SETDB1 cooperates with the CRL4B complex to promote colorectal cancer tumorigenesis by disrupting circadian rhythm

Abstract

The circadian rhythm is an intrinsic, self-regulatory physiological system in organisms that is hierarchically governed by core clock genes. Dysregulation of this rhythm is closely associated with tumor progression. However, the specific molecular mechanisms underlying circadian rhythm disruption–mediated colorectal cancer (CRC) remain largely elusive. The histone methyltransferase SET domain bifurcated 1 (SETDB1) exerts oncogenic functions in multiple solid tumors, prompting this study to investigate the molecular mechanisms through which SETDB1 regulates CRC progression. Our results demonstrate that SETDB1 acts as an oncogene in CRC by interacting with the Cullin 4B-RING E3 ligase (CRL4B) complex to enhance cancer cell proliferation, invasion, and stemness maintenance. Specifically, SETDB1 and the CRL4B complex assemble into a transcriptional repressor complex that binds to the promoters of core clock genes, such as PER3 and CRY1, to repress their transcription, thereby disrupting cellular circadian rhythms and ultimately accelerating CRC cell proliferation and metastasis. Collectively, our findings elucidate the pivotal role of the SETDB1/CRL4B complex in CRC pathogenesis and provide a potential therapeutic strategy targeting SETDB1 for patients with CRC with high CUL4B expression.

Introduction

Colorectal cancer (CRC) represents a major global health burden, ranking as the third most commonly diagnosed malignancy and the second leading cause of cancer-related mortality [1, 2]. The etiology of CRC involves a complex interplay of genetic predisposition, dietary habits, and lifestyle factors, making a comprehensive understanding of these relationships critical for its prevention and management [3–5].

Abnormal epigenetic modifications typically affect enzymatic activity, protein stability, and protein–protein interactions, thereby triggering the onset of various diseases, including tumors. Therefore, targeting abnormal epigenetic modifications to block tumor progression constitutes a core strategy in the development of precision cancer therapies and represents a novel approach for the clinical intervention of malignant tumors such as CRC [6–9]. Studies have demonstrated that aberrant histone methylation is prevalent in multiple types of tumors, including CRC, liver cancer, and breast cancer. The associated regulatory process is co-mediated by a variety of epigenetic regulators and transcription factors and contributes to the malignant progression of tumors via disruption of the balance of downstream target gene expression[10–12]. Recent evidence has highlighted the importance of the epigenetic regulator SET domain bifurcated 1 (SETDB1), a histone lysine methyltransferase, in CRC tumorigenesis [13–15]. SETDB1 is a vital epigenetic regulator that catalyzes histone H3 lysine 9 (H3K9) methylation—specifically, di- and tri-methylation—and recruits heterochromatin protein 1 (HP1) to mediate the conformational switch from euchromatin to heterochromatin [16–19]. SETDB1 also serves as a key transcriptional repressor to regulate several critical factors, including p53 [20, 21], homeobox A family genes [22, 23], and c-Myc [24, 25], and is involved in diverse developmental events, including cell fate determination [26, 27] and T-helper cell differentiation [28]. Consequently, SETDB1 dysregulation is implicated in a wide range of pathological conditions, from neurological disorders [29, 30] to carcinogenesis [31–34]. Studies in this context have established SETDB1 as an oncogene that promotes the development and malignant progression of multiple cancers, including those of the breast [35, 36], lung [37, 38], liver [39–41], and colorectum [13–15]. Therefore, obtaining a deeper understanding of the biological functions of SETDB1 in CRC is essential to elucidate the molecular basis of this disease and to evaluate its potential as a novel therapeutic target for clinical intervention.

The Cullin 4B-RING E3 ligase (CRL4B) complex, composed of Cullin 4B (CUL4B), damaged DNA binding protein 1 (DDB1), and small RING finger proteins (ROC1 or ROC2) as core components, represents a key constituent of the CRL family of ubiquitin ligases [42, 43]. Unlike its paralog CUL4A, CUL4B contains an N-terminal nuclear localization signal, which confers predominant nuclear localization and implicates it in chromatin-related functions [44, 45]. At the molecular level, the CRL4B complex catalyzes the monoubiquitination of histone H2A at lysine 119 (H2AK119ub1)—a modification associated with transcriptional repression [46]. Moreover, the CRL4B complex physically interacts with several epigenetic regulators, such as the polycomb repressive complex 2 (PRC2) complex [46], nucleosome remodeling and deacetylation (NuRD) complex [47], and the SUV39H1/HP1/DNMT3A complex [48] to cooperatively silence tumor suppressor genes through coordinated histone modification and DNA methylation. Notably, the CRL4B complex is involved in various biological processes, including DNA damage repair, cell cycle regulation, and transcriptional regulation. Through the degradation of specific cell cycle regulatory proteins, this complex modulates cell cycle progression and maintains cellular homeostasis [49]. Additionally, CUL4B plays a critical role in intestinal development; consequently, its depletion impairs the self-renewal capacity of intestinal stem cells. Mechanistically, the CRL4B complex regulates intestinal homeostasis by ubiquitinating its substrate IRGM1 at the K270 residue [50]. Consistent with this essential role, loss-of-function mutations in the X-linked CUL4B gene are linked to developmental disorders in humans, and Cul4b knockout results in embryonic lethality in mice, underscoring its non-redundant functions, therefore, in proliferation and development [51, 52]. Collectively, the CRL4B represents a critical epigenetic modulator involved in transcriptional regulation and cancer progression.

Circadian rhythms are fundamental biological cycles governed by a conserved transcriptional–translational feedback loop centered upon core clock genes [53, 54]. The core factor of this loop is the BMAL1/CLOCK transcription factor heterodimer, which activates the expression of clock-controlled genes such as the period (PER1, PER2, and PER3) and cryptochrome (CRY1 and CRY2) genes. Following translation, PER and CRY proteins accumulate in the cytoplasm, form complexes, undergo phosphorylation, and translocate into the nucleus, wherein they interact with the BMAL1/CLOCK heterodimer for repression in their transcriptional activity; thereby, a self-regulatory feedback cycle is completed. This tightly regulated core mechanism generates stable, approximately 24-h oscillations in clock gene expression and promotes the rhythmic transcription of a wide range of downstream target genes, whereby coordinating physiological and behavioral processes [55–57]. Therefore, the disruption of circadian rhythms is increasingly being implicated in tumorigenesis, with dysregulation of core clock components documented in multiple cancers, including breast cancer [58, 59], glioma [60, 61], leukemia [62, 63], and CRC [64, 65]. Notably, PER and CRY frequently behave as tumor suppressors, and their aberrant expression is associated with enhanced tumor proliferation and metastatic potential [66–69]. Thus, elucidating the regulatory networks and functional alterations of the circadian clock is essential for understanding the molecular mechanisms of carcinogenesis and may provide a foundation for the development of novel therapeutic approaches.

In this study, we investigated the oncogenic potential of SETDB1 and its underlying mechanisms in CRC. We showed that SETDB1 physically recruits the CRL4B complex, transcriptionally represses PER3 and CRY1, and disrupts the circadian rhythm to induce cell proliferation, invasion, and enhanced cancer stem cell (CSC)-like properties in CRC. Overall, our results indicate that the SETDB1/CRL4B complex is a novel regulator of circadian rhythms and represents a promising target for future treatment strategies for CRC.

Materials and methods

Antibodies and reagents

The antibodies and their respective sources are listed as follows: anti-SETDB1 (#2196S, Cell Signaling Technology), anti-FLAG (F1408, Sigma-Aldrich), anti-CUL4B (C9995, Sigma-Aldrich), anti-β-actin (A1978, Sigma-Aldrich), anti-vimentin (V6630, Sigma-Aldrich), anti-fibronectin (F3648, Sigma-Aldrich), anti-DDB1 (sc-25367, Santa Cruz Biotechnology), anti-ROC1 (ab2977, Abcam), anti-H2AK119ub1 (05-678, Millipore), anti-α-catenin (610193, BD Biosciences), anti-γ-catenin (610253, BD Biosciences), anti-E-cadherin (610181, BD Biosciences), and anti-N-cadherin (610920, BD Biosciences), anti-OCT4 (ab19857, Abcam), anti-NANOG (ab109250, Abcam), anti-KLF4 (#12173, Cell Signaling Technology), anti-C-MYC (ab32072, Abcam), anti-PER3 (12550-1-AP, Proteintech), anti-PER3 (sc-517227, Santa Cruz Biotechnology), and anti-CRY1 (bs-511441R, Bioss). Small interfering RNAs (siRNAs) and short hairpin RNAs (shRNAs) were obtained from Shanghai GenePharma (Shanghai, China). The protease inhibitor cocktail was purchased from Roche Applied Science (Penzberg, Germany).

Cell culture and transfection

All cell lines used in this study were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). HEK293T cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM). NCM-460 and HCT-15 cells were cultured in RPMI-1640 medium. RKO cells were cultured in ATCC-formulated Eagle’s Minimum Essential Medium. HT-29 and HCT116 cells were cultured in McCoy’s 5a Modified Medium. SW620 cells were cultured in Leibovitz’s L-15 medium, and LoVo cells were cultured in ATCC-formulated F-12K medium. All media were supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 U/mL streptomycin (Gibco). SW620 cells were incubated in a humidified incubator without CO₂ at 37 °C, whereas all other cells were maintained at 37 °C with 5% CO₂. All cell lines were confirmed to be mycoplasma-free. For transfection experiments, Lipofectamine RNAiMAX Transfection Reagent (Invitrogen, Carlsbad, CA, USA) was used to transfect siRNAs into the cells according to the manufacturer’s instructions. Each experiment was performed in triplicate, and the sequence with the highest interference efficiency was selected for subsequent use. Plasmid information is provided in Supplementary Table S1. Sequences of the siRNAs and shRNAs used are listed in Supplementary Table S2.

RNA-seq analysis

Total RNA was extracted from shSCR-, shSETDB1-, and shCUL4B-transfected HCT116 cells using TRIzol (Roche, Penzberg, Germany). The RNA samples were sent to Annoroad Gene Technology (Beijing, China) for library preparation and sequencing. Sequencing was performed on the DNBSEQ platform, generating an average of approximately 9.087 Gb of data per sample. The average mapping ratio to the reference genome was 80.76%, and the average mapping ratio to annotated genes was 95.66%. Raw reads were aligned to the human reference genome (hg38). Transcriptomic analysis was performed using the TopHat2 package. HTSeq-count v0.6.0 was used to quantify transcript abundance. DEGs were identified using a false discovery rate < 0.05 as the screening threshold. All RNA-seq data are available on https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE324627.

Real-time quantitative polymerase chain reaction (RT-qPCR) analysis

Total RNA was extracted from shSCR-, shSETDB1-, and shCUL4B-transfected RKO and HCT116 cells using the RNA-Quick Purification Kit (Esunbio, Shanghai, China). RNA was reverse-transcribed into cDNA using the PrimeScript™ RT Master Mix (TaKaRa Bio, Shiga, Japan). Quantitative real-time PCR was performed using FastStart Universal SYBR Green Master (Rox) (Sigma-Aldrich, St. Louis, MO, USA). Primer sequences used for amplification are listed in Supplementary Table S3. β-actin served as an internal reference gene for normalization. Each experiment was performed in triplicate with three independent biological replicates.

Colony formation assays

RKO (1 × 10³ cells per well) and HCT116 (1 × 10³ cells per well) cells were seeded into 6-well plates with equal volumes (2 ml) of medium following lentiviral transduction and cell line construction. After 10–14 days, cells were fixed with 4% paraformaldehyde (PFA) and stained with crystal violet for 30 min, and images were captured using a light microscope.

EdU incorporation assays

Cells were seeded into 96-well plates at a density of 1 × 10⁵ cells/mL with equal volumes of medium and cultured overnight. Upon reaching the logarithmic growth phase, cell proliferation was assessed using an EdU assay kit (RiboBio, Guangzhou, China) according to the manufacturer’s instructions.

Transwell assays

Following lentiviral transduction, RKO and HCT116 cells were seeded in serum-free media into the upper chamber of the Transwell inserts. The inserts were placed into 24-well plates containing medium with 10% fetal bovine serum. After incubation, cells were fixed with PFA, washed, and stained with crystal violet for 30 min, and images were obtained using a light microscope.

Sphere formation assay

RKO and HCT116 cells (5 × 10³ cells per well) were plated in 6-well ultra-low attachment plates and cultured in serum-free DMEM/F12 medium (HyClone) supplemented with B27 (50×; Invitrogen), 0.4% bovine serum albumin, 20 ng/mL bFGF, 10 ng/mL EGF, and 5 µg/mL insulin (Invitrogen). Fresh stem cell medium was added every other day. Spheres were observed on day 5 and continued to increase in size and cell number until day 15.

Flow cytometry

HCT116 cells were resuspended in sorting buffer (3% FBS [v/v] and 3 mM EDTA in phosphate-buffered saline (PBS)) before analysis. To identify CRC CSCs, the anti-CD44-FITC (397517, BioLegend) and anti-CD133-PE (297903, BioLegend) antibody, or appropriately isotype-matched control antibodies, were used. Samples were analyzed using a FACSVerse (BD) flow cytometer; data were analyzed using FlowJo9.2 software.

Mouse xenograft models

The number of animals per group was determined based on prior studies using similar in vivo models, ensuring sufficient biological replicates for robust and consistent phenotype assessment. The mice were randomly assigned to each experimental group. For the tumor initiation study, lentivirus-transduced HCT116 cells were subcutaneously injected into the groin of 6-week-old male nude mice at limiting dilutions of 5 × 10⁴, 5 × 10³, 5 × 10², or 1 × 10² cells, with five mice per group. Tumor size was monitored weekly. To evaluate tumor growth, HCT116 cells transduced with an empty vector, SETDB1-overexpressing, shSCR, shCUL4B, or shCUL4B + shSETDB1 constructs (3 × 10⁶ cells per mouse) were inoculated into the inguinal region of 6-week-old male nude mice. All mice were euthanized in accordance with ethical guidelines. Tumor tissues were harvested for further analysis. All evaluations of the animal experiments were conducted by researchers who were unaware of the specific groups of animals used, including the treatment methods and the result data. Tumor volume was calculated using the formula: (length × width²)/2. Animal handling and procedures were approved by the Experimental Animal Ethics Committee of the Cancer Hospital, Chinese Academy of Medical Sciences (NCC2023A154).

Immunoprecipitation and western blotting

Cells were washed three times with PBS at 4 °C and collected by centrifugation at 800 × g for 3 min at 4 °C. The cells were then lysed in lysis buffer (5 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.5% NP-40, and 0.25% sodium deoxycholate, supplemented with protease inhibitor cocktail) for 30 min at 4 °C, followed by centrifugation at 12,000 × g for 10 min. A total of 500 μg of protein was incubated with the appropriate antibody (2–3 µg) overnight at 4 °C. Protein A/G Sepharose CL-4B beads (Life Technologies, Carlsbad, CA, USA) were then added, and the samples were rotated for 2 h at 4 °C. After five washes, the beads were mixed with 2× loading buffer and boiled for 10 min at 95 °C to elute the proteins. Western blotting was subsequently performed. Proteins were separated using 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene difluoride (PVDF) membranes. Membranes were incubated with primary antibodies, followed by incubation with the appropriate secondary antibodies. Protein bands were detected using an enhanced chemiluminescence system (Thermo Scientific) according to the manufacturer’s instructions.

Immunoaffinity purification and LC–MS/MS

HCT116 cells stably expressing FLAG-SETDB1 were generated by transfecting cells with FLAG-tagged SETDB1 plasmids. Cell lysates were prepared from approximately 5 × 10⁸ cells and added to a FLAG affinity column to purify the FLAG protein complexes according to the manufacturer’s instructions. Fractions corresponding to the bed volume were collected, resolved on SDS-PAGE precast gels, silver-stained, and subjected to LC–MS/MS for protein identification and data analysis.

GST pull-down assays

GST fusion proteins were constructed and purified using glutathione-Sepharose 4B beads. In parallel, in vitro transcription and translation were performed using rabbit reticulocyte lysate (TNT Systems; Promega, Madison, WI, USA). For the GST pull-down assays, the purified GST fusion proteins were incubated with the in vitro translated products in binding buffer. The beads were subsequently washed, resuspended in SDS-PAGE loading buffer, and subjected to western blot analysis.

CUT&Tag-seq

Approximately 1 × 10⁴ cells were used for each CUT&Tag assay following the NovoNGS CUT&Tag 2.0 high-sensitivity kit protocol (Novoprotein, Shanghai, China). The procedure involved capturing live cells with magnetic beads, permeabilizing the cell membrane with digitonin, incubating the cells with primary antibodies against SETDB1 or CUL4B and the corresponding secondary antibody, cleaving the target regions via Mg²⁺-activated Tn5 transposase, and ligating adapters to both ends of the cleaved fragments. The complete sequencing library was then generated via PCR amplification. Whole-genome DNA sequencing was performed by Romics Corporation (Shanghai, China). All samples were processed with an IgG control to evaluate background signal. The overall mapping rates were 98.42% (CUL4B), 97.14% (IgG), and 99.64% (SETDB1). The unique mapping rates were 77.49% (CUL4B), 76.71% (IgG), and 92.92% (SETDB1). The FRiP scores were 35.83% (SETDB1) and 4.45% (CUL4B). Raw sequencing data were processed from the Illumina HiSeq 2000 and analyzed using MACS. Peak annotation, comparison, and visualization of SETDB1 and CUL4B binding sites were performed using IGV and an R package for ChIP peak analysis. All CUT&Tag-seq data are available on https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE324626.

ChIP and Re-ChIP

For ChIP assays, 1 × 10⁷ cells were crosslinked, sonicated, precleared, and incubated with the indicated antibodies. DNA was then extracted and precipitated. For Re-ChIP assays, immune complexes were eluted from the beads, diluted with ChIP dilution buffer, and subjected to a second immunoprecipitation reaction. Enrichment of target DNA sequences was analyzed via PCR or qPCR using primers specific for each gene promoter. Primer sequences are provided in Supplementary Table S4.

Time-series assay

HCT116 cells were seeded into 60-mm dishes and cultured for 24–48 h before being replaced with serum-free DMEM. Cells were then treated with 200 nmol/L dexamethasone at 37 °C for 2 h and collected at 4-h intervals from 24–48 h for subsequent analysis. JTK-CYCLE analysis was employed to identify rhythmically expressed transcripts. For each gene, expression values across the time course were formatted as a data matrix with time points in chronological order and biological replicates placed in adjacent columns. JTK_CYCLE was run with default parameters and a sampling interval of 2 h. The algorithm evaluated the statistical significance of rhythmic expression by comparing the data against a null distribution generated from random permutations, and computed the period, phase (lag), and amplitude for each transcript. Multiple hypothesis testing was controlled by calculating Benjamini–Hochberg false discovery rate (FDR)-adjusted q-values. Transcripts with an adjusted P-value (q-value) < 0.05 were considered significantly rhythmic.

IHC assays

All human tissue specimens (HColA030PG08-1, HCol-Can060PT-01, HCol-Muc060CS-01, and HOrgC180PG01-2) were obtained from Shanghai Outdo Biotech Company (Shanghai, China). The study has received approval from the ethical committee (Approval Number: SHYJS-CP-22I0015, SHYJS-CP-1501007, SHYJS-CP-1407013, and SHYJS-CP-2210041). Tissue sections were rapidly fixed in 4% paraformaldehyde (Sigma-Aldrich) at 4 °C overnight and embedded in paraffin for subsequent IHC staining according to a standard protocol. Sections were developed using 3,3’-diaminobenzidine as the chromogen. All IHC evaluations were performed by the pathologists who were blinded to all clinical and pathological information, including patient identity, treatment status, and outcome data. Immunohistochemical staining was semi-quantitatively evaluated using the histochemistry score (H-score). The H-score was calculated based on both the percentage of positively stained cells and the staining intensity, according to the formula: H-score = Σ(pᵢ × i), where i represents the staining intensity grade (0, negative; 1, weak, yellow; 2, moderate, brown; 3, strong, brownish-brown), and pᵢ denotes the percentage of cells stained at each intensity grade. The resulting H-score ranged from 0 to 300, with higher values reflecting greater overall staining positivity in terms of both intensity and extent.

Cell viability assays

HCT116 and RKO cells were seeded into 96‑well plates at a density of 1 × 10⁴ cells per well in 100 µL of complete culture medium and incubated overnight at 37 °C to allow attachment. Subsequently, 10 µL of Cell Counting Kit‑8 (CCK‑8) solution (Dojindo, Kumamoto, Japan) was added to each well. The plates were further incubated at 37 °C for 1 h. Cell viability was determined by measuring the absorbance of the formazan product at 450 nm according to manufacturer’s protocol. All experiments were performed in triplicate wells and independently repeated at least three times.

Statistical analysis

All statistical analyses were performed using GraphPad Prism (version 9.0; GraphPad Software Inc., San Diego, CA, USA). Sample sizes were determined based on preliminary experiments and literature precedents, with no formal power analysis performed. No animals or samples were excluded from the analysis. All inclusion and exclusion criteria were predefined prior to the commencement of the experiments, and no post‑hoc exclusions were applied. Data are presented as mean ± standard deviation (SD) for normally distributed data or as mean ± standard error of the mean (SEM) where indicated. Unless otherwise specified, comparisons between two groups were analyzed using two-tailed unpaired Student’s t-test, while comparisons among three or more groups were performed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple comparisons. Correlation analysis was performed using the Pearson correlation test. Colorectal cancer gene expression datasets were downloaded from the TCGA database and the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) under accession numbers: GSE10972, GSE37182, GSE87211, GSE21510, GSE117606, and GSE156451. Kaplan–Meier survival analyses were performed using the Kaplan–Meier Plotter online tool (https://kmplot.com/analysis) with auto-select best cutoff, and statistical significance was determined using two-sided log-rank tests. All statistical tests were two-sided, and P-values < 0.05 were considered statistically significant. Exact P-values are reported in the figure legends where applicable, with *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Effect sizes (Cohen’s d or partial η²) are provided in the corresponding raw data.

Results

SETDB1 promotes CRC progression as an oncogene

To clarify the critical role of SETDB1 in regulating CRC progression, we analyzed SETDB1 expression in CRC using data from The Cancer Genome Atlas (TCGA) database and Gene Expression Omnibus (GEO) datasets (GSE10972 and GSE37182). Our analysis revealed that SETDB1 expression was significantly elevated in CRC tissues compared with that in normal tissues (Fig. 1A). Kaplan–Meier survival analysis using the TCGA database showed that high SETDB1 expression was associated with poor prognosis (Fig. 1B). Consistent with the bioinformatics analysis results, SETDB1 was highly expressed in all CRC cell lines tested (Fig. 1C). Immunohistochemical (IHC) staining of clinical CRC samples further confirmed that SETDB1 was significantly overexpressed in CRC tissues (Fig. 1D).

Fig. 1. The histone methyltransferase SET domain bifurcated 1 (SETDB1) promotes colorectal cancer (CRC) progression as an oncogene.

A The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO; GSE10972 and GSE37182) datasets were analyzed to examine SETDB1 expression in CRC. B The association between SETDB1 expression and survival in CRC was analyzed using the Kaplan–Meier survival analysis based on TCGA datasets. C The mRNA and protein levels of SETDB1 in CRC cell lines were detected using real-time quantitative polymerase chain reaction (RT-qPCR) and western blotting, respectively. β-actin served as an internal control. D Representative images of immunohistochemistry (IHC) staining for SETDB1 in CRC tumors and adjacent normal tissues (HColA030PG08-1, n = 15). E SETDB1 was overexpressed (vector and SETDB1) and knocked down (shSCR and shSETDB1) in HCT116 and RKO CRC cell lines using lentivirus-mediated transduction. β-actin served as an internal control. Growth curve (F), colony formation (G), and 5-ethynyl-2’-deoxyuridine (EdU) (H) assays were performed to evaluate the effect of SETDB1 in HCT116 and RKO cells. I Transwell invasion assays were performed in SETDB1-overexpressing and SETDB1-knockdown HCT116 and RKO cells to assess invasive capacity. ImageJ software was used for quantitative analysis. RT-qPCR (J) and western blotting (K) were performed to detect epithelial–mesenchymal transition (EMT) markers in SETDB1-overexpressing and SETDB1-knockdown HCT116 and RKO cells. β-actin served as an internal control. Data are presented as mean ± standard deviation (SD) or standard error of the mean (SEM) for three independent experiments. Data were analyzed using a two-tailed unpaired Student’s t-test or one-way analysis of variance (ANOVA). *P < 0.05, **P < 0.01, ***P < 0.001.

To further evaluate the role of SETDB1 in CRC development and progression, we constructed SETDB1-overexpression and SETDB1-knockdown HCT116 and RKO cell lines (Fig. 1E and Supplementary Fig. S1A). Growth curve assays assessing the effect of SETDB1 knockdown on tumorigenicity showed that SETDB1 silencing significantly suppressed CRC cell proliferation (Fig. 1F). Colony formation assays showed that SETDB1 overexpression promoted CRC cell proliferation, whereas SETDB1 knockdown inhibited it (Fig. 1G). We then validated the proliferative effect of SETDB1 using 5-ethynyl-2’-deoxyuridine (EdU) incorporation assays. The results further confirmed these findings, showing that the proportion of EdU-positive cells was significantly increased in SETDB1-overexpressing CRC cells and significantly decreased in SETDB1-knockdown cells compared with that in the control cells (Fig. 1H). In addition, Transwell invasion assays demonstrated that SETDB1 overexpression significantly enhanced the invasive capacity of CRC cells, whereas SETDB1 knockdown reduced it (Fig. 1I). Consistent with these results, SETDB1 overexpression significantly promoted the migratory ability of CRC cells, whereas SETDB1 knockdown impaired it (Supplementary Fig. S1B). Epithelial–mesenchymal transition (EMT) is a hallmark of tumor progression and plays a crucial role in tumor metastasis. Therefore, we examined the expression of EMT markers in HCT116 and RKO cells after SETDB1 overexpression or knockdown. SETDB1 overexpression downregulated epithelial markers and upregulated mesenchymal markers, whereas SETDB1 knockdown increased the expression of epithelial markers and decreased that of mesenchymal markers (Fig. 1J, K). Taken together, these results indicate that SETDB1 is highly expressed in CRC tissues and promotes the proliferation, invasion, and migration of colorectal cancer cells.

To investigate how SETDB1 regulates CRC growth, RNA sequencing (RNA-seq) analysis was performed in the control group shScramble (shSCR) and SETDB1-knockdown group (shSETDB1). The results showed that, compared with that in the control group, a total of 1826 genes were upregulated, and 721 genes were downregulated (Supplementary Fig. S1C). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of these differentially expressed genes (DEGs) revealed that the downregulated genes were primarily enriched in the following pathways: B cell receptor signaling, cell adhesion molecules, estrogen signaling pathway, oxidative phosphorylation, cyclic adenosine monophosphate (cAMP) signaling, Ras signaling, and chemical carcinogenesis. In contrast, the upregulated genes were predominantly enriched in pathways including tumor necrosis factor (TNF) signaling, hypoxia-inducible factor 1 (HIF-1) signaling, interleukin-17 (IL-17) signaling, nuclear factor-κB (NF-κB) signaling, p53 signaling, and apoptosis (Supplementary Fig. S1C). These results suggest that SETDB1 is involved in CRC progression by regulating essential multiple cellular processes.

To further elucidate the role of SETDB1 in CRC, we verified the mRNA expression levels of potential oncogenes and tumor suppressor genes (TSGs). Consistent with the RNA-seq results, in SETDB1-knockdown cells, the expression of potential oncogenes (MMP9, CD44, EGFR, FGFR1, CCNE2, CCND1, MDM2, and ARID5A) was significantly downregulated, whereas the expression of potential TSGs (GADD45A, SIAH1, P53, FBXW7, ARID1A, VHL, CASP8, and BAD) was notably upregulated (Supplementary Fig. S1D). Collectively, these results indicate that SETDB1 promotes the initiation and progression of CRC by coordinately regulating key tumor-associated genes.

SETDB1 regulates the stemness of CRC cells

To further investigate the role of SETDB1 in the development and progression of CRC, we generated SETDB1-overexpressing and SETDB1-knockdown HCT116 and RKO CRC cell lines. Sphere formation assays revealed that, compared with that in the control cells, SETDB1 overexpression significantly enhanced sphere-forming efficiency and sphere diameter, whereas SETDB1 knockdown resulted in a notable reduction in both sphere number and size (Fig. 2A and Supplementary Fig. S2A). Flow cytometric analysis showed that SETDB1 overexpression elevated the fractions of CD44⁺ and CD133⁺ cells, while SETDB1 knockdown reduced the fractions of cells bearing these stemness‑associated surface markers (Supplementary Fig. S2B). Further analysis revealed that SETDB1 overexpression significantly elevated the mRNA and protein levels of stemness markers (OCT4, KLF4, C‑MYC, and NANOG), whereas SETDB1 knockdown markedly reduced their expression (Fig. 2B, C). These results indicate that SETDB1 is closely associated with the maintenance of stemness in CRC cells.

Fig. 2. SETDB1 sustains CRC stemness in vitro and in vivo.

A Sphere formation assays were performed in SETDB1-knockdown and SETDB1-overexpressing HCT116 and RKO cells. The diameter of Sphere was quantified. B, C The expression of cancer stemness markers was examined via RT-qPCR (B) and western blotting (C) in SETDB1-knockdown and SETDB1-overexpressing HCT116 and RKO cells. β-actin was used as an internal control for normalization of mRNA levels, and β-actin served as the loading control for western blotting. D, E SETDB1-knockdown HCT116 cells were subcutaneously injected into male nude mice (n = 5 per group). Tumor volume and final tumor weight were measured over time to assess the tumorigenic potential of cells at different inoculation doses. F Representative IHC images of SETDB1 and Ki-67 staining in subcutaneous xenografts derived from nude mice injected with SETDB1-knockdown HCT116 cells. G Tumor volume and final tumor weight in male nude mice (n = 4 per group) following subcutaneous injection with SETDB1-overexpressing HCT116 cells. H Establishment of the AOM–DSS-induced CRC mouse model. Representative images of intestinal tissues from three mice subjected to the AOM–DSS-induced colorectal CRC model. Visible tumor lesions were observed in the colonic mucosa of all three mice. These spontaneously formed tumor tissues were subsequently collected and used for the establishment of organoid cultures. I Tumors were isolated from the AOM–DSS-induced CRC mouse model, and tumor-derived organoids were established using in vitro culture. Setdb1 was subsequently knocked down in these organoids, followed by dynamic monitoring and quantification of organoid growth rate and volume. Representative images of the tumor organoids captured on days 1, 4, and 7. Data are presented as mean ± SD or SEM for three independent experiments. Statistical significance was determined using a two-tailed unpaired Student’s t-test or one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

To further evaluate the impact of SETDB1 on the proportion of colorectal CSCs in vivo, we established a nude mouse xenograft model. The proportion of colorectal CSCs in the SETDB1-knockdown group was significantly lower than that in the control group (Fig. 2D). Additionally, in the control group, tumor formation was observed in mice inoculated with as few as 100 cells, whereas no tumors were detected in the SETDB1-knockdown group, indicating that the tumor formation rate was significantly reduced following SETDB1 knockdown (Fig. 2E). Meanwhile, the tumor growth rate in the shSETDB1 group was significantly inhibited. Moreover, IHC analysis of tumor tissues demonstrated that SETDB1 knockdown significantly suppressed the expression of the proliferation marker Ki-67 in tumor tissues (Fig. 2F). Collectively, these findings suggest that SETDB1-knockdown significantly inhibits the stemness properties of CRC cells.

To further validate these findings in vivo, HCT116 cells overexpressing SETDB1 were subcutaneously injected into nude mice. Compared with that in the control group, SETDB1 overexpression contributed to a significant increase in subcutaneous tumor growth (Fig. 2G and Supplementary Fig. S2C). Additionally, we generated a CRC mouse model using Azoxymethane–Dextran Sodium Sulfate (AOM–DSS), to isolate tumor-derived organoids (Fig. 2H), and found that Setdb1 knockdown markedly inhibited organoid growth compared to that in the control (Fig. 2I). Collectively, these in vitro and in vivo findings demonstrate that SETDB1 enhances the stemness of CRC cells.

SETDB1 physically interacts with the CRL4B complex

To investigate the mechanism of SETDB1 in CRC, we performed immunoaffinity purification and liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis in HCT116 cells overexpressing FLAG-tagged SETDB1 to screen for candidate proteins that interact with SETDB1. As shown in our previous mass spectrometry results, components of the CRL4B complex were co-purified with SETDB1 [70]. This approach identified DDB1, a core component of the CRL4B complex, leading us to hypothesize that SETDB1 and the CRL4B complex coordinately exert regulatory functions in CRC (Fig. 3A). To further validate this interaction, we examined whether SETDB1 could interact with CUL4B, DDB1, and ROC1, key components of the CRL4B complex. Western blot analysis of the eluates from immunoaffinity purification confirmed the physical interaction between SETDB1 and CUL4B, DDB1, and ROC1 (Fig. 3B). Consistent with this result, co-immunoprecipitation assays also verified the interaction between SETDB1 and these key components of the CRL4B complex in HEK293T and HCT116 cells (Fig. 3C–E).

Fig. 3. Physical interaction between SETDB1 and the cullin 4B-RING E3 ligase (CRL4B) complex.

A Immunoaffinity purification coupled with silver staining and mass spectrometry was performed to identify SETDB1-interacting proteins. B Western blot analysis of candidate interacting proteins in HCT116 cells. C Co-immunoprecipitation using an anti-FLAG antibody was performed in SETDB1-overexpressing cells to verify protein–protein interactions. D, E Validation of endogenous protein–protein interactions in HEK293T and HCT116 cells. F Glutathione S-transferase (GST) pull-down assays using GST-fused SETDB1, CUL4B, DDB1, or ROC1 and in vitro transcribed/translated proteins. G, H GST pull-down assays using GST-fused C-terminal, N-terminal, and middle MBD domains of SETDB1 and in vitro transcribed/translated CUL4B and DDB1. I, J GST pull-down assays using GST-fused DID, Cullin, and C-terminal domains of CUL4B and in vitro transcribed/translated SETDB1. K, L GST pull-down assays using GST-fused BPA, BPB, or BPC domains of DDB1 and in vitro transcribed/translated SETDB1. M Schematic representation of the molecular interactions between SETDB1 and the CRL4B complex.

To explore whether direct interactions exist among SETDB1, CUL4B, DDB1, and ROC1, we performed glutathione S-transferase (GST) pull-down assays, which revealed that SETDB1 directly interacts with both CUL4B and DDB1 (Fig. 3F). To map the specific domains mediating these protein–protein interactions, we cloned and expressed different truncated variants of these proteins. Specifically, the C-terminal, N-terminal, and middle MBD domains of SETDB1 were constructed and individually expressed (Fig. 3G), followed by interaction assays with CUL4B and DDB1. The results showed that CUL4B interacted with the C-terminal domain of SETDB1, whereas DDB1 interacted with the MBD domain (Fig. 3H). Furthermore, various domain truncations of CUL4B were expressed and subjected to interaction assays. SETDB1 was confirmed to interact with the C-terminal domain (Fig. 3I, J). In addition, SETDB1 bound to the BPC domain of DDB1 (Fig. 3K, L). These findings reveal the direct interactions among SETDB1, CUL4B, and DDB1, providing a structural basis for subsequent functional studies (Fig. 3M).

SETDB1 cooperates with CUL4B to promote CRC cell proliferation, invasion, and stemness

Analysis of clinical CRC data from the TCGA database revealed that CUL4B expression in CRC tissues was significantly higher than that in adjacent normal tissues (Supplementary Fig. S3A). Moreover, high CUL4B expression was significantly correlated with patient survival, with elevated CUL4B expression associated with poorer outcomes in patients with CRC (Supplementary Fig. S3B). To further validate these findings, we collected tumor tissue samples and paired adjacent normal tissue samples from patients with CRC, followed by IHC assays to detect CUL4B expression. CUL4B was found to be significantly upregulated in CRC tissues compared with that in adjacent normal tissues (Supplementary Fig. S3C). Collectively, these results indicate that high CUL4B expression in CRC tissues promotes disease progression. To dissect the role of CUL4B in driving CRC progression, we employed lentiviral vectors to achieve stable CUL4B knockdown or ectopic overexpression in HCT116 and RKO cells (Supplementary Fig. S3D) and determined the effects of modulating CUL4B expression on cell proliferation using colony formation and EdU incorporation assays. Knockdown of CUL4B significantly suppressed colony formation and decreased the percentage of EdU-positive cells, whereas CUL4B overexpression notably enhanced colony formation and increased the proportion of EdU-positive cells (Supplementary Fig. S3E and S3F).

To gain deeper insights into the regulatory role of CUL4B, we performed RNA-seq analysis on shSCR and CUL4B-knockdown groups. Comparative analysis identified 797 upregulated and 864 downregulated genes in the CUL4B-knockdown group compared with the control group (Supplementary Fig. S3G). Subsequent KEGG pathway enrichment analysis of these DEGs showed that the downregulated gene set was primarily enriched in several key pathways, including the cell cycle, cellular senescence, p53 signaling pathway, mismatch repair, and IL-17 signaling. In contrast, the upregulated gene set was predominantly enriched in pathways such as circadian entrainment, neuroactive ligand–receptor interaction, calcium signaling, retinol metabolism, and cytokine–cytokine receptor interaction (Supplementary Fig. S3H). Taken together, these results suggest that CUL4B participates in CRC progression by regulating fundamental cellular functions.

Transwell invasion assays demonstrated that SETDB1 knockdown suppressed the invasive capacity of CRC cells, whereas CUL4B overexpression promoted invasion. Notably, concurrent knockdown of SETDB1 and CUL4B resulted in a greater reduction in cell invasion compared with silencing either gene alone. In contrast, simultaneous overexpression of both proteins augmented cell invasion by over twofold. Importantly, the pro-invasive effect induced by overexpressing one protein was abrogated when the other was knocked down (Fig. 4A, B). In addition, sphere formation assays showed that SETDB1 knockdown reduced tumor sphere diameter, whereas CUL4B overexpression increased it. Co-knockdown of SETDB1 and CUL4B led to a more marked decrease in sphere diameter than did single-gene knockdown, whereas co-overexpression of both proteins induced a greater increase in sphere size (Fig. 4C, D). Furthermore, SETDB1 knockdown significantly decreased the number of spheres formed by CRC cells, whereas overexpression of CUL4B robustly enhanced sphere formation. Notably, dual knockdown of SETDB1 and CUL4B resulted in a more pronounced inhibition of sphere formation than did silencing either gene individually. Importantly, the effect of overexpression of one protein on sphere number was almost completely abolished when the counterpart gene was knocked down (Supplementary Fig. S4A and S4B).

Fig. 4. SETDB1 and CUL4B promote CRC cell malignant phenotypes.

A, B HCT116 and RKO cells were transfected with the indicated lentiviruses, followed by Transwell invasion assays. Quantitative analysis was performed using ImageJ software. C, D Sphere formation assays were performed in HCT116 and RKO cells transduced with the indicated lentiviruses. The diameter of spheres was quantified. E Male nude mice (n = 5 per group) were subcutaneously injected with HCT116 cells (pre-transduced with the indicated lentiviruses). Tumor volume was monitored over time, and final tumor weight was measured at the experimental endpoint. F Representative IHC staining images of Ki-67 in subcutaneous xenograft tumors derived from HCT116 cells pre-transduced with the indicated lentiviruses. G Correlation analysis of SETDB1 and CUL4B mRNA expression levels in the TCGA CRC cohort. H Kaplan–Meier disease free survival analysis of patients with CRC stratified according to combined SETDB1 and CUL4B expression levels. Data are presented as mean ± SD or SEM for three independent experiments. Statistical significance was determined using a two-tailed unpaired Student’s t-test or one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

In vivo subcutaneous tumorigenesis assays showed that SETDB1 knockdown significantly inhibited tumor growth in mice, whereas dual knockdown of SETDB1 and CUL4B resulted in a more pronounced suppression of tumor progression (Fig. 4E and Supplementary Fig. S4C). IHC analysis indicated that, compared with that in the control cohort, CUL4B knockdown led to a notable reduction in Ki-67 expression, and additional silencing of SETDB1 further decreased Ki-67 levels (Fig. 4F). Analysis of publicly available TCGA datasets showed a positive correlation between SETDB1 and CUL4B expression, with increased expression of both proteins associated with poorer patient prognosis (Fig. 4G, H). Taken together, these findings indicate that SETDB1 acts synergistically with CUL4B to promote cell proliferation, invasion, and stemness maintenance in CRC.

Defining the transcriptional target landscape of the SETDB1/CRL4B complex

To systematically identify the transcriptional target genes of the SETDB1/CRL4B complex at the genome-wide level, we used the HCT116 cell line as the experimental model and performed Cleavage Under Targets and Tagmentation sequencing (CUT&Tag-seq) with antibodies specific for SETDB1 and CUL4B. Sequencing results identified 15,052 SETDB1-bound promoter peaks and 4179 CUL4B-bound promoters peaks (Fig. 5A, B). Notably, SETDB1 and CUL4B shared highly conserved binding motif profiles, providing further experimental evidence supporting the direct physical interaction between these proteins (Fig. 5C).

Fig. 5. The SETDB1/CRL4B complex mediates transcriptional regulation.

A Genomic distribution of SETDB1 and CUL4B binding peaks. B Heatmaps showing SETDB1 and CUL4B Cleavage Under Targets and Tagmentation sequencing (CUT&Tag-seq) signal intensity RPKM across their respective binding peaks in HCT116 cells. C Motif enrichment analysis of genomic regions co-occupied by SETDB1/CUL4B. D Venn diagram showing the overlap of binding peaks between SETDB1 and CUL4B. E Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of genes associated with SETDB1/CUL4B co-binding peaks. F Genome browser tracks showing SETDB1 and CUL4B CUT&Tag-seq signals at representative target gene loci. G Validation of genes identified via CUT&Tag-seq using quantitative chromatin immunoprecipitation (qChIP) assays in HCT116 cells. GAPDH served as an internal reference. H mRNA expression levels of selected target genes in HCT116 cells following SETDB1 or CUL4B knockdown. I Sequential ChIP (ChIP/Re-ChIP) assays were performed in HCT116 cells using the indicated primary antibodies. J Schematic diagram depicting the assembled and worked model of the SETDB1/CRL4B complex. Data are presented as mean ± SD or SEM for three independent experiments. Statistical significance was determined using a two-tailed unpaired Student’s t-test or one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

Building on these findings, we focused on the shared target gene promoter regions of the SETDB1/CRL4B complex, identifying 3555 overlapping promoters (Fig. 5D). KEGG enrichment analysis of the gene set corresponding to these shared promoters revealed significant enrichment in multiple key-signaling pathways that regulate tumor progression, including the cell cycle, Wnt signaling, p53 pathways, and circadian regulation pathways (Fig. 5E). Subsequent analysis showed that SETDB1 and CUL4B were notably enriched at the promoter regions of genes such as BMAL1, CLOCK, CRY1, CRY2, BHLHE40, BHLHE41, PER3, OCLN, TJP2, CASP7, GADD45A, and ARID5B (Fig. 5F). Quantitative chromatin immunoprecipitation (qChIP) assays further confirmed the recruitment of SETDB1 and CUL4B to the promoter regions of target genes associated with circadian regulation and tight junction pathways (Fig. 5G).

Previous studies have shown that CRY1, CRY2, PER3, and BHLHE40 are involved in circadian rhythm regulation in CRC [71, 72], whereas TJP2 exerts a pronounced tumor-suppressive effect in this disease [73]. Based on the finding that both SETDB1 and CUL4B can target and bind to the promoters of the aforementioned genes, we further investigated the transcriptional regulatory mechanism of the SETDB1/CRL4B complex in regulating CRY1, CRY2, PER3, BHLHE40, and TJP2. Knockdown of SETDB1 or CUL4B in HCT116 cells significantly upregulated the mRNA levels of CRY1, CRY2, BHLHE40, BHLHE41, PER3, TJP2, OCLN, CASP7, and GADD45A (Fig. 5H). Meanwhile, overexpression of SETDB1 reduced the protein-level expression of PER3 and CRY1, and knockdown of CUL4B reversed this inhibitory effect (Supplementary Fig. S4D). ChIP/Re-ChIP experiments confirmed that SETDB1 and CUL4B bound to the promoter regions of CRY1, CRY2, PER3, and BHLHE40 genes as a transcriptional co-repressor complex (Fig. 5I). Collectively, these findings establish that SETDB1 and CRL4B physically interact and functionally cooperate to form an extended transcriptional co‑repressive complex that suppresses downstream target gene expression (Fig. 5J).

The SETDB1/CRL4B complex disrupts the tumor circadian clock

qChIP assays showed that knockdown of SETDB1 or CUL4B significantly reduced the binding of histone modification markers H3K9me3 and H2AK119ub1 to the promoter regions of CRY1, CRY2, PER3, and BHLHE40 (Fig. 6A). These results indicate that the SETDB1/CUL4B complex inhibits the expression of downstream target genes at the transcriptional level via H3K9me3 and H2AK119ub1 modification. To clarify whether the SETDB1/CRL4B complex regulates cell proliferation and invasion through transcriptional repression of these target genes, we performed rescue experiments focusing on CRY1 and PER3, the two core circadian genes that exhibited the greatest sensitivity to complex regulation in our prior assays and are essential for circadian rhythm homeostasis. First, we verified the gene expression levels of SETDB1, CUL4B, PER3, and CRY1 in the constructed cell models (Supplementary Fig. S5A and S5B). Colony formation assays showed that knockdown of PER3 or CRY1 partially rescued the inhibition of cell proliferation induced by SETDB1 and CUL4B knockdown (Fig. 6B). Meanwhile, reconstitution of PER3 and CRY1 expression in SETDB1 or CUL4B overexpression cells attenuated the accelerated tumor growth (Supplementary Fig. S5C). In addition, Transwell invasion assays showed that PER3 and CRY1 knockdown reversed the inhibitory effect of SETDB1 and CUL4B knockdown on cell invasion (Fig. 6C). Similarly, reconstitution of PER3 and CRY1 expression attenuated the enhanced invasive capacity driven by SETDB1 and CUL4B overexpression (Supplementary Fig. S5D). Collectively, these findings demonstrate that transcriptional repression of PER3 and CRY1 by the SETDB1/CRL4B complex is critical for CRC cell proliferation and invasion.

Fig. 6. The SETDB1-CRL4B complex disrupts circadian rhythm regulation.

A Recruitment of the indicated proteins to the promoters of CRY1, CRY2, PER3, and BHLHE40 in HCT116 cells transfected with shSCR, shSETDB1, or shCUL4B. Purified rabbit IgG served as a negative control. B Colony formation assays were performed to evaluate the effects of inhibiting target gene expression in SETDB1-knockdown or CUL4B-knockdown HCT116 cells. C Transwell invasion assays were performed in SETDB1-knockdown or CUL4B-knockdown HCT116 and RKO cells following inhibition of target genes. D Schematic overview of the experimental workflow for time-series assays in CRC cells. Dexamethasone-mediated synchronization was performed in control, SETDB1-knockdown, and CUL4B-knockdown HCT116 cells for 24 h. E Cells were collected at 4-h intervals over a 24-h period (0–24 h), followed by qRT-PCR analysis of BMAL1, CLOCK, CRY1, CRY2, and PER3 mRNA levels. JTK-CYCLE analysis of rhythmic mRNA expression patterns, with ADJ.P indicating adjusted minimal p-values and AMP denoting amplitude. Data are presented as mean ± SD or SEM for three independent experiments. Statistical significance was determined using a two-tailed unpaired Student’s t-test or one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

Given the significance of PER3 and CRY1 in maintaining the stability of the circadian rhythm, we next to understand the role of the SETDB1/CRL4B complex in maintaining the circadian rhythm signals in CRC cells. To investigate this, HCT116 cells were synchronized for 24 h by serum starvation, followed by a 2-h treatment with dexamethasone to monitor clock gene expression oscillations (Fig. 6D). Loss-of-function experiments showed that knockdown of SETDB1 or CUL4B restored the oscillatory expression of BMAL1, CLOCK, CRY1, CRY2, PER1, and PER3, whereas PER2 expression was not markedly altered compared to that with control cells (Fig. 6E and Supplementary Fig. S5E). These findings indicate that the SETDB1/CRL4B complex disrupts the cellular circadian rhythm by transcriptionally repressing PER3 and CRY1 expression oscillations.

SETDB1 inhibitor affects the proliferation and invasion of CRC cells

To evaluate the therapeutic potential of targeting the SETDB1/CRL4B complex in CRC, we employed TTD‑IN‑1, a selective small molecular SETDB1 tandem tudor domain endogenous binder competitive inhibitor [74], in human CRC cell lines HCT116 and RKO. We first determined the half‑maximal inhibitory concentration (IC₅₀) via CCK‑8 assays, which yielded values of 17.31 µM (HCT116) and 42.96 µM (RKO), respectively (Supplementary Fig. S5F). Based on these results, subsequent functional experiments were carried out at a concentration of 10 µM, with DMSO as vehicle control. Cell counting and colony formation assays demonstrated that TTD‑IN‑1 markedly reduced proliferation in both cell lines (Supplementary Fig. S5G and S5H). Transwell invasion assays further demonstrated that inhibitor treatment markedly attenuated the invasive and migratory capacities of CRC cells (Supplementary Fig. S5I). Notably, the phenotypic alterations elicited by TTD‑IN‑1 closely paralleled those resulting from genetic ablation of SETDB1 or CUL4B, corroborating the on‑target specificity of the inhibitor. Collectively, our data indicate that pharmacological blockade of SETDB1 suppresses CRC cell proliferation and invasion, providing a compelling rationale for further exploring SETDB1 as a viable therapeutic target in colorectal cancer.

SETDB1 overexpression across multiple cancer types positions it as a prospective cancer biomarker

To explore the expression patterns of SETDB1 and CUL4B in CRC tissues, we performed IHC staining to detect protein levels in CRC samples of different histological grades and in normal colorectal tissues. SETDB1 and CUL4B expression levels in CRC samples were significantly upregulated and positively correlated with tumor histological grade (Fig. 7A, B). Combined analysis of the TCGA and GEO databases, along with verification in clinical samples, revealed that PER3 and CRY1 expression levels were significantly reduced in CRC tissues compared with that in normal tissues (Fig. 7C, D). Consistent with this, analysis of CRC samples in the GEO database (GSE21510, GSE117606, and GSE156451) revealed a significant negative correlation between the expression levels of CRY1 or PER3 and SETDB1 (Fig. 7E). Furthermore, data from TCGA indicate that low expression of CRY1 and PER3 is associated with poorer prognosis in CRC patients. More significantly, patients exhibiting both high SETDB1 expression and low CRY1 or PER3 expression demonstrate the worst clinical outcomes, underscoring the combined prognostic value of these markers (Fig. 7F). Further analysis of the TCGA database confirmed that SETDB1 was significantly upregulated across multiple cancer types (Supplementary Fig. S6A). Moreover, upregulation of CUL4B expression, paralleling that of SETDB1, has been documented in breast, colorectal, liver, prostate, and gastric cancers. Concurrently, CRY1 and PER3 levels are markedly reduced in these tumors compared with matched normal tissues (Supplementary Fig. S6B). These coordinated expression alterations across multiple cancer types reinforce the functional relevance of the SETDB1/CUL4B–CRY1/PER3 axis in circadian disruption and tumor development. In addition, compared with that in paired adjacent normal tissues, SETDB1 showed high expression in various cancers, including thyroid, breast, lung, esophageal, liver, colorectal, and pancreatic cancers (Fig. 7G, H). Furthermore, using Kaplan–Meier survival analysis, we found that high expression of SETDB1 is also closely related to poor prognosis of patients with liver cancer, ovarian cancer, and gastric cancer (Supplementary Fig. S6C). Collectively, these findings confirm that SETDB1 possesses oncogenic characteristics and may serve as a potential diagnostic biomarker for CRC.

Fig. 7. SETDB1 is overexpressed across multiple cancer types and represents a prospective cancer biomarker.

A, B Representative IHC staining of SETDB1 and CUL4B in normal and CRC tissues (histological grades I–III) (HCol-Can060PT-01 and HCol-Muc060CS-01, n = 60). C Representative IHC staining of CRY1 and PER3 in normal and CRC tissues (HCol-Can060PT-01 and HCol-Muc060CS-01, n = 60). D TCGA dataset and GSE87211 analysis of CRY1 and PER3 mRNA expression levels in normal versus CRC tissues. E Analysis of GSE21510, GSE117606, and GSE156451 dataset for the correlations in mRNA expression between SETDB1, CRY1, and PER3. The relative level of CRY1 or PER3 was plotted against that of SETDB1. F Kaplan–Meier survival analysis of COAD TCGA dataset for the relationship between survival time and CRY1, PER3, SETDB1/CRY1, or SETDB1/PER3 signature in CRC patients. G IHC staining of SETDB1 in paired tumor and adjacent normal tissues (HOrgC180PG01-2, n = 90); Quantification of SETDB1 H IHC staining intensity in paired tissues using ImageJ. I Schematic model summarizing the proposed regulatory mechanism by which the SETDB1/CRL4B complex modulates circadian rhythm during CRC tumorigenesis. Data are presented as mean ± SD or SEM. Statistical significance was determined using a two-tailed unpaired t-test or one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001.

Discussion

In this study, we demonstrate that SETDB1 functions as a potent oncogene by cooperating with the CRL4B complex to mediate transcriptional repression and disrupt the circadian rhythm in CRC cells, thereby promoting cell proliferation, invasion, and stemness (Fig. 7I). These findings suggest that the SETDB1/CRL4B complex may serve as a promising molecular therapeutic target for CRC.

SETDB1 functions as a histone methyltransferase that catalyzes di- and tri-methylation of histone H3 at lysine 9 (H3K9me2/3), thereby regulating core biological processes such as cell differentiation and proliferation, metabolism, and immune responses, including tumor immunity [13, 16, 18, 19, 75, 76]. During embryonic development, SETDB1, in combination with NSD1, marks heterochromatin regions through H3K9me3 and H3K36me3 modifications. Loss of SETDB1 activates genes normally repressed by H3K9me3, with H3K36me3-marked enhancer regions driving specific gene expression and cellular identity acquisition [26]. In lung cancer, aberrant SETDB1 expression leads to excessive accumulation of H3K9me3. SETDB1 loss activates epithelial gene programs, remodels the cytoskeleton, reduces proliferation and migration, and reverses malignant potential in lung adenocarcinoma [37]. In breast cancer, SETDB1 enhances the invasive capacity of breast cancer cells by directly inhibiting the expression of Snail, a key regulator during EMT [36]. At the translational and post-transcriptional modification (PTM) levels, SETDB1 increases the translation of c-Myc and CCND1 and methylates Polo-like kinase 3 (PLK3), thereby enhancing its kinase activity and further activating the HIF1α signaling pathway in breast cancer [25, 77, 78]. Additionally, in non-small cell lung cancer (NSCLC), SETDB1 methylates AKT, leading to constitutive activation of the AKT pathway, thereby promoting tumor cell proliferation, migration, and chemoresistance [79]. Accordingly, SETDB1 is implicated in the pathogenesis of multiple cancers. In this study, we observed that SETDB1 is highly expressed in CRC and promotes the proliferation and invasion of CRC cells and enhances CSC-like characteristics. Moreover, SETDB1 has been reported to repress p53 and p21 transcription to facilitate CRC development [80, 81], and to target FOSB, thereby inhibiting the FOSB/miR-22/BATF3/PD-L1 axis and promoting immune escape of CRC cells [82]. SETDB1 has also been reported to activate the STAT1–CCND1/CDK6 signaling axis by targeting STAT1, accelerating CRC cell cycle progression [77]. Consistent with this, our transcriptomic analysis revealed that SETDB1 knockdown upregulates TNF, HIF1, and TP53 signaling in CRC, indicating that SETDB1 plays a crucial role in tumorigenesis, immune evasion, and suppression of tumor immunity. At the PTM level, SETDB1 has been reported to methylate β-catenin in CRC, leading to aberrant activation of the WNT signaling pathway [83], while methylation of monocarboxylate transporter 1 (MCT1) enhances tumor glycolysis and M2-like polarization of tumor-associated macrophages, thereby promoting lactate shuttling and CRC progression [13]. SETDB1 is also recognized as an epigenetic checkpoint that suppresses tumor innate immunogenicity, highlighting its potential as a target for immunotherapy [84]. Collectively, these findings suggest that SETDB1 is a potent proto-oncogene in CRC and a promising biomarker for the diagnosis, treatment, and management of patients.

This study examined the molecular mechanisms of SETDB1 in CRC and revealed that SETDB1 recruits the CRL4B complex. The core scaffold protein in the CRL4B complex, CUL4B, is a member of the Cullin-RING ligase (CRL) family, the largest class of ubiquitin E3 ligases [43]. Unlike its homolog CUL4A, CUL4B contains a nuclear localization sequence that facilitates chromatin binding, where it catalyzes monoubiquitination of histone H2A at lysine 119 (H2AK119ub1) to mediate transcriptional repression, as demonstrated in our previous work [46]. Within the nucleus, CUL4B interacts with multiple transcriptional co-repressor complexes, such as the MTA1/NuRD complex, Sin3A/HDACs complex, NCoR/SMRT complex, CoREST complex, PRC2 complex, and the SUV39H1/HP1/DNMT3A complex [46–48]. Consistent with these interactions, CUL4B exerts oncogenic functions across various malignancies. In breast cancer, CUL4B is activated by the HIF1α signaling pathway in response to hypoxia, promoting tumor cell growth and invasion [47]. Simultaneously, the CRL4B complex can be recruited by multiple oncogenes to activate downstream mTOR and PPAR signaling pathways, reprogram fatty acid metabolism, and disrupt the circadian rhythm, thereby promoting the proliferation and metastasis of breast cancer cells [85–88]. In pancreatic cancer, CRL4B enhances CSC-like properties and disease aggressiveness [89], with similar pro-tumorigenic roles observed in gastric cancer [90], prostate cancer [91], and osteosarcoma [92]. Correspondingly, we found that CUL4B is upregulated in CRC and correlates with poor patient prognosis. Functionally, similar to SETDB1, CUL4B promotes the proliferation, invasion, and stemness of CRC cells. Notably, combined knockdown of SETDB1 and CUL4B resulted in a more pronounced suppression of malignant phenotypes compared to that with knockdown of either factor alone. Taken together, these findings establish the SETDB1/CRL4B complex as a critical driver in CRC progression and highlight its potential as a molecular target for CRC treatment. CSC-like characteristics represent a hallmark of aggressive tumor phenotypes [93–95]. Our results further revealed that the SETDB1/CRL4B complex significantly promotes not only tumor growth but also tumor-initiating capacity and CSC frequency in the nude mouse xenograft model, thereby underscoring its key role in promoting CRC stemness. Although the methods used in this study for detecting tumor stem cells are not perfect, still lack continuous sphere formation and ALDH activity detection methods. We plan to follow up on these detections in subsequent studies to further confirm the effect of this complex in promoting stem cell characteristics. Beyond these immediate experimental extensions, future investigations will aim to dissect the molecular mechanisms by which the SETDB1/CRL4B axis regulates core stemness‑associated pathways and to clarify whether its function is cell‑intrinsic or modulated by microenvironmental cues. Ultimately, exploring the dependency of CRC stem‑like cells on this transcriptional co‑repressor complex may offer novel opportunities for combination therapies designed to eradicate the CSC compartment, thereby improving long‑term clinical outcomes in CRC.

To further delineate the mechanistic role of the SETDB1/CRL4B complex in CRC cells, CUT&Tag-seq analysis revealed that the SETDB1/CRL4B complex coordinately regulates a series of downstream signaling pathways, including those involved in the cell cycle, metabolism, autophagy, and circadian rhythm. The circadian rhythm operates via a transcriptional–translational negative feedback loop to maintain 24-h physiological rhythms, and its disruption can lead to systemic phenotypic disorganization, disease, and tumorigenesis [53, 54, 67]. We found that the SETDB1/CRL4B complex co-occupies the promoter regions of the core circadian genes PER3 and CRY1, where it transcriptionally represses their expression by depositing the repressive histone marks H3K9me3 and H2AK119ub1. This expands the mechanistic framework of SETDB1 from a solitary epigenetic modifier to a dual‑functional machine, achieved through its assembly with CRL4B into a co‑repressive complex. Moreover, it offers a unifying mechanism that connects this complex both to upstream circadian disruption and to downstream tumor‑promoting outcomes, including enhanced stemness, proliferation, and invasion.

PER and CRY proteins are regulated by the CLOCK–BMAL1 heterodimer, and their expression accumulates during the day, after which they translocate to the nucleus at night, where they interact with CLOCK–BMAL1 to repress their own transcription [55, 56]. Dysregulation of these core repressors disrupts the entire circadian feedback loop and is a recognized contributor to cancer development. Beyond canonical cell‑cycle and metabolic effects, accumulating evidence implicates circadian disruption in the acquisition of tumor‑intrinsic stemness properties [64, 96, 97]. Aberrant circadian signaling rewires transcriptional networks that govern self‑renewal, epithelial–mesenchymal plasticity, and the expansion of tumor‑initiating cell populations, thereby fostering intratumoral heterogeneity and therapeutic resistance. This circadian‑to‑stemness axis is emerging as a critical node linking chrono disruption to aggressive tumor behavior.

Functioning as a TSG, PER3 suppresses cancer progression by repressing the AMPK/mTOR pathway in lung cancer [98]. In prostate cancer, low expression of PER3 stimulates BMAL1 expression, leading to phosphorylation of β-catenin and activation of the WNT/β-catenin pathway [99]. Meanwhile, low PER3 expression is closely associated with the development of CRC and can also be used as a prognostic marker in patients with CRC [100, 101]. Studies have shown that overexpression of PER3 inhibits Notch and β-catenin signaling pathways and reduces the chemoresistance and self-renewal capacity of CRC stem cells [102]. Similarly, CRY1, another key circadian regulator targeted by SETDB1/CRL4B, is considered a tumor suppressor and is frequently downregulated in CRC [71, 103]. Therefore, suppression of PER3 and CRY1 by the SETDB1/CRL4B complex provides a mechanistic explanation for circadian disruption in CRC cells and its contribution to tumorigenesis. Our study identifies this transcriptional co-repressor complex as a novel upstream regulator of circadian gene expression in CRC, thereby establishing a foundation for further investigation into the role of circadian dysregulation in CRC progression.

More importantly, disruption of circadian rhythm influences the tumor microenvironment surrounding CRC cells, promoting metastasis and other malignant behaviors by facilitating tumor immune evasion and thereby presenting a significant therapeutic challenge [64, 104]. In this study, we demonstrate that the SETDB1/CRL4B complex destabilizes the circadian rhythm in CRC by transcriptionally repressing PER3 and CRY1. Notably, inhibition of PER3 and CRY1 expression following SETDB1/CRL4B knockdown was sufficient to substantially rescue the oncogenic capacity of CRC cells. Collectively, these findings indicate that targeting the SETDB1/CRL4B complex restores circadian rhythm and attenuates tumor malignancy. Therefore, the SETDB1/CRL4B complex represents a viable target for modulating circadian rhythm of cancer cells and holds promising clinical potential for CRC therapy. Furthermore, we observed that the SETDB1/CRL4B complex is highly expressed not only in CRC but also in multiple other malignancies, including thyroid cancer and hepatocellular carcinoma, as well as breast and lung cancers discussed above, underscoring its broad relevance as a potential therapeutic target across cancer types.

We acknowledge that our current in vivo evidence relies predominantly on subcutaneous xenografts and AOM–DSS‑derived organoids, which only partially reflect the native tumor microenvironment, especially its immune and stromal elements, and that the absence of available CUL4B‑targeting inhibitors precludes immediate pharmacological co‑targeting studies, although our data provide a mechanistic rationale for such strategies. Furthermore, the lack of orthotopic or immunocompetent models, as well as the retrospective nature of our clinical prognostic analyses, underscores the need for prospective validation in independent cohorts. Notably, after adjusting for major clinicopathological covariates, multivariate Cox regression did not yield statistical significance for either SETDB1 (HR = 1.20, 95% CI: 0.8–1.9, P = 0.35) or CUL4B (HR = 1.00, 95% CI: 0.79–1.4, P = 0.75) as independent prognostic factors. Nevertheless, the hazard ratio for SETDB1 remains directionally consistent with a trend toward unfavorable prognosis, and the loss of significance in the multivariate model may partly reflect strong correlations between SETDB1 expression and covariates such as age and sex, which could obscure its independent effect. These observations highlight the importance of refined stratification in future studies; accordingly, we plan to investigate whether the SETDB1/CRL4B complex exerts context‑dependent functions across distinct CRC subtypes and patient backgrounds, which will help delineate its specific prognostic and biological roles. Collectively, addressing these limitations and exploring subtype‑specific mechanisms will be essential to consolidate the translational and mechanistic relevance of the SETDB1/CRL4B–circadian rhythm axis in colorectal cancer.

In summary, we identified the SETDB1/CRL4B complex as an important oncogenic driver in CRC development. SETDB1 inhibits the expression of multiple genes, including the core circadian regulators PER3 and CRY1, by recruiting the CRL4B complex and depositing histone marks H3K9me3 and H2AK119ub1. The SETDB1/CRL4B complex disrupts circadian rhythm, promotes tumor cell proliferation, invasion, and stemness, and thereby facilitates CRC initiation and progression. The SETDB1/CRL4B complex represents a promising therapeutic target for regulating circadian rhythm and treating CRC.

Author contributions

YW conceived this project. MZ, DY, JZ, TG, JC, TH, TM, YW, MH, DZ, YL, JZ, CG, XH, TZ, JG, ZT, and BY mainly conducted experiments. MZ and DY acquired data. MZ, DY, and YW analyzed data. MZ, DY, and YW wrote the manuscript.

Funding

This work was supported by grants to Yan Wang from the National Natural Science Foundation of China (42125707), the National Key Research and Development Program (2022YFA1103402), and the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (2025-I2M-XHJC-017, 2021-I2M-1-018, 2023-I2M-2-004); the National Natural Science Foundation of China (82303456 to Min Zhang, 82504167 to Jingyao Zhang, 42407568 to Baowen Yuan), the Fundamental Research Funds for the Central Universities, Peking Union Medical College (3332024054 to Jingyao Zhang, 3332024211 to Baowen Yuan), and the China Postdoctoral Science Foundation (2024M760260) to Jingyao Zhang.

Data availability

The data of this study are available from the corresponding author upon reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41388-026-03972-w.