Journal: Nature

Article Title: Replication-stress-induced chromatin loops protect fork stability

doi: 10.1038/s41586-026-10695-1

Figure Lengend Snippet: a , DNA FISH validation of a HU-specific loop in Hap1 cells. Left, representative images of left (green) and right (red) anchor probes (DAPI, blue). Right, the interprobe distance distribution. n = 108 (UT) and n = 119 (HU) S phase nuclei; n = 102 (UT), n = 81 (HU) non-S-phase nuclei. Statistical analysis was performed using two-sided Mann–Whitney U -tests; P = 0.0001 (S phase) and P = 0.4295 (non-S phase). Data were pooled from two independent experiments and normalized to their respective UT conditions (set to 1) to enable direct comparison across experiments. Scale bars, 5 μm. b , Representative HU-specific, G9a-dependent chromatin loops on chromosomes 9 (top) and 2 (bottom). The bottom tracks show H3K9me3 Rep-ChIC peaks, IZs and annotated genes. c , The fractions of IZs and TZs located within HU-unique loop bodies, anchors or neither. d , The aggregate average CTCF Rep-ChIC signal within HU-unique loops and ±10 kb around their anchors. Statistical analysis was performed using the two-sided Mann–Whitney U -test; P = 1.96 × 10 −16 . The results shown are from one biological replicate. e , The distribution of IZs and CTCF motifs within HU-specific loops and flanking regions. Consensus CTCF motifs at loop anchors enriched in the CTCF Rep-ChIC signal and their orientations (convergent versus tandem) are shown below. Heat maps of the CTCF Rep-ChIC signal around loop anchors (±1 kb) in UT and HU-treated MRC5 cells are also shown. f , g , APA of HU-specific loops in HCT116 CTCF-mAID2-mClover cells ( f ) and MRC5 cells ( g ). The enrichment relative to the background is indicated in the top-right corners. Results shown are from one biological replicate. h , Schematic of a loop with CTCF sites (top). Bottom, heat maps of IZs, H3K9me3, FANCD2 Rep-ChIC and the strand-specific fork pausing signal (TrAEL-seq) within HU-unique loops and flanking regions (±3.5 kb). C, Crick strand; W, Watson strand. i , Fork pausing signal within and flanking HU-unique loops with (+CTCF) or without (−CTCF) CBSs at the loop anchor, and UT-HU common loops. Results shown are a representative dataset out of two independent biological replicate. j , The average TrAEL-seq Watson (blue) and Crick (orange) strand profiles across HU-unique loops ±5 kb. All experiments used MRC5 cells unless otherwise stated.

Article Snippet: In experiments using HCT116 cells expressing CTCF-mAID2-mClover, the culture medium was replaced with medium containing 1 μM 5-Ph-IAA (HY-134653, MedChemExpress) for at least 2 h before the experiment to induce transient depletion of CTCF.

Techniques: Biomarker Discovery, MANN-WHITNEY, Comparison

Journal: Nature

Article Title: Replication-stress-induced chromatin loops protect fork stability

doi: 10.1038/s41586-026-10695-1

Figure Lengend Snippet: a , Distribution of TADs by replication timing (early, late, TTR: Timing Transition Regions) ± HU for two replicates (MRC5 cells). b, c , Insulation score ( b ) and CTCF Rep-ChIC signal ( c ) ± 10 kb of HU-unique loops overlapping (blue/top) or not overlapping (green/bottom) TAD borders. Aggregate average (top) and tornado plots (bottom) are shown. A1 and A2 indicate the loop anchors. Data shown are from one biological replicate. d , Top: CTCF-EdU PLA signal (red) in nuclei (blue). Bottom: Distribution of total PLA spot intensity per nucleus. S-phase nuclei analysed n = 986, 1416, 1556, 1524, 1204 and 846 (left to right) from one representative experiment which has been performed 2 times with similar results. ****P ≤ 0.0001, ns=non-significant (Kruskal–Wallis, Dunn’s test, P values from left to right: <0.0001, <0.0001, <0.0001, 0.3113). Experiments in HCT116-CTCF-mAID2-mClover cells. e , Western blot of CTCF depletion via 5-Ph-IAA, in HCT116-CTCF-mAID2-mClover cells. For gel source data, see Supplementary Fig. [n = 3] independent replicates. f , Heatmaps of average CTCF/FANCD2 Rep-ChIC, Fork-Deg-seq (WT vs. shBRCA2, in RPE1-shBRCA2 cells), γH2AX , and tumour SNVs across HU-unique loops with (+CTCF) or without (-CTCF) CTCF Binding Sites (CBSs) at the anchors for 3 biological replicates. g, h , Heatmaps of IZs (TrAEL-seq MRC5 IZ (g) or HCT116 IZ (h)), H3K9me3, FANCD2 Rep-ChIC, and strand-specific fork pausing signal (W for Watson strand and C for Crick strand, MRC5 cells) within HU-unique loops ± 3.5 kb in MRC5 ( g ) and HCT116-CTCF-mAID2-mClover ( h ) cells. i , Average TrAEL-seq Watson (blue) and Crick (orange) profiles ± 5 kb of HU-unique loops in HCT116-CTCF-mAID2-mClover cells (see Fig. ). All experiments in MRC5 cells unless otherwise stated.

Article Snippet: In experiments using HCT116 cells expressing CTCF-mAID2-mClover, the culture medium was replaced with medium containing 1 μM 5-Ph-IAA (HY-134653, MedChemExpress) for at least 2 h before the experiment to induce transient depletion of CTCF.

Techniques: Insulation, Western Blot, Binding Assay

Journal: Nature

Article Title: Replication-stress-induced chromatin loops protect fork stability

doi: 10.1038/s41586-026-10695-1

Figure Lengend Snippet: a , Rolling mean of z-scored DAPI intensity vs. S-phase progression from scEdU-seq tracks. Ribbon indicates s.d. b , Number of forks per cell vs. S-phase progression. Line is median; ribbon is 95% CI. c , Heatmap of maximum normalized scEdU-seq log counts binned per 40 kb along a 60 Mb region of chromosome 2, ordered by S-phase progression. Colour scale: Normalized read coverage across the region. d, e , DNA replication speeds at indicated loop anchors and background regions. Number of regions analysed: d, [n = 345 and 192 for UT and 0.5mMHU respectively] and e, n = 302, 283, 302, 214, 203 and 215 (left to right). Results shown are from one representative experiment which has been performed 2 times with similar results. f, g , Schematic of DNA fibre degradation assay in HCT116-CTCF-mAID2-mClover (top). IdU/EdU track length ratio distributions. Means ± s.d. shown (bottom). Number of tracks analysed: Panel f: n = 1015, 1058, 1057, 1057, 1116, 1080, 1117, 1079, 1101, 1086, 1069 and 1061 (left to right), pooled from 3 independent biological replicates and overlaid in three different colours in the plot. Panel g: n = 681, 726, 708, 724, 641, 669, 643, 676, 682, 684, 683, 666, 703, 676, 676, 689, 673, 686, 713 and 665 (left to right) pooled from 2 independent biological replicates and overlaid in two different colours. ****P ≤ 0.0001, **P ≤ 0.01, ns=non-significant (Kruskal–Wallis, Dunn’s test, panel f: All P values < 0.0001, panel g: P values: (left to right) >0.9999, <0.0001, <0.0001, <0.0001, <0.0001, 0.0095, 0.0061, <0.0001, >0.9999, <0.0001, 0.4485, >0.9999, 0.0019, >0.9999, >0.9999, >0.9999). h , Schematic of Fork-Deg-seq approach to map nucleolytic degradation at newly replicated regions. The diagram was created using BioRender; Taneja, N. https://BioRender.com/d7kf3t0 (2026). i , Fork-Deg-seq signal in WT and shBRCA2-induced RPE1 cells after 8 h 4 mM HU, alongside BrdU IP signal and replication timing for the indicated region on chromosome 3. Dotted square highlights a loop-poor region with enhanced Fork-Deg-seq enrichment. Loops of bidirectional replicon (fountain)- scale are shown in dark red; smaller loops are shown in light red. j , Aggregate heatmap of CTCF (CTCF-Rep-ChIC, in MRC5 cells), ForkDeg-seq signal (in RPE1-shBRCA2 cells), γH2AX , and Single Nucleotide Variant (SNV ) distribution within and in a +/− 1 kb region flanking the HU-specific loops identified in MRC5 cells. Schematics: loop body and flanking regions. Orange triangles mark loop anchors position based on CTCF-binding sites. Colour scale: Normalized read coverage across the region. All experiments in MRC5 cells unless otherwise stated.

Article Snippet: In experiments using HCT116 cells expressing CTCF-mAID2-mClover, the culture medium was replaced with medium containing 1 μM 5-Ph-IAA (HY-134653, MedChemExpress) for at least 2 h before the experiment to induce transient depletion of CTCF.

Techniques: Degradation Assay, Variant Assay, Binding Assay

Journal: Nature

Article Title: Replication-stress-induced chromatin loops protect fork stability

doi: 10.1038/s41586-026-10695-1

Figure Lengend Snippet: a , Schematic of the replication fork degradation DNA fibre assay in HCT116 CTCF-mAID2-mClover cells, involving CTCF depletion (dep; 5-Ph-IAA) and G9a inhibition (UNC0642) (top). Middle, representative fibres. Bottom, the IdU/EdU track length ratio. Data are mean ± s.d. From left to right, numbers of forks analysed per condition: n = 1,014, 1,015, 1,034, 1,039, 1,032, 1,053, 1,070 and 1,005, pooled from three independent replicates and overlaid in three different colours in the plot. Statistical analysis was performed using Kruskal–Wallis tests followed by Dunn’s test; from left to right, P < 0.0001, P < 0.0001, P < 0.0001, P < 0.0001, P < 0.0001, P > 0.9999, P > 0.9999, P > 0.9999. Scale bar, 5 μm. b , Representative locus (chromosome 16: 81.3–82.75 Mb). Top, Hi-C heat map (the red squares highlight the positions of the loop anchors). The Fork-deg-seq signal in HCT116-CTCF-mAID2-mClover cells UT or treated with G9ai (4 h), 5-Ph-IAA (4 h, CTCF-depleted) or both after 4 mM HU (5 h or 8 h). MRC5 CTCF and H3K9me3 Rep-ChIC signals are shown below, alongside IZs and fragile sites. The black arrowheads indicate high Fork-deg-seq signal. The shaded area shows a loop-dense region with reduced degradation; unshaded areas show enhanced Fork-deg-seq enrichment. c , BrdU-enriched 5-kb bins classified by HU-unique loop coverage: loop-poor (0–1 loop, left, n = 4,398 bins) and loop-dense (≥2 loops, right, n = 6,844 bins). The fold change in Fork-deg-seq signal relative to the WT is shown. Data are mean ± s.d. Statistical analysis was performed using two-sided Mann–Whitney U -tests; loop-free region, from top to bottom: P = 4.4 × 10 −33 , P = 2.4 × 10 −132 , P = 2.4 × 10 −132 , P = 4.1 × 10 −33 , P = 2.4 × 10 −132 , P = 2.7 × 10 −34 ; loop-dense region, from left to right: P = 1.4 × 10 −130 , P = 2.0 × 10 −130 , P = 1.8 × 10 −130 , P = 3.2 × 10 −1 , P = 7.8 × 10 −2 , P = 8.4 × 10 −1 . Results shown are from one biological replicate. d , Aggregate analysis of the mean ± s.d. Fork-deg-seq signal within HU-unique loops and 5 kb flanking regions after 3 h of treatment with 4 mM HU alone (top row, left four plots) or with mirin and DNA2i followed by 4 mM HU (bottom row). Ionizing radiation (10 Gy) was included as a control without further treatment (top right plot). The results shown are from one biological replicate. All of the experiments described in this figure were performed in HCT116 CTCF-mAID2-mClover cells, unless otherwise stated.

Article Snippet: In experiments using HCT116 cells expressing CTCF-mAID2-mClover, the culture medium was replaced with medium containing 1 μM 5-Ph-IAA (HY-134653, MedChemExpress) for at least 2 h before the experiment to induce transient depletion of CTCF.

Techniques: Inhibition, Hi-C, MANN-WHITNEY, Control

Journal: Nature

Article Title: Replication-stress-induced chromatin loops protect fork stability

doi: 10.1038/s41586-026-10695-1

Figure Lengend Snippet: a , Representative locus (chr8: 122.8-133.9 Mb). Top: Hi-C heatmap (red squares highlight positions of loop anchors). Below: Fork-Deg-seq signal (HCT116-CTCF-mAID2-mClover cells) untreated or treated with G9ai (4 h), 5-Ph-IAA (4 h, CTCF-dep) or both upon 4 mM HU (5 h or 8 h), alongside MRC5 CTCF and H3K9me3 Rep-ChIC signals, IZs and fragile sites. Black arrowheads indicate high Fork-Deg-seq signal. Shaded area: loop-dense region with reduced degradation; unshaded areas show increased degradation. b , Aggregate mean Fork-Deg-seq signal ± s.d. after 5 h 4 mM HU within HU-unique loops ± 5 kb in HCT116-CTCF-mAID2-mClover cells. Results shown are from one representative experiment which has been performed 2 times with similar results. c-e , Distribution of IZs at early [n = 2119] and late [n = 2193] replicating fragile sites. ( c ) HU-unique loops overlapping IZs (MRC5), ****P ≤ 0.0001, (Two-sided Mann–Whitney U test, p-value: 7.858e-38). ( d ) Fork-Deg-seq signal in HCT116 IZs, ****P ≤ 0.0001, (Two-sided Mann–Whitney U test, p-values: 1.222e-07, 6.247e-06, 1.161e-05, 6.247e-06 left to right). ( e ) Fork-Deg-seq signal in RPE1 IZs (RPE1-shBRCA2 cells), ****P ≤ 0.0001, (Two-sided Mann–Whitney U test, p-values: 3.119e-37 WT + HU and 2.080e-55 shBRCA2+HU). f , Representative electron micrographs showing a reversed fork with ssDNA gaps on both daughter strands and intact reversed arms (HCT116-CTCF-mAID2-mClover). P, parental strand; D, daughter strand; R, reversed arms. Scale bars: 250 nm or 1183 bp (main), 50 nm or 473 bp (insets). g , Ranked gap length distribution per condition. h , qPCR analysis of HU-unique loop formation in WT and TKO (SMARCAL1, ZRANB3, HLTF knockout) U2OS cells ± 4 mM HU. Means ± s.e.m. [n = 4 independent biological replicates]. ****P < 0.0001, ***P < 0.001, **P < 0.01, ns=non-significant (Ordinary one-way ANOVA, Tukey’s test, P value from top to bottom: a: <0.0001, 0.0026, 0.4735; b: 0.0003, 0.0095, 0.5794; c: 0.004, 0.1237, 0.3688; d: <0.0001, 0.0471, 0.1211; e: <0.0001, 0.0003, 0.0752; f: 0.0004, 0.0003, 0.5844; g: <0.0001, 0.0042, 0.4301; h: 0.0003, 0.0004, 0.935; i: <0.0001, 0.0019, 0.4734,; j: 0.0006, 0.0001, 0.6204; k: 0.0005, 0.0206, 0.4417; l: <0.0001, 0.0032, 0.3017; m: <0.0001, 0.0173, 0.269; n: 0.0013, 0.0074, 0.8887; o: 0.0013, 0.0074, 0.7701; p: 0.0003, 0.095, 0.1693). i , Top: Schematic showing that only H3K9me3 signal intensity overlapping with EdU was measured. Bottom: Distribution of H3K9me3 intensity in the region of interest and for the indicated conditions. Number of replication sites analysed per condition across two independent experiments: n = 148, 98, 111, 107, 103 and 84 from left to right. (*: p < 0.05, ****: p < 0.0001, ns: non-significant. Kruskal–Wallis test followed by Dunn’s test, P values: 0.0160, <0.0001, <0.0001, >0.9999, 0.7905, >0.9999, 0.0946, >0.9999 and 0.0156 (top to bottom)). j , Top: Schematic showing that only the H3K9me3 signal intensity just outside of the EdU track was measure (H3K9me3 intensity over a region covering 20% of the total length of the EdU track was measured on both side of the EdU track). Bottom: Distribution of H3K9me3 intensity in the region of interest and for the indicated conditions. Number of replication sites analysed per condition across two independent experiments: n = 98, 119, 107 and 103 from left to right. (***: p < 0.001, ****: p < 0.0001, ns: non-significant. Kruskal–Wallis test followed by Dunn’s test, P values: <0.0001, 0.7726, 0.4606, 0.0003, <0.0001 and <0.0001 (top to bottom). k , Total intensity distribution of CTCF-EdU PLA spots. S-phase nuclei analysed: n = 642 for all conditions imaged from one representative experiment, which has been performed 2 times with similar results. Red line marks mean. ****P ≤ 0.0001, **P ≤ 0.01, ns=non-significant (Kruskal–Wallis, Dunn’s test, P values from left to right: <0.0001, <0.0001, 0.0011, <0.0001, <0.0001, 0.1690).

Article Snippet: In experiments using HCT116 cells expressing CTCF-mAID2-mClover, the culture medium was replaced with medium containing 1 μM 5-Ph-IAA (HY-134653, MedChemExpress) for at least 2 h before the experiment to induce transient depletion of CTCF.

Techniques: Hi-C, MANN-WHITNEY, Knock-Out

Journal: Bioactive Materials

Article Title: Gynostemma pentaphyllum -derived extracellular vesicles alleviate skin aging by destabilizing STING

doi: 10.1016/j.bioactmat.2026.03.010

Figure Lengend Snippet: The extraction process and proposed mechanisms for GPEVs in treating UVB-induced Skin aging. Extracellular vesicles derived from Gynostemma pentaphyllum (GPEVs), isolated and purified from fresh whole herb via differential ultracentrifugation, exhibit anti-photoaging properties. UVB radiation triggers the up-regulation of STING, activating the TBK1-CTCF pathway and causing photoaging. GPEVs effectively promote STING degradation, reducing CTCF-mediated aging gene transcription, protecting against UVB-induced skin aging.

Article Snippet: The primary antibodies used included γ-H2AX and CTCF (1:1,000, Cell Signaling), p21 and STING (diluted 1:1,000, Cell Signaling).

Techniques: Extraction, Derivative Assay, Isolation, Purification

Journal: The Journal of Experimental Medicine

Article Title: A PI3Kδ-Foxo1-FasL signaling amplification loop rewires CD4 + T cell signaling and differentiation

doi: 10.1084/jem.20252154

Figure Lengend Snippet: Pik3cd E1020K reshapes the epigenetic landscape of CD4 + T cells. (A and B) Naïve CD4 T cells from WT and Pik3cd E1020K/+ mice were polarized under Th2 conditions and evaluated by ATACseq ( n = 3). A total of 71,040 peaks were detected. (A) Venn diagram of WT-specific, Pik3cd E1020K/+ -specific, and common peaks. (B) WT and Pik3cd E1020K/+ -specific peaks examined by motif enrichment analysis. Enrichment P values were plotted for both groups. Red: motifs specifically enriched in WT peaks; blue: motifs specifically enriched in Pik3cd E1020K/+ peaks; orange: motifs enriched in both groups. (C) Peak heatmap of CTCF CUT&Tag peaks from WT and Pik3cd E1020K/+ naïve, Th1, and Th2 cells organized into six clusters (1–6) specific to each indicated population, as described. (D) CTCF motif enrichment P value (top) and fold enrichment (bottom) in clusters 1–6. (E) DEGs (WT vs Pik3cd E1020K/+ ) and non-DEGs from bulk RNAseq data were compared in the indicated populations for percentages of genes showing differential CTCF peaks (WT versus Pik3cd E1020K/+ ). (F) Frequencies of CTCF peaks repressed, induced, or both induced and repressed in Pik3cd E1020K/+ Th2 cells (versus WT Th2) near DEGs (WT Th2 versus Pik3cd E1020K/+ Th2). (G) Th2 DEGs (WT Th2 versus Pik3cd E1020K/+ Th2) were organized into two categories: DEGs showing no change in CTCF (WT versus Pik3cd E1020K/+ ) and DEGs showing repressed CTCF peaks in Pik3cd E1020K/+ relative to WT. Pathway enrichment analysis (Enrichr; ) of TFTs (ChEA; ) was performed using these two categories of DEGs. Adjusted P values (−log 10 ) of the top 25 enriched TF signatures in each category plotted against each other. (H) Western blot evaluating CTCF and Zap70 in lysates from WT and Pik3cd E1020K/+ Th2-polarized cells, cultured in the presence or absence of Cal101 (10 nM) or rapamycin (200 nM). Data are representative of three independent experiments ( n = 3), quantified in . (I) Western blot evaluating CTCF and Zap70 in lysates from Th2-polarized NC and Foxo1 gRNA-Cas9–nucleofected WT CD4 T cells, compared with Pik3cd E1020K/+ cells. Data are representative of three independent experiments ( n = 3), . NC, negative control. Source data are available for this figure: .

Article Snippet: CUT&Tag utilized primary antibodies against CTCF (D31H2; Cell Signaling) and rabbit mAB IgG XP isotype control (DA1E; Cell Signaling).

Techniques: RNA sequencing, Western Blot, Cell Culture, Negative Control

Journal: The Journal of Experimental Medicine

Article Title: A PI3Kδ-Foxo1-FasL signaling amplification loop rewires CD4 + T cell signaling and differentiation

doi: 10.1084/jem.20252154

Figure Lengend Snippet: Altered chromatin accessibility and CTCF activity in Pik3cd E1020K/+ Th2 cells. Supporting data for . (A and B) Naïve CD4 T cells from WT and Pik3cd E1020K/+ mice underwent Th2 polarization and were examined by ATACseq. n = 3. Volcano plot showing WT-specific peaks in red and Pik3cd E1020K/+ -specific peaks in blue (fold change >1.5, P < 0.05) (B) CTCF CUT&Tag tracks of Id3 , Bcl2 , Tbx21 , and Eomes loci. (C) Quantification of CTCF protein expressed as a ratio of CTCF/Zap70. Supporting data for . n = 3 for each group, from three independent experiments. (D) Quantification of CTCF protein expressed as a ratio of CTCF/Zap70. Supporting data for . n = 3 for each group, from three independent experiments. Statistical comparisons were made using ratio paired t tests. *P < 0.05.

Article Snippet: CUT&Tag utilized primary antibodies against CTCF (D31H2; Cell Signaling) and rabbit mAB IgG XP isotype control (DA1E; Cell Signaling).

Techniques: Activity Assay