Fig 2: DLEU1 promotes KPNA3 expression by recruiting SMARCA1 in CRC. a DLEU1 knockdown impaired the expression of KPNA3 in HCT8 and SW480 cells. The expression levels of neighboring genes of DLEU1 were measured by RT-qPCR after DLEU1 knockdwon. b DLEU1 depletion decreased the protein levels of KPNA3 in HCT8 and SW480 cells. c DLEU1 enriched on the promoter (− 800 bp~ − 250 bp from TSS) of KPNA3 in HCT8 and SW480 cells. d DLEU1 knockdown inhibited the enrichment of H3K27Ac on KPNA3 promoter in HCT8 and SW480 cells. e SMARCA1 enriched on KPNA3 promoter in HCT8 and SW480 cells. f DLEU1 knockdown impaired the enrichment of SMARCA1 on KPNA3 promoter in HCT8 and SW480 cells. g DLEU1 overexpression promoted the enrichment of SMARCA1 on KPNA3 promoter in HCT8 and SW480 cells. h SMARCA1 co-localized with KPNA3 promoter in HCT8 cells while DLEU1 knockdown abrogated it. Green, KPNA3 promoter probe; Red, SMARCA1; Blue, DAPI. Scale bar, 10 μm. i SMARCA1 silence decreased the expression of KPNA3 in HCT8 and SW480 cells while SMARCA1 overexpression up-regulated that of KPNA3. j Higher expression of KPNA3 in SMARCA1 highly expressed CC samples. Scale bar, 100μm. k Overexpressing DLEU1 promoted KPNA3 expression while deletion of nt 1~ 400 abrogated it in HCT8 and SW480 cells. **P<0.01 and ***P<0.001. All data presented are shown as means ± SD collected from three independent experiments

Fig 3: DLEU1 interacts with SMARCA1 in CRC cells. a The expression of DLEU1 in cytoplasm and nucleus of HCT8 cells was measured by qRT-PCR. U6 serves as a nuclear control. GAPDH serves as a cytoplasmic control. b SMARCA1 was a potential interactive candidate of DLEU1. Biotin-labeled DLEU1 and intron control were incubated with HCT8 cell lysates, and the enriched products were eluted and separated by SDS-PAGE electrophoresis and silver staining. The differential band appearing in DLEU1 lane was analyzed by mass spectrum. c DLEU1 associated with SMARCA1 as shown by RNA pulldown and Western blot. Biotin-labeled DLEU1 and intron control were added into HCT8 cell lysates, and pulldown assays were performed. d DLEU1 was enriched by SMARCA1 in HCT8 and SW480 cell lysates. e SMARCA1 enriched DLEU1 in HCT8 cell lysates. SMARCA1 antibody was added into cell lysates and enriched RNAs were isolated. Then enriched DLEU1 was analyzed by PCR. f DLEU1 co-localized with SMARCA1 in HCT8 cells as shown by RNA FISH. Green, DLEU1; Red, SMARCA1; Blue, DAPI. Scale bar, 10 μm. g the region of nt 1~ 400 in DLEU1 was important for the interaction with SMARCA1. h DLEU1 (nt 1~ 400) associated with SMARCA1 directly as shown by RNA EMSA assays. i The region of nt 700~ 1050 is indispensable for the function of DLEU1 in colorectal cancer. Overexpression of DLEU1 with deletion of nt 1~ 400 cannot promoted proliferation and metastasis in CC. ***P<0.001. All data presented are shown as means ± SD collected from three independent experiments

Fig 4: The IGF1R-AS1 protein interactome reveals its interactions with chromatin remodeling complexes.a FISH analysis of IGF1R-AS1 in VCaP cells. (Left) IGF1R-AS1 FISH signal (magenta) in wild-type (WT) cells (top) and after siRNA knockdown (bottom). DAPI stains nuclei (blue). Scale bar: 5 μm. (Right) Quantification of IGF1R-AS1 foci distribution in WT cells (n = 4 biological replicates). Unpaired Student’s t-test, two-sided. p***=0.0009. b Workflow for detecting the IGF1R-AS1 interactome. c High-confidence interactions between IGF1R-AS1 and chromatin remodeling complex subfamily members are highlighted, with spectral counts indicating interaction strength. Additional interacting proteins are listed in Supplementary Data 2. d Binding of IGF1R-AS1 to SMARCA1 and SMARCA4 was detected in IGF1R-AS1 RNA pulldown assays. IGF1R sense RNA (IGF1R-S1) and streptavidin beads alone served as negative controls. e RIP-qPCR analysis of IGF1R-AS1 enriched by anti-SMARCA1 and anti-SMARCA4 antibodies in VCaP cells (n = 3 biological replicates). SChLAP1, a known SWI/SNF-interacting lncRNA, was included as a positive control. Unpaired Student’s t-test, two-sided. p****<0.0001. f, g Western blotting analysis of recombinant SMARCA4 proteins and SMARCA1 proteins pulled down by IGF1R-AS1 RNA in vitro binding assay. IGF1R-S1 and streptavidin beads served as negative controls. h The secondary structure model of IGF1R-AS1 is predicted by RNAfold webserver based on minimum free energy algorithm. i Western blotting analysis of SMARCA1 and SMARCA4 pulled down by full-length and truncated IGF1R-AS1. j, k Schematic representation of Flag-tagged full-length SMARCA4 (j) and SMARCA1 (k), and its truncations, and RIP-qPCR in VCaP cells showing IGF1R-AS1 enrichment by Flag-tagged full-length SMARCA4 (j) and SMARCA1 (k), and its truncations (n = 3 biological replicates). l, m GSEA analysis of RNA-seq data from SMARCA4 knockdown (l) and SMARCA1 knockdown (m) in VCaP cells. n–q qPCR analysis (n, o, n = 3 biological replicates) and western blotting analysis (p, q) of MYC mRNA and protein levels following SMARCA4 or SMARCA1 depletion in VCaP cells. Unpaired Student’s t-test in (n and o), two-sided. p****<0.0001. r, s VCaP cell proliferation curves following SMARCA4 (r, n = 3 biological replicates) or SMARCA1 (s, n = 3 biological replicates) depletion. Unpaired Student’s t-test, two-sided. p**** (r, siNC vs siSMARCA4-1#, day 5) < 0.0001, p*** (r, siNC vs siSMARCA4-2#, day 5) = 0.0002, p*** (s, siNC vs siSMARCA1-1#, day 5) = 0.0002, p*** (s, siNC vs siSMARCA1-2#, day 5) = 0.0004. Data in (a, e, j, k, n, o, r, and s) are shown as mean ± SEM. Experiments in (d, f, g, i, p, and q) were biologically repeated three times. Source data of this Figure are provided as Source data file.

Fig 5: IGF1R-AS1 modulates CTCF-mediated chromatin loops.a Multi-panel visualization of the enhancer regions upstream of MYC: (top) Log fold change of interaction frequencies (siNC/siAS1) at different regulatory elements, showing significant change at E3, n = 3 biological replicates; (middle) 3C-qPCR analysis demonstrating loss of chromatin looping between MYC promoter and the E3 enhancer locus upon IGF1R-AS1 depletion, n = 3 biological replicates; (bottom) genome browser tracks displaying ATAC-seq profiles in control and IGF1R-AS1-depleted cells, SMARCA4 binding by ChIP-seq, and H3K27ac HiChIP interactions connecting the MYC promoter to upstream enhancers. Unpaired Student’s t-test, two-sided. p**=0.0049. ns not significant. b CTCF HiChIP-seq heat maps within the MYC gene locus in VCaP cells treated with DMSO or the SWI/SNF degrader AU-15330 (1 μM) for 4 h (bin size = 50 kb). ChIP-seq read-density tracks for CTCF are overlaid for both conditions. HiChIP loops indicate read-supported cis interactions within the locus. Grey highlights mark key regulatory regions, including enhancers and the MYC promoter. IR interaction reads. Gene annotations are shown below the heat maps, highlighting the MYC locus and adjacent lncRNA genes. c ChIP-qPCR shows decreased binding of CTCF, cohesin (SMC1A), and chromatin remodelers (SMARCA1/A4) at the E3 enhancer after IGF1R-AS1 knockdown. n = 2 biological replicates. d CTCF HiChIP-seq heat maps within the MYC gene locus in VCaP cells in control and after IGF1R-AS1 knockdown. (bin size = 25 kb). Overlaid is ChIP-seq read-density track for CTCF in DMSO from (b). Grey highlights mark key regulatory regions. IR interaction reads. e Genome-wide aggregate peak analysis (APA) plots in CTCF HiChIP using square root VC normalization. APA values indicate ratio between signal at CTCF ChIP-seq peaks (middle pixel) and mean signal at bottom left 3 × 3 pixels. f APA values for each chromosome (chr1-22, chrX) in control and IGF1R-AS1 knockdown cells, showing that CTCF signal enrichment goes down in IGF1R-AS1 knockdown cells across all chromosomes (n = 23, chr1-22, chrX). Paired Wilcoxon signed-rank test, p****<0.0001. Boxplots are shown as the median (center line), interquartile range (box bounds, 25th to 75th percentile), and whiskers extending to 1.5× the interquartile range, and points beyond the whiskers represent outliers. g RIP-qPCR analysis of IGF1R-AS1 enrichment by anti-CTCF antibody in VCaP cells (n = 3 biological replicates). Unpaired Student’s t-test, two-sided. p****<0.0001. h Western blotting analysis of CTCF pulled down by full-length and truncated IGF1R-AS1. i Co-immunoprecipitation (co-IP) of SMARCA4 with CTCF in VCaP cells upon IGF1R-AS1 depletion. j A model demonstrating IGF1R-AS1 promotes MYC transcription by mediating SWI/SNF-CTCF interaction and facilitating chromatin loops between MYC enhancer and promoter. Data in (a and g) are shown as mean ± SEM. Experiments in (h and i) were biologically repeated three times. Source data of this Figure are provided as Source data file.