All mouse experiments were approved by the Laboratory Animal Ethics Committee of Tongji Hospital and conducted in accordance with relevant guidelines and regulations.

Fluorescein isothiocyanate (FITC)-dextran 4000 (FD-4), lipopolysaccharide (LPS), 3-methyladenine (3-MA), wortmannin, and rapamycin were purchased from Sigma-Aldrich (St. Louis, MO). Tumor necrosis factor (TNF)-α and interferon (IFN)-γ were obtained from PeproTech (Rocky Hill, NJ). Dulbecco’s modified Eagle’s medium and fetal bovine serum were procured from GIBCO (Carlsbad, CA). Negative control lentivirus (LV-Vector), Bmal1-expressing lentivirus (LV-Bmal1), negative control-shRNA lentivirus (LV-shNC), and Bmal1-shRNA lentivirus (LV-shBmal1) were purchased from GeneChem (Shanghai, China). Primary antibodies against Bmal1 (1:1000), Bax (1:1000), and cleaved caspase-3 (1:500) were purchased from Abcam (Cambridge, UK). Bcl-2 (1:1000) was purchased from Abclonal (Wuhan, China). ZO-1 (1:5000), Occludin (1:5000), Claudin-1 (1:1000), and β-actin (1:5000) were purchased from Proteintech (Wuhan, China). Antibodies against LC3B (1:1000), Beclin-1 (1:1000), and P62 (1:1000) were purchased from Cell Signaling Technology (Danvers, MA).

The Caco-2 cell line was purchased from the China Center for Type Culture Collection (Wuhan, China). These cells were routinely cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin solution in a humidified atmosphere (95% air/5% CO₂) at 37°C. The cultured medium was replaced every 2 days.

Caco-2 cells were infected with lentivirus to overexpress Bmal1 (LV-Bmal1) or knockdown Bmal1 (LV-shBmal1). After 72 hours of infection, the cells were then screened for 2 weeks using 5 μg/ml puromycin. Selected cells were identified as stable cell lines and used for the next experiments. The shBmal1 sequence was GCAGAATGTCATAGGCAAGT.

To visualize autolysosomes/autophagosomes, mRFP-GFP-LC3 lentivirus was transfected into Caco-2 cells. After appropriate interventions, the cells were fixed with 4% paraformaldehyde and observed with a Nikon Eclipse Ti laser confocal microscope (Nikon, Tokyo, Japan) after being sealed. Autophagosomes were characterized by co-localized green fluorescent protein (GFP) and red fluorescent protein (RFP) fluorescence, whereas autolysosomes showed RFP dots without GFP.

The integrity and permeability of the Caco-2 monolayer were determined by estimating transepithelial electrical resistance (TEER) and FD-4 permeability as described previously (34, 35). Briefly, Caco-2 cells were seeded at a density of 1 × 10⁵/cm² cells on 0.4-μm 12-well hanging cell culture inserts (Corning, NY) and cultured for 21 days, with fluid changes every 2 days for the first 7 days and daily for the last 14 days. When cells were raised to day 21, Transwell plates were slowly washed 3 times with Hank’s balanced salt solution (HBSS) pre-warmed to 37°C. TEER values were measured using an EMD Millipore Millicell-ERS2 Volt-Ohm Meter (Millipore). The entire process was carried out at a 37°C constant temperature to guarantee the stability and accuracy of the value. Three different areas were selected for measuring the TEER value of each well. The final TEER value for each well was the average of the measurements for each well minus the blank control value multiplied by the area of each well of the Transwell plate. To measure the permeability of Caco-2 monolayers, monolayers pretreated with or without 10 ng/ml TNF-α/IFN-γ for 6 hours were washed twice with HBSS, and then 200 μL of FD-4 solution (1 mg/mL in HBSS) was added. After 2 hours of incubation, the concentration of FD-4 in samples extracted from the substrate chamber was evaluated using a fluorescent 96-well plate reader (Biotek, Vermont) at an excitation wavelength of 480 nm and an emission wavelength of 525 nm.

To investigate the effect of CRD on colitis, 120 male C57BL/6 mice were randomly divided into 4 groups: (a) wild type (WT) group; (b) DSS: WT group; (c) CRD group; and (d) DSS: CRD group. WT group mice were kept in a normal light-dark cycle environment (12 hours light, 12 hours dark) and CRD group mice were kept under constant light for 4 weeks (36). A concentration of 3% DSS (MP Biomedicals, Solon, OH) was added to the drinking water of mice for 7 days to construct the colitis model according to previous studies (37). In the DSS: CRD group, mice were kept under constant light for four weeks and 3% DSS was added to the drinking water at the beginning of the fourth week for seven days. Finally, all mice were sacrificed at the end of the fourth week.

To investigate the association between Bmal1 and autophagy and their role in colitis in vivo, we used Bmal1 knockout male C57BL/6 mice provided by Shulaibao Biotechnology (Wuhan, China). The strain information for Bmal1-/- mice is provided at https://www.jax.org/strain/009100. We divided the mice into 4 groups: (a) WT group; (b) DSS: WT group; (c) DSS: Bmal1-/- group, and (d) DSS: Bmal1-/- + rapamycin group (2 mg/kg/d, intragastric administration for 7 days), each group consisted of six mice.

The mice colitis model was modeled as described above. All mice were sacrificed at the end of the seventh day. Mice were weighed daily and checked for blood in the feces.

The daily observation indices during mouse colitis modeling mainly included changes in body weight, fecal properties, and blood in the feces. The disease activity index (DAI) was calculated as follows: DAI = (weight loss score + fecal trait score + blood in the feces)/3. The scale of the DAI is as follows: 0 points indicates no weight loss, normal fecal traits, and no blood in the feces; 1 point indicates weight loss of 1% to 5%, loose stools, and stool occult blood; 2 points indicate weight loss of 5% to 10%, loose stools, and stool occult blood; 3 points indicate weight loss of 10% to 15%, loose stools, and bloody stools; and 4 points indicate weight loss of ≥ 15%, loose stools, and bloody stools.

A histological evaluation and scoring of mouse colon tissue samples were performed according to criteria that have been previously determined (38).

The intestinal permeability of mice was measured according to a published protocol (34).

RNA from cells or colon tissues was isolated using the FastPure Cell/Tissue Total RNA Isolation Kit and converted to cDNA using the HiScript II First Strand cDNA Synthesis Kit according to standard protocols. ChamQ Universal SYBR qPCR Master Mix was used for quantitative reverse transcription-polymerase chain reaction (qRT-PCR) amplification. The above-mentioned reagents were supplied by Vazyme (Nanjing, China). The qRT-PCR primers used in this study were synthesized and provided by Tsingke Biotechnology (Beijing, China), and the sequences are shown in Table 1 .

To extract cellular or mouse colon proteins after different treatment steps, samples were lysed in RIPA lysis buffer (Servicebio Technology, Wuhan, China) with a protease inhibitor cocktail (MedChemExpress, NJ). Protein concentrations were quantified using the BCA Protein Kit (Servicebio Technology). The supernatant of the lysate was mixed with the upwelling buffer (Servicebio Technology) and boiled at 100°C for 10 minutes. A total of 30 μg of total cellular protein or 50 μg of total tissue protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to a polyvinylidene difluoride membrane (Millipore). The membrane was incubated overnight at 4°C in a configured specific primary antibody after being blocked with 5% skim milk for 1.5 hours. On the following day, the membrane was washed 3 times with tris-buffered saline Tween-20 for 10 minutes each time, followed by incubation with horseradish peroxidase-conjugated secondary antibody for 1.5 hour at room temperature. The membrane was then washed 3 times with tris-buffered saline Tween-20 for 10 minutes each time, and finally, protein bands were visualized using an ECL kit (NCM Biotech, Suzhou, China).

Mouse or human colon tissue was fixed in 4% paraformaldehyde, followed by paraffin embedding and sectioning. The prepared sections were used for subsequent immunohistochemical or immunofluorescence staining. To perform immunohistochemical staining, the sections were deparaffinized, hydrated, performed antigen retrieval, quenched by endogenous peroxidase, and blocked before dropwise addition of specific primary antibodies to the sections.

After overnight incubation of the specific primary antibody at 4°, IgG-HRP secondary antibody was added to the slide, and DAB was performed to observe the visible brown staining.

Colon sections were processed in the same way as for immunohistochemistry. The secondary antibody used was CoraLite594-conjugated goat anti-rabbit IgG (H + L) antibody (Proteintech), and the nuclei were stained with 4’,6-diamidino-2-phenylindole (Servicebio Technology).

To perform cell immunofluorescence staining, the cells were fixed with 4% paraformaldehyde for 20 minutes at room temperature for Caco-2 cells or with methanol for 10 minutes at −20°C for Caco-2 monolayers. After permeabilization and blocking, the cells were incubated with specific primary antibodies overnight at 4°. The following day, the cells were incubated with CoraLite594-conjugated goat anti-rabbit IgG (H + L) secondary antibody for 1 hour in the dark. Subsequently, the nuclei were stained with 4’,6-diamidino-2-phenylindole, and the cells were observed by fluorescence microscopy.

The apoptosis rate of treated cells was assessed using flow cytometry. The Apoptosis Detection Kit (Yeasen, Shanghai, China) was used in this experiment. The cells were washed twice with phosphate-buffered saline before resuspension with a binding solution. This was followed by the addition of 5 μl of annexin V-FITC and 10 μl of propidium iodide (PI) to the cell mixture and incubation for 15 minutes at room temperature. Finally, the stained cells were evaluated by a FACScan flow cytometer (Becton Dickinson, San Diego, CA).

Treated cells were stained to assess their apoptosis rate using the One-Step TUNEL Apoptosis Detection Kit (Yeasen) according to the manufacturer’s instructions.

Transmission electron microscopy was used to visualize subcellular structures. Collected cells or colon tissue were fixed with 2.5% glutaraldehyde for 1 hour at room temperature, followed by fixation with osmium tetroxide and embedding in Spurr’s Epon. Ultra-thin sections (60 nm) were then cut and stained with uranyl acetate for 3 minutes. The sections were observed with a Hitachi 7500 electron microscope and photographed with a digital camera. The numbers of autophagosomes and autolysosomes were counted and plotted as histograms.

GraphPad Prism Software Version 8.0 (La Jolla, CA) was used to analyze the data. The data are shown as the mean ± standard deviation (SD). Significant differences between groups were analyzed by Student’s t-test, one-way analysis of variance (ANOVA) or two-way ANOVA, as appropriate. A value of p < 0.05 was considered statistically significant.

To study the effects of CRD on UC, we modeled CRD and CRD combined with acute colitis (DSS: CRD) by subjecting mice to constant light environment for 4 weeks and administering %3 DSS or no DSS for 7 days at the beginning of the fourth week. We found that mRNA expression levels of the core biological clock genes Bmal1, Clock, Per1, Per2, Cry1, and Cry2 in mouse colon tissue oscillated in a 24-hour cycle in the WT group, indicating clock rhythmicity. However, rhythmicity was disrupted in the colon tissue of mice raised in a constant light environment, which indicated successful construction of the CRD model ( Figure 1A ). Among the six biological clock genes in the CRD group compared with the WT group, Bmal1 mRNA expression was the only one that was significantly decreased at all time points. Western blot showed that Bmal1 protein levels in the CRD group were significantly lower than those in the WT group at zeitgeber time (ZT) 0 and ZT12 ( Figure 1B , Supplementary Figure 1A ), which is consistent with the findings of a previous study (39). We then examined the effect of CRD on acute colitis. We found that CRD significantly exacerbated DSS-induced acute colitis, as shown by a more pronounced weight loss ( Figure 1C ), a higher disease activity index ( Figure 1D ), shortened colon length ( Figures 1E, F ), marked disruption of epithelial crypt structure, and more immune cell infiltration ( Figures 1G, H ). However, CRD alone in the absence of colitis did not significantly affect the appearance or histological structure of the mouse colon ( Figures 1E–H ).

Previous studies have shown that the genes encoding the TJ proteins Claudin-1 and Occludin are expressed in a daily rhythm in WT mice (12). Additionally, excessive IECs apoptosis and intestinal barrier dysfunction are key factors in the pathogenesis of UC (40). We found that CRD did not significantly affect IEC apoptosis and intestinal barrier integrity in mice that were not subjected to DSS treatment. This finding was confirmed by the similar intestinal permeability ( Figure 2A ), apoptosis-related proteins (cleaved caspase-3, Bax and Bcl-2), TJ protein levels ( Figures 2B, C , Supplementary Figures 2A, B ), Periodic Acid-Schiff (PAS) staining ( Figure 2D ) and Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labelling (TUNEL) staining in WT group and CRD group mice ( Figure 2E and Supplementary Figure 2C ). However, DSS: CRD group mice showed significantly higher intestinal permeability ( Figure 2A ), decreased TJ protein levels ( Figures 2B, C , Supplementary Figure 2A, B ), and more apoptotic IECs ( Figures 2B, E and Supplementary Figure 2C ) compared with DSS: WT group mice. DSS: CRD group mice also had fewer Goblet cells ( Figure 2D ), which protect the intestinal epithelial barrier function by secreting mucus (41). These results suggested that CRD aggravated impairment of intestinal barrier function and the apoptosis rate of IECs in the DSS-induced colitis model.

As mentioned above, Bmal1 mRNA expression was reduced in the CRD group at all 5 time points, and Bmal1 protein levels were reduced at ZT0 and ZT12 compared with the WT group ( Figures 1A, B and Supplementary Figure 1A ). Furthermore, previous studies have shown that Bmal1 is the central transcriptional activator of the mammalian biological clock, and its disruption or absence leads to complete loss of circadian rhythms in organisms (24). Therefore, we hypothesized that decreased Bmal1 expression plays a major role in CRD affecting the pathogenesis of UC. Our qRT-PCR, Western blot and immunofluorescence staining results showed that Bmal1 mRNA and protein levels of colon tissue in the DSS group were significantly lower than those in the WT group ( Figures 3A–D and Supplementary Figure 3A ). All colon tissues used for intercomparison were collected at the same time. Furthermore, our in vitro experiments showed that Bmal1 expression in Caco-2 cells was decreased after 12 hours of incubation with 10 ng/ml TNF-α/IFN-γ ( Figures 3E–G ). These results suggested that reduced Bmal1 expression may be involved in the pathogenesis of UC.

As shown above, mice in the CRD group had decreased Bmal1 expression, and aggravated impairment of intestinal epithelial barrier function and increased apoptosis were observed in DSS: CRD group compared with DSS: WT group. To further investigate the potential regulatory effects of Bmal1 on intestinal epithelial barrier function, we transfected Caco-2 cells using LV-Bmal1 and LV-Vector, and verified Bmal1 mRNA expression by qRT-PCR ( Figure 4A ) and western blot ( Figure 4C ). In addition, mRNA and protein levels of Claudin-1, Occludin and ZO-1 were higher in the LV-Bmal1 group than in the LV-Vector group ( Figures 4B, C and Supplementary Figure 4A ). Measurements of TEER values in Caco-2 monolayers showed that Bmal1 overexpression enhanced the barrier function of Caco-2 monolayers ( Figure 4D ). Additionally, TEER values of Caco-2 monolayers expressing high Bmal1 levels declined to a lesser extent after the administration of TNF-α/IFN-γ ( Figure 4E ). This finding suggested that Bmal1 overexpression attenuated TNF-α/IFN-γ-induced impairment of the monolayer membrane barrier function in Caco-2 cells. FITC-dextran flux experiments also showed that FITC-dextran was less likely to pass through monolayers in the LV-Bmal1 group than in the LV-Vector group, regardless of the presence of TNF-α/IFN-γ ( Figure 4F ). Immunofluorescence staining showed that overexpression of Bmal1 in Caco-2 monolayers prevented the decrease in expression and disruption of localization of TJ proteins induced by TNF-α/IFN-γ ( Figure 4G ). Western blot showed that Bmal1 overexpression alleviated LPS-induced decreased TJ protein expression in Caco-2 cells ( Figure 4H and Supplementary Figure 4C ). Previous studies have reported that excessive epithelial cell death can lead to disruption of the intestinal barrier (42). Therefore, we evaluated the effect of Bmal1 on epithelial cell apoptosis. We knocked down Bmal1 in Caco-2 cells by transfecting lentivirus. Western blot, flow cytometry, and TUNEL staining suggested that Bmal1 knockdown induced apoptosis in Caco-2 cells and elevated cleaved caspase-3 Bax protein levels but decreased Bcl-2 protein level. ( Figures 4I–L and Supplementary Figures 2B, D ). However, Bmal1 overexpression alleviated the LPS-induced apoptosis in Caco-2 cells ( Figures 4M–O and Supplementary Figures 4E, F ), which is an in vitro model for mimicking IEC epithelial barrier damage and apoptosis in vivo (43).

The above-mentioned results suggest that Bmal1 can regulate the barrier function of IECs by modulating the expression of TJ proteins and apoptosis of IECs.

Specific deletion of Bmal1 in astrocytes affects autophagy and protein degradation kinetics (44), and Bmal1 can mediate the process of intervertebral disc degeneration by regulating autophagy (32). In addition, autophagy maintains intestinal barrier function by reducing IEC death and regulating TJ proteins (29, 30). To determine whether Bmal1 regulates IEC barrier function by modulating autophagy, we used transmission electron microscopy to visualize subcellular structures. The number of autophagosomes was significantly higher in the LV-Bmal1 group than in the LV-Vector group ( Supplementary Figure 5A and Figure 5A ). Western blot showed that the LC3II:I ratio and Beclin-1 levels were higher in the LV-Bmal1 group than in the LV-Vector group. However, P62 protein levels were significantly lower in the LV-Bmal1 group than in the LV-Vector group ( Figures 5B, C ). The western blot findings were confirmed by immunofluorescence staining of LC3B and P62 in the LV-Bmal1 and LV-Vector groups ( Supplementary Figure 5B ). These results suggested that Bmal1 overexpression in Caco-2 cells activated autophagy. To investigate the effect of Bmal1 on autophagic flux in Caco-2 cells in the presence of LPS, we transfected the cells with the mRFP-GFP-LC3 lentiviral reporter gene. Yellow spots fused by red (mRFP) and green spots (GFP) indicated autophagic vesicles, while autophagic lysosomes were shown as red spots because the acidic environment quenched the fluorescence of GFP. We found that the numbers of autophagosomes and autolysosomes were significantly higher in the LPS + LV-Bmal1 group than in the LPS group. However, when the autophagy inhibitors 3-MA and wortmannin were added, autophagic vesicles and autophagic lysosomes were reduced ( Figures 5D, E ). These findings suggested that the activation of autophagy by Bmal1 was abolished, which was also confirmed by transmission electron microscopy ( Supplementary Figure 5C and Figure 5F ). Western blot showed that the LC3II:I ratio and Beclin-1 levels were decreased and P62 levels were increased by the administration of autophagy inhibitors, and lower levels of TJ proteins (Claudin-1, Occludin, and ZO-1) and higher levels of apoptosis-related proteins (cleaved caspase-3 and Bax) can also be observed in the wortmannin- or 3-MA-treated group than in the LPS + LV-Bmal1 group ( Figure 5G and Supplementary Figure 5D ). Flow cytometry showed that the inhibition of apoptosis by Bmal1 overexpression was partly abolished by autophagy inhibitors ( Supplementary Figure 5E ). Furthermore, we also observed that the protective effect of overexpression of Bmal1 on Caco-2 monolayers was partly deprived when TNF-α/IFN-γ stimulation was applied to Caco-2 monolayers pretreated with autophagy inhibitors wortmannin or 3-MA ( Figures 5H–J and Supplementary Figure 5D ). The above-mentioned results indicated that the protective effect of Bmal1 on the barrier function of Caco-2 cells was weakened by the administration of autophagy inhibitors exogenously. In summary, Bmal1 regulated the barrier function of IECs by activating autophagy.

We investigated the role of Bmal1 in UC using the Bmal1-/- mouse strain. Western blot analysis of proteins extracted from whole colon tissue showed that Bmal1 was ablated in colon tissue ( Supplementary Figure 6A ).Our results showed that Bmal1-/- and WT groups had similar colon lengths ( Supplementary Figures 6B, C ), and there was no significant difference in the histological structure or distribution of Goblet cells between the groups of mice. Additionally, the proliferative activities of the two groups were approximately the same ( Supplementary Figure 6D ). These results suggested that knockout of Bmal1 did not significantly affect mouse colonic function under baseline conditions.

We assessed the severity of colitis and found that the DSS: Bmal1-/- group showed significantly greater weight loss ( Figure 6A ) and a higher DAI score ( Figure 6B ) than the DSS: WT group. The length of the colon was significantly shorter in the DSS: Bmal1-/- group than in the DSS: WT group ( Figures 6C, D ). Histopathological staining showed significant morphological differences in the colonic epithelium of DSS: Bmal1-/- group mice, as shown by more extensive epithelial detachment, ulceration, disruption of crypt structures, and thickening of muscular tissue than in the DSS: WT group. These findings resulted in significantly higher histological scores in DSS: Bmal1-/- group than in the DSS: WT group ( Figures 6E, F ). We also found that the use of the autophagy agonist rapamycin (DSS: Bmal1-/- + rapamycin group) alleviated colitis exacerbated by Bmal1 knockout ( Figures 6A–F ). We then further confirmed the role of the Bmal1–autophagy–gut barrier function axis in the pathogenesis of colitis. The DSS: Bmal1-/- group showed more severe impaired autophagy than the DSS: WT group, as shown by the results of transmission electron microscopy ( Supplementary Figure 7A and Figure 6G ), a decreased LC3II:I ratio and Beclin-1 protein levels, increased P62 protein levels ( Figure 6H and Supplementary Figure 7C ) and LC3B immunofluorescence staining of colon tissue sections ( Supplementary Figure 7B ). DSS: Bmal1-/- group mice had significantly higher intestinal permeability ( Supplementary Figure 7D ), lower TJ protein (Claudin-1, Occludin, and ZO-1) levels ( Figures 6I, J and Supplementary Figures 7E, F ), fewer Goblet cells ( Figure 6K ), higher protein levels of cleaved-caspase-3, Bax ( Figures 6I and Supplementary Figure 7E ) and more apoptotic IECs compared with DSS: WT group ( Figure 6L and Supplementary Figure 7G ). Furthermore, we also observed that all of the above-mentioned conditions were relieved with the use of rapamycin (DSS: Bmal1-/- + rapamycin group) ( Figures 6G–L and Supplementary Figures 7A–G ). In conclusion, we demonstrated that Bmal1 deficiency exacerbated DSS-induced colitis by impairing intestinal barrier function through impaired autophagy.