Colorectal cancer (CRC) is one of the leading causes of cancer-related mortality globally, with over 20% of cases linked to chronic inflammation [1]. Colitis-associated colorectal cancer (CAC) represents a particularly aggressive subtype of CRC that occurs in patients with inflammatory bowel disease (IBD) who have been in a chronic inflammatory environment for a long time [2, 3]. The largest cohort study in China showed that homologous recombination pathway gene mutations are another major genetic risk factor for heterogeneous clinical phenotypes of colorectal cancer [4]. Unlike sporadic CRC, which typically progresses through the adenoma-dysplasia-carcinoma sequence, CAC is characterized by the accumulation of somatic mutations that facilitate the progress transition from inflammatory mucosa to dysplasia to carcinoma [5], and patients with CAC are often diagnosed at an advanced stage and with a poor prognosis [6]. Thus far, the management of colorectal cancer has encountered substantial challenges, particularly due to the development of drug resistance in treatments targeting the disease [7,8,9,10]. Studies have shown that mitochondria in colorectal cancer stem cells are a target for drug resistance [11]. Inflammatory damage accelerates the onset and progression of CAC [12]. This inflammatory signaling is mediated through dynamic crosstalk between cancer cells and tumor microenvironment (TME) cells. Therefore, exploring the intrinsic connection between cancer cells and TME cells is also crucial for a better understanding of carcinogenesis.

The TME consists of various types of non-malignant stromal cells, including macrophages, neutrophils, lymphocytes, endothelial cells, and cancer-associated fibroblasts (CAFs). Tumor-associated macrophages (TAMs) play a pivotal role in tumor progression, metastasis, and recurrence following treatment [13, 14]. The plasticity and heterogeneity of macrophages allow them to be classified along the M1-M2 polarization axis [15,16,17]. Tumor-associated macrophages usually exhibit an M2-like phenotype with pro-tumor functions, whereas M1 macrophages have anti-tumor functions. Consequently, reprogramming TAMs to adopt an M1-like phenotype represents a promising strategy to promote tumor regression [18, 19]. Various approaches exist for selecting the M1 phenotype from TAMs or reprogramming TAMs from an M2 to an M1 phenotype, including TLR agonists, monoclonal antibodies targeting M1 phenotype suppressor proteins, and other compounds [20]. Targeted therapy usies utilizing small molecule compounds can specifically target genes in tumor cells, precisely acting on cancer cells to inhibit proliferation and reduce damage to normal cells, providing new insights for cancer treatment. Numerous studies have demonstrated that small molecule monomers derived from traditional Chinese medicine can exert therapeutic effects by targeting specific proteins. Vitexin, a flavonoid compound, has been reported to possess protective effects on the intestines. Our previous research indicated that vitexin significantly modulated macrophage polarization in the intestines of azoxymethane (AOM)/ dextran sodium sulfate (DSS) mice and exhibited protective effects on the intestines [21]. Despite significant advancements in our understanding of tumorigenesis, the complexities underlying the interactions between cancer cells and macrophages following vitexin exposure, particularly within the immune microenvironment, remain inadequately elucidated. The role of vitamin D receptors (VDR) in the pathogenesis of colorectal cancer and colitis has been well established. Notably, patients suffering from both inflammatory bowel disease and colorectal cancer often exhibit vitamin D/VDR deficiencies. Early downregulation of VDR has been observed in the onset of colitis, correlating with the development of larger and more numerous tumors in VDR-deficient models of CAC [22, 23]. Recent investigations have indicated that vitamin D activity becomes dysregulated in advanced cancer stages, although it is known to modulate the interactions between immune and cancer cells, thereby inhibiting the production of pro-inflammatory cytokines [24]. Despite extensive research on vitamin D, critical questions regarding the biological role of intestinal VDR during the transition from colitis to CAC remain unresolved. Therefore, it is important to search for novel VDR agonists and thus explore their therapeutic role and mechanisms in inflammatory cancers. Similarly, the changes occurring in macrophages during the transition from inflammatory bowel disease to CAC are complex. Within the tumor microenvironment, both VDR and macrophages are implicated in the progression of CAC, but whether they are closely linked or functionally independent remains unclear. Considering the multiple functional roles of the VDR in the development of CAC, the cellular and molecular mechanisms by which the VDR regulates macrophages and thereby protects the host are important.

In this study, we established a mouse model of chronic colitis progressing to colorectal cancer (mid-term CAC) and demonstrated that myeloid-specific VDR knockout mice exhibit low VDR expression in the colon, which negatively correlates with the presence of M2-type macrophages. We confirmed that vitexin can target VDR and promote its transcriptional functions. We used a co-culture of macrophages and cancer cells to simulate the vitexin regulation of macrophages in the tumor microenvironment in a VDR-dependent manner and found that this regulation is achieved through the VDR/phenazine biosynthesis-like domain-containing protein (PBLD) pathway. In addition, we used a mouse model with myeloid-specific VDR knockout to study the protective effect of vitexin on VDR during inflammation and tumorigenesis. Our findings elucidate novel pathways and candidate therapeutic agents for addressing the transformation of chronic intestinal inflammation into colorectal cancer by regulating the polarization of macrophages in the tumor microenvironment.

Materials and methodsMaterials

Vitexin (HPLC purity > 98%), and Calcitriol were purchased from Sigma-Aldrich. AOM were purchased from Sigma Aldrich. DSS were purchased from MP Biomedicals. Pronase were purchased from Roche. For in vitro experiments, vitexin was dissolved in DMSO and Calcitriol in water, and for in vivo studies, vitexin was dissolved in 1% carboxymethylcellulose sodium (CMC-Na) solution and Calcitriol in saline. Recombinant human protein VDR was constructed at KMD Bioscience (Tianjin, China) and recombinant human protein VDR-LBD was constructed and sequenced at DetaiBio. (Nanjing, China). APC anti-Mouse F4/80 Antibody and FITC anti-Human/Mouse CD11b Antibody were purchased from MULTI SCIENCES. PE/Cyanine7 anti-mouse CD206 (MMR) Antibody, PE anti-Nos2 (iNOS) Antibody, PE/Cyanine7 anti-human CD206 (MMR) Antibody and PE anti-human CD86 Antibody were purchased from BioLegend. PcDNA3.1(+)-VDR-3×FLAG, pGL4.10-CYP24A1 promoter, Wide-Type, pCMV3-flag-ratVDR, pCMV3-flag-ratVDR- T287A were constructed and sequenced by OBio. (Shanghai, China). Extraction kits for cell membrane and nucleus fractions were performed by beyotime (Shanghai, China).

Animals

SPF-grade Lyz2-Cre (JAX 004781) was purchased from the Jackson Laboratory, VDRfl/fl was purchased from GemPharmatech (Jiangsu, China), and Lyz2Cre-VDRfl/fl (VDRΔMΦ) mice were generated by crossing VDRfl/fl (VDRflox) mice with Lyz2-Cre (Figure S2, Supporting Information). The purchased VDRFl/wt transgenic male mice will be co-housed with wild-type female C57BL/6J background mice at a ratio of 1:2 for mass breeding. After the newborn mice reach 2 weeks of age, they will be numbered and their tails will be clipped for genetic identification, with the identified PCR primers (Table S1, Supporting Information).

SPF-grade male C57BL/ 6 J mice (18–22 g) were purchased from Guangzhou University of Chinese Medicine Laboratory Animal Centre. All mice were crossed with C57BL/6 for at least 10 generations. The animals were housed in the SPF-grade laboratory animal room of the Guangzhou University of Chinese Medicine and were given normal feed and free drinking water every day, with a relative humidity of 55 ± 5%, an indoor temperature of 22 ± 2 ℃, and 12 h of light. The bedding was changed every other day. The animals were acclimatized to the laboratory for at least 2 weeks before the start of the study. All animal experiments were conducted and analyzed in a blinded randomization manner.

AOM-DSS induced mid-stage colorectal cancer model

Mice were injected intraperitoneally with 10 mg/kg AOM (A5486, Sigma), followed by two cycles of 1.5% DSS in the drinking water on days 8–15 and 22–29 (referenced to previously published [25] literature but adapted). Disease progression was monitored by body weight measurements and DAI examination, and dissection was determined on day 36. The methodology for the determination of DAI is described in detail in Supplementary Table 2. At the end of the experiment, mice were executed, colon harvested and colon weight and length measured. Tumor load was quantified post-mortem by macroscopic examination of the colon.

Cell culture

THP-1 (Human Acute Monocytic Leukemia Cells) and RAW 264.7 (Mouse Mononuclear Macrophages Cells), is the most commonly used inflammatory cell model. CT26.WT (Mouse Colorectal Carcinoma Cells) are undifferentiated cells induced by N-nitroso-N-methylurea (NNMU). HCT 116 (Human Colorectal Carcinoma Cells) was isolated from male patients with colon cancer. Human Embryonic Kidney Cells (HEK) 293T cells, used as tool cells for plasmid transfection. All cells were purchased from Procell (Wuhan, China). CT26.WT, HCT116 cells were cultured in RPMI-1640 medium, and RAW264.7, HEK293T were cultured in DMEM medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 100 U/ml penicillin-streptomycin. THP-1 cells were cultured in RPMI-1640 supplemented with 10% heat-inactivated FBS and 100 U/ml penicillin-streptomycin, with the addition of 0.05 mM β-mercaptoethanol, and PMA activation (Sigma, USA) THP-1 cells were used as human macrophage studies.

PBLD siRNA

THP-1 cells were transfected with PBLD or control siRNA (Tsingke, China) using Lipofectamine® 6000 Transfection Reagent (Beyotime, China) in Opti-MEM I Reduced Serum Medium for 72 h. Then, the transfected cells were used for further research.

Cellular RNA sequencing

Total RNA was extracted from THP-1 cells using TRIzol (Accurate Biotechnology, China). Library sequencing was performed by Gene Denovo Biotechnology (Guangzhou, China) on an Illumina HiseqTM 2500/4000. Bioinformatics analysis was performed using Omicsmart, a real-time interactive online platform for data analysis (http://www.omicsmart.com).

PCR assay

Gene expression profiles were analyzed by macrophage differentiation PCR arrays (Wcgene Biotech, Shanghai, China) according to the manufacturer’s protocol. β-actin and Gapdh were used as endogenous controls. Data were normalized to the reference gene based on cycle threshold (Ct) values. log2 (fold change) was calculated based on the 2−ΔΔCt method.

Immunofluorescence staining and immunohistochemistry

The method according to a previous study [26]. Tissue wax blocks were hydrated by serial dewaxing, while cells were fixed by 4% paraformaldehyde. Then, after permeabilization with 0.05% Triton X-100 and sealing with 5% BSA, the cells were incubated with primary antibody at a dilution of 1:200 at 4 °C overnight. Fluorescently labeled secondary antibodies were used the following day. The stains were then counterstained with DAPI nuclear stain.

The operation of immunohistochemistry is carried out as described above. Following the procedures of primary antibody at a dilution of 1:200 at 4 °C overnight. After incubating with the secondary antibody the next day, DAI staining was performed. The sections were then stained with hematoxylin to visualize nuclei. The final processing steps included dehydration in graded alcohols, clearing in 100% xylene, and mounting. Images were taken with a fluorescence microscope equipped with a digital camera (Nikon, Tokyo, Japan).

Quantitative real-time PCR (qRT-PCR)

The method according to previous studies [27, 28]. Total RNA was isolated from cells using TRIzol (Accurate Biotechnology).The concentration of total RNA was determined using a Nanodrop 2000, and 1 µg of total RNA was converted to cDNA by reverse transcription according to the PrimeScript™ RT kit instructions in a CFX96 Touch The target gene was detected in a CFX96 Touch Real-Time RCR Detection System (Bio-Rad) detector using a SYBR Green PCR master mix kit with GAPDH as an internal reference, and the primer sequences of the target gene were designed by Shanghai Sangong Biotechnology Co. The primer sequences of target genes were designed by Shanghai Bioengineering Biotechnology Company. The relative expression of target genes was calculated by using the 2-ΔΔCt method for quantification, and the correlation analysis was carried out. Primer sequences are shown in Supplementary Table S3.

Western blot

The method is according to previous study [29]. In brief, proteins were isolated from cells and tissues and assayed by BCA assay. Proteins were separated by 8-12% gel electrophoresis separation kit (Beyotime, China) and transferred to polyvinylidene fluoride (PVDF) membranes using a semi-dry transfer apparatus (Bio-Rad). The membranes were blocked with 5% BSA for half an hour, and then the PVDF membranes were incubated with specific primary antibodies. The antibodies used in this study were as follows: VDR (Signalway Antibody, China, Cat#38397, 1:1000); PBLD (Proteintech, China, Cat#68317-1-Ig, 1:1000); PCNA (Affinity, China, Cat#AF0239, 1:1000); GAPDH (Affinity, China, Cat#AF7021, 1:3000); Histone (Affinity, China, Cat#BF9211, 1:3000). Immunoreactive bands were detected the following day with horseradish peroxidase-conjugated secondary antibodies and visualized by enhanced chemiluminescence. Analyses were performed using Image J analysis software and standardized against their respective controls.

Flow cytometry analysis

The method according to a previous study [30]. Cells were collected and resuspended in 500 µL of phosphate-buffered saline (PBS). Following a 30-minute incubation at room temperature with surface flow antibodies, the cells were centrifuged at 900 rpm for 4 min. The supernatant was discarded, and the cells were washed with PBS before being treated with a cell membrane-disrupting solution (Thermo Fisher, diluted 1:3) for 60 min. Afterward, the cells were washed again in PBS and incubated with polarization-associated flow antibodies for 1 h at room temperature. The cells were subsequently washed and resuspended in PBS containing 1% bovine serum albumin (BSA). Following another wash, the cells were resuspended in PBS with 1% BSA and analyzed using a NovoCyte Quanteon flow cytometer. Data were processed using NovoExpress software, with specific labeling for F4/80, CD11b, iNOS (M1), and CD206 (M2) flow-through antibodies.

VDR binding assayThe binding of vitexin to VDR was determined by several methods

cellular thermal shift assay (CETSA), drug affinity responsive target stability (DARST), surface plasmon resonance assay (SPR), isothermal titration microcalorimetry assay (ITC), pull-down assay and immunofluorescence co-localization assay.

For CETSA, the CETSA method was performed according to the literature [31]. THP-1 cells were inoculated in a 100 mm dish and subsequently treated with FBS-free medium containing either 0.1% DMSO or 100 µM vitexin for 2 h the following day. Post-treatment, the cells were digested with trypsin, collected, and washed before being resuspended in 1 mL of PBS supplemented with a protease inhibitor. A 90 µL aliquot of this suspension was transferred into 0.2 mL PCR tubes and subjected to heating in a PCR machine for 3 min at temperatures ranging from 43 to 67 °C. 20 µL of RIPA buffer was added to each tube, and the precipitated proteins were removed by centrifugation at 15,000 × g for 20 min at 4 ℃ after thorough mixing. Subsequent supernatant manipulation was based on general Western blot experiments.

For DARTS, the DARTS method was performed according to the literature [32]. THP-1 cells were inoculated in a 100 mm dish at an appropriate density. The following day, the cells were harvested and lysed in 500 µL of RIPA buffer containing a protease inhibitor. The resulting lysate was aliquoted equally into 5 Ep tubes and subsequently diluted tenfold in TNC buffer (50 mM Tris-HCl, pH 8.0, 50 mM NaCl, 10 mM CaCl2). The samples were then incubated with 0.1% DMSO and varying concentrations of vitexin (0, 2.5, 5, and 20 µM) for 1 h at room temperature with gentle shaking. The samples were then proteolytically cleaved with protease (2.5 µg/mL) for 10 min. Subsequent operations were based on general Western blot experiments.

For SPR experiments, the interaction between vitexin and VDR (ligand-binding domain, LBD) protein was analyzed using a Biacore T200 system (GE Healthcare, Uppsala, Sweden). Recombinant VDR (LBD) at a concentration of 40 µg/mL was used to achieve non-covalent immobilization on the surface of the activated chip. The final level of immobilized VDR-LBD was approximately 12,000 reaction units (RU). Subsequently, different concentrations of vitexin (ranging from 6.25 µM ~ to 200 µM) were injected at a flow rate of 30 µL/min, and 1× PBST (1.37 M NaCl; 26.8 mM KCL; 81 mM Na2HPO4; 17.6 mM KH2PO4; Ph7.2-7.4, 0.05% Tween 20) was used as the running buffer. The results were analyzed using Biacore evaluation software (T200 version 1.0) and curves were fitted in 1:1 binding mode.

For the ITC experiments, the potential interaction between vitexin and VDR (LBD) was determined using the Nano-ITC instrument (TA instruments, USA) at 25 °C. A solution of 10 µM VDR (LBD) and 200 µM vitexin were dissolved in phosphate-buffered saline (PBS) containing 5% DMSO, and stirred at 250 rpm. Twenty titrations of 2.5 µL each were performed for each titration experiment. The heat of dilution of VDR protein was determined by titrating it into PBS. Data analysis was performed using the NanoAnalyze software package (TA Instruments). The total heat exchange during each injection of VDR into the vitexin solution was fitted to an independent model with variable parameters.

For pull-down experiments, Beaver-Beads streptavidin and biotinylated vitexin were utilized in this study. Specifically, 100 µL of biotinylated vitexin glycoside was added to 10 µL of streptavidin-agarose beads and incubated for 2 h at 25 °C. Controls included biotin, unbiotinylated vitexin, and untreated beads. The full-length prokaryotic proteins of the VDR were constructed at KMD Bioscience (Beijing, China), while the prokaryotic proteins corresponding to the VDR-LBD were constructed at Detai Bioscience (Nanjing, China). And mutant VDRs (T287) of the LBD structural domain was achieved by transfecting HEK-293 cells with the encoding plasmid under Lipofect 2000. Lysates were prepared from HEK-293 cells of the constructs and then the lysates were mixed into treated streptavidin-agarose beads. The mixture was incubated at 25 °C for 3 h with gentle shaking. The samples were then spun and washed three times. The samples were boiled with 5× loading buffer and loaded onto 10% polyacrylamide gels for western blot analysis.

For immunofluorescence co-localization experiments, THP-1 cells were seeded at a density of (1 × 103 cells/well) were inoculated in confocal dishes, and 60 µL of biotin-vitexin (20 µM), and biotin (20 µM) intervened in the incubation for 24 h. The cyto-fluorescence staining process was started. The cells were first washed once with PBS and then fixed with 4% paraformaldehyde for 10 min. the cells were washed three times with PBS, each time for 3 min, and then closed with 5% BSA for 30 min, and diluted with FITC-avidin as well as VDR antibody, and then incubated at 4 ℃ overnight. At the end of the incubation, the cells were washed three times with PBS for 3 min each time, a fluorescent secondary antibody (Alexa Fluor Plus 594) was added, and the cells were incubated for 1 h. The nuclei were washed three times with PBS for 3 min each time, and the nuclei were stained by DAPI for 5 min. After the washing with PBS, the co-localization of the green light and the red light in the cells by laser confocal microscopy was observed.

Molecular docking

The crystal structure of the LBD structural domain (PDB:1QBD) of human VDR was obtained from the Protein Data Bank. Initial structures of ligands and receptors for docking were prepared with MGLTools 1.5.6 (The Scripps Research Institute, CA, USA). Molecular docking was performed with AutoDock Vina 1.0.2. The binding affinity of each docking pose of the oyster glycosides was recovered by the MM/GBSA method in the AmberTools18 software package [33]. Finally, key residues for protein-ligand interactions were identified based on the breakdown energy calculations for each residue.

ChIP-seq and PCR

THP-1 cells were seeded at a density of 4 × 106 in 100 mm dishes. After 24 h, cell attachment was induced by treatment with 100 ng/ml of phorbol 12-myristate 13-acetate (PMA). Subsequently, the cells were preincubated with either dimethyl sulfoxide (DMSO) or vitexin at a concentration of 20 µM for an additional 24 h. Chromatin immunoprecipitation (ChIP) was performed in accordance with the protocol outlined in the Thermo Fisher manual (Catalog No. 26157). Following treatment, THP-1 cells were harvested and crosslinked using 1% paraformaldehyde for 15 min, after which 125 mM glycine was added to quench the crosslinking reaction. Chromatin was sheared into 200–500 bp DNA by sonication using a Biorupter (Diagenode, UCD-200). After reserving 5.0 µl of the sheared chromatin for input to the control samples, the rest of the sheared chromatin was rotationally incubated with 5 µg of VDR (CST, USA, Cat#12550, 1:50) or IgG at 4 °C overnight. The next day, samples were added to protein A/G dynamic beads and incubated at 4 °C for 2 h. The DNA-protein complexes were washed twice with dilution buffer and then eluted from the dynamic beads with SDS buffer. Purified ChIP DNA was obtained after reverse cross-linking, proteinase K digestion, phenol/chloroform extraction, and ethanol precipitation. For ChIP-sequencing, input DNA and ChIP DNA were used to generate sequencing libraries using the Illumina DNA Sample Preparation Kit according to the manufacturer’s protocol. Briefly, the DNA samples were end-repaired, then ligated with barcode adapters, and finally amplified and purified. These DNA libraries were sequenced in an Illumina. nextSeq500 sequencer, based on the 35nt pair-end sequencing protocol. Chromatin immunoprecipitation analyses and RNA sequencing were performed with the help of Wuhan GeneBen Biotechnology Co Ltd (Wuhan, China). The promoter primers Cyp24a1 and PBLD were subsequently constructed and detected by RT-PCR.

Luciferase reporter gene assay

HEK293T cells were seeded in 6-well plates at 1 × 105 cells/well, wall-adhered, and transiently transfected with pGL4.10, PGL4.10-CYP24A1, pcDNA3.1-VDR-3×FLAG, and TK-luc using the Lipofectamine 3000 reagent for 6 h. The transfection solution was discarded and added to the high and medium in turn, low concentrations of vitexin (20, 10, and 5 µM) and Calcitriol (50 nM) after induction for 48 h. The transfection solution was washed twice with PBS, and the dual-luciferase reporter gene assay system was used to measure the dual-luciferase activity and to study the effect of vitexin on the transcriptional activation of the CYP24A1 promoter.

Histological assessment

Colon tissues fixed in 4% paraformaldehyde were paraffin-embedded and sectioned were used for histological analysis. The sections were stained with hematoxylin and eosin (H&E) for routine histological evaluation. For immunohistochemistry, sections were deparaffinized and hydrated. Endogenous peroxidase was blocked with 3% H2O2 for 30 min. Sections were then incubated with primary antibody at a dilution of 1:100 overnight at 4 °C. Secondary antibodies were incubated at 1:200 for 1 h at room temperature and immunoreactivity was detected by diaminobenzidine (DAB). All sections were counterstained with hematoxylin.

Statistical analysis

All in vitro data represent at least three independent experiments and in vivo data represent at least six independent experiments. All experimental data are expressed as mean ± SEM (standard error of the mean). Statistical analyses were performed using GraphPad Pro Prism 8.0 (GraphPad, San Diego, CA). One-way analysis of variance (ANOVA) followed by Tukey’s test was used to analyze differences between groups of data. A p value of < 0.05 was considered a significant difference.