BRD9 promotes the progression of gallbladder cancer via CST1 upregulation and interaction with FOXP1 through the PI3K/AKT pathway and represents a therapeutic target

Introduction

Gallbladder cancer (GBC) is one of the most common and aggressive malignant tumors of the human biliary system and originates from epithelial cells of the gallbladder mucosa [1]. Gallbladder stones and polyps with chronic inflammation are important risk factors of GBC [2]. Currently, the only treatment for GBC is surgical removal [3]. In the early stages of GBC, patients are often asymptomatic; therefore, most patients diagnosed with GBC often miss the best opportunity for surgical treatment and have a very poor prognosis [4,5,6]. There is, therefore, an urgent need to identify molecular markers, and deep research of the molecular mechanisms of GBC progression could facilitate the development of novel targeted drugs and improve the overall prognosis.

Bromodomain proteins (BRDs) are evolutionarily highly conserved in most tissues [7]. BRDs are widely located in protein-protein interaction modules with different catalytic and scaffolding functions and can play important roles in gene expression by selectively recognizing and specifically binding to acetylated lysine residues of histones [8]. When tumors develop, BRD-containing proteins are frequently dysregulated, involved in gene fusions, and produce different, usually oncogenic, proteins with many oncogenic mutations located on BRDs [9].

Further investigation of many structural domains of BRDs is required, among which is the SWI/SNF complex of bromodomain-containing protein 9 (BRD9) that is involved in the gene transcription, DNA repair, and cell differentiation [10, 11]. The high mutation rate of SWI/SNF subunits in human tumors and the overexpression of BRD9 in different tumor types suggest that BRD9 is an important potential target in the study of anti-tumor therapeutic agents [12]. BRD9 is a key regulator of acute myeloid leukemia tumorigenesis and has been confirmed to play a strong oncogenic part in blood cells [13]. In patients suffering melanoma, studies have shown that both the mRNA and protein levels of BRD9 are overexpressed and working in connection with poor prognosis [14]. Similarly, BRD9 is a synthetic lethal target of malignant transverse tumors and synovial sarcomas, allowing tumors to have driver mutations in and interact with the core components of the SWI/SNF complex [15]. However, the role and molecular mechanisms of BRD9 in GBC are poorly understood.

We found that BRD9 overexpression was significantly linked to GBC development. Mechanistically, we found that BRD9 regulates cystatin 1 (CST1) expression by binding the transcription factor forkhead BOX P1 (FOXP1) to the CST1 promoter. In addition, activation of the CST1/PI3K/AKT pathway was critical for regulating the proliferation of GBC cells. These results demonstrate the role of BRD9 in GBC, its potential as a diagnostic marker, and that I-BRD9 could be a new promising drug agent for GBC.

Materials and methods

Patients and clinicopathological data

Gallbladder cancer tissues and adjacent non-tumor tissues (paraneoplastic at least 2 cm from the tumor margin) were obtained from the Department of General Surgery, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine (Shanghai, China). None of these patients had received radiotherapy, chemotherapy, or immunotherapy before surgery. Informed consent was obtained from all patients, all experiments were approved by the Research Ethics Committee of Xinhua Hospital, Shanghai Jiao Tong University School of Medicine (Shanghai, China), and all affected tissue specimens were confirmed by pathological diagnosis. The cancer staging of each patient was based on the American Joint Committee on Cancer Staging 8th edition manual.

Cell culture and reagents

The cell lines used in this study included human embryonic kidney cells (HEK-293T), gallbladder cancer cell lines (NOZ, GBC-SD, OCUG1, EH-GB1, SGC-996, and ZJU-0430), and all of the above cell lines were obtained from the cell bank of the Shanghai Institutes for Biological Sciences of the Chinese Academy of Sciences (Shanghai, China), and the cells were cultured in DMEM high-sucrose medium (Gibco) supplemented with 10% fetal calf serum (Gibco). The authenticity of the cell lines involved in the experiments was confirmed by short tandem repeat (STR) analysis. DMEM high glucose medium (Gibco) added with 10% fetal bovine serum (Gibco), and all cells were cultured in a humid chamber at 37 °C with 5% CO2. I-BRD9 was purchased from MedChemExpress, and 740Y-P (PI3K activator, #HY-P0175; 20 μM, treatment time 24 h) was purchased from MedChemExpress, and all the reagents were dissolved in DMSO.

Cell transfection

To knock down the expression of BRD9 gene, 3 pairs of siRNAs were synthesized using gene editing technique (see annex for sequences). In addition to this, the genes MXI1, SRY, Arid3a, ZNF354C, and FOXP1 were knocked down using different siRNAs for them (see Appendix for the sequences) According to the protocols provided by the reagent manufacturers, rfect reagent (Baidai, China) was used as a transfection reagent to transfect NOZ, and GBC-SD, and the respective transfection was verified with qRT-PCR Efficiency. The siRNA was diluted to 50 nM with 50 uL Opti-MEM (Gibco), followed by a 5 uL dilution of RFect (Baidai) with 50 uL Opti-MEM (Gibco). The two dilutions were then mixed and allowed to stand at room temperature for 15 min. The mixture was then added to a cell culture dish containing 1 mL of Opti-MEM (Gibco), and the medium was replaced with fresh medium after 6 h. Ectopic CST1 overexpression lentivirus was constructed based on the full-length sequence of human CST1, and the empty vector was used as a control. Lentiviral infection was performed according to the manufacturer’s instructions.

Quantitative real-time PCR

Total RNA was extracted from tissue samples, 293T, NOZ, OCUG1, EH-GB1, GBC-SD, SGC-996, and ZJU-0430 cells, respectively, using Trizol reagent (Invitrogen, USA), and cDNA was generated by reverse transcription using Takara’s Prime Script RT master mix. SYBR Green (Takara, RR420A, Japan) was used to detect the expression of target genes on a StepOnePlus™ real-time PCR system. The primer sequences used for amplification are shown in the Table. In this experiment, GAPDH was used as an internal reference for normalization.

Western blot analysis

The total protein extraction and western blotting procedure can be referred to the literature [1]. In brief, proteins were separated using RIPA lysis buffer (Beyotime) and protein quantification was performed for each protein sample using the BCA method (Beyotime, China). Proteins were then separated using SDS-PAGE gels and subsequently transferred to PVDF membranes (Millipore). They were blocked with 5% skimmed milk for 1 h at room temperature, followed by washing three times for 5 min each, after which they were incubated with primary antibody at 4 °C overnight. The following day the blot was co-incubated with goat anti-rabbit enzyme-labeled secondary antibody (Beyotime) for 1 h at room temperature, followed by chemiluminescence detection of the immunoreactive bands and visualization of the results using Gel Doc 2000 (Bio-Rad). All antibodies used are listed in Table 1.

Table 1 Information of antibody.

Full size table

Cell proliferation assay and IC50 assay

Cell proliferative capacity and IC50 assay of NOZ and GBC-SD cells were performed using Cell Counting Kit-8 (CCK-8) (YEASEN). NOZ and GBC-SD cells were inoculated into 96-well plates with a system of 100 µL per well and a cell density of 1000 cells/well and incubated overnight. The next day, the original medium was discarded and fresh medium containing different concentrations of I-BRD9 was added. After incubation for 24 and 48 h respectively, the original medium was discarded, 10 µL of CCK-8 reagent and 100 µL of complete medium were added to each well, and the cells were incubated for 2 h at 37°C, protected from light, and the IC50 value was measured at 450 nm with an enzyme marker and the cell proliferation curve was plotted.

Colony formation assay

Colony formation assay was performed as previously described [16]. Briefly, NOZ and GBC-SD cells were inoculated in 6-well plates at a density of 1*103 cells/well. The following day, they were transfected with siRNA targeting BRD9 or treated with different concentrations of I-BRD9, and after 24 h of incubation, the medium was changed to fresh medium and incubated for 2 weeks. After 2 weeks, the cells were washed three times with PBS, fixed with 4% paraformaldehyde for 15 min, and stained with 0.1% crystal violet for 30 min. The cells were counted using the Image J software.

Dual-luciferase assay

Wild-type and mutant (CST1 promoter truncated at different positions) plasmids were cloned on pGL3 vector, and after 24 h of cell culture, cells were collected and assayed for luciferase activity according to the manufacturer’s protocol (Dual-Luciferase Reporter Gene Assay Kit, YEASEN). The results were normalized.

Chromatin immunoprecipitation (ChIP)

The interaction between FOXP1 protein and CST1 was verified by ChIP experiment, and corresponding operations were performed according to the instructions of ChIP detection kit (Beyotime). We then validated the enriched fragments using DNA gel electrophoresis.

CO-IP

Cell lysates were extracted and then lysates were warmed with Flag affinity beads for wash buffer to wash the agarose beads. Interacting proteins were detected by western blot.

Subcutaneous xenograft model in nude mice

This study was approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. 4-week-old male nude mice were purchased from Shanghai Laboratory Animal Center, Chinese Academy of Sciences (Shanghai, China) and fed under specific pathogen-free (SPF) conditions. We followed the ethical principle of reducing the number of animals used and used an appropriate sample size of animals for the experiments. In order to avoid human subjectivity biasing the observation of experimental results, a blind experiment method was used. The experiment was conducted using completely randomized grouping method for grouping. Twenty mice were randomly divided into two groups: si-group and drug treatment group. The 10 mice in si-group were randomly divided into two groups (5 mice/group). One group was injected with 5 × 106 NOZ cells in the right axilla of nude mice, and the other group was injected with 5 × 106 LP-ShBRD9 NOZ cells. The length and width of the tumor were measured with a cursor caliper every week and the volume was calculated (1/2 * width2 * length). Four weeks after injection, the mice were euthanized, and the tumor was removed for follow-up study. Ten mice were randomly divided into two groups (5 mice/group), and 5 *106 NOZ cells were injected into the right axilla of the two groups of nude mice to construct a tumor xenograft model, and after the tumor grew for 7 d, the nude mice in the two groups were treated by oral gavage every 2 d with the drug-loaded group (10% EtOH, 30% PEG400, 60% MCT [0.5% methyl cellulose, 0.5% Tween 80]) and I-BRD9 group (30 mg/kg dissolved in I-BRD9 10% EtOH, 30% PEG400, 60% MCT) by gavage [17], and the tumor growth was evaluated weekly by vernier calipers, and after four weeks, the mice were executed and the tumors as well as the tissues of heart, liver, lungs, and kidneys were collected for further examination. In addition, mouse ophthalmology was taken for routine blood and liver and kidney function tests.

Statistical analysis

Statistical analysis was performed using GraphPad Prism (V9.5.1, Prism) software. The experimental data followed a normal distribution. Significant differences between two groups were analyzed using Student’s t test and one-way ANOVA between multiple groups. Results were expressed as mean ± standard deviation (SD). p-values < 0.05 were considered statistically significant differences. In addition, we used Pearson χ2 test to correlate BRD9 expression with clinicopathologic parameters. One-way survival analysis was performed using the Kaplan–Meier test.

Results

Expression of BRD9 is elevated in GBC and correlated with poor prognosis in patients with GBC

We analyzed of the Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/geo/) datasets, GSE139682 and GSE7663, and found that the expression of BRD9 was higher in gallbladder tumor tissues compared to precancerous tissues (Fig. 1A). To clarify the specific relationship between the expression of BRD9 and GBC tissues, we performed qRT-PCR using surgically resected specimens from 38 patients with GBC, collected from our institution. The results showed that BRD9 expression was significantly upregulated in GBC tissues compared with the corresponding non-cancerous tissues (Fig. 1B, C). This result was confirmed by western blotting that found the expression of BRD9 in GBC tissues was obviously higher than that in paired adjacent non-cancerous tissues (Fig. 1D). Based on the fold change values, we divided patients into BRD9 high expression (fold change ≥ 2) and low expression (fold change <2) groups. High expression of BRD9 was closely associated with histology (P < 0.001); lymph node metastasis (P = 0.014); and tumor, node, metastasis stage (P = 0.002) evaluated by Pearson’s χ2 test (Table 2). Kaplan–Meier analysis showed that patients with high BRD9 expression had significantly shorter postoperative overall survival than those with low BRD9 expression. The expression of BRD9 in GBC cell lines was confirmed by qRT-PCR and western blotting which demonstrated that GBC cell lines NOZ, OCUG1, EH-GB1, SGC-996, ZJU-0430 and GBC-SD showed significantly higher BRD9 mRNA and protein expression compared with 293T cells (Fig. 1F, G). Collectively, the above results indicate that BRD9 may be correlated with the development and prognosis of GBC and warrants further investigation.

**Fig. 1: BRD9 expression is elevated in GBC and correlates with poor prognosis in GBC.**

Table 2 Association of BRD9 expression with the clinicopathological characteristics of GBC.

Full size table

BRD9 promotes GBC cell proliferation

To further elucidate the role of BRD9 in GBC, we knocked down BRD9 using siRNA. All three BRD9-siRNAs significantly downregulated BRD9 in NOZ and GBC-SD cells. Three different BRD9 siRNAs significantly reduced BRD9 protein levels (Fig. 2A). siRNA2 and siRNA3 showed good silencing effects at both the mRNA and protein levels, therefore, these siRNAs were used in subsequent experiments. CCK-8 results showed that BRD9 knockdown caused a significant decrease in cell proliferation (Fig. 2B). We performed colony formation and EdU assays to detect the colony formation ability and DNA replication activity of tumor cells, respectively, which showed that BRD9 knockdown reduced the colony formation ability and proportion of EdU-positive cells in NOZ and GBC-SD cells compared to negative controls (Fig. 2C, D).

**Fig. 2: BRD9 Promotes gallbladder cancer cell proliferation.**

In the GBC subcutaneous tumor model, tumors grew slower in the shBRD9 group than in the control group (Fig. 2E). In addition, tumor volume (Fig. 2F) and weight (Fig. 2G) were notably decreased in the shBRD9 group. In summary, BRD9 promotes GBC cell proliferation both in vitro and in vivo.

The BRD9-selective inhibitor I-BRD9 inhibits GBC cell proliferation and suppresses GBC growth in vivo and has a favorable drug safety profile

Our results suggested that BRD9 inhibition suppresses tumor cell proliferation, therefore, we assessed a selective inhibitor of BRD9 to replace the use of siRNA. Treatment with the BRD9 selective inhibitor I-BRD9 had significant time- and dose-dependent effects on the proliferation of NOZ and GBC-SD cells (Fig. 3A). The I-BRD9 IC50 values for NOZ and GBC-SD cells for 48 h were 5.05 and 3.76 µM, respectively, and these concentrations were used for subsequent experiments. Western blotting verified that I-BRD9 significantly reduced BRD9 protein expression (Fig. 3B). Colony formation assays manifested that I-BRD9 significantly reduced the number and size of NOZ and GBC-SD cell clones (Fig. 3C). In addition, the proportion of EdU-positive cells was significantly reduced in the I-BRD9 group compared to that in the control (Fig. 3D). These experiments confirmed that I-BRD9 has anti-tumor effects in vitro. We then investigated whether I-BRD9 had the same effect in vivo using the GBC mouse tumor model and found that I-BRD9-treated mice had significantly slower tumor growth than control group (Fig. 3E). Furthermore, both the tumor volume (Fig. 3F) and weight (Fig. 3G) were significantly reduced. HE staining revealed that the appearance and morphology of the major organs, heart, liver, spleen, lungs, and kidneys of I-BRD9-treated mice did not show significant damage (Fig. 3H). Routine blood tests indicated that I-BRD9 caused no significant systemic toxicity (Additional file 1: Supplementary Table S3). These results indicate that I-BRD9 caused the inhibition of the proliferation of GBC cells in vitro and tumor growth in vivo without any obvious drug toxicity. Collectively, we propose that BRD9 exhibits druggable susceptibility in GBC and that I-BRD9 may play a crucial part in clinical diagnosis and treatment.

**Fig. 3: BRD9-selective inhibitor I-BRD9 inhibits GBC cell proliferation and suppresses GBC growth in vivo with a favorable drug safety profile.**

Activation of the PI3K/AKT pathway reverses the inhibitory effect of I-BRD9 on GBC cell proliferation

To further identify the potential mechanism underlying the anti-tumor effects of I-BRD9 in GBC, RNA-seq was performed on total RNA from NOZ cells treated with I-BRD9 or DMSO for 48 h. Figure 4A shows the volcano plot for the differently expressed genes identified in the comparison between treatment with I-BRD9 and the untreated controls. Gene Ontology (GO) analysis identified that differentially expressed genes were enriched in various terms including “signal transduction,” “cell adhesion,” “membrane,” “plasma membrane,” and “protein binding” (Fig. 4B). KEGG results indicated that BRD9 expression in GBC cells may be associated with the PI3K-AKT pathway (Fig. 4C). We used a cell- permeable PI3K activator, 740Y-P, to investigate the response of GBC cells to I-BRD9 treatment. Colony formation assays showed that 740Y-P reversed the reduction in the number of colonies in NOZ and GBC-SD cells induced by I-BRD9 (Fig. 4D), and the inhibitory action of I-BRD9 on the proliferative capacity of GBC cells (Fig. 4E). Western blot analysis for several key proteins in the PI3K-AKT pathway showed that the expression of p-AKT and p-PI3K was downregulated in the I-BRD9-treated group compared with the control, whereas treatment with 740Y-P rescued this downregulation (Fig. 4F). Collectively, the inhibitory effect of I-BRD9 on the proliferation of GBC cells may be mediated via the PI3K-AKT pathway.

**Fig. 4: Activation of the PI3K/AKT pathway reverses the inhibitory effect of I-BRD9 on gallbladder cancer cell proliferation.**

BRD9 promotes gallbladder carcinogenesis by activating the PI3K-AKT pathway through the upregulation of CST1 expression

To further explore the oncogenic mechanism of BRD9, we observed that CST1 expression was significantly downregulated after I-BRD9 treatment (Fig. 4A). qRT-PCR and western blotting results manifested that I-BRD9 reduced the mRNA and protein levels of CST1 in both NOZ and GBC-SD cells (Fig. 5B). Accordingly, the overexpression of CST1 (Fig. 5C) reversed the inhibitory effect of I-BRD9 on NOZ and GBC-SD cell proliferation (Fig. 5D). The use of I-BRD9 when overexpressing CST1 increased the number of clonal masses in tumor cells compared to I-BRD9 alone (Fig. 5E). Western blot analysis was used to determine the alternation among the levels of several key proteins of the PI3K-AKT pathway after treatment with DMSO, I-BRD9, CST1 OE plasmid, or I-BRD9 + CST1 OE plasmid. CST1 overexpression restored the reduction in expression of CST1, p-AKT, and p-PI3K induced by BRD9 inhibition (Fig. 5F). These results suggest that BRD9 upregulates CST1 in GBC cells, which further activates the PI3K-AKT pathway to promote their proliferation.

**Fig. 5: BRD9 promotes gallbladder carcinogenesis by activating the PI3K-AKT pathway through the promotion of high CST1 expression.**

BRD9 upregulates CST1 expression through the transcription factor FOXP1

No relevant studies have reported a direct interaction between BRD9 and DNA, therefore, we investigated whether BRD9 regulates CST1 expression by recruiting host transcription factors. To further investigate which transcription factors regulate CST1, we knocked down MXI1, SRY, Arid3a, ZNF354C, and FOXP1 in NOZ cells using siRNAs (Fig. 6A) and analyzed the levels of CST1 using qRT-PCR. The results showed that knockdown of FOXP1 significantly downregulated CST1 expression, whereas knockdown of the other genes did not cause significant changes (Fig. 6B). Collectively, these results suggest that FOXP1 is involved in the regulation of CST1 expression.

**Fig. 6: BRD9 can upregulate CST1 expression through the transcription factor FOXP1.**

We used co-IP experiments to further analyze whether BRD9 interacts with FOXP1 and demonstrated an interaction between BRD9 and FOXP1 (Fig. 6C). To explore whether FOXP1 plays a transcriptional regulatory role, we used JASPAR database to predict transcription factor binding sites (TFBSs) for FOXP1 in the promoter region of CST1 with a transcription factor binding probability of over 88% (Fig. 6D). The prediction results showed that there may be two relevant binding sites for FOXP1, TFBS1 (−1664 to −1650bp) and TFBS2 (−1245 to −1231bp), within the promoter (−2000 to +200 bp) region of CST1 (Fig. 6E). To confirm whether FOXP1 directly binds to sites in the CST1 promoter region, we performed ChIP experiments in NOZ cells, designed amplification primers for TFBS1 and TFBS2, and performed PCR on the ChIP products. Agarose gel electrophoresis showed that the PCR amplification of the ChIP products at the TFBS2 site was significantly greater than that for TFBS1, indicating that TFBS2 may be the main binding site for FOXP1 in the CST1 promoter region (Fig. 6F). To further confirm that TFBS2 is the major binding site of FOXP1, we mutated TFSB1 and TFSB2 sites to MUT1 and MUT2, respectively (Fig. 6G). Dual-luciferase reporter assays showed that luciferase activity was significantly elevated after overexpression of FOXP1, and luciferase activity of the CST1 promoter in the wild type was obviously higher than that of the MUT1 mutant, whereas the expression of luciferase activity did not change significantly after MUT2 mutation compared to the wild-type (Fig. 6H). This suggests that FOXP1 positively regulates the expression of the downstream gene CST1 in NOZ cells and that the TFSB2 site is the major binding site for FOXP1.

A new model for BRD9 regulation of GBC proliferation

In summary, BRD9 is an oncogenic bromodomain protein. It was found that BRD9 can promote the binding of transcription factor FOXP1 to the CST1 promoter region to promote the transcription process. It further activates the PI3K-AKT pathway, thus promoting the growth and proliferation of gallbladder cancer. And BRD9-specific small molecule inhibitor I-BRD9 can inhibit the above process. Therefore I-BRD9 is expected to be a clinically effective therapeutic drug (Fig. 7).

Discussion

BRD is a highly conserved structural domain, and proteins containing this structural domain belong to the BRD protein family, including BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, and BRD9 [18]. Repeat mutations in subunits of the SWI/SNF complex have been shown to be potential targets for a variety of tumor therapies [19]. BRD9 is a key component of the SWI/SNF (BRG1/BRM associated factor) chromatin remodeling complex, and some studies have demonstrated its involvement in gene transcriptional regulation and tumor development [20]. We postulated that BRD9 may be an important target for the diagnosis and treatment of GBC. Our study demonstrates that both BRD9 mRNA and protein levels were upregulated in clinically collected GBC tissues compared with those in neighboring non-cancerous tissues. Combined with these pieces of evidence, we hypothesized that BRD9 may be a promising biological target point for GBC therapy.

In this study, we found that BRD9 knockdown inhibited the proliferation of GBC cells. Because BRD9 expression is upregulated in a variety of tumors, inhibition of BRD9 has the potential to improve the therapeutic efficacy of cancer treatments. JQ1 [21] and I-BET762 [22] can inhibit the bromodomain of BET family members such as BRD4. However, despite its anticancer effects in clinical trials, the possibility that JQ1 may exhibit toxicity and resistance during clinical administration cannot be excluded [23]. Compound 16 was the first reported nonselective inhibitor of BRD9, has mixed bromodomain pharmacology, and inhibitory effects on both BRD9 and BRD4 [24]. Bromine structural domains have high structural similarity; therefore, to avoid off-target effects and improve specificity, we aimed to identify a selective BRD9 inhibitor. I-BRD9 is a selective inhibitor of BRD9 that has high selectivity for the bromostructural domains of the BET family, as well as a series of receptor sites, ion channels, and enzymes [25]. I-BRD9 is 700-fold more selective than BET family. In retrospect, studies have shown that I-BRD9 inhibited the growth and proliferation of colon cancer cells in vitro in a dose-dependent manner, inhibited tumor growth in vivo [17], and significantly inhibits proliferation and promotes apoptosis in acute myeloid leukemia cells [26]. We demonstrated that I-BRD9 effectively inhibited the proliferative activity of NOZ and GBC-SD cells. Besides, I-BRD9 inhibited tumor growth in a time- and dose-dependent manner.

CST1 [27] is mainly present in the submandibular gland, gallbladder, and uterus but is also abundantly expressed in highly malignant tissues [28]. CST1 is closely associated with the proliferation of several malignant tumors, such as lung cancer, breast cancer, pancreatic malignancy, esophageal squamous cell carcinoma, and colorectal cancer [27]. CST1 is closely associated with the activation of the PI3K-AKT pathway [29, 30]. Therefore, we postulated that BRD9 activates the PI3K-AKT pathway by affecting the expression of CST1, which further promotes the growth and proliferation of GBC cells. The PI3K-AKT signaling pathway is involved in a large number of physiological processes and plays a significant part in the control and regulation of cell survival, cellular metabolism, and inflammatory factor recruitment, as well as being an important signaling pathway for tumorigenesis and development [31, 32].

Proliferation experiments showed that the inhibitory effect of I-BRD9 on the proliferation of NOZ and GBC-SD cells was reversed by the PI3K activator 740Y-P. Next, we demonstrated that I-BRD9 inhibited the expression of CST1, p-PI3K, and p-AKT, and that this inhibitory effect could be reversed by 740Y-P, confirming that the expression of CST1 and BRD9 has a positive correlation. A previous study revealed that CST1 regulates the proliferation and migration of breast cancer cells via the PI3K-AKT pathway [30]. Our results indicated that CST1 overexpression reversed the inhibitory effect of I-BRD9 on the proliferation of NOZ and GBC-SD cells, and western blot assays illustrated that CST1 overexpression reversed I-BRD9-mediated downregulation of pPI3K and pAKT expression.

No study has yet reported that BRD9 can directly interact with DNA, therefore, we investigated whether BRD9 could regulate CST1 expression by recruiting related transcription factors. Jiang et al. screened relevant motifs that were enriched in the BRD9 binding site and were missing after BRD9 inhibition, such as MXI1, SRY, Arid3a, ZNF354C, and FOXP1, suggesting that these transcription factors may work synergistically with BRD9 to promote the downstream transcription of CST1 [33]. We demonstrated that knockdown of FOXP1 affected CST1 mRNA expression.

FOXP1 is a member of the FOX transcription factor superfamily that is widely expressed in human tissues and includes a DNA-binding domain, a leucine zipper, and a zinc-finger structure [34]. FOXP1 can increase the proliferation, self-renewal, and metastasis of osteosarcoma cells by inhibiting the transcription of P21 and RB [35]; FOXP1 is also able to promote diffuse large B-cell lymphoma progression by regulating S1PR2 transcription [36]. We hypothesized that BRD9 regulates CST1 transcription through FOXP1 and thus influences GBC progression. Subsequently, we predicted two relevant binding sites for FOXP1 in the promoter region of CST1 and confirmed that FOXP1 binds mainly at the TFBS2 site (−1245 to −1231) to regulate the expression of CST1. However, the specific mechanism by which BRD9 regulates FOXP1 remains to be unknown. It is worth noting that this study still has some limitations. Although this article elucidated that BRD9 may be closely related to gallbladder cancer development as well as the specific mechanism of action, and I-BRD9 may inhibit the proliferation of gallbladder cancer. However, the use of I-BRD9 still has a long way to go from animal model to actual clinical application.

In addition, more experimental models should be utilized to further validate the therapeutic targets of BRD9 and the cancer inhibitory effects of I-BRD9 in the future to ensure that it can be more safely and effectively applied to clinical treatment and improve patient prognosis. In the future, we can also use case data to conduct retrospective studies, based on preclinical basic research results, carry out translational research, improve treatment methods and optimize drug dosage, so that I-BRD9 can be applied to clinical translational therapy more quickly and improve the prognosis of patients.

In conclusion, our study identified BRD9 as an oncogenic protein that promotes the growth and proliferation of GBC cells. High BRD9 expression is associated with poor prognosis. BRD9 promotes the binding of the transcription factor FOXP1 to the CST1 promoter and activates the PI3K-AKT pathway to facilitate GBC progression, however, I-BRD9 treatment reversed this process. Our results suggest that BRD9 may serve as a GBC biomarker and that I-BRD9 holds promise as an effective chemotherapeutic agent for GBC treatment.