant, the cells were resuspended in 50 μL antibody buffer and incubated with 1 μg of primary antibodies (Anti‐H3K9ac and Anti‐H3K27ac). The mixture was incubated overnight at 4°C with gentle rotation. The next day, after centrifugation and removal of the supernatant, 50 μL of Dig‐wash buffer and 0.5‐1 μg of secondary antibody were added. The mixture was rotated at room temperature for 1 h, followed by three gentle washes with Dig‐wash buffer. Then, 2 μL of PA/G–Tnp and 98 μL of Dig‐300 buffer were added, and the mixture was incubated at room temperature for 1 h, followed by another three washes with Dig wash buffer. After discarding the supernatant, 40 μL of Dig‐300 buffer and 10 μL of 5 × TTBL were mixed, and the mixture was gently resuspended and incubated at 37°C for 1 h. DNA was extracted from the fragmented DNA using DNA Extract Beads and amplified through the appropriate cycles. The resulting library was purified using 2 × AMPure XP (Beckman Coulter). The library concentration was determined according to the Qubit dsDNA HS Kit (Invitrogen) protocol. The final CUT&Tag libraries were subjected to paired‐end sequencing on the NovaSeq 6000 platform (Illumina) according to the manufacturer's instructions. Embryoid Bodies (EBs) Spontaneous differentiation was induced by forming EBs through an 8‐day suspension culture on non‐adherent dishes. The EBs were then transferred to gelatin‐coated plates and cultured for another 5 days in mESC medium without LIF to promote further differentiation. ULI‐NChIP‐seq ULI‐NChIP‐seq was performed on 200–500 cells per sample with three biological replicates. Protein A beads (10 μL) were incubated with 1 μg of H3K4me3, H3K9me3, or H3K27me3 antibody at 4°C for at least 2 h. Cells were lysed in Nuclei Extraction Buffer and treated with MNase Master Mix at 25°C for 8 min. After stopping the reaction with EDTA and Nuclear Break Buffer, samples were brought to 180 μL with ChIP buffer, with 1/50 reserved as input. The remaining lysate was incubated with antibody‐coupled beads at 4°C for at least 4 h, then washed twice each with Low Salt and High Salt Wash Buffers. DNA was eluted at 65°C, purified by phenol‐chloroform extraction and isopropanol precipitation, then processed using the KAPA Hyper Prep kit. Libraries were sequenced on Illumina NovaSeq with 150‐bp paired‐end reads. ChIP‐seq Data Analysis ChIP‐seq data were analyzed following standard bioinformatics pipelines. Sequencing reads were trimmed using Trim Galore (v0.6.10) and aligned to the mouse reference genome (mm10) using Bowtie2 (v2.4.4). High‐confidence alignments (mapping_quality ≥ 20) were retained using Sambamba (v1.0.1), and PCR duplicates were removed using Picard (v3.1.0). Blacklist regions were filtered to eliminate artifacts. Normalized coverage tracks (bigWig) were generated using bamCoverage from deepTools (v3.5.4) with RPKM normalization. Background correction was performed by subtracting input signals using bigwigCompare. Peak calling was conducted using MACS2 (v2.2.7.1). Signal intensities were calculated using computeMatrix and visualized with plotHeatmap and plotProfile. Pearson correlation and PCA analyses assessed replicate consistency. ChIP‐seq signals were visualized using IGV (v2.16.0). Quantitative Fluorescence In Situ Hybridization (Q‐FISH) Q‐FISH experiments were performed using Cy3‐labeled PNA probes (0.7 μg/mL, Panagene) for telomere detection. Cells were first cultured in the presence of demecolcine (Sigma) for 4 h, followed by hypotonic swelling in 0.03 m sodium citrate at 37°C for 5 min. After two fixation changes, the cell suspension was dropped onto clean, cold microscopy slides and baked overnight at 65°C for 8–12 h. The slides were then soaked in PBS for 15 min, fixed in 4% formaldehyde for 2 min, and washed three times with PBS (5 min each). 15 μL of PNA probe was added to the slides and covered with a coverslip, then hybridized at 85°C for 10–15 min, followed by incubation at 18–25°C for 3 h. The coverslip was carefully removed with forceps, and the slides were washed in staining wash solution 1 (containing deionized formamide, BSA, and Tween 20) for 20 min, then in staining wash solution 2 (containing 5% dextrose) for 2 min. Finally, slides were counterstained with DAPI solution diluted in PBS at 18–25°C in the dark for 25 min. Images were acquired using an LSM 510 Meta confocal microscope equipped with a 63×/1.4 Oil DIC objective. Dual‐Luciferase Reporter Assay To assess the regulatory effect of DFA on HDAC1 promoter activity, the promoter sequence of mouse HDAC1 (spanning from −2000 to +1 bp) was cloned into the pGL6 reporter vector to generate the pGL6‐HDAC1 construct. 46C mESCs were co‐transfected with the pGL6‐HDAC1 plasmid and a Renilla luciferase internal control plasmid using Lipofectamine 3000. At 24 h post‐transfection, the culture medium was replaced with PDF medium, while cells maintained in PD medium served as the control group. Following an additional 24 h of incubation, luciferase activities were measured using the TransDetect Dual‐Luciferase Reporter Assay Kit (FR201, TransGen Biotech). The activity of the Firefly luciferase was normalized to that of the Renilla luciferase to determine the relative promoter activity. All assays were performed in at least three independent biological replicates to ensure statistical significance. Chimera Generation The microinjection technique was outlined in our previous report []. In brief, 8 to 10 single cells were injected into 3.5 dpc blastocysts obtained from ICR mice. The injected blastocysts were then transplanted into uterine horns of 2.5 d.p.c. pseudopregnant female mice. Chimeric offspring were subsequently identified based on coat color. For the generation of E12.5 chimeric embryos, two‐cell stage embryos were flushed from the oviducts of pregnant ICR mice using HCZB medium and transferred to G1‐plus medium covered with mineral oil, then cultured in vitro to the blastocyst stage. 8–10 GFP+ cells cultured in PDF condition were injected into each blastocyst‐stage embryo. After injection, the embryos were recovered in G1‐plus medium for 1–2 h, then 10–15 chimeric embryos were transferred into surrogate mice and dissected at the E12.5 developmental stage. H&E Staining for E12.5 Mouse Embryos E12.5 mouse embryos were dissected and immediately fixed in 4% PFA at 4°C overnight, washed three times with PBS, then dehydrated through a graded ethanol series (70%‐100%), 1 h per concentration. Samples were cleared in xylene three times (30 min each), infiltrated with xylene:paraffin (1:1) mixture and pure paraffin (58–60°C, 1 h per step), then embedded in paraffin blocks. Paraffin blocks were sectioned at 4–7 μm thickness, floated on a 40–45°C water bath, mounted onto slides, and baked at 65°C for 2 h. Sections were deparaffinized in xylene (twice, 10–15 min each), rehydrated through graded ethanol (100%–70%) to distilled water. Sections were stained with hematoxylin for 3 min, differentiated in 1% acid alcohol for 3–5 s, and blued in running water for 5–10 min. Sections were dehydrated in 95% ethanol for 1 min, then stained with eosin for 15 s to 1 min. Following staining, sections were dehydrated through graded ethanol (70%‐100%), cleared in xylene (twice, 5 min each), and mounted with neutral resin. Images were captured under a microscope. Staining showed nuclei in blue‐purple and cytoplasm in pink to red. Gentle handling is required as E12.5 embryonic tissue is small and soft. Ethical Statement All animal experiments involving the use of animals were approved by the Institutional Animal Care and Use Committee of Anhui University (No. IACUC(AHU)‐2024‐053). Accession Number Our RNA‐Seq dataset has been deposited in GEO under accession number GSE286230. Statistical analysis The number of biological replicates is indicated in each legend. All data are reported as the mean ± SD. The data were visualized with GraphPad Prism 8. Unpaired, two‐tailed Student's t‐tests were employed for comparisons between two groups. One‐way ANOVA or two‐way ANOVA was utilized for comparisons among multiple groups. A p‐value of less than 0.05 was considered statistically significant. Contributor Information Wenqiang Liu, Email: liuwenqiang@tongji.edu.cn. Min Zhou, Email: dminzhou@ustc.edu.cn. Shou‐Dong Ye, Email: shdye@126.com.