Vascular restenosis following paclitaxel-coated balloon therapy is attributable to NLRP3 activation and LIN9 upregulation

Abstract

Lower limb arterial occlusive disease is treated with intraluminal devices, such as paclitaxel (PTX)-coated balloons (PCBs); however, post-procedural restenosis remains a significant challenge. NLRP3 activation is known to play a significant role in atherosclerosis, but its involvement in restenosis following PCB intervention remains to be investigated. We identified that NLRP3 was differentially expressed in lower-limb arterial tissues sourced from healthy controls and patients with arterial occlusive disease. Through cell experiments, we confirmed that PTX is involved in the activation of NLRP3. Subsequently, we demonstrated that NLRP3 activation promotes the proliferation and migration of vascular smooth muscle cell (VSMC), thereby reducing their sensitivity to PTX. NLRP3 activation also stimulates the secretion of the inflammatory cytokine interleukin IL-1β. RNA sequencing of IL-1β-treated VSMC revealed the upregulation of BRD4 and LIN9. Further mechanistic investigations confirmed that IL-1β facilitates BRD4 recruitment, leading to enhanced LIN9 expression. The transcription factor LIN9 binds to the promoter region of the cell-cycle regulator AURKA, thereby promoting its transcription and subsequently upregulating the expression of the cell proliferation-associated molecule FOXM1. These processes ultimately mediate the proliferation, migration, and PTX resistance of VSMC. Additionally, we discovered that JQ1 inhibited the overexpression of the above molecules, and exhibited a synergistic effect with PTX. Our conclusions were validated through in vivo experiments in Sprague-Dawley rats. Collectively, our findings provide insights into the molecular mechanisms underlying restenosis following PCB therapy, and suggest that the combined use of JQ1 and PTX devices may represent a promising therapeutic strategy.

Graphical Abstract

Introduction

Paclitaxel-coated balloon (PCB) dilation has become the primary treatment for atherosclerosis obliterans (ASO) of the lower extremities. Despite promising results, PCB therapy is associated with many limitations in the real world, and restenosis remains a primary impediment to the long-term efficacy of these interventions [1]. In one trial, 18% of patients with femoropopliteal artery in-stent restenosis treated with PCB therapy developed recurrent restenosis after 12 months [2]. This trial indicated that PCB therapy only delayed the recurrence of restenosis, and failed to eliminate in-stent restenosis. Indeed, the issue of prolonging the postoperative patency of vascular stents has become a research hot spot.

The migration and proliferation of vascular smooth muscle cells (VSMCs) play important roles in the restenosis process, and these cell processes are often driven by inflammation [3]. Our prior work verified that the Nod-like receptor family, pyrin domain-containing 3 (NLRP3) protein leads to atherosclerosis and restenosis, and confirmed that interleukin IL-1β, a key downstream molecule of NLRP3, is involved in disease progression [4, 5]. Interestingly, some studies have directly confirmed that PCB induce the infiltration of a variety of inflammatory cells in the blood vessel walls; substantial inflammatory infiltration was observed between 56 and 180 days after PCB therapy, and the increase in inflammatory cells occurred in a dose-dependent manner [6, 7]. And breast cancers with high IL-1β expression show decreased sensitivity to PTX [8]. Additionally, vascular smooth muscle cell (VSMC) incubated with a low dose of PTX show clear signs of regeneration [9]. All of the above studies converge on a common issue: the pivotal role of inflammation in the development of restenosis. However, it remains to be investigated whether the phenomenon observed in the field of tumor therapy—where inflammation promotes cell proliferation while concurrently reducing sensitivity to PTX—also occurs after PCB therapy for arterial occlusive disease.

IL-1β, as an effector molecule released by NLRP3, can enhance the migration and proliferation of VSMCs. However, few studies have investigated whether the expression of IL-1β regulates the sensitivity of VSMC to PTX. We treated VSMC with IL-1β and performed RNA sequencing to screen for downstream molecules, and we identified BRD4 and LIN9 as important downstream molecules of IL-1β. BRD4 is a member of the Bromodomain and Extra-terminal Domain (BET) protein family, and participates in the progression of atherosclerosis [10]. LIN9 is an important regulator of the cell mitotic process, and plays an important role in the transcriptional regulation of the expression of cell cycle-related genes [11]. Hence, we wished to determine whether IL-1β exerts its influence on VSMC proliferation and migration via the BRD4/LIN9 axis as well as to elucidate the potential role of IL-1β in modulating the sensitivity of VSMCs to PTX.

The transcription factor LIN9 binds to specific gene promoter sequences, and controls the transcription and subsequent functions of downstream genes. We found that LIN9 is enriched in the promoter regions of Aurora kinase A (AURKA), and promotes the DNA transcription of AURKA; moreover, the high expression of AURKA upregulated the expression of the downstream gene Forkhead Box M1 (FOXM1). Interestingly, both these genes have been unequivocally implicated in facilitating cancer-cell proliferation and mediating resistance to PTX in multiple cancer types, and the upregulation of FOXM1 expression not only underscores its significance in promoting cancer-cell growth but also highlights its potential involvement in the proliferation and survival of VSMCs following vascular injury [12, 13].

Consequently, we conducted an in-depth exploration of the intricate mechanisms governing the effects of inflammatory activation on the proliferation, migration, and PTX sensitivity of VSMC. Our ultimate objective is to unveil novel therapeutic targets that hold promise for mitigating restenosis after PCB therapy.

Materials and methods

Human femoral artery tissue specimens

Arterial tissues were obtained from patients who had undergone open surgery at the First Affiliated Hospital of Sun Yat-sen University, patient’s treatment period was from September 2022 to December 2023. Specimens of normal femoral arteries were obtained from organ transplant donors, while specimens of femoral arteries afflicted with peripheral vascular disease were obtained from amputees who had undergone percutaneous transluminal angioplasty PTA or PCB dilation. None of the donors had a history of connective tissue diseases, tumors, or systemic infections. Written informed consent was obtained from all patients or their family members prior to participation in this study. The ethical standards set forth in the 1975 Declaration of Helsinki were followed in this study. The study protocol was approved by the institutional review committee of Sun Yat-sen University (approval No: [2022]668), and conformed to the ethical guidelines of the Office of Research Compliance and Human Research Protection Program.

Cultivation and stimulation of VSMCs

Primary arterial smooth muscle cells were isolated and cultured from the normal femoral arteries of organ transplant donors. The cells were subjected to immunostaining with specific antibodies (to α-SMA and SMA22α) to identify the cell type. Experiments on primary VSMCs were typically conducted when the cells were in their optimal state, which was generally between the 2nd (R2) and 8th (R8) passages of culture. The cells were maintained in Dulbecco modified Eagle medium (Gibco, Karlsruhe, Germany) supplemented with 10% fetal bovine serum (Gibco). For NLRP3 activation, VSMC were pretreated with or without lipopolysaccharide (LPS) (MCE, HY-D1056, Shanghai, China) 1000 ng/mL for 24 h and 5 mM ATP (MCE, HY-B2176) for 2 h, and different concentrations of PTX (0.001–10µmol/L; MCE, HY-B0015) were used to stimulate the VSMC. To verify the expression of NLRP3 in VSMC after prolonged exposure to PTX, we selected the minimum effective dose (0.01µmol/L) of PTX to treat primary VSMC. In passages 2 to 8 of cell subcultures, PTX was administered with exposure durations ranging from approximately 3 to 30 days in each group. In subsequent experiments, VSMC was pretreated with or without different concentrations of IL-1β (5–40 ng/mL; MCE, HY-P78459A) with or without JQ1 (50–1000 nmol/L; MCE) for 24 h. Similarly, some cells were treated with thiostrepton (1µM; MCE, HY-13030) to inhibit FOXM1.

Western blotting

VSMCs and tissues from various experimental groups were harvested and lysed in radioimmunoprecipitation assay buffer supplemented with phosphatase and protease inhibitors (MCE). After quantification of the protein concentrations, the cell lysate or protein samples (30 ug/lane) were separated using Future-PAGE^™ on 8–12% gels (ACE Biotechnology, Jiangsu, China) and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA). The membranes were blocked with 5% non-fat dry milk in Tris-buffered saline for 1 h and incubated with primary antibodies overnight at 4 °C. The following primary antibodies were used: mouse anti-GAPDH (Proteintech,60004-1-lg, Wuhan, China), rabbit anti-AURKA (Abcam, ab108353, Cambridge, UK); rabbit anti-LIN9(Proteintech,17882-1-AP), rabbit anti-FOXM1(Proteintech,13147-1-AP), rabbit anti-BRD4(Proteintech,28286-1-AP), rabbit anti-NLRP3 (Proteintech, 68102-1-lg); rabbit anti-Caspase1 (Cell Signaling Technologies, 24232,Beverly, MA, USA), and rabbit anti-IL-1β (Cell Signaling Technologies,63124), rabbit anti-Cleaved caspase3 (Cell Signaling Technologies,9664).

Real-time polymerase chain reaction

Total RNA was extracted from VSMCs by using Trizol reagent (Invitrogen, Carlsbad, CA, USA). Each RNA sample was reverse transcribed into cDNA by using the Evo M-MLV RT Kit with gDNA Clean for PCR and applying the SYBR Green Premix Pro Tap HS qPCR Kit (Accurate Biology, Hunan, China) to obtain the relative levels of target gene mRNA transcripts. The testing instrument was the Roche LightCycler 480 Real-Time PCR System (Rocher, Basel, Swiss Confederation). The primer sequences are provided in Supplementary (Table 1).

Histology, immunohistochemistry, and immunofluorescence

Hematoxylin and eosin (HE) staining was used for morphological analysis. We performed immunohistochemical (IHC) analysis to assess the differential expressions of NLRP3 and LIN9 in different groups. After antigen retrieval and blocking, the following primary antibodies were diluted in a suitable antibody diluent and applied to the sections, which were then incubated overnight at 4 °C: NLRP3 (1:100 dilution) and LIN9 (1:100 dilution; SC-130571, Santa Cruz Biotechnology, Dallas, USA). Next, a suitable secondary antibody conjugated with a detection system was applied to the slices. Finally, 3,3′-diaminobenzidine was applied for color development, and the nuclei were counterstained with hematoxylin. The sections were imaged using an optical microscope at 10×–20× magnification. The quantitative results of IHC were calculated with Image J software as follows: integrated optical density/area = mean density.

Immunofluorescence assays were performed in a similar way to the IHC assays. Cells were incubated overnight at 4 °C with primary antibodies to the following: LIN9 (1:100), BRD4 (1:50), AURKA (1:100), FOXM1 (1:200), and α-SMA (1:500). Next, a suitable secondary antibody conjugated with a detection system was applied to the slices or cells. Finally, the slides were examined under a fluorescence microscope or a confocal microscope, and Image J software was used to quantify fluorescence intensity or colocalization.

VSMC proliferation assay

VSMCs were plated in a 96-well plate at a density of 5,000 cells per well. After the cells had adhered to the wells, the experimental drugs were added in various sequences following experimental design proposal. The CCK-8 kit (Dojindo Laboratories, Kumamoto, Japan) was used according to the manufacturer’s protocols. At specific time points (24, 48, and 72 h), 10µL CCK-8 solution was added to the wells. Following incubation at 37 °C for 2 h, the absorbance at 450 nm was measured using a plate reader.

VSMC migration assay

Transwell chambers with 8.0 μm pores (Corning, New York, USA) were used for the cell migration assays. In brief, VSMC pretreated with different drugs were plated onto the upper chambers, and serum-free medium was added to the lower chambers. After incubation at 37 °C for 24 h, the chambers were fixed with 4% paraformaldehyde. A cotton swab was used to remove the non-migrated cells from the upper surface, and the migrated cells were stained with 0.3% crystal violet for 15 min. Image J software was used to count the cells that had migrated to the lower chambers.

Cell scratch assay

Cells were grown in 6-well plates to 70–80% confluence, and a 200-µL pipette tip was used to gently scratch the surface of the cell layer. The cells were treated with or without different drugs at different concentrations, and their migration was evaluated. The degree of scratch healing was observed using images captured at 0 h and 24 h.

5-Ethynyl-2′-deoxyuridine assay

VSMCs were seeded in 24-well culture plates and allowed to adhere overnight. The cells were then treated under different experimental conditions. A 5-ethynyl-2′-deoxyuridine (EDU) labeling solution was prepared by dissolving the EDU reagent in the culture medium at the recommended concentration (BeyoClick™ EdU Cell Proliferation Kit, c0071s; Beyotime Biotechnology, Shanghai, China). Following the manufacturer’s instructions, the click reaction mixture was added to the permeabilized cells, and the cells were incubated in the dark for 30 min at room temperature. A fluorescence microscope (DMI8; Leica Corporation, WETZLAR, Germany) was used to visualize the cells at 20× magnification and capture the fluorescence signals. The percentage of EDU-positive cells was quantified using Image J software to assess cell proliferation and DNA synthesis.

Enzyme-linked immunosorbent assay

After being treated with the relevant experimental drugs for 24 h, the VSMC culture supernatant from each group was collected for the determination of the IL-1β concentration. After centrifugation, the supernatant was collected and tested with human enzyme-linked immunosorbent assay (ELISA) kits (Cusabio Biotech, Wuhan, China). The optical density of each well was measured using a microplate reader at the appropriate wavelength. The data were analyzed, and the concentrations of the analytes in the samples were calculated using a standard curve.

Transfection of small interfering RNA

VSMCs at 70–80% confluence was transfected with 100 nmol/L si-NLRP3, 50 nmol/L si-LIN9, 50 nmol/L si-AURKA, and si-control (RiboBio, Guangzhou, China) using Lipofectamine RNAi MAX (Invitrogen), according to manufacturers’ protocols, for 24 h. The transfection efficiency was assessed before subsequent cellular experiments by gene expression analysis and protein analysis. The sequences of the small interfering RNAs (siRNAs) used are provided in Supplementary (Table 2).

Cell cycle assay

Cell cycle analysis was performed using the cell cycle analysis kit from Beyotime Biotechnology. VSMCs were treated with different drugs, harvested from the culture medium, and washed with phosphate-buffered saline to remove the remaining debris and medium. Cell concentration within an appropriate range (Cell count, above 2 × 10⁶/ml) was selected for flow cytometry analysis. The cells were fixed with 70% ethanol overnight at 4 °C, to preserve their morphology, and then washed with phosphate-buffered saline. A solution of propidium iodide was prepared for staining, and the stained cells were examined using a flow cytometer (CytoFLEX, Beckman Coulter, California, USA). The raw data were analyzed using Flow Jo software.

Co-immunoprecipitation and chromatin immunoprecipitation assays

VSMC in a 100-mm culture dish was lysed using a suitable lysis buffer supplemented with protease inhibitors and phosphatase inhibitors. Specific antibodies, depending on the target protein, were immobilized on suitable solid support with protein A + G Magnetic Beads (Beyotime Biotechnology). Precleared lysates were added to the antibody-coated beads and incubated overnight at 4 °C with gentle agitation to allow antigen-antibody complex formation. The established positive control (input group) and negative control (IgG group) were grouped, and the remaining steps were carried out according to the standard procedure for western blotting. Similarly, cells treated according to pre-designed experimental conditions were collected and crosslinked with chromatin by adding formaldehyde directly to the culture medium to achieve a final concentration of 1%. All experimental steps were performed according to the instructions provided with the chromatin immunoprecipitation (ChIP) assay kit (Beyotime Biotechnology). Cell lysates were sonicated to shear the chromatin into fragments of 200–600 base pairs, and then centrifuged to remove cellular debris. The supernatant containing chromatin was collected and incubated overnight at 4 °C with the relevant antibodies under gentle rotation. Protein A + G Agarose beads were added, and the incubation was continued for an additional 2–4 h at 4 °C. Elution and de-crosslinking with reagents were performed, followed by reverse transcription polymerase chain reaction (RT-PCR) to amplify specific DNA regions of interest. The sequences of AURKA are provided in Supplementary (Table 3).

Dual luciferase assay

Dual-luciferase reporter assays were performed to evaluate the direct binding of LIN9 to the AURKA promoter. The promoter-binding regions were predicted based on the human transcription factor targets website, and vector construction and mutant identification were performed (Puzon Gene Technology Company, Guangzhou, China). Next, 293T cells were transfected with the reporter plasmid (LIN9-pcDNA3.1, WT-pGL3, and Mut-pGL3) containing the firefly luciferase gene or the control plasmid containing the Renilla luciferase gene, by using a transfection reagent (LipoTrans^™, Puzon Gene Technology Company, Guangzhou, China). At 6 h after the transfection of the plasmids into 293T cells, the culture medium was replaced with fresh medium. Samples were collected for analysis at 48 h after transfection with the Dual-Luciferase^® Reporter Assay System (Promega Company, Wisconsin, USA). The ratio of firefly luciferase activity to Renilla luciferase activity was calculated for each sample.

Electrophoretic mobility shift assay (EMSA)

293T cells are lysed in a buffer containing protease inhibitors and reducing agents, the lysed samples are vortexed, kept on ice for 20 min, the labeled probe was added to the solution, mixed thoroughly, and incubated at room temperature (20–25 °C) for 20 min before undergoing electrophoresis analysis. Subsequently, 1.5 µL of loading buffer was added, mixed, and immediately loaded onto the gel with 0.5× TBE used as the electrophoresis buffer. Further steps included electro-transfer and UV cross-linking using a 254 nm UV wavelength at 120 mJ/cm² for 60 s. Subsequent procedures involved incubation and washing with sealing and washing solutions at 37–50 °C. Finally, the ECL Plus Reagent working solution was carefully applied to the surface of the nylon membrane for exposure, (EMSA Kit, Axl-EMSA-100, Axl-Bio, China); AURKA probe (Axl-Bio, China).

Proximity ligation assay (PLA)

This assay about BRD4/LIN9 and AURKA/FOXM1 protein in VSMC was conducted according to manufacturer’s instructions with mouse anti-BRD4 (Proteintech,67374-2-lg) and rabbit anti-LIN9 (Proteintech,17882-1-AP); mouse anti-AURKA (Proteintech,66757-1-lg) and rabbit anti-FOXM1 antibodies (Proteintech,13147-1-AP), Duolink In Situ PLA Probe Anti-MOUSE MINUS (DUO92004); Anti-Rabbit PLUS ( DUO92002, Sigma), Duolink In Situ Detection Reagents Red (DUO920008, Sigma-Aldrich) and washing buffers (DUO82049, Sigma).

Rat carotid artery balloon injury model

The use of animals for the following experiments was reviewed and approved by the institutional review committee of Sun Yat-sen University (approval No: [2022]593). The rats were sourced from LaiDi Biotechnology (Guangzhou, China). We used a model establishment method described in the literature [14]. In brief, 40 male Sprague-Dawley rats (weighing above 300 g) were divided equally into 5 groups: control group (operation without drug treatment), vehicle group (operation followed by treatment with dimethyl sulfoxide and pluronic gel), PTX group (local PTX delivery in the injured carotid artery), JQ1 group (JQ1 gel applied around the blood vessel), and PTX + JQ1 group (combined treatment with both methods of drug administration). PTX injection was usually prepared by dissolving 2 mg/kg PTX in a mixture of polyoxyethylene castor oil and ethanol solvent mixture to obtain a final working concentration of 90 µg/30µL [15, 16]. PTX solution (30µL) was instilled into the injured common carotid artery and touching for 30 min. JQ1 (500 µg per rat) or dimethyl sulfoxide dissolved in Pluronic F-127 gel (Beyotime Biotechnology) to achieve a final volume of 300µL, was applied to the outside of the balloon-injured segment of the carotid artery [17]. This is a widely adopted and long-standing technique. At two weeks following the surgery, the rats were anesthetized using an intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg), and the carotid arteries were harvested for further analysis. The specific tissue collection methods follow relevant literature reports [18].

RNA-seq and data processing

To analyze the differential gene expression profile in VSMC upon treatment with IL-1β, we divided VSMC into distinct groups: 3 samples served as the control group, and 3 samples were treated with IL-1β (10 ng/mL) for 24 h. Total RNA was extracted from the treated cells, and sequencing libraries were constructed. These experiments were conducted by Shu Pu Biotechnologies LLC (Shanghai, China), and the specific workflow was as follows: the total RNA extracts were subjected to oligo(dT) enrichment, and used for library construction with the KAPA Stranded RNA-Seq Library Preparation Kit (Roche). The quality of the libraries was assessed using the Agilent 2100 Bioanalyzer. Subsequently, the libraries were sequenced using the Illumina NovaSeq 6000 sequencing platform. Transcription abundance was estimated using StringTie software, and Fragments Per Kilobase of transcript per Million mapped reads (FPKM) values were calculated at the gene and transcriptome levels using R software.

Statistical analysis

Results were expressed as mean ± SEM, and the t-test was used to compare two groups, one-way ANOVA was used to compare between multiple groups, p < 0.05 indicated statistical significance, GraphPad v9.0 software (San Diego, CA, USA) was used to analyze and make graphs.

Results

NLRP3 is highly expressed in the femoral artery tissues of patients with postoperative restenosis

We aimed to investigate the potential role of NLRP3 by analyzing its expression in normal arteries and in arterial tissues from patients with post-PTA and post-PCB therapy restenosis. IHC analysis of femoral arterial tissues from these 3 groups revealed that NLRP3 expression was significantly higher in the PTA and PCB groups than in the normal control group (Fig. 1A). Interestingly, the highest NLRP3 expression was observed in the PCB group, and quantitative analyses of protein and mRNA extracts from the tissues yielded consistent results (Fig. 1B, C). These results indicate the crucial role of NLRP3 overexpression in postoperative restenosis of the lower limb vasculature.

Fig. 1

Differential expression of NLRP3 in femoral artery tissue. Human vascular tissues were obtained from healthy donors and patients with a history of PTA or PCB. A HE staining and IHC assays were employed to assess the differences in NLRP3 expression between normal arteries and arteries derived from the PTA and PCB groups (scale bars: HE, 200 μm; IHC, 100 μm; n = 6 per group). B Differential expression of NLRP3 protein in arterial tissues was assessed using western blot analysis (N means sample ID; n = 6 per group). C The mRNA levels of NLRP3 were detected using qPCR (n = 6 per group). Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

NLRP3 activation promotes VSMC proliferation and migration, and decreases sensitivity to PTX

After identification and confirmation of the extracted VSMC, we performed cell experiments (Supplementary Fig. 1A). First, we established different concentration gradients to select the appropriate concentration for the PTX inhibition experiments. Concentrations ranging from 0.01µmol/L to 10µmol/L could inhibit VSMCs proliferation without inducing apoptosis (Supplementary Fig. 1B, C). Next, to investigate the differential expression of NLRP3 between the PCB and PTA groups, we treated VSMC with the minimum effective dose of PTX (0.01µmol/L) for an extended period, in passages 2 to 8 of cell subcultures (R2-R8). We aimed to determine whether PTX is involved in the activation of NLRP3, which requires dual signal stimulation; therefore, we co-administered LPS (1000 ng/mL) along with PTX (0.01µmol/L). We observed that compared to LPS alone, combined LPS and PTX treatment was associated with a higher expression of NLRP3, as determined using qRT-PCR and western blot assays, particularly in VSMC of the R6–R8 generations (Fig. 2A, B).

In the previous part of our study, we validated that NLRP3 is activated in VSMC subjected to long-term low-dose PTX stimulation. However, as primary VSMC exhibit a rapid decline in viability after the R9 generation, it is challenging to establish PTX-resistant cell lines. Therefore, considering the results of our preliminary experiments, we selected the combination of LPS (1000 ng/mL) and ATP (5 mM) to construct a cell model for NLRP3 activation and conducted relevant experiments. Pre-treatment with LPS for 24 h followed by a 2 h ATP treatment was performed. Western blot and qRT-PCR assays demonstrated that this combination treatment was more effective than its individual components in promoting the activation of NLRP3 and its downstream molecules caspase-1 and IL-1β, leading to their conversion into their mature forms (Fig. 2C, Supplementary Fig. 1D). Furthermore, ELISA of the supernatant of the treated culture medium revealed an increase in the expression level of IL-1β (Fig. 2F). In vitro cell experiments showed that the combination of LPS and ATP significantly enhanced VSMC proliferation and migration (Fig. 2D, E and G). Furthermore, we observed a decrease in the sensitivity of VSMC to PTX under the influence of LPS and ATP stimulation (Fig. 3G).

Fig. 2

NLRP3 activation promotes the proliferation and migration of VSMC. A, B Western blot analysis was conducted to examine the protein expression levels of NLRP3 in 8 generations of SMCs following combined treatment with LPS and PTX, and the mRNA levels of NLRP3 were detected using qPCR (n = 3, ###, p < 0.001, versus R0 of no LPS treatment group; ***, p < 0.001, versus R0 of LPS treatment group). C VSMCs were pre-stimulated with LPS for 24 h, followed by a 2 h stimulation with ATP, before western blot analysis to assess the expression of NLRP3 and its downstream molecules in the cells (n = 3). D Transwell assays were performed to analyze the migratory capacity of the cells under different treatment conditions (scale bar: 100 μm; n = 3). E EDU assay was conducted to analyze the proliferative capacity of VSMCs under different treatment conditions (scale bar: 50 μm, n = 3). F The supernatants of the VSMC culture media were collected and centrifuged, followed by ELISA to measure the level of IL-1β (n = 5). G Cell scratch assay was performed to evaluate the impact of different treatments on cell migration (n = 3). (*p < 0.05, **p < 0.01, ***p < 0.001)

Inhibiting the expression of NLRP3 and its downstream molecules restores the sensitivity of PTX

To confirm the crucial role of NLRP3 activation in VSMC proliferation and migration, we chose to knock down NLRP3 using siRNA (Fig. 3A, B). As the expression level of NLRP3 decreased, the downstream expressions of caspase-1 and IL-1β were downregulated (Fig. 3C, D; Supplementary Fig. 1E). Furthermore, after NLRP3 knockdown, the pro-proliferative and pro-migratory effects of LPS combined with ATP were abolished, and the inhibitory effect of PTX on VSMC proliferation was restored (Fig. 3E, F, H; Supplementary Fig. 1F). These findings suggest that the activation of NLRP3 and the subsequent release of IL-1β may contribute to VSMC proliferation and migration, and reduce the response of VSMC to PTX.

Fig. 3

Inhibiting the expression of NLRP3 and its downstream molecules restores the sensitivity of PTX. A, B After transfection of SMC with siRNA for 24 h, western blot and qPCR experiments were conducted to observe the efficiency of NLRP3 knockdown using si-NLRP3 (n = 3). C After transfection to knock down NLRP3, VSMCs were treated with LPS and ATP, and the inhibition of NLRP3 and its downstream molecules was assessed using western blot analysis (n = 3). D ELISA was performed to measure the expression level of IL-1β in the cell culture supernatant after NLRP3 knockdown followed by treatment with LPS and ATP. E, F After transfection to knock down NLRP3, SMCs were treated with LPS and ATP. Transwell assay (scale bar: 100 μm) and EDU assay were performed to observe whether si-NLRP3 inhibited the pro-migratory effects of LPS and ATP on SMC (scale bar: 50 μm; n = 3). G, H The median inhibitory concentration (IC50) values of different treatments on SMCs were calculated using a cell viability assay to evaluate the impact of NLRP3 activation on PTX inhibition (n = 5). Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

IL-1β promotes proliferation and migration of VSMC while reducing their sensitivity to PTX

We found that IL-1β promotes VSMC proliferation in a dose-dependent manner. Specifically, at a concentration of 10 ng/mL, it significantly enhanced cell proliferation (Fig. 4A). We further found that an effective dose of IL-1β (10 ng/mL) not only promotes SMC proliferation and migration but also reduces the inhibitory effect of PTX on these cells (Fig. 4B, C). In contrast to the control cells, the majority of the cells treated with PTX exhibited cell cycle arrest in the G0/G1 phase. However, we also noted a substantial increase in the proportion of cells in the S and G2 phases upon exposure to IL-1β (Fig. 4D). These findings suggest that IL-1β mitigates the inhibitory effects of PTX on cellular proliferation, and this evidence suggests that NLRP3 regulates SMC proliferation, migration, and sensitivity to PTX through the action of IL-1β.

Fig. 4

IL-1β promotes the proliferation and migration of SMCs and reduces their sensitivity to PTX. A Cell viability was assessed to evaluate the effect of different concentrations of IL-1β on SMC proliferation after 24 h and 48 h (n = 5, ###, p < 0.001, versus control group for 24 h; ***, p < 0.001, versus control group 48 h). B, C To assess the impact of IL-1β on PTX sensitivity, the EDU assay and Transwell assay after treatment with IL-1β alone or in combination with PTX ( n = 3). D Cells were treated with IL-1β for 24 h, followed by an additional 24-h treatment with PTX, by cell cycle analysis the proportions of cells at different phases of the cell cycle were quantified (n = 5). E A volcano plot was employed to visualize the differential gene expression in VSMCs between the control group (n = 3) and the group treated with IL-1β (10ng/mL for 24 h, n = 3). F A heatmap depicting the differential expression of relevant genes in SMCs following treatment with IL-1β. G A histogram summarizing the pathways associated with the differentially expressed genes identified in this study. Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

IL-1β recruits BRD4 to upregulate LIN9 expression in VSMCs

To elucidate the precise molecular mechanisms influenced by IL-1β, we exposed VSMCs to IL-1β (10ng/mL) for 24 h and conducted RNA sequencing to identify differentially expressed genes, which we further investigated in subsequent experiments. The analysis of differential gene expression between the control group and the IL-1β treatment group revealed a pronounced upregulation of BRD4 and LIN9 in the IL-1β group (Fig. 4E–G). This result was also confirmed in cell experiments following treatment with IL-1β (Supplementary Fig. 2A, 2B). In human femoral arterial tissues, we observed an upregulation of LIN9 expression in patients undergoing PTA and PCB therapy for restenosis after intravascular intervention (Fig. 5A). Notably, the PCB group exhibited the highest levels of LIN9 expression, and quantitative analysis of both the protein and mRNA extracts from the arterial tissues also revealed similar expression differences (Fig. 5B, C). We conducted additional investigations to assess the influence of various conditions on the expression of BRD4 and LIN9. We subjected VSMCs to 24 h pretreatment with IL-1β (10ng/mL) and PTX (0.01µmol/mL) and monitored the expression levels of BRD4 and LIN9. Our findings revealed that PTX administration effectively downregulated the expression of both BRD4 and LIN9. However, under combined treatment with IL-1β and PTX, the expression levels of BRD4 and LIN9 were found to be higher than those in the control group (Fig. 5D, E).

Following recruitment, BRD4 is known to promote the expression of downstream cell cycle-related molecules through RNA polymerase II [19]. To explore the association between BRD4 and LIN9, we conducted cellular co-localization studies, which revealed their co-localization within the nucleus. Additionally, co-immunoprecipitation (co-IP) experiments and proximity ligation assay (PLA) were performed to validate the interaction between BRD4 and LIN9 (Fig. 5F–H).

Fig. 5

IL-1β recruits BRD4 to upregulate LIN9 expression in SMCs. A IHC experiments were conducted to validate the differential expression of LIN9 in femoral artery tissue, (scale bars: HE, 200 μm; IHC, 100 μm; n = 6 per group). B, C Protein and mRNA of LIN9 extracts from femoral artery tissue were used to perform western blotting and qRT-PCR experiments, (N means sample ID; n = 6 per group). D, E VSMCs were treated with IL-1β alone or in combination with PTX, and then examined using western blotting and qRT-PCR to observe the changes in BRD4 and LIN9 expression (n = 3). F Fluorescence co-localization experiments were performed to further investigate the interaction between BRD4 and LIN9 (DAPI, blue; BRD4, green; LIN9, red), images were acquired using confocal microscope, ( scale bar 10 μm). G The interaction between BRD4 and LIN9 within the cells was examined using immunoprecipitation (IP) assay. (H) In the PLA experiment, the interaction between BRD4 and LIN9 was observed (DAPI, blue; PLA spots, red; scale bar 50 μm). Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

Suppression of LIN9 expression diminishes VSMC proliferation and migration, and reinstates sensitivity to PTX

To delve deeper into the involvement of the cell cycle regulator LIN9 in vascular restenosis, we transfected cells with siRNA to specifically reduce the expression of LIN9 (Supplementary Fig. 3A). Following a 24 h period of cell transfection, the cells were treated with IL-1β, which resulted in the notable inhibition of LIN9 expression and weakened IL-1β upregulation (Fig. 6A, B). In the cell experiments, we found that compared to the negative control group, the si-LIN9 group showed a significant reduction in the IL-1β-induced migration and proliferation of SMCs. Additionally, si-LIN9 demonstrated a remarkable synergistic effect when combined with PTX, augmenting the inhibitory impact of PTX on VSMC (Fig. 6C, D; Supplementary Fig. 3B).

Next, we used the competitive BRD4 inhibitor JQ1 to explore the ability of BRD4 to mitigate the proliferative and migratory effects induced by IL-1β, and potentially restore sensitivity to PTX. Initially, no significant effect was observed on the proliferation of normal control VSMCs exposed to different doses of JQ1 (Supplementary Fig. 3C). However, when VSMCs were pretreated with IL-1β, the CCK-8 assay showed that JQ1 could effectively mitigate the IL-1β-induced promotion of VSMC proliferation (Fig. 6E). In the cell cycle experiment, we observed that the combination of JQ1 and PTX resulted in a synergistic effect, leading to a higher proportion of cells arrested in the G1 phase and a greater suppression of cell proliferation compared to treatment with JQ1 alone (Fig. 6F). We subjected the cells to a 24 h pretreatment with IL-1β, followed by treatment with varying concentrations of PTX in the presence or absence of JQ1 for an additional 24 h. Remarkably, we observed a more pronounced inhibitory effect of PTX on SMC proliferation when JQ1 was present, suggesting a synergistic interaction (Fig. 6G). The suppressive effect of JQ1 on VSMCs, along with its synergistic interaction with PTX, was similarly detected in the Transwell assay (Supplementary Fig. 3D). In the protein and mRNA quantification analyses of LIN9, we found that the expression level of LIN9 was significantly reduced in the JQ1 + PTX + IL-1β group as compared to the IL-1β + PTX group. However, no significant difference in LIN9 expression was observed between the JQ1 group and the control group (Fig. 6H).

Fig. 6

Suppression of LIN9 expression diminishes VSMC proliferation and migration, and reinstates sensitivity to PTX. A, B VSMCs were transfected with siRNA for 24 h and then treated with IL-1β. The protein and RNA expression levels of LIN9 were measured (n = 3). C After transfection with LIN9 siRNA, VSMCs were treated with different doses of PTX for 24 h, and the changes in the inhibitory efficiency of PTX were observed (n = 6). D Cell scratch assay was performed to analyze the inhibitory effect of LIN9 knockdown on IL-1β-induced cell migration and the modulation of cell sensitivity to PTX (scale bar: 200 μm; n = 3). E CCK-8 assay was conducted to evaluate the inhibitory effect of different doses of JQ1 on the IL-1β-induced proliferation of SMCs. The cells were first treated with IL-1β (10 ng/mL) for 24 h, followed by treatment with varying doses of JQ1 for an additional 24 h (n = 6, ###, p < 0.001, versus control group; ***, p < 0.001, versus IL-1β treatment group). F Flow cytometry analysis was performed to assess the effect of JQ1 on the cell cycle and to evaluate the modulation of cell sensitivity to PTX by JQ1 (n = 5). G CCK-8 assay was conducted to analyze the restoration of SMC sensitivity to PTX by JQ1 (n = 6). H Western blot and qRT-PCR experiments were performed to analyze the effects of different treatments on the protein and RNA expression levels of LIN9 (n = 3). Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

LIN9 binds to the promoter region of AURKA to promote its expression

As a transcription factor, LIN9 exerts its regulatory effects by binding to downstream target genes. In this study, we employed bioinformatics tools (Database of Human Transcription Factor Targets) to predict potential downstream target genes, and integrated them with the findings in the literature on cell proliferation and PTX resistance-related genes [12]. Ultimately, we identified AURKA as a promising downstream molecule for subsequent investigations. Initially, we subjected cells to various drug treatments and assessed the expression of AURKA (Fig. 7A, B). We conducted LIN9 knockdown experiments followed by IL-1β treatment in SMCs. Interestingly, we observed that the expression level of AURKA did not increase upon LIN9 knockdown (Fig. 7C). Simultaneously, we treated cells with JQ1 to downregulate the expression of LIN9, the combination of IL-1β + PTX + JQ1 treatments resulted in a decreased protein expression of AURKA compared to the IL-1β + PTX group (Fig. 7D, E). This result indicated that the presence of LIN9 is essential for the IL-1β-induced activation of AURKA.

Three binding sites of LIN9 to AURKA are predicted to be located at p (-1919/-1908), p (+ 34/+45), and p (+ 66/+77). Primers were designed for each site, and ChIP-qPCR was performed to validate the binding of LIN9 to AURKA. We observed relatively high enrichment of LIN9 at the binding sites located at p (+ 34/+45) and p (+ 66/+77) (Fig. 7F). Dual luciferase reporter analysis revealed that in comparison to the control group, LIN9 facilitated the expression of AURKA at this binding site. Notably, upon introducing mutations to this specific binding site, we observed a significant reduction in the expression level of AURKA (Fig. 7G). The enrichment of the LIN9 and AURKA promoter regions was modulated following treatment with different drugs (Fig. 7H). We further validated the direct binding of LIN9 to BRD4 at two binding sites (A and B) using Electrophoretic mobility shift assay (EMSA) experiments (Fig. 7I).

Fig. 7

LIN9 binds to the promoter region of AURKA to increase its expression. A, B Western blotting and qRT-PCR experiments were conducted to analyze the effects of IL-1β and PTX on the protein and mRNA expression levels of AURKA and FOXM1 (n = 3). C After knocking down LIN9, western blot experiments to observe whether the promoting effect of IL-1β on downstream molecules was blocked (n = 3). D, E The effects of different drug treatments on the protein expressions of AURKA and FOXM1 were evaluated using western blot assay (n = 3). F Primer design for AURKA was conducted based on the predicted binding sites, and ChIP followed by qRT-PCR experiments were performed to validate the enrichment levels of each binding site (n = 3). G The relative luciferase activity of plasmids containing the wild-type AURKA promoter (WT) or its mutants was measured in 293T cells transfected with plasmids overexpressing LIN9 or control plasmids. The upper sequences show the LIN9 binding motif (blue) and its mutant (red; n = 3). H The differences in the enrichment levels of LIN9 at the AURKA promoter region were observed under different drug treatments (n = 3). Data are presented as mean ± SEM (ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001). I The EMSA was performed to confirm the binding of LIN9 to the A means p (+ 34/+45) and B means p (+ 66/+77) binding sites of AURKA

AURKA regulates smooth muscle cell proliferation and migration, reducing cell sensitivity to PTX

Finally, to validate the impact of differential AURKA expression on VSMC function, we conducted relevant experiments. Initially, we transfected siRNA to knock down AURKA, and observed that the knockdown inhibited the promoting effect of IL-1β on AURKA (Supplementary Fig. 4A, Fig. 8A). In cellular experiments, we found that the proliferation- and migration-promoting effects of IL-1β on VSMC were suppressed after AURKA knockdown. Additionally, when si-AURKA was used in combination with PTX, the inhibitory effects on VSMC proliferation and migration were stronger than those observed with PTX alone (Supplementary Fig. 4B, Fig. 8B). In comparison to PTX monotherapy, co-treatment with si-AURKA and PTX resulted in a reduced proportion of cells in the G2 phase and an increased proportion of cells in the G0/G1 phase, suggesting a synergistic effect between the two agents (Fig. 8C). In a rescue experiment, we found that the migratory activity of VSMC, which was suppressed under LIN9 knockdown, was restored by AURKA overexpression. Furthermore, cells that were initially sensitive to PTX, under the condition of LIN9 knockdown, regained resistance to PTX upon AURKA overexpression (Fig. 8D and E; Supplementary Fig. 4C).

Fig. 8

AURKA mediates the proliferation and migration of VSMC and regulates sensitivity to PTX. A VSMC transfected to knock down AURKA, and western blot experiments and qRT-PCR analysis were conducted to examine the expressions of AURKA and FOXM1 (n = 3). B Transwell assays were performed to analyze the migratory capacity of VSMC after si-AURKA transfection as well as to determine the synergistic effect of AURKA knockdown and PTX treatment (scale bar: 100 μm; n = 3). C Flow cytometry analysis was conducted to assess the impact of AURKA knockdown on the cell cycle and determine the regulatory effect of combined AURKA knockdown and PTX treatment on the cell cycle (n = 5). D In the rescue experiment, we performed plasmid overexpression of AURKA, siRNA knockdown of LIN9, included a negative control, and co-transfected both agents into VSMC (n = 3). E In the rescue experiment, the differential inhibitory effects of PTX on VSMC were observed after different treatments (n = 6). Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

The regulatory role of AURKA is mediated through FOXM1

In our previous investigations, we observed that IL-1β not only upregulates the expression of AURKA but also stimulates the expression of FOXM1. AURKA plays a crucial role in recruiting and activating target proteins to regulate cell proliferation and migration. Hence, we aimed to explore the potential interaction between AURKA and FOXM1. The interaction between AURKA and FOXM1 was confirmed through co-IP experiments (Fig. 9A). Moreover, fluorescence co-localization analysis demonstrated a clear association between AURKA and FOXM1 in the cell nucleus (Fig. 9B). The PLA experiment further confirmed the mutual interaction between AURKA and FOXM1 (Fig. 9C). To ascertain the upstream-downstream relationship between these two factors, we performed targeted knockdown of AURKA and concomitantly treated the cells with thiostrepton, a specific inhibitor of FOXM1. Remarkably, the reduction in AURKA expression led to a concomitant decrease in FOXM1 expression, mirroring the effects observed with thiostrepton treatment (Fig. 9D). To assess the influence of FOXM1 on VSMC, we employed thiostrepton to inhibit FOXM1, and subsequently stimulated the cells with IL-1β in the presence or absence of PTX. Notably, thiostrepton mitigated the migration-promoting effect of IL-1β on VSMC. Strikingly, when thiostrepton was used in combination with PTX, it exerted a synergistic inhibitory effect on VSMC migration (Fig. 9E). In a rescue experiment, we overexpressed FOXM1 in AURKA-knockdown cells and observed a reinstatement of the migratory capacity in the previously inhibited VSMC. Notably, the cells that were originally responsive to PTX displayed a diminished inhibitory rate upon FOXM1 overexpression (Fig. 9F–H).

Fig. 9

The regulatory role of AURKA is mediated through FOXM1. A Co-IP experiments confirmed the mutual interaction between AURKA and FOXM1. B Fluorescence co-localization assay to evaluate the interaction between AURKA and FOXM1. Analysis of the average fluorescence intensity showed a high level of co-localization (DAPI, blue; AURKA, green; FOXM1, red). The images were acquired using confocal microscope(scale bar 10 μm). C In the PLA experiment, the interaction between AURKA and FOXM1 was observed (DAPI, blue; PLA spots, red; scale bar 50 μm). D The protein levels of FOXM1 were assessed using western blot analysis following treatment with thiostrepton (1 μm, for 24 h) and the siRNA knockdown of AURKA (n = 3). E VSMC were pretreated with IL-1β, followed by the inhibition of FOXM1 expression by using thiostrepton. Transwell assay was conducted to assess the migratory capacity of VSMC, along with the evaluation of PTX sensitivity (scale bar: 100 μm, n = 3). F In the rescue experiment, cells from various groups were subjected to AURKA knockdown, FOXM1 overexpression, a negative control, and co-transfection of both for 24 h, followed by the assessment of molecular protein levels (n = 3). G The effect of AURKA knockdown and simultaneous FOXM1 overexpression on the inhibitory effect of PTX on cell proliferation was analyzed using the CCK-8 assay (n = 6). H The migratory capacity of cell was assessed using the Transwell assay under the conditions of AURKA knockdown and simultaneous overexpression of FOXM1 (scale bar: 100 μm; n = 3). Data are presented as mean ± SEM (*p < 0.05, **p < 0.01, ***p < 0.001)

PTX and JQ1 exhibits a synergistic effect in inhibiting vascular neointima formation

To evaluate the effect of Lin9 inhibition on neointimal hyperplasia following balloon injury, we administered the Lin9 inhibitor JQ1 in the carotid arteries of Sprague-Dawley rats. After a 14-day period, we assessed the extent of neointimal hyperplasia after combined or individual treatment with JQ1 and PTX, by using morphometric analysis of HE-stained sections. Our findings revealed that both JQ1 and PTX alone were capable of suppressing neointimal hyperplasia in the carotid artery. However, the combined administration of JQ1 and PTX demonstrated a more pronounced inhibitory effect, resulting in a reduced intimal area-to-medial area ratio (Fig. 10A).

Fig. 10

PTX and JQ1 exhibits a synergistic effect in inhibiting vascular neointima formation. A HE staining of carotid artery balloon-injury model in SD rats after different treatments; analysis of lumen area, neointimal area, medial area, and neointima-to-media ratio, (Scale bar, 200 μm). (n = 8; ns, p > 0.05; *p < 0.05, **p < 0.01, ***p < 0.001)

JQ1 inhibits the expression of Nlrp3/Lin9/Foxm1 in vascular neointimal hyperplasia

Immunofluorescence staining of carotid artery tissues from Sprague-Dawley rats demonstrated elevated expression levels of Nlrp3, Lin9, and Foxm1 in the neointimal tissue of the vehicle and PTX-alone groups, with the PTX group displaying the highest expression. Treatment with JQ1 suppressed the expressions of Nlrp3, Lin9, and Foxm1 in the carotid artery tissues, thereby mitigating the PTX-induced upregulation of these molecules (Fig. 11A–C). To further validate the expression differences of the three proteins in the carotid artery tissues of SD rats, we conducted Western blotting experiments on the carotid artery tissues of SD rats and obtained consistent experimental results (Fig. 11D).

Fig. 11

JQ1 inhibits the expression of Nlrp3/Lin9/Foxm1 in vascular neointimal hyperplasia. A Immunofluorescence analysis was conducted to evaluate the differential expression of Nlrp3 in the neointimal layer of the carotid arteries of SD rats following different treatment regimens (DAPI, blue; Nlrp3, green; α-SMA, red; scale bar: 50 μm; n = 8). B The variations in the fluorescence intensity of Lin9 expression were examined across different treatment groups (DAPI, blue; Lin9, green; α-SMA, red; scale bar: 50 μm; n = 7). C The variations in the fluorescence intensity of Foxm1 expression were examined across different treatment groups (DAPI, blue; Foxm1, green; α-SMA, red; scale bar: 50 μm; n = 7). D The expression differences of Nlrp3/Lin9/Foxm1 in the carotid artery tissues of SD rats were analyzed using western blot analysis (n = 3). Data are presented as mean ± SEM (ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001)

Discussion

In this study, we unveiled the activation of NLRP3 and its downstream effector molecule LIN9 in the pathogenesis of post-peripheral vascular intervention restenosis. Through comprehensive tissue analyses, we revealed a significant upregulation of NLRP3 expression in the tissues of patients afflicted with peripheral arterial occlusive disease, with a particularly pronounced elevation observed in individuals with a documented history of PCB therapy. This finding aligns with the most recent research elucidating the comprehensive involvement of NLRP3 in the pathogenesis of arterial atherosclerosis [20]. By comparing the disparities in NLRP3 expression between the PTA and PCB groups in an in vitro experiment, we substantiated the hypothesis that PTX functions as a potent stimulus that elicits the activation of the NLRP3 inflammasome. This discovery aligns with previous findings in the field of tumor therapy that the PTX-mediated activation of NLRP3 facilitates tumor cell proliferation, thereby giving rise to the emergence of PTX resistance [21]. Our experimental findings showed that the long-term exposure of VSMC to a low PTX dose of 0.01µmol/L (which was still effective according to the dose-determination experiment) served as a signal or stimulus to promote NLRP3 activation, and this effect surpassed the effect observed in VSMC treated solely with LPS. This finding extends previously reported data, which revealed that inflammatory cells accumulate in the vascular wall after PCB therapy in Sprague-Dawley rat models [7].

Primary cells can undergo a limited number of passages, and achieving NLRP3 activation through PTX necessitates a prolonged incubation period. Consequently, we devised a cellular model to expedite NLRP3 activation by utilizing a combination of LPS and ATP, thereby reducing the required time. The activation of NLRP3 promotes the maturation of caspase-1, resulting in the cleavage of pro-IL-1β and the subsequent generation of mature IL-1β. In our in vitro experiments, we discovered that pre-stimulation with LPS (signal 1), followed by ATP (signal 2), augments NLRP3 activation, thereby facilitating SMC proliferation and migration. This phenotype is a crucial factor in the development of restenosis following treatment for peripheral vascular disease [22, 23]. Of greater significance, we observed a correlation between NLRP3 activation and an increase in the median inhibitory concentration (IC50) of PTX in VSMC. This discovery potentially unveils a pivotal factor in the development of postoperative restenosis after peripheral vascular surgery, particularly in the context of PCB treatment.

NLRP3 promotes the proliferation and migration of VSMC as well as inhibits their sensitivity to PTX, through the activation of its downstream molecule IL-1β. By subjecting IL-1β-treated VSMC to RNA sequencing analysis, we identified BRD4 and its ligand LIN9 as promising candidates for investigating the molecular mechanisms involved. Our selection was based on the substantial differential expression of these two molecules between the study groups and the existing knowledge gap regarding the downstream molecular pathways of BRD4, despite its established importance in atherosclerosis [10, 24]. Our study extends our knowledge of the molecular mechanisms by which IL-1β modulates VSMC proliferation and migration through the BRD4/LIN9 pathway, thereby contributing to PTX resistance. This phenomenon has only been reported in a limited number of studies within the field of tumor treatment [25, 26]. IL-1β promotes the recruitment of BRD4 to the cell nucleus through its interaction with the IL-1 receptor; BRD4 interacts with acetylated histones and transcription factors, such as LIN9, and acts as a molecular scaffold between chromatin and the transcriptional machinery to regulate gene expression [27]. In line with our experimental findings, we observed that IL-1β stimulation enhances the expression of both BRD4 and LIN9. Furthermore, our comprehensive assessment of LIN9 expression levels in arterial tissues provided further support, affirming the consistency of our results.

LIN9, as a cell cycle regulator, plays a crucial role in tumor cell proliferation and drug resistance [28]. We made similar observations in VSMC co-treated with IL-1β and PTX. The elevated expression of LIN9 indicates enhanced proliferation and migration of VSMC as well as reduced sensitivity to PTX. Intriguingly, we observed a marked cessation in both cell proliferation and migration upon the inhibition of LIN9 expression by using JQ1 or siRNA. Notably, the combination of LIN9 suppression and PTX administration synergistically enhanced the inhibitory effect, yielding a more pronounced outcome. Hence, our experimental findings indicate that JQ1-related compounds hold promise as a viable adjunctive therapy with PCBs for the management of vascular restenosis.

As a nuclear transcription factor, LIN9 exerts its function by binding to the promoter regions of downstream molecules. Through network analysis tools and preliminary RNA sequencing analyses, AURKA emerged as a candidate molecule for our research. AURKA is an essential protein for the normal progression of the cell cycle, but its overexpression is closely associated with tumor cell proliferation and drug resistance [29, 30]. Consistent with our observations, VSMC exposed to the inflammatory cytokine IL-1β displayed heightened AURKA expression. Conversely, treatment with PTX alone resulted in decreased AURKA expression in cell. Intriguingly, the IL-1β-induced upregulation of AURKA was diminished upon LIN9 knockdown, underscoring the pivotal role of LIN9 as a key upstream modulator of AURKA. The dual-luciferase reporter assay, ChIP-qPCR and ESMA experiments provided compelling evidence to substantiate the involvement of LIN9 in promoting AURKA transcription through its enrichment at the AURKA promoter region. Notably, the degree of enrichment was found to be dynamically regulated by distinct treatment modalities. Importantly, the targeted depletion of AURKA led to diminished cellular proliferation and migration, along with an augmented suppressive response of VSMC to PTX under inflammatory conditions, thereby highlighting a synergistic interplay.

The modulation of the cell cycle by AURKA involves the recruitment or activation of specific proteins through phosphorylation events [31, 32]. AURKA has been reported to facilitate tumor cell proliferation and resistance by inhibiting FOXM1 degradation; in addition, FOXM1 upregulation has been documented in narrowed arteries after balloon injury in mice [12, 13]. Considering these findings, it is crucial to address the potential interrelationship between AURKA and FOXM1 concerning VSMC proliferation, migration, and PTX resistance. Through an analysis of FOXM1 expression levels in IL-1β- and PTX-treated VSMC, we observed a consistent trend that mirrored AURKA expression. Additionally, the co-IP, fluorescence co-localization experiments and PLA experiments provided further evidence of the AURKA-FOXM1 interaction. Subsequent functional recovery experiments confirmed the reciprocal relationship between AURKA and FOXM1. In our subsequent in vitro investigations, the thiostrepton-mediated knockdown of FOXM1 inhibited VSMC functionality, and a further potentiation of this effect was observed in the presence of PTX, resulting in the suppressed proliferation and migration of cell. These findings underscore the role of AURKA in recruiting FOXM1 to contribute to the mechanisms underlying VSMC proliferation, migration, and resistance to PTX.

In our in vivo experiments utilizing a rat balloon injury model, we observed that both JQ1 and PTX reduced neointimal proliferation. Moreover, the combined administration of JQ1 and PTX demonstrated a synergistic effect, leading to the enhanced suppression of neointimal proliferation. These findings align with the results obtained from our in vitro cell experiments. The inhibitory effect of JQ1 on VSMC has previously been reported to mitigate pulmonary arterial hypertension and suppress arterial VSMC remodeling [33, 34]. In this study, we demonstrated a synergistic effect of JQ1 with PTX, enhancing the inhibitory action of PTX on VSMC. Quantitative analysis of key molecules in carotid artery smooth muscle demonstrated the elevated expressions of Nlrp3, Lin9, and Foxm1 in proliferating vascular tissue. These molecules were downregulated in the JQ1 group. The protein level analysis of the carotid artery tissues from rats under different treatment conditions also yielded consistent results. Intriguingly, we also observed that PTX, despite its inhibitory effect on neointimal proliferation, paradoxically induced an increase in the Nlrp3, Lin9, and Foxm1 expression levels. Our findings corroborate the conclusions drawn from observations made in human tissues, demonstrating that the retention of PTX within the vascular wall can augment inflammatory cell infiltration in later stages, regardless of the initial high concentration of PTX. Notably, similar phenomena have been observed in other experimental studies [7, 21, 35]. Furthermore, we observed that JQ1, functioning as a downstream mediator in our experimental setting, exhibited inhibitory effects on the expression of the upstream molecule NLRP3, possibly through an alternative pathway involving the regulation of P65/NF-KB signaling [36].

While this study has validated PTX as one of the stimuli promoting NLRP3 activation, mediating cell proliferation and migration while reducing PTX sensitivity, there are limitations that need to be acknowledged. Firstly, the study is limited by the culture lifespan of primary smooth muscle cells, preventing the establishment of PTX-resistant cell lines akin to tumor cells. Secondly, NLRP3 activation is a result of various cellular interactions, such as macrophage polarization. Therefore, in the next phase, we aim to co-culture macrophages and smooth muscle cells to investigate the regulation of NLRP3 activation on PTX sensitivity.

Conclusion

In summary, we have elucidated the molecular mechanisms underlying postoperative restenosis in peripheral blood vessels. PTX was found to induce the activation of NLRP3 and its downstream molecule IL-1β, leading to the upregulation of the cell cycle-regulatory molecules LIN9, AURKA, and FOXM1 through BRD4 recruitment. This activation mediated the proliferation and migration of VSMC and contributed to the development of PTX resistance. Furthermore, we confirmed that JQ1, a BRD4 inhibitor, could block the activation of the inflammatory pathway and exhibit a synergistic effect with PTX inhibition. These findings provide promising potential therapeutic targets for the comprehensive treatment of peripheral vascular diseases.

Qinghui Kan, Zhanli Peng and Kangjie Wang assisted in the experiment and approved the final manuscript. Tang Deng, Zhihao Zhou and Ridong Wu performed the bioinformatics. All authors read and approved the final manuscript.

All experiments involving rats were conducted in accordance with the ethical guidelines and procedures approved by the Ethics Committee of the first affiliated hospital of Sun Yat-Sen University. All experiments complied with the Statement on animal use in biomedical research.