Abstract

   Promoting tumor cell senescence arrests the cell cycle of tumor cells
   and activates the immune system to eliminate these senescent cells,
   thereby suppressing tumor growth. Nevertheless, PD‐L1 positive
   senescent tumor cells resist immune clearance and possess the ability
   to secret various cytokines and inflammatory factors that stimulate the
   growth of tumor cells. Consequently, drugs capable of both triggering
   senescence in tumor cells and concurrently diminishing the expression
   of PD‐L1 to counteract immune evasion are urgently needed. Here, a
   berberine derivative B68 is developed, which specifically induces tumor
   cell senescence by targeting BMI1. B68 also involves the degradation of
   PD‐L1 by targeting CSN5, thereby disrupting the immunosuppressive
   PD‐1/PD‐L1 interaction and enabling rapid clearance of senescent tumor
   cells. This approach simultaneously inhibits tumor progression and
   activates T cell immunity, as evidenced by the robust antitumor
   response following B68‐induced immunization of senescent cancer cells.
   Moreover, the synergistic effect of B68 with anti‐CTLA4 therapy further
   enhances antitumor immunity, and its ability to induce senescence in
   cancer cells triggers a strong protective response by dendritic and
   CD8^+ T cells. These findings provide a scientific basis for developing
   a new tumor treatment strategy based on senescence induction and lay
   the foundation for further preclinical research.

   Keywords: BMI1, colorectal cancer, CSN5, PD‐L1, tumor cell senescence
     __________________________________________________________________

   B68 uniquely induces tumor cell senescence by directly targeting BMI1,
   which blocks the cell cycle and inhibits tumor growth, and also targets
   CSN5 to reduce the expression of PD‐L1, which blocks PD‐1/PD‐L1
   immunosuppression, thereby enhancing the ability of the immune system
   to eliminate senescent tumor cells and exert antitumor effects,
   providing new research ideas and strategies for the treatment of
   colorectal cancer.

   graphic file with name ADVS-12-2413122-g007.jpg

1. Introduction

   Colorectal cancer (CRC) is the third most common cancer worldwide and
   the second leading cause of cancer‐related death.^[ [62]^1 , [63]^2 ^]
   Currently, surgical resection and chemotherapy are the main options for
   CRC treatment and may improve the 5‐year survival rate of patients.^[
   [64]^3 , [65]^4 ^] However, these chemotherapeutic drugs primarily
   induce tumor cell apoptosis, which does not yield satisfactory
   therapeutic outcomes and is accompanied by significant drug resistance
   and toxic side effects.^[ [66]^5 , [67]^6 , [68]^7 ^] Since the
   progression of tumors to malignant cancers necessitates overcoming the
   constraints of senescence, activating the function of tumor suppressor
   genes or triggering senescence signals could induce senescence in tumor
   cells.^[ [69]^8 , [70]^9 ^] Furthermore, senescent tumor cells within
   the body can be eliminated by immune cells or other pathways, thereby
   effectively suppressing tumor progression.^[ [71]^8 , [72]^10 ^] Thus,
   inducing senescence in tumor cells may represent a potential
   therapeutic approach for cancer treatment.

   Cancer cells are typically exposed to a variety of known stressors that
   can trigger senescence, including oncogenic signals, replicative
   stress, hypoxia, reactive oxygen species, nutrient deprivation, and
   exposure to cytokines present in the tumor microenvironment, such as
   TGF‐β.^[ [73]^11 , [74]^12 ^] In addition, cytokines produced by immune
   cells can also induce senescence within tumors, such as IFNγ produced
   by Th1 cells. Senescent cancer cells themselves do not promote tumor
   growth due to their low or nonproliferative capacity. However, they can
   alter the tumor microenvironment through the senescence‐associated
   secretory phenotype (SASP).^[ [75]^13 , [76]^14 ^] Notably, the SASP of
   tumor cells can recruit and activate CD4^+ and CD8^+ T cells, thereby
   inducing antitumor protection.^[ [77]^15 ^] Furthermore, senescent cell
   vaccines may suppress tumor growth by inducing cancer cells to enter a
   state of senescence and be cleared by the immune system.^[ [78]^16 ^]
   Additionally, in some cases, senescent cell vaccines may help reverse
   the resistance of tumors to traditional chemotherapy by activating the
   immune system to clear drug‐resistant tumor cells.^[ [79]^17 ^]
   Therefore, cancer cell senescence has become a new frontier in drug
   development.

   However, senescent cells can also accelerate tumor development by
   secreting inflammatory cytokines, growth factors and MMPases, which
   promote tumor aggressiveness, metastatic ability, and angiogenesis,
   thereby providing them with the nutrients and oxygen needed for
   growth.^[ [80]^18 ^] Furthermore, studies have indicated that slightly
   senescent cells heterogeneously express the immune checkpoint protein
   programmed death‐ligand 1 (PD‐L1) and that the accumulation of
   PD‐L1‐positive senescent cells increases with the degree of aging.^[
   [81]^19 ^] As a T cell immune checkpoint molecule, PD‐L1 deactivates
   tumor‐infiltrating lymphocytes (TILs) that express Programmed cell
   death 1 (PD‐1) on the cell surface. The interaction between PD‐L1 and
   PD‐1 inhibits T cell immune activation and effector responses.^[
   [82]^20 , [83]^21 ^] Disruption of the PD‐1/PD‐L1 axis and reactivation
   of TILs are considered promising targets for cancer immunotherapy.^[
   [84]^22 ^] Since PD‐L1‐negative cells are sensitive to T cell
   surveillance, PD‐L1‐positive cells are resistant even in the presence
   of the SASP. Additionally, studies have shown that although cellular
   senescence is a stable state of cell cycle arrest, this phenomenon is
   not entirely irreversible, especially in therapy‐induced senescent
   cells, some of which can escape the state of cell cycle arrest, leading
   to cancer recurrence.^[ [85]^23 ^] Thus, in the context of cancer
   therapeutics, the discovery of drugs capable of inducing senescence in
   tumor cells while simultaneously activating the immune system through
   the downregulation of PD‐L1 expression could significantly enhance the
   immune response against cancer cells and facilitate the clearance of
   senescent cells.^[ [86]^24 ^]The realization of such a strategy would
   introduce a novel and more potent approach to cancer treatment.

   To date, a variety of prosenescence drugs, such as GRN163L, nutlin, and
   MIRA‐1, have been developed, some of which have entered clinical
   trials. Nevertheless, their systemic toxicity and drug resistance
   greatly limit their clinical application.^[ [87]^25 , [88]^26 ^]
   Although some reports indicate that natural products can decrease the
   expression of PD‐L1 in tumor cells, their clinical application is
   unknown.^[ [89]^27 , [90]^28 ^] Drugs that can both induce cellular
   senescence and reduce the expression of PD‐L1 in tumor cells have
   rarely been reported. B68 is a compound modified from quaternary
   isoquinoline alkaloids extracted from natural plants of the
   Berberidaceae, Ranunculaceae, Rutaceae, and other families. Berberine
   has been used for the treatment of diseases such as diabetes,
   gastroenteritis, polycystic ovary syndrome, etc.^[ [91]^29 ^] Previous
   studies have shown that berberine can inhibit the deubiquitination
   activity of CSN5, reduce the expression of PD‐L1 in cancer cells, and
   promote antitumor immunity.^[ [92]^30 ^] In this study, we found that
   B68, a derivative modified from berberine, targets CSN5 with an effect
   similar to that of berberine and can reduce the level of PD‐L1. In
   addition, B68 induces senescence in cancer cells by targeting BMI1,
   which berberine cannot do. Our study identified a drug that can promote
   the senescence of tumor cells while simultaneously decreasing PD‐L1
   expression in senescent cells, which will provide new prospects for the
   treatment of refractory cancer cells.

2. Results

2.1. B68 Notably Promoted Cell Senescence in Colorectal Cancer Cells

   Considering the limitations of berberine in its antitumor effects, we
   performed a series of structural modifications to berberine and
   obtained six products, namely, berberine (B1), B2, B3, B68, B5, B6, and
   B7.^[ [93]^31 ^] Through the CCK‐8 assay, we found that B68 had a
   significant inhibitory effect on tumor growth in RKO cells compared
   with that of berberine (B1), with an IC[50] of 1.03 µm (Figure
   [94]S1A,B, Supporting Information). Moreover, RKO and HCT116 cells are
   more sensitive to B68 than other CRC cells (Figure [95]S1C–E,
   Supporting Information), and B68 (0–20 µm) has almost no toxic effect
   on normal human intestinal epithelial cells (NCM460) (Figure [96]S1F,
   Supporting Information). To further assess the growth inhibition effect
   of B68 on HCT116 and RKO cells, a colony formation assay was performed.
   The results showed that B68 treatment significantly suppressed the
   colony‐forming ability of RKO and HCT116 cells (Figure [97]S1G,H,
   Supporting Information). Next, an EdU assay and calcein‐AM/PI
   double‐staining assay were performed to analyze the effect of B68 on
   CRC cell proliferation or cell death. As shown in Figures [98]S1I,J and
   [99]S2A–D (Supporting Information), B68 significantly inhibited the
   proliferation of RKO and HCT116 cells with little induction of
   apoptosis, which is consistent with the flow cytometry results.
   Moreover, B68 notably increased the percentage of RKO and HCT116 cells
   in G0/G1 phase, from 33.4% ± 1.3% in controls to 80.9% ± 6.4% and 37.8%
   ± 2.3% to 60% ± 1.2%, respectively (Figure [100]S2E,F, Supporting
   Information). Furthermore, we utilized scratch assays to determine the
   migratory potential of LOVO and CT26 cells. As shown in Figure
   [101]S3A,B (Supporting Information), the application of B68 resulted in
   a pronounced reduction in cell migration. Additionally, in RKO and
   HCT116 cells, B68 significantly altered the cell morphology from
   spindle‐shaped and round to flat and enlarged, and vacuoles were
   observed, especially in HCT116 cells, suggesting that B68 might induce
   cellular senescence (Figure [102]S2G, Supporting Information).
   Subsequent SA‐β‐gal activity assays and H3K9me3 immunofluorescence
   experiments also confirmed that B68 significantly promoted RKO and
   HCT116 cell senescence (Figure [103] 1A–D). It is well documented that
   cell cycle arrest in senescent cells is predominantly mediated through
   activation of the p53/p21 signaling pathway. Therefore, to further
   determine the mechanism of B68‐induced senescence, we evaluated p53 and
   its downstream target p21. As shown in Figure [104]1E–H, B68 robustly
   upregulated the transcription of p53 in CRC cells, consequently
   promoting the expression of p21. These findings suggest that B68 may
   initiate a senescence program in CRC cells via activation of the
   p53/p21 pathway. A prominent feature of cellular senescence is the
   acquisition of the SASP. We further explored whether B68 affects
   cellular senescence and SASP regulatory factors. As shown in
   Figure [105]1I and Figure [106]S4 (Supporting Information), similar to
   bleomycin, a positive control drug, B68 significantly increased the
   levels of IL‐6, BAX, BCL2, CXCL3, GM‐CSF, CXCL8, p16, p21, spink1, and
   AREG. Taken together, these results indicate that, unlike berberine,
   the modified berberine derivative B68 can inhibit cell growth by
   inducing senescence in tumor cells.

Figure 1.

   Figure 1
   [107]Open in a new tab

   B68 induces senescence in colon cancer cells. A) RKO and HCT116 cells
   were cultured with different concentrations of B68, B1, and bleomycin
   (BLEO) for one week, and cellular senescence was assessed by SA‐β‐gal
   activity staining, and quantification of SA‐β‐gal activity staining is
   shown in B). Scale bars, 200 µm (n = 3, error bars represent SEM,
   mean ± SEM, one‐way ANOVA). C) Representative images of H3K9me3
   staining of RKO and HCT116 cells after 7 days of treatment with B68.
   Scale bars, 100 µm. D) Quantitative results of H3K9me3 staining in RKO
   and HCT116 cells (n = 3, error bars represent SEM, mean ± SEM, one‐way
   ANOVA). E–H) The protein expression levels of p21 and p53 were detected
   by immunoblotting after the treatment of RKO and HCT116 cells with
   different concentrations of B68 for 7 days. The results of the
   quantification of the p21 and p53 protein levels are shown below
   (n = 3, error bars represent SEM, mean ± SEM, one‐way ANOVA). I)
   Quantitative RT‐PCR was conducted to assess the expression levels of
   mRNAs associated with the SASP following a 7‐day treatment of HCT116
   cells with B68 at a concentration of 15 µm and bleomycin at 10 mg mL^−1
   (n = 3, error bars represent SEM, mean ± SEM, one‐way ANOVA). n.s, not
   significant; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

2.2. B68 Induces Senescence to Suppress Colorectal Cancer Growth in Xenograft
Mice

   To further determine the antitumor effect of B68‐induced senescence,
   HCT116 and RKO xenograft nude mouse models were established. Before the
   formal experiment, our preliminary experiments indicated that
   8 mg kg^−1 B68 had significant antitumor effects (Figure [108]S5A–C,
   Supporting Information). Therefore, 8 mg kg^−1 B68 was used in
   subsequent animal experiments. Notably, B68 markedly suppressed HCT116
   and RKO tumor growth, tumor size and tumor weight, whereas berberine
   had no therapeutic effect (Figures [109]S6A–D and [110]S7A–D,
   Supporting Information). In addition, after intraperitoneal injection
   of B68 for 14 days, no significant alterations in average body weight
   were observed among the treatment groups (Figures [111]S6E and
   [112]S7E, Supporting Information). B68 treatment significantly
   increased the percentage of β‐gal‐positive senescent cells in tumor
   tissue, whereas the control and berberine treatments did not (Figures
   [113]S6F and [114]S7F, Supporting Information). To evaluate whether the
   antitumor effect of B68 is attributable to senescence induction,
   cellular senescence and the expression of the DNA damage sensors
   γ‐H2AX, p16, p53, and p21 were detected in tumor tissues. As shown in
   Figures [115]S6G,H and [116]S7G,H (Supporting Information), the
   expression of γ‐H2AX, p16, p53, and p21 was also markedly increased in
   the B68 treatment group in both the HCT116 and RKO xenograft mouse
   models. Furthermore, immunohistochemistry (IHC) showed that the level
   of Ki‐67 (a marker of proliferation) was decreased, while the level of
   cleaved caspase‐3 (a marker of apoptosis) was increased in B68‐treated
   tumor xenograft mice. Moreover, compared with the control treatment,
   B68 did not markedly alter the pathological changes in major organs,
   including heart, liver, spleen, lung, and kidney (Figures [117]S6I and
   [118]S7I, Supporting Information). Collectively, these findings
   indicate that B68 can suppress the growth of colorectal cancer cells by
   inducing senescence in tumor cells.

2.3. B68 Suppresses Tumor Growth in AOM/DSS‐Induced Colon Cancer

   The AOM/DSS model is currently the most widely used chemically induced
   model for ulcerative colitis (UC). Clinical symptomatology,
   morphological, and pathological observations have proven that this
   model is highly similar to the process of cancer in human ulcerative
   colitis and can be effectively used for the evaluation of antitumor
   drug efficacy, allowing for the observation of its biological effects
   in a short period of time. To determine whether B68 can attenuate the
   formation of CRC, we established an AOM/DSS model and treated the mice
   with 8 mg kg^−1 B68 daily via intraperitoneal injection. The specific
   experimental procedure is shown in Figure [119] 2A. Mice in the AOM/DSS
   group had a higher tumor burden in the colonic tissue, whereas B68
   treatment significantly inhibited the tumors induced by AOM/DSS.
   Subsequently, our research revealed that B1 is capable of suppressing
   the development of colorectal cancer, albeit to an extent that is
   intermediate between the control group and the B68 treatment group, and
   notably less pronounced than the effects observed in the B68 group
   (Figure [120]2B). Quantitatively, treatment with B68 reduced the
   percentages of large (>4 mm in diameter), medium (2–4 mm in diameter)
   by 85.2% and 66.7%, respectively, whereas treatment with B1 led to a
   reduction in the percentages of large (>4 mm in diameter), medium
   (2–4 mm in diameter), and small (<2 mm) tumors by 44.4%, 39.4%, and
   23.5%, respectively, which indicates that the therapeutic efficacy of
   B68 is significantly superior to that of B1 (Figure [121]2E).
   Interestingly, the B68 mice presented a longer colon length and a
   heavier body weight than the AOM/DSS mice (Figure [122]2C,D),
   suggesting that AOM/DSS mice are more likely to develop tumors, as
   DSS‐induced shortening of colon length in mice is an indicator of the
   severity of inflammation, a potential cause of colon carcinogenesis.
   Furthermore, H&E findings indicated that treatment with B68 in mice
   also substantiated the effectiveness of B68 therapy, indicating that
   the therapeutic effects of B68 is superior to B1 (Figure [123]2F).
   Moreover, IHC also showed that B68 could increase the expression of the
   protein senescence‐related markers γ‐H2AX, p16, p53, and p21.
   Concurrently, the expression of Ki‐67, which is associated with
   proliferation, decreased, whereas the expression of cleaved caspase‐3,
   which is related to apoptosis, significantly increased
   (Figure [124]2G,H). Interestingly, following treatment with B68, there
   was a significant increase in the number of CD8^+ cells within the
   tumor, while the levels of PD‐L1 and Foxp3 were significantly reduced.
   The levels in the group treated with B1 were intermediate between those
   in the control and B68 groups (Figure [125]S8A,B, Supporting
   Information). Additionally, B68 did not significantly change the tissue
   indices of the heart, liver, spleen, lungs, or kidneys compared with
   those in the control treatment group (Figure [126]S8C, Supporting
   Information). Taken together, these results suggest that B68 suppresses
   AOM/DSS‐induced carcinogenesis by promoting tumor cell senescence.

Figure 2.

   Figure 2
   [127]Open in a new tab

   B68 significantly inhibited the development of colorectal cancer
   induced by AOM/DSS. A) Azoxymethane (AOM‐12.5 mg kg^−1) was first
   injected into C57BL/6J (male) mice, followed by 2% or 2.5% glucose
   sodium sulfate (DSS) in the drinking water for 1 week and then regular
   drinking water for 2 weeks; this process was repeated three times as
   described in the “Animal experiment methods” section. B) Representative
   images of the colon and rectum after AOM/DSS treatment depicting the
   extent of the tumor load. C) The length of mouse colon in the control
   group, model group, AOM/DSS+B68 group and AOM/DSS+B1 group (n = 5,
   error bars represent SEM, mean ± SEM, Student's t‐test). D) Changes in
   the body weights of mice in the control, model, AOM/DSS+B68 and
   AOM/DSS+B1 groups during the experimental period (n = 5, error bars
   represent SEM, mean ± SEM, Student's t‐test). E) The number and
   diameter of tumors in the control group, model group, AOM/DSS+B68
   group, and AOM/DSS+B1 group were measured at the end of the treatment
   (n = 5, error bars represent SEM, mean ± SEM, Student's t‐test). F)
   Representative micrographs of H&E staining of colon tumors in control,
   model, AOM/DSS+B68 and AOM/DSS+B1 mice after AOM/DSS treatment (scale
   bar, 100 µm or 500 µm). G) Representative images of IHC staining of
   cleaved caspase 3, Ki‐67, p16, p21, p53, and γ‐H2AX in the colon of
   model, AOM/DSS+B68 and AOM/DSS+B1 mice; scale bar, 100 µm. H)
   Quantitative statistical results of the immunohistochemical staining
   shown in Figure [128]2G (error bars represent SEM, mean ± SEM,
   Student's t‐test). n.s, not significant; *p < 0.05, **p < 0.01,
   ***p < 0.001, ****p < 0.0001.

2.4. B68 Induces Senescence in Colorectal Cancer Cells by Targeting BMI1

   To explore the molecular mechanism of B68‐induced cell senescence, we
   performed transcriptome sequencing of B68‐treated RKO cells. By KEGG
   pathway enrichment analysis (Figure [129] 3A), the top three enriched
   pathways were identified as the cell cycle, p53 signaling pathway, and
   cellular senescence pathway, which was consistent with the senescence
   phenotype induced by B68 and further confirmed that B68 triggers
   cellular senescence. In the B68‐treated group, the mRNA levels of 1050
   genes increased, and the mRNA levels of 1221 genes decreased. Among
   these genes, several senescence‐associated genes, including CXCL8,
   SPINK1, TP53, MYC, and FOXO3, were upregulated, whereas others,
   including CCNA2, FOXM1, PIK3R1, and CDK8, were downregulated
   (Figure [130]3B). Additionally, GO analysis confirmed the activation of
   the cellular senescence pathway (Figure [131]3C). Considering that
   B68‐induced cellular senescence leads to changes in the levels of
   senescence‐associated proteins, we also conducted a proteomic analysis
   of B68‐treated cells. As anticipated, the cell cycle, p53 signaling
   pathway, and cellular senescence pathway were the most enriched
   pathways (Figure [132]3D,E). We subsequently focused on genes
   responsible for the activation of senescence‐associated proteins and
   observed significant upregulation or downregulation of a series of
   senescence‐associated proteins, such as CCND3, RBL2, CCNE2, and CDK2
   (Figure [133]3F). In conclusion, these results further indicate that
   B68 mediates the death of tumor cells through senescence.

Figure 3.

   Figure 3
   [134]Open in a new tab

   B68 exerts antitumor effects by targeting BMI1‐induced senescence in
   colorectal cancer cells. A) KEGG analysis of differentially expressed
   mRNAs in B68‐treated RKO cells. B) The volcano plot illustrates the
   changes in mRNA levels after B68 treatment in RKO cells (red dots
   represent significant upregulation, and blue dots represent significant
   downregulation), with marked genes related to senescence and cell cycle
   (fold change ≥2, p value ≤0.05). C) Gene Ontology (GO) analysis of
   differentially expressed mRNAs in RKO cells after B68 treatment. D,E)
   KEGG enrichment of the potentially differentially expressed proteins
   associated with senescence and cell cycle in B68‐treated RKO cells. F)
   Heatmap showing the expression of the top 30 differentially expressed
   proteins in B68‐treated RKO cells, with senescence‐ and cell
   cycle‐related genes labeled (fold change ≥1.2, unique peptides ≥2). G)
   DARTS analysis of potential target proteins that bind to B68. H)
   Pull‐down analysis of potential target proteins that bind to B68. I)
   Venn analysis results of differentially expressed proteins that appear
   in the DARTS, pull‐down and proteomics results. Detection of SA‐β‐gal
   activity J), p21 and p53 protein expression K) following BMI1 gene
   knockdown in RKO and HCT116 cells. L) H3K9me3 expression in RKO and
   HCT116 cells after BMI1 gene knockdown was detected by
   immunofluorescence, and the results of the quantitative analysis are
   shown on the right (M). Scale bars, 100 µm (n = 3, error bars represent
   SEM, mean ± SEM, one‐way ANOVA). ****p < 0.0001.

   To explore the targets of B68‐induced tumor cell senescence, we
   utilized drug affinity responsive target stability (DARTS) to identify
   protein targets that directly interact with B68. The workflow of DARTS
   is shown in Figure [135]S9A (Supporting Information). DARTS predicted
   33 potential target proteins (unique peptide 2, with an intensity ratio
   of 1.5) (Figure [136]3G). We subsequently used B68 as a chemical probe
   to explore potential targets involved in the senescence of RKO cells.
   First, we synthesized biotin‐B68, which results in a senescent
   phenotype in RKO cells. An in vitro pull‐down assay was conducted using
   Bio‐B68 (or Bio as a negative control) with cancer cell lysates (Figure
   [137]S9B, Supporting Information). However, in the SDS‐PAGE analysis,
   no differential bands were detected by silver staining (Figure [138]S9C
   Supporting Information,). Notably, the absence of specific bands does
   not indicate a lack of target pulling; hence, we subjected the entire
   lane to mass spectrometry analysis. The results revealed 15 potential
   target genes of B68 through pull‐down analysis (Figure [139]3H). By
   analyzing the results of the DARTS and pull‐down experiments, we
   identified four aging‐related genes: UBAP2L, YBOX1, BMI1, and S100A9.
   Subsequently, we conducted interference experiments on these four genes
   and discovered that the senescent phenotype was observed exclusively
   upon BMI1, which is consistent with the intersecting results from
   DARTS, proteomics and pull‐down assays related to senescence
   (Figure [140]3I; Figure [141]S9D–H, Supporting Information).
   Furthermore, changes in the expression of BMI1 in RKO and HCT116 cells
   were detected by immunoblotting (IB). Treatment with B68 significantly
   reduced BMI1 expression in both cell lines in a time‐ and
   concentration‐dependent manner. Additionally, B68 had no effect on the
   BMI1 mRNA level, suggesting that B68 induces senescence in tumor cells
   by degrading BMI1 (Figure [142]S10, Supporting Information). Next, we
   knocked out BMI1 in both RKO and HCT116 cells. After gene knockout,
   both cell lines exhibited a pronounced senescence phenotype, with
   significant upregulation of SA‐β‐gal activity or p53 or p21
   (Figure [143]3J,K; Figure [144]S9I–L, Supporting Information).
   Consistent results were also obtained from H3K9me3 immunofluorescence
   assays, which demonstrated that BMI1 knockout significantly promoted
   the senescence of RKO and HCT116 cells (Figure [145]3L,M).
   Additionally, after the knockout of BMI1 followed by treatment with
   B68, B68 did not further promote cellular senescence in HCT116 cells
   (Figure [146] 4A,B).

Figure 4.

   Figure 4
   [147]Open in a new tab

   B68 induces senescence in colorectal cancer cells by targeting BMI1. A)
   After BMI1 knockdown, B68 did not further increase SA‐β‐Gal activity in
   HCT116 cells. B) Quantitative results of SA‐β‐Gal staining are shown on
   the right (n = 3, error bars represent SEM, mean ± SEM, one‐way ANOVA).
   C) CETSA was used to determine the thermal stabilization of the BMI1
   interaction with B68 at a series of temperatures from 62 to 83 °C, and
   the quantification of BMI1 is shown in (D) (n = 3, error bars represent
   SEM, mean ± SEM, Student's t‐test). E) Stability of different
   concentrations of B68 on BMI1 at 65 °C. F) Quantitative results of BMI1
   intensity in (E) (n = 3, error bars represent SEM, mean ± SEM,
   Student's t‐test). G) Effect of B68 on the stability of the BMI1
   protein at different protein hydrolase concentrations. H) Quantitative
   results of BMI1 intensity in (F) (n = 3, error bars represent SEM,
   mean ± SEM, Student's t‐test). I) Effect of different concentrations of
   B68 on BMI1 protein stability at a pronase‐to‐protein ratio of 1:300.
   The quantitative results for the BMI1 protein are shown on the right
   (J) (n = 3, error bars represent SEM, mean ± SEM, Student's t‐test).
   (K) Molecular docking of B68 with BMI1. L) Cellular MST assay of
   GFP‐tagged BMI1 upon overexpression of wild‐type BMI1 and disrupted
   mutants (Arg123 or Glu120). n.s, not significant; *p < 0.05,
   **p < 0.01, ***p < 0.001, ****p < 0.0001.

2.5. ARG123 of BMI1 is Critical for Binding to B68

   To ascertain whether B68 binds directly to the BMI1 protein, we
   employed cellular thermal shift assay (CETSA) technology, a technique
   grounded in the principle that the thermal stability of the target
   protein is enhanced upon ligand binding. Given that prolonged
   protein‐drug incubation in lysates could lead to protein‐metabolite
   interactions, we opted for a 1‐minute incubation to detect direct
   binding. As depicted in Figure [148]4C,D, B68 notably increased BMI1
   accumulation across temperatures ranging from 37 to 83 °C, implying
   that B68 directly interacts with BMI1 by modulating its thermal
   stability. To corroborate the CETSA findings, we subjected the lysate
   to B68 at varying concentrations at 65 °C and observed increased BMI1
   stability with increasing B68 levels (Figure [149]4E,F). In line with
   the CETSA results, BMI1 accumulation increased with increasing B68
   concentration at an enzyme‐lysate ratio of 1:300 (Figure [150]4G–J). To
   determine the BMI1 amino acid residues involved in B68 binding, we
   conducted a docking simulation via Maestro software (Schrödinger,
   version 9.0) to model the binding mode of B68 to BMI1 (Figure [151]4K).
   Subsequently, we performed microscale thermophoresis (MST) to analyze
   the direct binding between B68 and exogenous, GFP‐tagged BMI1 protein.
   As anticipated, B68 strongly bound to BMI1, with a Kd value of 91.4 nm.
   Figure [152]4L illustrates the formation of two stable hydrogen bonds
   between the backbone of BMI1 at positions Glu120 and Arg123 with B68.
   We then conducted cellular MST assays on overexpressed GFP‐tagged BMI1,
   including wild‐type and disruptive mutants at positions Glu120 and
   Arg123. Although the mutation of Glu120 to alanine resulted in a Kd
   nearly identical to that of the wild type, the mutation of Arg123 to
   alanine completely abrogated BMI1‐B68 complex formation.

2.6. B68 Induces the Immunogenicity of Cellular Senescence and Promotes
Antitumor Immunity

   Previous studies have shown that senescent cancer cells can elicit
   antitumor protection when used in a vaccination setting, a
   characteristic attributed to their SASP. Next, we initially assessed
   the expression of SASP in response to B68 in MC38 cells. As the
   transcriptomic results showed, compared to those in nonsenescent
   controls, an upregulation of the interferon (IFN) transcriptional
   signature, including the major gene components involved in MHC‐I
   antigen processing and presentation, was observed in senescent cells
   (Figure [153] 5A). The upregulation of the majority of
   senescence‐associated genes, particularly the master transcriptional
   regulators of MHC‐I‐dependent immune responses B2 m, Nlrc5, and Tap2,
   was confirmed through qRT‐PCR analysis of separate biological
   replicates (Figure [154]5C). Senescent cells upregulate MHC‐I antigen
   presentation, and to determine whether the transcriptional upregulation
   of MHC1‐related genes is reflected in an increase in antigen
   presentation, we examined the expression of H‐2K^b/D^b in CT26 and MC38
   cells and HLA‐A/B/C in RKO and HCT116 cells via flow cytometry. The
   results showed that the expression of H‐2K^b/D^b and HLA‐A/B/C
   significantly increased in senescent cells treated with B68, suggesting
   that B68 enhances the ability of senescent cells to process and present
   MHC‐I antigens (Figure [155]5B; Figure [156]S11A–C, Supporting
   Information). After demonstrating the enhanced immunogenicity of
   senescent cells, we turned our attention to cancer research with the
   ultimate goal of using senescent cancer cells to elicit protective
   antitumor immune responses. To determine whether the secretions from
   senescent cells reflect an enhancement in immune recruitment in vivo,
   we injected mice subcutaneously with either vehicle (PBS) or senescent
   MC38 cells and performed histological analysis on the injection sites.
   Interestingly, one week post‐injection, the subcutaneously injected
   senescent MC38 cells were clearly visible as large pleomorphic and
   anaplastic cells surrounded by a rich influx of immune cells (CD45^+),
   primarily myeloid (CD11b^+) and T cells (CD3^+) (Figure [157]5D,E).
   Following the recruitment of immune cells, the next crucial step in
   triggering an adaptive immune response includes the capture of antigens
   by dendritic cells (DCs), as well as the activation and maturation of
   DCs. To test this possibility, we cocultured dendritic cells with
   untreated or senescent MC38 cells and examined the surface expression
   of the DC activation and maturation markers CD80, CD86, and MHC‐II
   (I‐A/I‐E) on the DCs. The cells induced to senesce with bleomycin
   served as a positive control for DC activation. We found that
   coculturing dendritic cells with senescent MC38 cells, as well as B68,
   both increased the expression of DC maturation markers CD80, CD86, and
   I‐A/I‐E (Figure [158]5F,G). Given that senescent cells exhibit
   increased antigenicity and adjuvanticity, we hypothesize that senescent
   cancer cells could be utilized to promote an immune response against
   cancer. Here, we initially immunized B68‐induced senescent MC38 cells
   to assess the preventive effects against subsequent proliferation upon
   rechallenge with MC38 cells. One week after subcutaneous immunization,
   all the experimental groups received a subcutaneous injection of
   1.5 × 10^5 untreated MC38 cells on the contralateral side
   (Figure [159]5H). We observed that immunization with sen‐MC38 provided
   robust antitumor protection (Figure [160]5I,J). Taken together, our
   observations suggest that immunization with B68‐induced senescent
   cancer cells can promote an antitumor immune response that contributes
   to tumor prevention.

Figure 5.

   Figure 5
   [161]Open in a new tab

   Immunization with senescent tumor cells promotes antitumor effects and
   immune surveillance. A) RNA‐seq analysis of the mRNA expression of
   genes related to antigen presentation mechanisms and immunity in MC38
   cells in the control and B68 groups. B) Flow cytometry detection of
   H‐2K^b/D^b expression in B68‐treated MC38 and CT26 cells. C) After
   senescence was induced in MC38 cells with B68, the mRNA expression of
   the p16, p21, p53, B2 m, Tap2, and Nlrc5 genes was detected by qRT‐PCR
   (n = 3 independent experiments, error bars represent SEM, mean ± SEM,
   one‐way ANOVA). D) C57BL/6 mice were subcutaneously injected with
   either vector (PBS) or senescent MC38 cells. On day 7, the skin at the
   site of tumor inoculation was subjected to H&E staining and
   immunohistochemistry, with a scale = 100 µm. The quantitative results
   of immunohistochemistry are shown on the right E) (n = 3, error bars
   represent SEM, mean ± SEM, one‐way ANOVA). F) Flow cytometry analysis
   of CD80, CD86, and MHC‐II (I‐A/I‐E) expression in DCs cocultured with
   senescent MC38 cells and quantification of CD80, CD86, and MHC‐II
   (I‐A/I‐E) expression are shown on the right G) (n = 3, error bars
   represent SEM, mean ± SEM, one‐way ANOVA). H) C57BL/6J mice were
   subcutaneously injected with PBS or senescent MC38 cells on the left
   side, and on day 7, MC38 cells were injected on the right side, after
   which the growth of tumors on the right side of the mice was observed.
   I,J) Individual tumor growth curves of PBS‐treated mice or mice
   immunized with senMC38 cells. Tumor growth (number of animals that
   developed tumors out of the total) and tumor latency (mean ± SEM of the
   day on which the tumor appeared) are indicated for each group. One‐way
   ANOVA was performed, and the error bars represent SEM, mean ± SEM).
   n.s, not significant; **p < 0.01, ***p < 0.001, ****p < 0.0001.

2.7. B68 Mediates PD‐L1 Degradation by Targeting CSN5

   Considering that berberine can modulate the expression of PD‐L1 to
   exert an antitumor immune effect, we also investigated whether compound
   B68 can reduce PD‐L1 levels. As shown in Figure [162]S12A–L (Supporting
   Information), the WB, flow cytometry, and immunofluorescence results
   all showed that B68 dose‐dependently reduced the expression of PD‐L1 in
   RKO and HCT116 cells, indicating that B68 also has an antitumor immune
   effect by reducing PD‐L1. Subsequently, we tested whether B68 could
   diminish the ability of cancer cells to bind to PD‐1. After incubating
   B68‐treated RKO cells with recombinant PD‐1 protein and fluorescent
   antibodies, we used confocal microscopy to detect the interaction
   between PD‐L1 and PD‐1. The reduced green fluorescence in B68‐treated
   RKO cells indicated that B68, by lowering PD‐L1 levels, diminished the
   binding capacity of cancer cells to PD‐1 (Figure [163]S13A,B,
   Supporting Information). Similarly, coculture experiments of
   B68‐treated RKO and HCT116 cells with T cells demonstrated that B68 can
   reduce the viability of RKO and HCT116 cells in a dose‐dependent
   manner, indicating that B68 can exert an antitumor immune effect by
   reducing the expression of PD‐L1 (Figure [164]S13C–F, Supporting
   Information). To further explore the mechanism by which B68 affects the
   expression of PD‐L1, we measured the expression levels of PD‐L1 mRNA in
   colorectal cancer cells after treatment with B68 or treated tumor cells
   in combination with the protein synthesis inhibitor cycloheximide (CHX)
   to assess the changes in PD‐L1 protein expression. These findings
   indicate that, similar to berberine, B68 does not directly affect the
   transcriptional level of PD‐L1 but modulates its expression through
   posttranslational modifications (Figure [165]S14A–H, Supporting
   Information). Regardless of whether B68, like berberine, mediates the
   degradation of PD‐L1 via the ubiquitin‒proteasome pathway, the results
   showed that the accelerated degradation of PD‐L1 in RKO cells induced
   by B68 can be reversed by MG132, while CQ, 3MA and Baf cannot, and the
   increased ubiquitination of PD‐L1 by B68 confirmed this point
   (Figure [166] 6A–I; Figure [167]S14I,J, Supporting Information).
   Finally, molecular docking and microscale thermophoresis (MST)
   experiments revealed that B68 may mediate the degradation of PD‐L1 by
   targeting the Glu76 site of the CSN5 protein (Figure [168]6J–K).
   Additionally, our immunoprecipitation (IP) experiments confirmed the
   interaction between CSN5 and PD‐L1 (Figure [169]S14K,L, Supporting
   Information), further confirming that B68 mediates the
   ubiquitination‐mediated degradation of PD‐L1 by targeting CSN5.^[
   [170]^32 ^]

Figure 6.

   Figure 6
   [171]Open in a new tab

   B68 targets CSN5 to decrease the expression of PD‐L1 on tumor cells.
   A–H) Immunoblotting for PD‐L1 expression in RKO and HCT116 cells
   treated with B68 in combination with the proteasome inhibitor MG132
   (1 µm, 10 µm), chloroquine (20 µm, 40 µm), the lysosomal inhibitor
   bafilomycin A1 (200 nm, 800 nm) or the autophagy inhibitor
   3‐methyladenine (1 mm, 5 mm). The quantitative results of PD‐L1
   expression are shown in the figure below (n = 3, error bars represent
   SEM, mean ± SEM, one‐way ANOVA). I) Immunoprecipitation (IP) analysis
   of PD‐L1 ubiquitination in RKO cells treated with B68 and IB analysis
   with a ubiquitin antibody. The cells were treated with MG132 before
   ubiquitination analysis was performed. J) Molecular docking of B68 with
   CSN5. K) The binding of CSN5 to B68 was determined by MST after the
   overexpression of GFP‐tagged wild‐type CSN5 or disruptive mutants
   (Ser481 and Glu76). n.s, not significant; *p < 0.05, **p < 0.01,
   ****p < 0.0001.

2.8. B68 Suppresses MC38 Tumor Growth by Promoting Tumor Cell Senescence and
Enhancing T Cell Infiltration

   Considering that berberine can exert antitumor immune effects by
   regulating PD‐L1, we then conducted animal experiments on C57BL/6J mice
   with a normal immune system (Figure [172] 7A). By analyzing tumor
   volume, weight, and size, we found that both B68 and B1 significantly
   inhibited the growth of the MC38 xenograft, with B68 showing better
   therapeutic effects than berberine treatment (Figure [173]7B–E). To
   explore whether the superior therapeutic effect of B68 compared to B1
   is caused by the induction of tumor cell senescence, we measured
   indicators related to senescence and PD‐L1 in the tumor tissue,
   respectively. As shown in Figure [174]7F–H, compared with the control
   group, both berberine and B68 significantly reduced the expression of
   PD‐L1. However, B68 increased the percentage of
   β‐galactosidase‐positive senescent cells in the tumor tissue, whereas
   berberine did not, indicating that B68 exerts dual antitumor effects by
   inducing tumor cell senescence and reducing tumor PD‐L expression. To
   verify this hypothesis, we subsequently conducted IHC and WB analyses
   on the tumor tissue and found that both B68 and B1 significantly
   reduced the expression of PD‐L1 in the tumor tissue, as well as the
   expression of immune‐suppressive factors such as Foxp3. Moreover, by
   analyzing the senescence‐associated markers γ‐H2AX, p16, p21, and p53,
   we found that only B68 increased the protein expression of γ‐H2AX, p16,
   p21, and p53, which was consistent with the results in nude mice
   (Figure [175]7I,J). We also assessed the changes in PD‐L1, p21, p53,
   and ubiquitinated protein levels in tumor tissues from MC38
   subcutaneous tumor‐bearing mice treated with B68 (8 mg kg^−1) or B1
   (8 mg kg^−1). As shown in Figure [176]7K,L and Figure [177]S15A
   (Supporting Information), only B68 was able to increase the protein
   expression of the senescence markers p21 and p53. Moreover, both B68
   and B1 were found to reduce PD‐L1 protein levels by increasing the
   ubiquitination levels of PD‐L1.

Figure 7.

   Figure 7
   [178]Open in a new tab

   B68 is capable of inducing senescence in colorectal cancer cells and
   exerting antitumor activity by diminishing PD‐L1 expression on tumor
   cells. A) MC38 cells were injected into C57BL/6 mice on day ‐5, and B68
   (8 mg kg^−1) and B1 (8 mg kg^−1) were administered as indicated. B) Ex
   vivo observation of the tumors. C) Tumor weight, D) tumor volume, and
   E) weight of the treated mice. F) After the experiment, the tumors were
   removed, and frozen sections were prepared for SA‐β‐Gal staining, scale
   bar, 100 µm. G–J) Representative IHC staining results for PD‐L1, CD8,
   cleaved caspase‐3, Foxp3, Ki‐67, p16, p21, p53, and γ‐H2AX in tumor
   tissues and quantitative results of immunohistochemical analysis are
   shown on the right. Scale bar, 100 µm (error bars represent SEM,
   mean ± SEM, Student's t‐test). K) Western blot analysis of p21, p53 and
   PD‐L1 expression in the tumor tissues of C57BL/6J mice. L) The results
   of the quantitative analysis are shown in the right (error bars
   represent SEM, mean ± SEM, one‐way ANOVA). n.s, not significant;
   *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

   Moreover, the immunohistochemical results also showed no obvious
   toxicity in the heart, liver, spleen, lung, or kidney
   (Figure [179]S15B, Supporting Information), indicating that B68 could
   exert an inhibitory effect on colorectal cancer both through senescence
   and by reducing the expression of tumor PD‐L1.

2.9. B68 Can Enhance the Anti‐CTLA4‐Mediated Inhibition of CRC Effects

   Considering that B68 can induce cellular senescence while reducing the
   expression of PD‐L1 in tumor cells and that combination therapy with
   anti‐PD‐L1 and anti‐CTLA4 has better therapeutic effects, we
   hypothesize that the combination of B68 with anti‐CTLA4 may further
   enhance B68 antitumor activity. Next, we sought to determine whether
   B68 could synergize with anti‐CTLA4 to inhibit tumor growth in vivo by
   using an MC38 mouse cancer model. A schematic representation of the
   model is depicted in Figure [180] 8A. The results revealed that B68
   significantly suppressed tumor growth without affecting body weight,
   which was better than the effect of anti‐PD‐1 and anti‐CTLA4,
   suggesting that B68 suppresses tumor growth through senescence in
   addition to its immune checkpoints (Figure [181]8B–D). We further found
   that cotreatment with B68 and anti‐CTLA4 significantly reduced the size
   of the tumors compared to treatment alone or anti‐CTLA4 monotherapy
   (Figure [182]8E,F). Consequently, the combination therapy led to a
   further increase in granzyme B production within the tumor
   (Figure [183]8G,H). Myeloid‐derived suppressor cells (MDSCs) and
   regulatory T cells (Tregs) are recognized as the primary
   immunosuppressive elements in the tumor microenvironment and contribute
   to tumor immune evasion by suppressing T lymphocyte responses. Flow
   cytometry analysis also revealed that the combination treatment
   significantly decreased the accumulation of MDSCs, identified as
   CD11b^+Gr1^+ cells, and Tregs, characterized by a CD4^+CD25^+ Foxp3^+
   phenotype, compared to the single treatment group (Figure [184]8I–L).
   In addition, IHC showed decreased levels of PD‐L1 and ki‐67 (a marker
   of proliferation) and increased caspase 3 levels in tumor‐xenografted
   mice, indicating that B68 triggered significant apoptosis in
   tumor‐bearing mice. Moreover, an analysis of the senescence‐related
   markers γ‐H2AX, p16, p21, and p53 revealed that, as expected, B68
   increased the expression of these proteins, and H&E staining revealed
   that B68 had no significant toxic effects on C57BL/6J mice
   (Figure [185]8M,N; Figure [186]S16A–C, Supporting Information). In
   brief, these findings collectively demonstrate that B68 inhibits the
   progression of CRC by inducing cellular senescence and reducing the
   levels of PD‐L1 in the tumor microenvironment and can exert synergistic
   antitumor effects when combined with CTLA‐4 antibody therapy, thereby
   further enhancing the response against cancer.

Figure 8.

   Figure 8
   [187]Open in a new tab

   B68 in combination with CTLA4 antibody effectively inhibited tumor
   growth in vivo in immunoreactive mice. A) C57BL/6 mice harboring MC38
   cell‐derived tumors were subjected to various treatments, including
   PBS, anti‐PD‐L1, anti‐CTLA4, B68 alone, and a combination of B68 with
   anti‐CTLA4. On day 5, MC38 cells were implanted into C57BL/6 mice,
   followed by the administration of B68, anti‐PD‐L1, or anti‐CTLA4 at the
   indicated time points. Subsequently, the tumor morphology B), tumor
   mass C), body weight D), and tumor volume E) were monitored and
   measured over a 14‐day period. F) Xenograft tumor growth curves of
   C57BL/6J mice in each treatment group (n = 5, error bars represent SEM,
   mean ± SEM, Student's t‐test). Flow cytometric analysis was utilized to
   quantify the presence of GzmB^+ cells F), Gr‐1^+CD11b^+ myeloid cells
   G), and Foxp3^+CD25^+ regulatory T cells H) among the groups treated
   with PBS, anti‐PD‐L1, anti‐CTLA4, B68 alone, B68 combined with
   anti‐PD‐L1, or B68 combined with anti‐CTLA4, and the quantitative
   results of the flow cytometry analysis are shown on the right (n = 5,
   error bars represent SEM, mean ± SEM, Student's t‐test). M)
   Representative immunohistochemical images of p21 in the tumor tissue of
   mice in each group, and the quantitative results of
   immunohistochemistry are shown on the right. Scale bar, 100 µm (n = 3,
   error bars represent SEM, mean ± SEM, Student's t‐test). n.s, not
   significant; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

2.10. The Protein Expression Levels of BMI1, CSN5, and PD‐L1 in Human Tumor
Tissue

   To confirm the validity of the aforementioned model in human tumor
   tissues, we evaluated the protein levels of CSN5, BMI1, and PD‐L1 in
   both paracancerous and tumor tissues extracted from six colon cancer
   patients. The findings, as depicted in Figure [188] 9A,B, revealed a
   marked increase in the expression levels of these proteins in the
   cancerous tissues compared with the surrounding noncancerous tissues.
   To delineate the correlation between the expression patterns of CSN5,
   BMI1, and PD‐L1 in the tumor tissues of colorectal cancer patients and
   their prognostic implications, we analyzed overall survival data from
   the TCGA colorectal cancer database. The data analysis indicated that
   patients exhibiting elevated levels of PD‐L1 and CSN5 had poorer
   clinical outcomes than did those who received PD‐1 antibody therapy,
   who exhibited an extended survival period (Figure [189]9C).
   Furthermore, leveraging the TIMER 2.0 database, we scrutinized the
   interplay between the expression of PD‐L1 and CSN5 and the degree of
   immune cell infiltration within colon (COAD) and rectal adenocarcinomas
   (READ), respectively. This analysis revealed an inverse relationship
   between the expression of these proteins and the infiltration of CD8^+
   T cells within the tumor immune microenvironment (Figure [190]9D,E).
   Consistent with our hypotheses, CSN5 was inversely correlated with
   pivotal elements of the cancer‐immunity cycle, including the release of
   tumor antigens, the priming and activation phase, the trafficking of
   immune cells to the tumor site, including the recruitment of CD4^+ T
   cells, macrophages, dendritic cells, and NK cells, as well as the T
   cell‐mediated recognition of cancer cells in COAD and READ, with a
   pronounced negative correlation in COAD (Figure [191]9F,G). Overall,
   elevated BMI1 expression was correlated with an adverse prognosis in
   colorectal cancer patients. The overexpression of CSN5 or PD‐L1 was
   associated with tumor‐induced immunosuppression, while patients
   receiving PD‐1 antibody treatment exhibited prolonged survival.

Figure 9.

   Figure 9
   [192]Open in a new tab

   Clinical relevance of BMI1, PD‐L1, or CSN5 in colorectal cancer
   tissues. A) Representative fluorescence images of BMI1, CSN5, or PD‐L1
   in paracancerous tissue (Normal) and tumor tissue (Tumor) of CRC
   patients. B) Quantification of BMI1, CSN5, or PD‐L1 expression in (A)
   (n = 3, error bars represent SEM, mean ± SEM, Student's t‐test). C) The
   survival of CRC patients stratified by the expression of CSN5 or PD‐L1
   was compared by two‐sided log‐rank analysis. D,E) Scatter plot showing
   the correlation between PD‐L1 or CSN5 expression and infiltrating CD8^+
   T cells in COAD and READ patients detected by TIMER 2.0. F,G)
   Correlations between CSN5 and the steps of the cancer immune cycle in
   COAD and READ patients. **p < 0.01, ***p < 0.001, ****p < 0.0001.

3. Discussion

   Cellular senescence refers to the attainment of a stable state of cell
   cycle arrest, which effectively inhibits the further proliferation of
   damaged or aged cells and plays a critical role in tumor suppression.^[
   [193]^33 ^] Although cellular senescence is generally considered
   irreversible, under specific conditions, particularly in tumor cells,
   reentry into the cell cycle is possible. Consequently, researchers have
   proposed a strategy that involves inducing senescence in tumor cells
   followed by the clearance of these senescent cells, which can
   effectively control the proliferation of premalignant cells and thus
   prevent the formation and progression of malignant tumors.^[ [194]^34 ,
   [195]^35 , [196]^36 , [197]^37 ^] Here, we found that B68, a derivative
   modified from berberine, can induce senescence in colorectal cancer
   cells by targeting BMI1, exerting an antitumor effect. Moreover, B68
   can reduce the expression of PD‐L1 on the surface of senescent cells by
   targeting CSN5, thereby blocking the direct interaction between PD‐1 on
   T cells and PD‐L1 on tumor cells, activating the immune
   microenvironment within the tumor, increasing the number of effective
   CD8^+ T cells, reducing the number of MSDCs and Treg cells, and
   effectively eliminating senescent tumor cells. Therefore, B68 can both
   induce tumor cell decay and inhibit tumor growth by regulating the
   immune system.

   In the context of cancer treatment, the relationship between cellular
   senescence and tumor development is complex and multifaceted.^[
   [198]^38 , [199]^39 ^] On the one hand, cellular senescence can be
   considered a mechanism for tumor suppression. For instance, the
   activation of oncogenes within cells may trigger a state known as
   oncogene‐induced senescence (OIS), in which cells enter an irreversible
   state of growth arrest, thus preventing the formation of tumors.^[
   [200]^40 , [201]^41 ^] Consistent with our findings, B68 significantly
   upregulated the expression of p21 and p53 when acting on colorectal
   cancer cells, whereas B68 treatment significantly increased the
   percentage of β‐gal‐positive senescent cells in tumor tissue, causing
   cycle arrest. On the other hand, senescent cells influence the
   surrounding microenvironment by secreting the SASP, which contains
   proinflammatory cytokines, growth factors, and enzymes that may
   facilitate tumor invasion and metastasis, as well as promote tumor
   angiogenesis, supplying the necessary nutrients and oxygen for tumor
   growth.^[ [202]^42 , [203]^43 ^] Our research revealed that after B68
   treatment, the levels of SASP‐related factors, such as IL6, GM‐CSF,
   CXCL3, and AREG, in colorectal cancer cells also significantly
   increased, indicating a certain degree of risk of immune suppression.
   Manuel Serrano's team has shown that chemotherapy can induce the
   senescence of cancer cells within the tumor, and PD‐L2 is highly
   upregulated in different types of cancer cells when they are induced to
   become senescent. Senescent cancer cells can protect themselves from
   the immune system by upregulating PD‐L2, while also recruiting
   immunosuppressive cells, creating an inhibitory environment that
   weakens the ability of lymphocytes to kill cancer cells. When senescent
   cells are cleared by immunotherapy, chemotherapy becomes more
   effective, providing a therapeutic strategy for the combination of
   chemotherapy and anti‐PD‐L2.^[ [204]^44 ^] Moreover, some studies have
   reported that PD‐L1 is heterogeneously expressed in senescent cells and
   that PD‐L1‐positive senescent cells accumulate in vivo with aging,
   which can further prevent clearance by the immune system.^[ [205]^24 ^]
   Previous studies have shown that berberine can suppress the
   deubiquitinating enzyme CSN5, reduce the expression of PD‐L1 on the
   surface of tumor cells, activate CD8^+ T cells in the tumor
   microenvironment, and exert a tumor‐clearing effect.^[ [206]^30 ^] The
   berberine derivative B68 in this study, similar to berberine, can also
   reduce the expression of PD‐L1 on tumor cells and exert an antitumor
   immune effect. Therefore, our cancer treatment strategy involves
   inducing senescence in tumor cells and simultaneously reducing PD‐L1
   expression on the surface of these senescent cells. This dual approach
   aims to activate the immune system to eliminate senescent cells without
   toxic side effects. Compared with the combination of chemotherapy and
   PD‐L2 antibodies, our strategy offers significant advantages, such as
   avoiding the harm caused by secondary medications and the potential
   side effects associated with chemotherapy and PD‐L2 antibody
   treatments.

   BMI1, known as B‐cell specific Moloney murine leukemia virus
   integration site 1, is part of Polycomb repressive complex 1 (PRC1) and
   plays a central role in regulating gene silencing and chromatin
   structure, which is crucial for the self‐renewal of stem cells.^[
   [207]^45 , [208]^46 ^] In this study, we predicted and verified BMI1 as
   a potential target of B68 via DARTS/MS, pull‐down, CETSA, and MST
   methods. During cellular senescence, the expression of BMI1 is often
   reduced, affecting cell cycle arrest and the production of the SASP.^[
   [209]^47 ^] Next, we knocked down BMI1 in the colorectal cancer cell
   lines RKO and HCT116, and upon knockdown, the cells exhibited features
   of senescence, including upregulated expression of p53 and p21,
   indicating that downregulation of BMI1 significantly promoted cellular
   senescence. Additionally, B68 treatment did not induce further
   senescence in RKO cells with BMI1 gene knockout, indicating that B68
   induces senescence through a BMI1‐dependent pathway.

   Transcriptome sequencing of B68‐induced senescent cells revealed
   significant upregulation of antigen presentation mechanisms related to
   MHC‐I molecules, including the main transcriptional regulatory factor
   Nlrc5, in senescent cells. This finding suggested that B68‐induced
   senescence may increase the immunogenicity of tumor cells. It has been
   reported that the use of senescent cancer cells as vaccines can
   stimulate antitumor protective effects.^[ [210]^48 , [211]^49 ^] We
   demonstrated the ability of B68‐induced senescent MC38 cells to
   activate protection against subsequent rechallenge with untreated MC38
   cells. As expected, B68‐induced senescent cells activate immune cells,
   including antigen‐presenting cells and CD8^+ T cells, to provide
   specific immune protection against tumors. The advantage of this
   vaccine lies in the enhanced immunogenicity of senescent cells, which
   can more effectively activate the adaptive immune response, thereby
   resulting in a significant inhibitory effect on tumor growth in the
   MC38 animal model.

   In summary, our research demonstrated that B68 possesses a dual
   mechanism of action in combating tumorigenesis. It facilitates tumor
   cell senescence, thereby inhibiting tumor cell proliferation, and
   simultaneously triggers an immune response by modulating the expression
   of PD‐L1 on senescent cells through the CSN5 pathway. This dual action
   not only targets the direct elimination of senescent tumor cells but
   also harnesses their potential as a therapeutic vaccine. Our findings
   offer novel insights and therapeutic strategies that could
   revolutionize the treatment of colorectal cancer by integrating
   senescence induction with immune modulation.

4. Experimental Section

Cell Lines and Culture

   Colorectal cancer cells (RKO, HCT116, CT26, LOVO, and SW620) were
   purchased from the Cell Bank of Shanghai Institute of Cell Biology,
   Chinese Academy of Sciences. MC38 cells were purchased from American
   Type Culture Collection (ATCC). The colon normal immortalized
   epithelial cell line NCM460 was obtained from InCell (San Antonio, TX)
   and cultured according to the manufacturer's instructions. Jurkat cells
   were donated by Kongming Wu's research group, Department of Oncology,
   Tongji Medical College, Huazhong University of Science and Technology
   (Wuhan, China). RKO cells were cultured in MEM; MC38 and HCT116 cells
   were cultured in DMEM and McCoy's 5A medium; LOVO cells were cultured
   in F‐12K medium; CT26 and NCM460 cells were cultured in RPMI‐1640; and
   SW620 cells were cultured in L15 medium. All media were purchased from
   Meilun (Dalian, China). Then, 100 mg mL^−1 streptomycin (Meilunbio,
   China), 100 U mL^−1 penicillin (Meilunbio, China) and 10% fetal bovine
   serum (Biological Industries, Cromwell, CT, USA) were added to the
   media, and the cells (excluding SW620 cells cultured in an air
   incubator) were placed in an incubator at 37 °C with 5% CO[2].

Mice

   Female or male C57BL/6J mice and BALB/c nude mice, 6–8 weeks old,
   weighing 18–20 g, were purchased from Shanghai Jihui Laboratory Animal
   Breeding Company Limited (Shanghai, China) and housed at the Shanghai
   University of Traditional Chinese Medicine (SUTCM) Animal Center for
   animal experiments. The animals were housed in an SPF environment with
   45% humidity, constant temperature (23 ± 2°C), and 12 h of light per
   day, and their food and drinking water were routinely sterilized by
   autoclaving.

Antibody

   All antibodies used are listed in Table [212]S1 (Supporting
   Information).

Cell Viability Assay

   The cells were inoculated in 96‐well plates (Corning, USA) at a density
   of 8 × 10^3 per well, and after the cells had adhered, the drug
   concentration was adjusted to the concentration required for the
   experiment. The drug‐containing medium was discarded, and 100 µL of
   CCK‐8 solution (Meilunbio, China) was added after 24 h. The cells were
   incubated at 37 °C for 1–4 h in the dark. The absorbance value (OD
   value) was measured at 450 nm by using an enzyme marker (BioTek, USA)
   to calculate the cell viability. Afterward, the absorbance value (OD
   value) was measured at 450 nm via an enzyme labeling instrument
   (BioTek, USA), and the cell viability was calculated.

Apoptosis Assay

   The cells in the logarithmic growth phase were inoculated in 6‐well
   plates at a density of 5 × 10^5 per well, and after the cells had
   completely adhered, the drug was added for 24 h at the corresponding
   concentration. The cells were subsequently digested with trypsin,
   collected in centrifuge tubes, washed twice with PBS (Meilunbio,
   China), and stained with an Annexin V‐FITC/PI apoptosis double staining
   kit (BD Biosciences, USA). Then, 1× binding buffer was used to dilute
   Annexin V‐FITC and PI to 100 µL well^−1, which were mixed thoroughly
   and incubated for 15 min at room temperature in the dark. Three hundred
   microliters of 1 × binding buffer was added before detection, and the
   mixture was filtered through a cell filtration sieve to prevent cell
   clumps from clogging the instrument. The results were detected via flow
   cytometry (Beckman Coulter Cytoflex, USA), and the data were processed
   and analyzed via FlowJo software.

Cell Cycle Analysis

   The logarithmic growth phase cells were digested into a cell suspension
   and inoculated in 6‐well plates at a density of 5 × 10^5 per well.
   After the cells had adhered, the appropriate concentration of drugs was
   added, and the cells were treated for 24 h. All the cells were
   collected in centrifuge tubes and centrifuged at 300 × g for 5 min. The
   supernatant was discarded, and the cells were washed once with PBS
   buffer, mixed and fixed by adding 70% ice‐cold ethanol and left at 4 °C
   overnight. For detection, the ethanol was discarded, the samples were
   washed once with PBS, prepared pyridinium iodide staining solution
   (Beyotime, China) was added, and the samples were incubated for 30 min
   at 37 °C in the dark. The results were subsequently detected via flow
   cytometry (Beckman Coulter Cytoflex, USA) with a maximum excitation
   wavelength of 488 nm. The data were analyzed via ModFit LT 5.0.

EdU Cell Proliferation Experiment

   Cells in the logarithmic growth phase were inoculated into 96‐well
   plates at 8 × 10^3 cells well^−1. After the cells had adhered, drugs at
   the corresponding concentrations were added. The experiment was carried
   out according to the instructions of the EdU test kit (Beyotime,
   China). Finally, Operetta CLS High Content Cell Imaging Analysis
   (PerkinElmer, USA) was used to photograph and record the statistical
   data (the maximum excitation wavelength of Azide 488 was 495 nm, and
   the maximum emission wavelength was 519 nm).

Cell Colony Formation Experiment

   After digestion of the cells in the logarithmic growth phase, the cells
   were resuspended, 400 cells per well were inoculated in a 12‐well
   plate, and the culture was continued for 14 days or until the number of
   cells in most single clones was greater than 50. The medium was changed
   every 3 days during the process, and the cell status was observed.
   After cloning, the cells were washed with PBS once, 0.5 mL of 4%
   paraformaldehyde was added to each well for 30 min, and the cells were
   washed with PBS 3 times. Each well was stained with 0.5 mL of crystal
   violet solution for 15 min, then washed with PBS buffer and dried.
   Colonies of more than 10 cells were counted under a microscope to
   calculate the colony formation rate.

Cell Activity Assay

   The logarithmic growth phase cells were inoculated in a 96‐well plate
   at 8 × 10^3 cells per well, and after the cells had adhered, the
   corresponding concentration of drugs was added. The culture medium was
   aspirated, and the cells were washed once with PBS. Then, 100 µl of
   Calcein AM assay working solution (Beyotime, China) was added to each
   well, the plates were incubated for 30 min at 37 °C in the dark, and
   photos were taken for recording via an Operetta CLS high‐content cell
   imaging system.

Cell Scratch Experiment

   A total of 6 × 10^5 cells were inoculated in a 6‐well plate, and the
   cell confluence reached 100% after overnight culture. On the second
   day, a vertical scratch was made with a 20 µL tip. After the scratch
   was completed, the cells were washed 3 times with sterile PBS to remove
   the nonadherent cells, and the medium was then replaced with fresh
   serum‐free medium. The cells were placed in a 5% CO[2] incubator at
   37 °C. After 16 h, the cells were removed, the width of the scratches
   was observed and measured with a microscope, and photos were taken. The
   distance between the cells was statistically analyzed with ImageJ
   software.

SA‐β‐Gal Activity Detection

   The cells in good condition were inoculated in 12‐well plates. After
   the cells had adhered, they were incubated with different
   concentrations of drugs for 2 days, and the drug‐containing medium was
   removed and replaced with complete medium for 5 days. After 5 days, the
   culture plate was removed, the plate was washed with PBS, 1 mL of
   SA‐β‐gal staining fixing solution was added, and the plate was fixed at
   room temperature for 15 min. The fixing solution was removed, and the
   plate was washed with PBS. SA‐β‐gal dye solution (Beyotime, China) was
   prepared according to the instructions, and 1 mL was added to each
   well. The 12‐well plates were sealed with a parafilm membrane
   (PARAFILM, USA) to prevent evaporation and incubated at 37 °C
   overnight.

Western Blot and Immunoprecipitation

   The cells were inoculated into a 6‐well plate at a density of 5 × 10^5
   per well. After the cells had adhered, they were treated with different
   drug concentrations for 2 days, after which the cells were collected.
   The collected cells were centrifuged for 3 min at 400× g. After the
   media were discarded, the cells were washed twice with PBS and lysed on
   ice for 15 min. The supernatant was collected, and the protein
   concentration was determined via the BCA (Beyotime, China) method.
   Then, 5 × sample buffer was added, mixed, placed in a metal bath, and
   heated at 95 °C for 10 min. The corresponding loading amount was
   calculated, and SDS‐PAGE was performed. After electrophoresis, the
   proteins were transferred to a PVDF membrane (Millipore, USA). After
   membrane transfer, the membrane was incubated with 5% skim milk powder
   at room temperature for 1 h. Primary antibody was added, and the
   membrane was incubated overnight at 4 °C. The membrane was then
   incubated with secondary antibody at room temperature for 1 h, after
   which images were taken with an imaging system (Bio‐Rad, USA).

   For the inhibitor experiments, the cells were treated with B68 in the
   absence or presence of the proteasome inhibitor MG132 (133407‐82‐6,
   MedChemExpress, USA) or the lysosome inhibitors bafilomycin (Baf,
   88899‐55‐2, MedChemExpress, USA), chloroquine (CQ, 54‐05‐7,
   MedChemExpress, USA), and 3‐methyladenine (3‐MA, 5142‐23‐4,
   MedChemExpress, USA). For cycloheximide (CHX, 66‐81‐9, MedChemExpress,
   USA) chase studies, cells were treated with B68 for 16 h, and
   25 µg ml^−1 CHX was added at the indicated time points. The subsequent
   processing was consistent with the above procedure.

   Protein ubiquitination was analyzed by immunoprecipitation. After
   transfection for 48 h, the cells were collected, an appropriate amount
   of precooled cell lysis buffer (including protease inhibitor) was added
   (MedChemExpress, China), and the cells were lysed on ice for 30 min.
   The cells were then centrifuged at 13 500 × g at 4 °C for 30 min to
   obtain the supernatant of the lysate. IgG or PD‐L1 antibody was added
   to the supernatant of the lysate, which was subsequently incubated with
   rotation at 4 °C overnight. Then, 30 µL of protein A/G beads (Santa
   Cruz Biotechnology, USA) were added and incubated at 4 °C for 3 h.
   Then, 500 µL of IP lysate was added to rinse the magnetic beads, which
   were then washed 5 times. Then, an equal volume of 2× loading buffer
   was added to the magnetic beads, the mixture was heated at 95 °C for
   10 min in a metal bath, and the samples were stored in a ‐20 °C
   refrigerator.

RNA Extraction and Real‐Time PCR Analysis

   Total RNA was isolated from the collected cells via RNAiso Plus
   (Takara, Shiga, Japan), followed by reverse transcription of total mRNA
   (2 µg) into cDNA via a Prime Script RT reagent kit (Yeasen; Shanghai,
   China). Quantitative PCR was performed via SYBR Green (Roche, Mannheim,
   Germany) and a LightCycler96 qPCR system (Roche, Mannheim, Germany).
   The relative expression of target genes was calculated via the ^△△CT
   method. The sequences of the primers used are shown in Table [213]S2
   (Supporting Information).

Detection of Membrane Proteins by Flow Cytometry

   After the cells were treated with different concentrations of drugs,
   they were collected, washed twice with PBS, and incubated with an
   anti‐PD‐L1 antibody at 4 °C for 30 min in the dark. After incubation,
   the contaminated solution was centrifuged and discarded. The cells were
   washed and suspended in PBS. The expression of PD‐L1 was detected via
   flow cytometry (Beckman Coulter Cytoflex, USA).

Immunofluorescence Staining Assay

   After 24 h of treatment, the media in the 12‐well plates were
   discarded, the cells were washed twice with PBS buffer, and then the
   cells were fixed for 15 min with 4% paraformaldehyde. Then, the cells
   were washed 3 times with PBS and incubated with 5% BSA at room
   temperature for 1 h. Then, the primary antibody was added, and the
   cells were incubated overnight on a shaking bed at 4 °C. Then, the
   fluorescently labeled secondary antibody was added, and the samples
   were incubated at room temperature for 1 h in the dark and washed with
   5% BSA (Beyotime, China) 3 times. An anti‐fluorescence quencher
   containing DAPI (Beyotime, China) was added before loading, and the
   expression level of PD‐L1 on the surface of the cell membrane was
   detected with a Cytation 5 instrument (BioTek, USA).

T Cell‐Mediated Tumor Cell Killing Assay

   RKO and HCT116 cells were inoculated in 12‐well plates at a density of
   3 × 10^5 per well, and after the cells had adhered, different
   concentrations of drugs were added and incubated for 24 h. Then,
   activated T cells (Jurkat cells stably transfected with human PD‐1)
   were added for coculture with the tumor cells. T cells were cocultured
   for 24 h at a ratio of tumor cells:T cells = 1:8. After 24 h, the cells
   were washed twice with PBS buffer, fixed with 4% paraformaldehyde
   (Meilun Biotechnology, China), and discarded. After 24 h, the T cells
   were washed twice with PBS buffer, fixed with 4% paraformaldehyde
   (Meilun Biotechnology, China), discarded, washed three times with PBS
   buffer, stained with 0.5 mL of crystal violet staining solution for
   15 min, washed with PBS buffer, and then imaged with Cytation 5. The
   area of the remaining cells was calculated via ImageJ. The cells were
   then stained with 0.5 mL of crystal violet staining solution for
   15 min.

Proteomic Analysis

   After drug treatment, RKO cells were subjected to the SDT lysis method,
   the supernatant was removed, and the mixture was subjected to enzyme
   digestion with trypsin (Promega Corporation, USA) at 37 °C for 16–18 h.
   The collection tube was replaced with a new tube at the end of enzyme
   digestion, 50 µL of TEAB (100 mm) (Thermo Fisher Scientific, USA) was
   added, and the mixture was centrifuged to collect the filtrate. The
   filtrate was collected by centrifugation. The peptides were desalted
   via a C18 cartridge (Shimadzu GL, Japan), and the peptides were
   lyophilized and redissolved via the addition of 0.1% formic acid
   (Thermo Fisher, USA) for BCA (Beyotime, China) quantification. The
   TMTpro 16plex (Thermo Fisher Scientific, USA) label was reconstituted
   with acetonitrile (Thermo Fisher Scientific, USA), added to the
   samples, and incubated for 1 h in the dark, followed by the addition of
   5% hydroxylamine (Thermo Fisher Scientific, USA) to terminate the
   reaction. The samples were combined and dried under vacuum using a
   centrifugal concentrator. The samples were then desalted via a C18
   cartridge; the samples were then graded and analyzed by LC‐MS/MS.

Drug Affinity Target Stabilization Tests (DARTS)

   The basic principle of DARTS is that the binding of small molecule
   ligands to target proteins can stabilize the target proteins and
   increase their resistance to hydrolysis by protein hydrolases. After
   the proteins bind to the small molecule drug, the mixed protein after
   the affinity reaction was enzymatically hydrolyzed, and the enzyme
   hydrolysis product was separated by gel electrophoresis. Compared with
   the blank group, there was an increase in the protein fragments bound
   to the drug. The prepared protein samples were divided equally.
   Different concentrations of B68 were added, and an equal volume of DMSO
   was added to the control group. Protease hydrolysis was subsequently
   carried out to detect differences between the control group and the
   treatment group proteins by immunoblotting and Coomassie brilliant blue
   staining; the samples were prepared by the collection of the gel bands
   of the target proteins, which were analyzed by LC‐MS/MS.

Pull Down the Experiment

   The prepared cell lysate was divided evenly and incubated on a rotator
   at 4 °C for 4 h with Bio, Bio‐B68, 10 × B68 or 20 × B68. Then, magnetic
   beads (Sigma, USA) were added and incubated at 4 °C for 4 h,
   centrifuged at 1000× g at 4 °C for 3 min, and washed with IP lysis
   buffer (Beyotime, China) 5 times to prepare protein samples.
   Electrophoresis was carried out, and the obtained gel was silver
   stained, cut and decolorized. The samples were processed according to
   the proteomics sample preparation method and analyzed by LC‐MS/MS.

Molecular Modeling

   The amino acid sequence of BMI1 was sourced from UniProt. The 3D
   structure of BMI1 was predicted via AlphaFold and Molecular Operating
   Environment (MOE), and a molecular docking model of B68 with the 3D
   structure of BMI1 was constructed in the MOE. Regularized proteins were
   used to identify important amino acids in the predicted binding
   pockets. After energy minimization via the prepared ligand protocol,
   all conformations of B68 were induced to fit into the selected active
   sites via the MOE DOCK module. The docking compounds were scored on the
   basis of their binding patterns at the binding sites.

MST Experiment

   A plasmid carrying the GFP target protein was purchased from Guannan
   Biology (Hangzhou, China), and sequence information for the target
   protein was obtained from NCBI. The BMI1‐GFP and CSN5‐GFP plasmids were
   cis‐transfected into 293T cells for expression, and the cells were
   collected after 48 h. The supernatant was collected after lysis. The
   compounds were mixed with the supernatant separately to prepare sixteen
   gradient concentrations to be tested, and the mixed samples were
   aspirated using a capillary tube and then detected with a MonolithTM
   NT.115 MST (NanoTemper, Germany). The mass action equation in
   NanoTemper software was applied to calculate the Kd value.

Cell Transfection Experiments

   siRNAs targeting BMI1, UBAP2L, YBOX1, and S100A9 were purchased from
   GenePharma (Shanghai, China), and the sequences used are listed in
   Table [214]S3 (Supporting Information). A negative control (NC) was
   used as a control. After the RKO cells had adhered, the double‐stranded
   siRNAs were transfected with Liposome 2000 (Invitrogen, Carlsbad, CA),
   and the medium was replaced with fresh medium after 6–8 h of
   transfection. After 48 h, the cells were subjected to secondary
   transfection, and after incubation, β‐galactosidase staining was
   performed. The plasmids used for transfection were pcDNA3.1‐Ub and
   VP048 pCMV‐MCS‐ 3*Flag‐SYVN1, which were purchased from GenePharma
   (Shanghai, China). The other steps and reagents used were similar to
   those used for siRNA transfection. BMI1‐deficient RKO and HCT116 cells
   were generated by targeting the sgRNA sequence of BMI1 via the
   lentivirus‐Cas9 system. The sgRNA sequences used are shown in Table
   [215]S3 (Supporting Information).

Cell Thermal Shift Assay

   RKO cells were collected, lysed with IP lysis buffer (Beyotime, China)
   for 30 min at 13 500× g, and centrifuged for 15 min. The supernatant
   was collected, divided equally into two portions, and then treated with
   200 µm B68 and an equal volume of DMSO. The plates were rotated and
   incubated for 1 min at room temperature. Each sample was divided into
   100 µL, heated at different temperatures (37 °C, 62 °C, 65 °C, 68 °C,
   71 °C, 74 °C, 77 °C, 80 °C, and 83 °C) for 3 min, and then cooled to
   room temperature. After centrifugation at 4 °C and 13 500 × g for
   15 min, the supernatant was collected and prepared for SDS‐PAGE
   analysis.

Assessment of BMDC Antigen Capture

   The femurs and tibias of C57BL/6J mice were harvested, and the bone
   marrow was isolated, rinsed, and centrifuged at 400× g for 5 min. The
   bone marrow was incubated with red blood cell lysis buffer for 5 min,
   after which the cells were sequentially filtered through 100 and 70 µm
   meshes to remove clumps. A total of 5 × 10^4 cells per well were seeded
   in a 6‐well plate and cultured in RPMI 1640 medium supplemented with
   mouse‐derived GM‐CSF (25 ng ml^−1, Peprotech 315‐03) and IL‐4
   (10 ng ml^−1, Peprotech 214‐14). On days 2 and 4, 3/4 of the medium was
   replaced, and the medium was replenished with GM‐CSF and IL‐4. On day
   6, the dendritic cells (DCs) were harvested and replated to promote
   further maturation. On day 10, the dendritic cells were collected and
   cocultured with senescent cancer cells at a 1:1 ratio for 24 h. The
   expression of DC activation and maturation markers was subsequently
   analyzed via flow cytometry.

Animal Experiment

   All 6–8 weeks old C57BL/6J mice and BALB/c nude mice weighing 18–20 g
   were purchased from Shanghai Jihui Laboratory Animal Breeding Company
   Limited (Shanghai, China) and were generated following the ethical
   obligations of the Department of Laboratory Animal Science, Shanghai
   University of Traditional Chinese Medicine. MC38 cells in the
   logarithmic growth phase were counted by resuspending the cells in
   saline. A total of 8 × 10^5 cells were inoculated subcutaneously into
   C57BL/6J mice. When the average tumor volume reached 50 mm^3, the mice
   were randomly divided into different groups: blank group, control
   group, and B68 and B1 treatment groups. The treatment group was
   injected with 8 mg kg^−1 B68 per day, and the control group was
   injected with the same volume of saline. Mouse body weight and tumor
   size were recorded every two days.

   The axilla of the BALB/c nude mice were inoculated with RKO and HCT116
   cells (10 × 10^5 cells), as described above, and the mice were randomly
   divided into three groups: control group, B68 treatment group, and B1
   treatment group. Mice in the treatment group were injected with
   8 mg kg^−1 B68 or 8 mg kg^−1 B1 daily. Mice in the control group were
   injected with the same volume of saline. Mouse body weight and tumor
   size were recorded every two days.

   For the combined treatment of B68 and CTLA4 antibodies, the mice were
   randomly divided into five groups as described above: control group
   (saline), PD‐1 antibody‐treated group, CTLA4 antibody‐treated group,
   B68‐treated group, B68 and CTLA4 antibody combination group,
   administration concentration PD‐1 antibody (100 µg/100 µL), CTLA4
   antibody (100 µg/100 µL), B68‐treated group (8 mg kg^−1). B68 was
   injected every day until the average tumor volume reached ≈50 mm3, and
   PD‐1 and CTLA4 antibodies were administered every five days. In
   addition, mouse body weight and tumor size were recorded every two
   days.

   At the end of the experimental period, the tumor tissues of the mice
   were removed, photographed, recorded, and weighed. Tumor tissues from
   each group were analyzed by flow cytometry, immunohistochemistry, WB
   strips and β‐galactosidase staining, and the viscera were stained with
   H&E.

   Two drugs, azoxymethane (AOM) and dextran sulfate sodium salt (DSS),
   were commonly used to establish classical mouse models of colitis
   carcinogenesis and can reflect the process of intestinal carcinogenesis
   more realistically. For the AOM/DSS animal model experiments, 6–8 weeks
   old C57BL/6J mice (male) were used. First, mice were induced via
   intraperitoneal injection of 12.5 mg kg^−1 AOM (Sigma, St. Louis, MO).
   After one week of receiving normal drinking water, the mice were
   allowed to drink water containing 2.5% DSS (MP Biomedicals, USA) for
   one week to induce colitis. The water was then replaced with normal
   drinking water for two weeks. The above steps were repeated for three
   consecutive cycles (DSS drinking water‐normal drinking water) to mimic
   the cancerous process of human ulcerative colitis. During the third
   cycle, the concentration of DSS was adjusted from 2.5% to 2%. On day
   65, the mice were sacrificed, and several indicators, including
   intestinal appearance, length, tumor incidence, and mouse body weight,
   were measured.

   For the animal vaccine experiments, for prophylactic immunization
   against cancer cells, senescent cells were resuspended in PBS, and
   1.5 × 10^5 senescent cells were injected subcutaneously into the left
   side of 6–8 weeks old C57BL/6J mice. After seven days, 1.5 × 10^5
   cancer cells were injected subcutaneously into the animals, and mouse
   body weight and tumor size were recorded every two days (DELIXI
   ELECTRIC, China).

   To analyze the adjuvant properties of the cancer cells, 6–8 weeks old
   C57BL/6J mice were subcutaneously injected with the carrier (PBS), and
   1.5 × 10^5 senescent cells were resuspended in 100 µL of PBS. After
   seven days, the subcutaneous inoculation sites were dissected, and
   histological images were analyzed.

Clinical Tissue Samples

   All patients were treated at Longhua Hospital, Shanghai University of
   Traditional Chinese Medicine (Shanghai, China). All samples were
   collected after written informed consent was obtained from patients
   and/or their parents. The study protocol was approved by the Ethics
   Committee of Longhua Hospital affiliated to Shanghai University of
   Traditional Chinese Medicine.

Data Processing

   The data were expressed as mean ± SD or mean ± SEM, with P < 0.05
   considered to indicate statistical significance. The data were collated
   and analyzed using GraphPad Prism 8.0.1. Statistical analysis was
   performed using Student's t test for comparisons between two groups and
   one‐way ANOVA for comparisons among multiple groups. ImageJ was used
   for gray intensity quantification of protein bands and mean
   fluorescence intensity measurements, and FlowJo was used to analyze the
   flow correlation data. The data were analyzed with GraphPad Prism
   8.0.1.

Conflict of Interest

   The authors declare no conflict of interest.

Author Contributions

   H.H., Q.W., D.Y., and X.T. contributed equally to this work. S.L.,
   W.Z., B.L., and H.X. performed conceptualization, original draft,
   methodology, review and editing, funding acquisition and supervision.
   H.H., Q.W., D.Y., and X.T. analyzed the data, carried out the
   experiments, generated the figures and wrote the paper. M.G., H.X.,
   S.T., Q.Z., M.X., X.G., H.Z., C.X., L.L., and S.X. participated in part
   of the experiments. K.C., W.Z., and X.‐W.L. performed
   conceptualization, methodology, and supervision.

Supporting information

   Supporting Information
   [216]ADVS-12-2413122-s001.docx^ (20.1MB, docx)

Acknowledgements