Journal: Journal of Cell Science

Article Title: Interplay between nuclear survivin and the PRC2 complex and its impact on H3K27me3-directed transcriptional repression

doi: 10.1242/jcs.264572

Figure Lengend Snippet: Hypoxia increases expression of EZH2, H3K27me3 and survivin. (A) Immunoblots of WCEs from U2OS, HeLa and MRC5 lines cultured under normoxic or hypoxic environments (24 h). Blots were immunoprobed with anti-EZH2, anti-H3K27me3 and anti-survivin antibodies. Anti-Hif1a used to prove the hypoxic state had been induced, and anti-tubulin was used as a loading control. (B–D) Quantification of immunoblots represented in A from three independent experiments demonstrating that EZH2, H3K27me3 and survivin are all more abundant under hypoxia. Data presented are means±s.d. * P <0.05, ** P <0.01, *** P <0.001 (two-way ANOVA with Tukey's multiple comparisons post test).

Article Snippet: Primary antibodies were diluted 1:1000 in TBST with 5% milk, unless otherwise stated, and were against: tubulin (Sigma, B512, T5168), β-actin (Invitrogen MA1-140), TBP (CST, 8515), survivin (C60, CST 71G4B7, TBST 2% milk; or 6E4), H3K27me3 (Abcam, ab192985; TBST-2% BSA), GST (Cytivia, RPN1236V), EZH2 (CST, D269 or Proteintech 21800-1-AP), Hif1α (Novus Biologics, NB100-449).

Techniques: Expressing, Western Blot, Cell Culture, Control

Journal: Journal of Cell Science

Article Title: Interplay between nuclear survivin and the PRC2 complex and its impact on H3K27me3-directed transcriptional repression

doi: 10.1242/jcs.264572

Figure Lengend Snippet: Survivin and EZH2 interact. (A) Immunoprecipitation was carried out using whole MRC5 extracts using anti-survivin (C60), anti-EZH2, mouse IgG antibodies (negative control). Co-immunoprecipitation was assessed with the alternative antibodies. Co-immunoprecipitation of EZH2 with survivin was evident when anti-EZH2 was used to immunoprecipitate but not when the anti-survivin (C60) antibody was used. (B) GST pulldown assay was carried out with WCEs prepared from RPE cells expressing GST (negative control), GST–survivin and various GST-tagged survivin truncations, (numbering indicating amino acids), used as bait. (C) Quantification of interactions represented in B. EZH2 binds mainly to the first 90 amino acids of survivin. Data are mean±s.d. from three independent experiments. *** P <0.001; **** P <0.0001; ns, not significant (one-way ANOVA with Dunnett's post hoc test). (D) Immunoprecipitation was carried out as in A but using anti-H3K27me3 specific antibodies, rather than anti-EZH2. Co-immunoprecipitation of survivin and H3K27me3 was evident in reciprocal samples. (E) The GST pulldown experiment as in B was repeated using RPE cell lysates with GST or GST–survivin, and interaction with H3K27me3 determined by immunoblotting. (F) Quantification of data represented in E, normalised to the GST or GST–survivin. Data are mean±s.d., n =3. *** P <0.001 (unpaired two-tailed Student's t -test). Blots in A and D are representative of three repeats. Inputs are 7.5%.

Article Snippet: Primary antibodies were diluted 1:1000 in TBST with 5% milk, unless otherwise stated, and were against: tubulin (Sigma, B512, T5168), β-actin (Invitrogen MA1-140), TBP (CST, 8515), survivin (C60, CST 71G4B7, TBST 2% milk; or 6E4), H3K27me3 (Abcam, ab192985; TBST-2% BSA), GST (Cytivia, RPN1236V), EZH2 (CST, D269 or Proteintech 21800-1-AP), Hif1α (Novus Biologics, NB100-449).

Techniques: Immunoprecipitation, Negative Control, GST Pulldown Assay, Expressing, Western Blot, Two Tailed Test

Journal: Journal of Cell Science

Article Title: Interplay between nuclear survivin and the PRC2 complex and its impact on H3K27me3-directed transcriptional repression

doi: 10.1242/jcs.264572

Figure Lengend Snippet: Survivin knockdown increases H3K27me3 abundance. (A) U2OS and MRC5 cells were incubated with control or survivin-specific siRNA for 48 h. Lysates were immunoblotted with antibodies against the indicated proteins. (B,C) Quantitative analysis of immunoblots, normalised to β-actin loading for (B) U2OS, and (C) MRC5 cells. No change was seen in EZH2 expression but H3K27me3 was increased in both lines. Data are means±s.d. from three independent experiments. * P <0.05; ** P <0.01; *** P <0.001, ns, not significant (two-way ANOVA with Tukey's multiple comparisons post test).

Article Snippet: Primary antibodies were diluted 1:1000 in TBST with 5% milk, unless otherwise stated, and were against: tubulin (Sigma, B512, T5168), β-actin (Invitrogen MA1-140), TBP (CST, 8515), survivin (C60, CST 71G4B7, TBST 2% milk; or 6E4), H3K27me3 (Abcam, ab192985; TBST-2% BSA), GST (Cytivia, RPN1236V), EZH2 (CST, D269 or Proteintech 21800-1-AP), Hif1α (Novus Biologics, NB100-449).

Techniques: Knockdown, Incubation, Control, Western Blot, Expressing

Journal: Journal of Cell Science

Article Title: Interplay between nuclear survivin and the PRC2 complex and its impact on H3K27me3-directed transcriptional repression

doi: 10.1242/jcs.264572

Figure Lengend Snippet: Survivin and EZH2 in ihPSCs. (A) Three pluripotent stem cell lines, CGT-RCIB 10, ReBL Pat and iAT1 were grown in normoxia and immunostained for EZH2 (red), endogenous survivin (green), and counterstained with NucBlue to show the nucleus (blue). Scale bars: 50 µm. (B) There is colocalisation of EZH2 and survivin in the nuclei as shown by the intensity profiles along the yellow line in A (FIJI software). Results representative of N =3 independent repeats. (C) CGT-RCIB 10 cells were incubated with control or survivin-specific siRNA for 24 h. Lysates were immunoblotted with antibodies against the indicated proteins. (D) Quantitative analysis of bands in immunoblots in C, normalised to the β-actin loading control. Although survivin was only partially knocked down, H3K27me3 abundance increased significantly. Data are normalized to control siRNA treatment and are means±s.d. from n =3 plotted. * P <0.05; ** P <0.01; ns, not significant (two-way ANOVA with Tukey's multiple comparisons post test). (E) qPCR analysis was carried out for the genes indicated from CGT-RCIB 10 cells treated with either control or survivin-specific siRNA (24 h). Data are normalized to control siRNA treatment and means±s.d. from N =3 plotted. * P <0.05; ** P <0.01; *** P <0.001; ns, not significant (two-way ANOVA with Tukey's multiple comparisons post test). (F) qPCR analysis of major satellite transcripts from CGT-RCIB 10 cells exposed to control or survivin-specific siRNA. A significant reduction in major satellite expression occurred in the absence of survivin. Data are normalized to control siRNA treatment and means±s.d. from n =3 plotted. ** P <0.01 (unpaired two-tailed Student's t -test).

Article Snippet: Primary antibodies were diluted 1:1000 in TBST with 5% milk, unless otherwise stated, and were against: tubulin (Sigma, B512, T5168), β-actin (Invitrogen MA1-140), TBP (CST, 8515), survivin (C60, CST 71G4B7, TBST 2% milk; or 6E4), H3K27me3 (Abcam, ab192985; TBST-2% BSA), GST (Cytivia, RPN1236V), EZH2 (CST, D269 or Proteintech 21800-1-AP), Hif1α (Novus Biologics, NB100-449).

Techniques: Software, Incubation, Control, Western Blot, Expressing, Two Tailed Test

Journal: Journal of Cell Science

Article Title: Interplay between nuclear survivin and the PRC2 complex and its impact on H3K27me3-directed transcriptional repression

doi: 10.1242/jcs.264572

Figure Lengend Snippet: Hypoxia increases expression of EZH2, H3K27me3 and survivin. (A) Immunoblots of WCEs from U2OS, HeLa and MRC5 lines cultured under normoxic or hypoxic environments (24 h). Blots were immunoprobed with anti-EZH2, anti-H3K27me3 and anti-survivin antibodies. Anti-Hif1a used to prove the hypoxic state had been induced, and anti-tubulin was used as a loading control. (B–D) Quantification of immunoblots represented in A from three independent experiments demonstrating that EZH2, H3K27me3 and survivin are all more abundant under hypoxia. Data presented are means±s.d. * P <0.05, ** P <0.01, *** P <0.001 (two-way ANOVA with Tukey's multiple comparisons post test).

Article Snippet: The following primary antibodies were used at 1:200: anti-survivin (Cell Signaling Technologies 71G4B7 or 6E4), anti-H3K27Me3 (Abcam, Ab192985 ), anti-EZH2 [Cell Signalling Technologies (CST), D269].

Techniques: Expressing, Western Blot, Cell Culture, Control

Journal: Journal of Cell Science

Article Title: Interplay between nuclear survivin and the PRC2 complex and its impact on H3K27me3-directed transcriptional repression

doi: 10.1242/jcs.264572

Figure Lengend Snippet: Survivin and EZH2 interact. (A) Immunoprecipitation was carried out using whole MRC5 extracts using anti-survivin (C60), anti-EZH2, mouse IgG antibodies (negative control). Co-immunoprecipitation was assessed with the alternative antibodies. Co-immunoprecipitation of EZH2 with survivin was evident when anti-EZH2 was used to immunoprecipitate but not when the anti-survivin (C60) antibody was used. (B) GST pulldown assay was carried out with WCEs prepared from RPE cells expressing GST (negative control), GST–survivin and various GST-tagged survivin truncations, (numbering indicating amino acids), used as bait. (C) Quantification of interactions represented in B. EZH2 binds mainly to the first 90 amino acids of survivin. Data are mean±s.d. from three independent experiments. *** P <0.001; **** P <0.0001; ns, not significant (one-way ANOVA with Dunnett's post hoc test). (D) Immunoprecipitation was carried out as in A but using anti-H3K27me3 specific antibodies, rather than anti-EZH2. Co-immunoprecipitation of survivin and H3K27me3 was evident in reciprocal samples. (E) The GST pulldown experiment as in B was repeated using RPE cell lysates with GST or GST–survivin, and interaction with H3K27me3 determined by immunoblotting. (F) Quantification of data represented in E, normalised to the GST or GST–survivin. Data are mean±s.d., n =3. *** P <0.001 (unpaired two-tailed Student's t -test). Blots in A and D are representative of three repeats. Inputs are 7.5%.

Article Snippet: The following primary antibodies were used at 1:200: anti-survivin (Cell Signaling Technologies 71G4B7 or 6E4), anti-H3K27Me3 (Abcam, Ab192985 ), anti-EZH2 [Cell Signalling Technologies (CST), D269].

Techniques: Immunoprecipitation, Negative Control, GST Pulldown Assay, Expressing, Western Blot, Two Tailed Test

Journal: Neoplasia (New York, N.Y.)

Article Title: BPGM as an intrinsic brake to constrain metastasis through phospho-epigenetic-mediated carnitine biosynthesis suppression

doi: 10.1016/j.neo.2026.101299

Figure Lengend Snippet: 2,3-BPG-CDK1-EZH2-H3K27me3 Axis: BPGM’s epigenetic circuit breaker for cellular migration. (A) Integrated functional metabolomics analysis revealed BPGM-altered metabolites clustered in methyl donor group. Bubble size: metabolites count. (B) Hypothesis of molecular mechanism underlying BPGM regulated BBOX1 expression by post transcriptional modification (PTM). (C) Silencing BPGM significantly reduced the protein level of H3K27me3, while overexpressing BPGM increased its level. Cells stably expressing shBPGM/BPGM and its control cells (shCtrl/Ctrl) were used to detect protein level by western blotting. (D) ChIP assays disclosed that the fragments of BBOX1 and MMP9 promoter precipitated by anti-H3K27me3 antibody were increased upon overexpressing BPGM. SK-HEP-1 cells stably expressing BPGM and its control cells (Ctrl) were employed to ChIP assay. The antibody precipitated DNAs were amplified by qPCR. 5 % of the total DNAs were amplified to serve as the control for DNA content. Values shown are signal of α-H3K27me3-precipitated DNA relative to the input and the mean value of the control group was normalized as 1. (E) Overexpressing BPGM significantly increased the protein level of EZH2 but decreased the protein level of p-EZH2-T 345 in tumor cells. (F) The molecular docking of 2,3-BPG and CDK1. Predicted structure of 2,3-BPG binding with CDK1. Key contact residues: Thr14, Arg127, Arg170. (G) Overexpressing BPGM significantly increased the protein level of p-CDK1-T 14 in tumor cells. Cells stably expressing BPGM (BPGM-OE) and its control cells (Ctrl) were used to detect protein level by western blotting. (H) 2,3-BPG treatment enhanced the phosphorylation of CDK1 at thr14 in tumor cells. The indicated concentration of 2,3-BPG was incubated with the lysate of trophoblasts and tumor cells for 30 minutes followed by western blotting. (I-J) RO-3306 treatment enhanced the phosphorylation of CDK1 at thr14 and reduced the phosphorylation of EZH2 at thr345 in tumor cells. The tumor cells were treated with the indicated concentration of RO-3306 for 12 hours followed by western blotting. (K) The model deciphers the role of BPGM in regulating BBOX1 and MMP9 expression. Error bar: mean ± SEM. P -values are labeled above the bar chart.

Article Snippet: The antibodies used included mouse antibody against β-actin (BM0627, Boster, Wuhan, China), rabbit antibody against BPGM (17173-1-AP, Proteintech), EZH2 (F0281, Selleck), phospho-EZH2 (Thr345) (TA3584S, Abmart, Shanghai, China), phospho-CDK1 (Thr14) (AP1465, Abclonal, Wuhan, China), ubiquitin (10201-2-AP, Proteintech), HIF1α (36169, Cell Signaling Technology, CST, Beverly, MA, USA), H3K4me3 (91264, Active Motif), H3K79me3 (cat 49-1020, Thermos Fisher), H3K9me3 (61014, Active Motif), H3K27me3 (91168, Active Motif) and Histone 3 (F0057, Selleck).

Techniques: Migration, Functional Assay, Expressing, Modification, Stable Transfection, Control, Western Blot, Amplification, Binding Assay, Phospho-proteomics, Concentration Assay, Incubation, Labeling

Journal: Neoplasia (New York, N.Y.)

Article Title: BPGM as an intrinsic brake to constrain metastasis through phospho-epigenetic-mediated carnitine biosynthesis suppression

doi: 10.1016/j.neo.2026.101299

Figure Lengend Snippet: Working model of BPGM-mediated metabolic-epigenetic regulation axis and its gatekeeper role in tumor metastasis. In low-metastatic tumors, higher oxygen levels activate KDM4A, which removes repressive H3K9me3 marks at the BPGM promoter, thereby promoting BPGM transcription. Elevated BPGM expression increases the production of 2,3-BPG, which stabilizes EZH2 and enhances SAM-dependent H3K27me3 deposition. This epigenetic remodeling leads to transcriptional silencing of BBOX1 , a key gene involved in carnitine biosynthesis, consequently suppressing fatty acid oxidation and inhibiting tumor metastasis. In contrast, under hypoxic conditions commonly found in high-metastatic tumors, KDM4A activity is diminished, resulting in the accumulation of H3K9me3 at the BPGM promoter and subsequent downregulation of BPGM expression. This disruption of the BPGM-mediated regulatory axis abrogates its anti-metastatic function. Notably, preclinical studies revealed that pharmacological inhibition of BBOX1 with Meldonium restores the metabolic-epigenetic barrier, effectively impeding metastatic progression.

Article Snippet: The antibodies used included mouse antibody against β-actin (BM0627, Boster, Wuhan, China), rabbit antibody against BPGM (17173-1-AP, Proteintech), EZH2 (F0281, Selleck), phospho-EZH2 (Thr345) (TA3584S, Abmart, Shanghai, China), phospho-CDK1 (Thr14) (AP1465, Abclonal, Wuhan, China), ubiquitin (10201-2-AP, Proteintech), HIF1α (36169, Cell Signaling Technology, CST, Beverly, MA, USA), H3K4me3 (91264, Active Motif), H3K79me3 (cat 49-1020, Thermos Fisher), H3K9me3 (61014, Active Motif), H3K27me3 (91168, Active Motif) and Histone 3 (F0057, Selleck).

Techniques: Expressing, Activity Assay, Disruption, Inhibition

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Ezh2 Regulates Brain Vasculature Development and Integrity. a Genetic strategy for conditional knockout of Ezh2 in endothelial cells. Top: Schematic representation of the breeding strategy to generate endothelial cell-specific Ezh2 knockout mice ( Ezh2 cKO). Ezh2 floxed mice were crossed with mice expressing Cre recombinase under the control of the Tek promoter (Tek-Cre), leading to deletion of Ezh2 in endothelial cells. Bottom: Illustration of genetic recombination in WT ( Ezh2 flox/flox ; Tek-Cre negative) and Ezh2 cKO ( Ezh2 flox/flox ; Tek-Cre positive) mice. b Whole-brain clearing and visualization. Representative images of whole brains from WT and Ezh2 cKO mice before and after tissue clearing. The transparent brains allow for deep tissue imaging of the vasculature. c Representative immunofluorescence images confirming endothelial cell-specific recombination in Ezh2 flox/flox ; Tek-Cre; G/R mice. Brain sections were stained for the endothelial cell marker CD31 (grey), tdTomato (red, indicating Cre-mediated recombination), and ZsGreen (green, indicating lack of Cre-mediated recombination in the Ezh2 cKO model). DAPI (blue) counterstains nuclei. The ‘all merge’ panel shows the overlay of all channels. The tdTomato expression is robustly co-localized with CD31 + endothelial cells. And ZsGreen expression is minimal within CD31 + endothelial cells, indicating successful Ezh2 deletion in these cells. Scale bar = 50 μm. d Quantification of tdTomato + endothelial cells. Bar graph showing the percentage of CD31 + endothelial cells that are also tdTomato-positive in brains, confirming the efficiency and specificity of Cre-mediated recombination in endothelial cells. Data are presented as individual values with the mean. *** p < 0.001 (Mann-Whitney U test). e EZH2 expression in brain endothelial cells. Immunofluorescence images showing EZH2 expression (grey) and tdTomato (red) in endothelial cells (indicated by white arrows) or non-endothelial cells (indicated by white arrowheads) within brain sections from WT and E zh 2 cKO mice. DAPI (blue) stains nuclei. Note the significant reduction of EZH2 signal in tdTomato-positive endothelial cells of Ezh2 cKO mice and no change in non-endothelial cells. f Quantification of EZH2 expression in tdTomato + cells. Bar graph quantifying the percentage of tdTomato-positive cells that also express EZH2, demonstrating the successful knockdown of Ezh2 in endothelial cells of Ezh2 cKO mice. Data are presented as mean ± SD. **** p < 0.001 (Mann-Whitney U test). g Specificity of Ezh2 deletion. Bar graph showing the percentage of tdTomato-negative cells expressing EZH2, indicating that EZH2 expression is largely unaffected in non-endothelial cells in Ezh2 cKO mice. Data are presented as individual values with the mean. ns: not significant (Student’s t-test). h 3D vascular reconstruction and analysis. Representative 3D volume renderings (left), XY slices (middle), and XZ slices (right) of brain vasculature from WT and Ezh2 cKO mice, reconstructed from light-sheet microscopy data. White and yellow squares indicate regions magnified in the adjacent panels to show detailed vascular morphology. i - l Quantitative analysis of brain vasculature. Scatter plots as individual values with the mean showing the quantitative analysis of various vascular parameters: ( i ) Vessel Volume 3D (total volume occupied by vessels, µm³), ( j ) Number of Segments (count of individual vessel segments), ( k ) Total Length (total length of all vessels, µm), and ( l ) Mean Radius (average radius of vessels, µm) in WT and Ezh2 cKO brains. Statistical significance was determined by Mann-Whitney U test. ** p < 0.01; *** p < 0.0001

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Knock-Out, Expressing, Control, Imaging, Immunofluorescence, Staining, Marker, MANN-WHITNEY, Knockdown, Microscopy

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Endothelial-specific Ezh2 Knockout in Mice and its Impact on Endothelial EZH2 Expression and Tight Junction Protein Claudin-5. a Generation of endothelial-specific Ezh2 conditional knockout (cKO) mice. Schematic illustrating the breeding strategy to generate Ezh2 cKO mice. Ezh2 floxed mice ( Ezh2 fl/fl ), carrying LoxP sites flanking the Ezh2 gene, were crossed with mice expressing Cre recombinase under the control of the Tek (Tie2) promoter (Tek-Cre). This cross results in the deletion of Ezh2 specifically in endothelial cells in the Ezh2 fl/fl ; Tek-Cre genotype. b Genotyping of Ezh2 alleles by PCR. Representative gel electrophoresis image showing the PCR products for genotyping the Ezh2 floxed allele (f/f) and the wild-type (w/w) allele. DNA from homozygous floxed (f/f), heterozygous (w/f), and wild-type (w/w) mice are shown. c Genotyping of Tek-Cre transgene by PCR. Representative gel electrophoresis image showing the PCR products for genotyping the Tek-Cre transgene. Bands indicate the presence of the Cre transgene in Tek-Cre positive mice, while its absence in WT mice is shown as a negative control. d Immunofluorescence staining for EZH2 in brain endothelial cells. Representative immunofluorescence images of brain sections from wild-type (WT) ( Ezh2 fl/fl ) and Ezh2 cKO ( Ezh2 fl/fl ; Tek-Cre) mice. Sections were stained for CD31 (red), an endothelial cell marker; EZH2 (green), the target protein; and DAPI (blue), a nuclear stain. White arrows point to endothelial cells and white arrowheads point to non-endothelial cells. Note the robust EZH2 expression in CD31-positive endothelial cells in WT mice, and the significantly reduced or absent EZH2 signal in endothelial cells of Ezh2 cKO mice. e Quantification of EZH2 expression in CD31 + endothelial cells. Bar graph representing the percentage of CD31 + endothelial cells that exhibit EZH2 expression in WT versus Ezh2 cKO mice. Data are presented as individual values with the mean. *** p < 0.001 (Mann-Whitney U test), demonstrating efficient knockout of Ezh2 in endothelial cells of cKO mice. f Immunofluorescence staining for Claudin-5 in brain vasculature. Representative confocal microscopy images of brain sections from WT and Ezh2 cKO mice, stained for Claudin-5 (red), a key tight junction protein, and CD31 (green). DAPI (blue) stains nuclei. Note the change in Claudin-5 staining patterns in the vasculature of Ezh2 cKO mice compared to WT. g Quantification of mean intensity of Claudin-5. Scatter plot as individual values showing the mean fluorescence intensity of Claudin-5 staining in the brain vasculature of WT and Ezh2 cKO mice. Each dot represents an individual mouse. ** p < 0.01 (Mann-Whitney U test), indicating a significant alteration in Claudin-5 expression or localization in the absence of endothelial Ezh2

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Knock-Out, Expressing, Control, Nucleic Acid Electrophoresis, Negative Control, Immunofluorescence, Staining, Marker, MANN-WHITNEY, Confocal Microscopy, Fluorescence

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Endothelial Ezh2 is essential for the maintenance of blood-brain barrier integrity and vascular density in the adult mouse brain. a Genetic strategy for inducible conditional knockout of Ezh2 in endothelial cells. Top: Schematic representation of the breeding strategy to generate tamoxifen-inducible, endothelial cell-specific Ezh2 knockout mice ( Ezh2 iKO). Ezh2 floxed mice ( Ezh2 fl/fl ) were crossed with mice expressing a tamoxifen-inducible Cre recombinase under the control of the Tek (Tie2) promoter (Tek-CreERT2). Bottom: Illustration of the genetic recombination process. Upon tamoxifen administration, CreERT2 translocates to the nucleus, leading to the excision of the Ezh2 gene specifically in endothelial cells. b Genotyping of Ezh2 alleles by PCR. Representative gel electrophoresis image showing the PCR products for genotyping the Ezh2 floxed allele (f/f) and the wild-type (w/w) allele. DNA from homozygous floxed (f/f), heterozygous (w/f), and wild-type (w/w) mice are shown. c Genotyping of Tek-CreERT2 transgene by PCR. Representative gel electrophoresis image showing the PCR products for genotyping the Tek-CreERT2 transgene. Bands indicate the presence of the CreERT2 transgene in Tek-CreERT2 positive mice, while its absence in WT mice is shown as a negative control. d Tamoxifen treatment regimen. Timeline illustrating the daily tamoxifen treatment protocol for adult mice (8–10 weeks old). Tamoxifen was administered for 5 consecutive days, followed by a 2-day recovery period, and then brains were harvested for histological analysis. e Immunofluorescence staining for EZH2 in brain endothelial cells after tamoxifen induction. Representative confocal microscopy images of brain sections from wild-type (WT) ( Ezh2 fl/fl ) and Ezh2 iKO ( Ezh2 fl/fl ; Tek-CreERT2) mice after tamoxifen treatment. Sections were stained for CD31 (red), an endothelial cell marker; EZH2 (green), the target protein; and DAPI (blue), a nuclear stain. White arrows point to endothelial cells and white arrowheads point to non-endothelial cells. Note the robust EZH2 expression in CD31-positive endothelial cells in WT mice, and the significantly reduced or absent EZH2 signal in endothelial cells of Ezh2 iKO mice post-tamoxifen treatment. f Quantification of EZH2 expression in CD31 + endothelial cells. Bar graph representing the percentage of CD31 + endothelial cells that exhibit EZH2 expression in WT versus Ezh2 iKO mice following tamoxifen induction. Data are presented as individual values with the mean. *** p < 0.001 (Mann-Whitney U test), demonstrating efficient inducible knockout of Ezh2 in endothelial cells of iKO mice. g Immunofluorescence staining for Claudin-5 in brain vasculature. Representative confocal microscopy images of brain sections from WT and Ezh2 iKO mice after tamoxifen treatment, stained for Claudin-5 (red), a key tight junction protein, and CD31 (green). DAPI (blue) stains nuclei. Note the change in Claudin-5 staining patterns in the vasculature of Ezh2 iKO mice compared to WT after Ezh2 deletion. h Quantification of mean intensity of Claudin-5. Scatter plot as individual values showing the mean fluorescence intensity of Claudin-5 staining in the brain vasculature of WT and Ezh2 iKO mice. *** p < 0.001 (Mann-Whitney U test). i Representative confocal images of the forebrain cortex stained for vascular and perivascular markers: CD31/AQP4 (astrocyte endfeet), GLUT1 (glucose transporter 1), and ZO1 (tight junction protein 1). Comparisons are shown between WT, Ezh2 cKO (constitutive), and Ezh2 iKO (inducible) mice. Scale bars = 100 μm (low mag) and 50 μm (high mag). j - l Quantitative analysis of vascular parameters: ( j ) percentage of AQP4 coverage on vessels, ( k ) vascular density based on GLUT + vessel area (%), and ( l ) relative vessel density based on ZO1 + area (%). Data reveal significant vascular rarefaction and loss of barrier-associated markers in both cKO and iKO models compared to WT. Data are presented as individual values with the mean. Statistical significance was determined by Mann-Whitney U test ( f , h ) or one-way ANOVA with Tukey’s post hoc test ( j – l ). ** p < 0.01, **** p < 0.0001; ns, non-significant

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Knock-Out, Expressing, Control, Nucleic Acid Electrophoresis, Negative Control, Immunofluorescence, Staining, Confocal Microscopy, Marker, MANN-WHITNEY, Fluorescence

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Endothelial E zh 2 deficiency exacerbates sepsis-induced mortality, neurological impairment, and brain injury. a Schematic workflow of the experimental design. Adult mice received tamoxifen for inducible Ezh2 deletion, followed by Cecal Ligation and Puncture (CLP) surgery. Behavioral assessments were performed prior to sacrifice at 3 days post-infection (dpi) for RNA sequencing (RNA-seq) of the cortex, hippocampus, and midbrain. b Representative surgical photographs illustrating the CLP procedure: ligation of the cecum, puncture with a needle, and extrusion of cecal contents (squeeze). c , d Kaplan-Meier survival curves comparing Sham, Ezh2 cKO+Sham, Ezh2 iKO+Sham, WT-CLP, and ( c ) Ezh2 cKO-CLP or ( d ) Ezh2 iKO-CLP mice over 3 days. Loss of Ezh2 significantly reduces survival probability following polymicrobial sepsis (n=12 per group). e Representative clinical photographs of mice 6 hours post-induction (hpi), showing increased lethargy and hunched posture in Ezh2 iKO-CLP mice compared to WT-CLP and Sham controls. f Schematic of the rotarod test used to evaluate motor coordination and balance. g , h Quantification of rotarod performance, including ( g ) rotating rod distance ( m ) and h in Ezh2 iKO-CLP mice. i Mouse Clinical Assessment Score for Sepsis (M-CASS) at 3 dpi, indicating increased disease severity in the iKO group. j Representative track maps from the open field test (OFT) and k Quantification of total distance traveled (cm), showing reduced locomotor activity and exploratory behavior in Ezh2 iKO-CLP mice. l , n Representative images of H&E-stained brain sections highlighting the ( l ) cortex and n parenchyma at 3 dpi. Arrowheads indicate apoptotic/pyknotic nuclei (condensed, fragmented nuclei) characteristic of cell death. m , o Quantification of the percentage of apoptotic cells in the ( m ) cortex and o parenchyma, demonstrating increased neurodegeneration in the absence of endothelial Ezh2 . Data are presented as individual values with the mean. Statistical significance was determined by Log-rank (Mantel-Cox) test for survival ( c , d ), one-way ANOVA with Tukey’s post hoc test ( g , h , k , i ), or unpaired Student’s t-test ( m , o ). * p < 0.05, ** p < 0.01, **** p < 0.0001; ns, non-significant. Scale bars = 50 μm

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Ligation, Infection, RNA Sequencing, Activity Assay, Staining

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Endothelial Ezh2 deficiency exacerbates neuronal damage and triggers robust glial activation in the septic brain. a Representative images of Toluidine Blue staining in the brain (cortex and hippocampus) at 3 dpi. Higher magnification panels reveal distinct morphological changes in Sham, WT-CLP, and Ezh2 iKO-CLP groups. Black arrowheads highlight dark, shrunken, and degenerated neurons (pyknotic cells), while white arrowheads indicate healthy neurons. b Quantification of the percentage of neural death cells in the cortex, showing a significant increase in neuronal loss in Ezh2 iKO-CLP mice compared to WT-CLP. c Representative confocal immunofluorescence images of the cortex stained for markers of glial activation: GFAP (astrocytes, green), Olig2 (oligodendrocytes, green), and IBA1 (microglia, red), with DAPI (nuclei, blue). d – f Quantitative analysis of glial cell densities: ( d ) GFAP + cells per mm 2 , ( e ) Olig2 + cells per mm 2 , and ( f ) IBA1 + cells per mm 2 . The data demonstrate that sepsis-induced glial activation and proliferation are significantly exacerbated in the absence of endothelial Ezh2 , suggesting a link between vascular dysfunction and neuroinflammation. Data are presented as individual values with the mean. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test. ** p < 0.01, *** p < 0.001, **** p < 0.0001. Scale bars = 200 μm (low mag), 100 μm (medium mag), and 50 μm (high mag)

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Activation Assay, Staining, Immunofluorescence

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Ezh2 deficiency exacerbates neuronal loss and modulates glial activation in the midbrain following sepsis-induced injury. a Representative images of Toluidine blue (Nissl) staining in the midbrain sections of Sham, WT-CLP, and Ezh2 iKO-CLP groups at 3 days post-infection (dpi). Successive panels show increasing magnifications of the injured area. White arrowheads indicate healthy neurons with distinct Nissl bodies; black arrowheads indicate degenerating neurons characterized by pyknotic nuclei and cytoplasmic shrinkage. Scale bars = 200 μm, 100 μm, 50 μm, and 20 μm (left to right). b Quantitative analysis of the percentage of neural death cells in the midbrain across the three groups. Data show a significant increase in neuronal death in the Ezh2 iKO-CLP group compared to the WT-CLP group. c Representative immunofluorescence images of the midbrain at 3 dpi. Brain sections were triple-stained for Tyrosine Hydroxylase (TH, red), DAPI (blue), and specific glial markers (green): GFAP (astrocytes, left), Olig2 (oligodendrocytes, middle), and IBA1 (microglia, right). Scale bars = 100 μm (low mag) and 50 μm (high mag). d - f Quantitative comparison of the number of ( d ) GFAP + reactive astrocytes, ( e ) Olig2 + oligodendrocytes, and ( f ) IBA1 + activated microglia per mm 2 in the midbrain. Note that while CLP induction triggers significant glial activation and oligodendrocyte loss, Ezh2 deletion further modulates these populations compared to WT-CLP mice. Data are presented as individual values with the mean ( n = 12 per group). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test, **** p < 0.0001

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Activation Assay, Staining, Infection, Immunofluorescence, Comparison

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Ezh2 deficiency exacerbates neuronal death and modulates glial responses in the hippocampus following sepsis-induced brain injury. a Representative images of Toluidine blue staining in the hippocampal region of Sham, WT-CLP, and Ezh2 iKO-CLP mice at 3 days post-injury (dpi). Increasing magnifications (left to right) show the structural integrity of the hippocampal layers. Black arrowheads indicate damaged or pyknotic neurons characterized by shrunken cell bodies and condensed nuclei; white arrowheads indicate healthy neurons. Scale bars as indicated (200 μm, 100 μm, 50 μm, 25 μm). b Quantification of the percentage of neural death cells in the hippocampus. CLP significantly induces neuronal death, which is further exacerbated in the Ezh2 iKO group. c Representative immunofluorescence images of the hippocampus at 3 dpi stained for glial markers (green or red) and counterstained with DAPI (blue) for nuclei. GFAP (green): Marker for astrocytes. Olig2 (green): Marker for oligodendrocytes. IBA1 (red): Marker for microglia/macrophages. Scale bars = 100 μm (low mag) and 50 μm (high mag). d - f Quantitative analysis of glial cell populations per mm 2 : ( d ) GFAP + cells: CLP induces reactive astrogliosis, which is significantly increased in Ezh2 iKO mice compared to WT-CLP mice. e Olig2 + cells: Sepsis-induced loss of oligodendrocytes is significantly more severe in the Ezh2 iKO group compared to the WT-CLP group. f IBA1 + cells: While CLP increases microglial density in WT mice, this response is significantly attenuated in Ezh2 iKO mice. Data are presented as individual values with the mean ( n = 12 per group). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: Staining, Immunofluorescence, Marker

Journal: Journal of Neuroinflammation

Article Title: Endothelial-specific Ezh2 deficiency exacerbates blood-brain barrier dysfunction and neuroinflammation in sepsis-associated encephalopathy

doi: 10.1186/s12974-026-03798-z

Figure Lengend Snippet: Comparative transcriptomic analysis across brain regions reveals regional-specific molecular signatures and regulatory networks mediated by EZH2 in the septic brain. (a1-a3) Schematic diagrams indicating the three brain regions harvested for RNA-seq analysis: (a1) Cortex, (a2) Midbrain, and (a3) Hippocampus. (b1-b3) Volcano plots representing differentially expressed genes (DEGs) in the (b1) cortex, (b2) midbrain, and (b3) hippocampus of Ezh2 iKO-CLP mice compared to WT-CLP controls. Red and blue dots denote significantly up-regulated and down-regulated genes, respectively (Fold Change > 2, p < 0.05). Key genes such as Cldn5 , Tfn , and Slc2a1 are highlighted. (c1-c3, d1-d3) Gene Ontology (GO) enrichment analysis of (c1-c3) up-regulated and (d1-d3) down-regulated genes across the three regions. Enriched terms include “inflammatory response,” “response to cytokine,” “apoptotic process,” and “vascular development,” indicating widespread neurovascular and inflammatory dysregulation. (e1-e3, f1-f3) Pathway enrichment analysis for (e1-e3) up-regulated pathways and (f1-f3) down-regulated pathways, highlighting alterations in metabolic and signaling circuits. (g1-g3) Hierarchical clustering heatmaps showing the expression profiles of the top DEGs across individual replicates for each brain region. The consistency across replicates (Cortex iKO 1–3 vs. WT 1–3) demonstrates the robustness of the transcriptomic shift following Ezh2 deletion. (h1-j3) Integrative network analysis (IPA) illustrating predicted regulatory interactions and hub gene networks in the (h1-j1) cortex, (h2-j2) midbrain, and (h3-j3) hippocampus. Networks highlight central nodes involved in (h) inflammatory signaling, (i) cell survival/death pathways, and (j) blood-brain barrier maintenance and cytoskeletal organization. Red-dashed circles identify key hub genes (e.g., Cldn1 , Mmp9 , Tnf ) that drive the pathological phenotype in the absence of endothelial Ezh2

Article Snippet: For inducible endothelial Ezh2 knockout ( Ezh2 iKO) mice, homozygous Ezh2 flox mice were crossed with Tek-CreERT2 mice (Jackson Laboratory, Stock No: 030597) to generate Ezh2 fl/fl ; Tek-CreERT2 mice (Fig. a).

Techniques: RNA Sequencing, Expressing