Integrative bioinformatics, network toxicology, and molecular docking elucidate molecular mechanisms of ATBC-induced sarcoma progression with experimental validation

Abstract

Background

Acetyl tributyl citrate (ATBC), a widely used plasticizer, has raised health concerns due to its potential environmental persistence and human exposure, but its toxicological effects on sarcoma remain unclear.

Methods

We employed an integrated approach combining network toxicology, molecular docking, survival analysis, and experimental validation to systematically investigate ATBC’s impact on sarcoma.

Results

Multi-database screening identified 102 overlapping targets. Protein–protein interaction analysis revealed six hub genes: TLR4, ESR1, PPARG, SIRT1, NFKB1, and PTGS2. Functional enrichment analysis indicated significant involvement in cancer-related pathways, including PI3K–Akt, HIF‑1, and AGE–RAGE signaling. Molecular docking predicted potential interactions between ATBC and these targets, with binding energies ranging from -1.15 to -2.99 kcal/mol. Survival analysis associated high PPARG expression with poor prognosis, while high TLR4 and ESR1 expression correlated with better survival. In vitro experiments showed ATBC promoted proliferation and migration of sarcoma cells. qPCR results confirmed ATBC downregulated TLR4 and ESR1 and upregulated PPARG, aligning with clinical prognostic trends.

Conclusions

These findings suggest ATBC may exert tumor-promoting effects by modulating core targets and pathways, highlighting its potential role in sarcoma progression and the importance of environmental health risk assessment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40360-026-01141-z.

Introduction

Acetyl tributyl citrate (ATBC) is an organic ester produced by esterifying citric acid. Owing to its high chemical stability and good material compatibility, it serves as a common plasticizer within polyvinyl chloride and related polymeric substances [1]. Currently, ATBC is considered a low-toxicity plasticizer, with adverse effects typically occurring only under very high exposure conditions. The European Food Safety Authority established a no-observed-adverse-effect level of 100 mg/kg body weight per day and derived a tolerable daily intake of 1.0 mg/kg body weight for ATBC [2, 3]. As a result, ATBC is authorized by European Union legislation for unrestricted application in food-contact plastics. Additionally, The U.S. FDA approves this substance as a pharmaceutical excipient, functioning both as a drug carrier and a coating for tablets and capsules [4, 5].

However, as ATBC finds broader application, its high migration potential has drawn increasing attention. The migration of ATBC from plastic products, including food packaging and medical devices, is attributed to the absence of covalent integration with the polymer matrix, thereby facilitating its mobilization into the surrounding environment [6]. As a result, humans can be exposed to ATBC through numerous routes, such as consuming tainted food and water, inhaling polluted air and dust, skin contact, using certain medical devices, and taking medications with ATBC coatings. An FDA report indicates that daily intake of the substance from some pills or capsules can reach 20.2 mg [7]. Alves et al. reported that ATBC is eliminated via urine within about two days, whereas it can be deposited in nail tissue for periods as long as six months [8]. These findings suggest that ATBC is ubiquitous in the human environment. Its relatively high solubility enhances bioavailability, facilitating entry into the body and contributing to its persistence and bioaccumulation in humans. Given the expanding applications of ATBC and its marked migration-leaching behavior, there is an urgent need to develop new strategies for comprehensive health risk evaluations and research to elucidate its underlying toxicological mechanisms.

Recent studies have raised widespread concern regarding the potential health risks posed by ATBC to humans. Multiple toxicological investigations have revealed its endocrine-disrupting effects, including anti-estrogenic and anti-androgenic activities, disruption of thyroid hormone homeostasis, and adverse impacts on glucose and lipid metabolism. These effects may be mediated through mechanisms such as disruption of metabolic pathways, triggering of inflammatory responses, or compromise of cellular homeostasis [9–12]. Furthermore, animal experiments indicate that when mothers are exposed to ATBC during pregnancy, it can cause transgenerational disturbances in the glucose metabolism of their offspring [13], underscoring the importance of evaluating the association between ATBC exposure and long-term health risks in humans. While current research has primarily focused on metabolic disorders resulting from ATBC-induced endocrine disruption, its potential role in promoting tumor progression remains poorly understood and warrants further investigation.

Sarcoma is a group of heterogeneous malignant neoplasms derived from mesenchymal tissue, comprising more than 70 subtypes that are broadly classified into soft tissue sarcoma (STS) and osteosarcoma. Although rare, sarcomas exhibit diverse pathological types and are associated with a high postoperative recurrence rate [14]. Advanced sarcoma carries a dismal prognosis, evidenced by a five-year overall survival rate of merely 30–40% [15]. Advances in sequencing technologies have prompted recent studies to focus on gene mutations in sarcoma patients, which play important roles in tumor initiation and progression [16–18]. Nevertheless, the etiology of most sarcomas remains unclear, and their increasing annual incidence poses a significant medical challenge [19]. Studies have also suggested that genetic factors, dietary habits, and environmental exposures may play a role in sarcoma [20–22]. The rising incidence of sarcoma has been partly attributed to environmental factors, with long-term exposure to chemicals such as plasticizers suspected of having a promoting effect [23]. Therefore, in-depth investigation into the molecular mechanisms by which the plasticizer ATBC may promote sarcoma development significantly to the formulation of strategies for sarcoma prevention.

Integrating network toxicology with molecular docking provides an effective approach for uncovering toxicological mechanisms. Originally developed in 2007, network pharmacology has established itself as an important methodology based on systems biology and bioinformatics to decipher molecular mechanisms of drug action [24]. It focuses on complex interaction networks among drugs, pathways, targets, and diseases, and is particularly suitable for systematic interpretation of multi-target drug mechanisms in areas such as psychiatry, oncology, and anti-infective therapy [25]. Building on this foundation, network toxicology extends the principles of network pharmacology and network biology by incorporating bioinformatics, big data analysis, genomics, proteomics, and metabolomics to deeply investigate compound toxicity and biological interactions [26]. This method constructs association networks among compounds, toxic effects, and targets to visualize and systematically analyze multi-component, multi-target toxicity, thereby predicting disease mechanisms from toxicant exposure [27]. Rooted in computer-aided drug design, molecular docking is a pivotal computational method for evaluating ligand-target interactions by computing their binding energy, where lower values indicate more stable binding conformations and more favorable interactions [28].

This study employed an integrated approach combining network toxicology, molecular docking, survival analysis, and in vitro experiments to assess the impact of ATBC on sarcoma. We screened potential carcinogenic targets of ATBC across multiple databases and applied topological algorithms to identify core targets. Based on these findings, we developed a network-based predictive model and employed network toxicology methods to systematically identify and analyze toxic pathways associated with potential ATBC-induced sarcoma risk. This approach aims to determine the carcinogenic potential of ATBC and to elucidate the molecular mechanisms underlying its promotion of sarcoma development and progression. To evaluate the influence of the core target genes on overall survival in sarcoma patients, we constructed prognostic landscapes. Subsequent in vitro experimental validation demonstrated that ATBC exposure promotes sarcoma cell proliferation and migration (Fig. 1). Through this comprehensive investigation, we aim to uncover the molecular basis of potential ATBC-associated carcinogenic effects and identify potential targets for mitigating its adverse effects. Moreover, the research provides a critical reference for the toxicity evaluation of frequently encountered chemicals and lays the groundwork for diagnosing related exposure-associated diseases.

Fig. 1.

Flowchart of the mechanistic analysis of ATBC-promoted sarcoma

Materials and methods

Toxicity analysis of ATBC

We evaluated the potential toxicity of the target compounds using the toxicity prediction tool ProTox (available at https://tox.charite.de/protox3/). Basic toxicological properties of the compounds of interest (ATBC) were obtained from the ProTox platform.

Acquisition of ATBC targets

The initial screening for potential target proteins of ATBC involved obtaining its standard chemical structure and SMILES identifier from the PubChem database. Using “acetyl tributyl citrate” as the search term, potential targets were screened from the Swiss Target Prediction, Super-PRED, TargetNet, and Comparative Toxicogenomics Database (CTD) platforms, with species limited to “Homo sapiens”. Following structural verification and consistency checks of the retrieved targets, the predicted targets were standardized via the UniProt database, resulting in a merged and deduplicated set of potential ATBC targets (Table 1).

Table 1.

Target source information for ATBC and Sarcoma

Source	Website	Number of targets	Type	Filtration Options
SwissTargetPrediction	http://swisstargetprediction.ch/	13	ATBC	Probability > 0
TargetNet	http://targetnet.scbdd.com/	138	ATBC	Probability > 0
SuperPred	https://prediction.charite.de/subpages/target_prediction.php	119	ATBC	Probability ≥ 50%
CTD	https://www.ctdbase.org/	47	ATBC	-
GeneCards	https://www.genecards.org/	2639	Sarcoma	Relevance score ≥ median
OMIM	https://www.omim.org/	73	Sarcoma	-
PharmGKB	https://www.clinpgx.org/	36	Sarcoma	-
TTD	https://db.idrblab.net/ttd/	31	Sarcoma	-

Note: SwissTargetPrediction: Swiss Drug Target Prediction Database; TargetNet: Target Netting; SuperPred: Prediction of Drug Targets and Off-targets; CTD: Comparative Toxicogenomics Database; GeneCards: The Human Gene Database‌; OMIM: Online Mendelian Inheritance in Man; PharmGKB: Pharmacogenetics and Pharmacogenomics Knowledge Base; TTD: Therapeutic Target Database

Acquisition of sarcoma-related targets

Potential therapeutic targets for sarcoma were compiled through a systematic search of the GeneCards, OMIM, PharmGKB, and Therapeutic Target Database (TTD) platforms, using “sarcoma” as the primary search term. To ensure high relevance between the selected genes and sarcoma, genes obtained from GeneCards were filtered based on their “Relevance score”, with the median score set as the threshold. Only genes scoring above the median were retained. Data from all four databases were integrated, duplicates were removed, and a non-redundant set of sarcoma-related targets was established (Table 1). Subsequently, a Venn diagram analysis was utilized to determine the intersection between the predicted ATBC targets and the sarcoma-related target set. These common targets were recognized as the potential therapeutic targets through which ATBC may mediate its underlying mechanism in sarcoma.

Protein-protein interaction (PPI) network construction and core target screening

The overlapping target genes were imported into the STRING database using the species parameter “Homo sapiens” and a medium confidence interaction score threshold of 0.400 as the minimum required criterion. A protein-protein interaction (PPI) network was then constructed and its topological parameters were calculated using the Network Analyzer module within Cytoscape software (version 3.10.1). A multi-algorithm integration strategy was implemented via the CytoHubba plugin, incorporating eight topological analysis methods—MCC, MNC, EPC, Degree, Closeness, Betweenness, Radiality, and Stress—to screen the top 10 key genes ranked by each algorithm. The results from all eight algorithms were further integrated, leading to the identification of six core regulatory targets with multi-algorithm consensus.

Functional enrichment and pathway analysis of targets

Bioinformatic analysis of the intersecting targets was conducted using the Metascape database, which involved performing Gene Ontology (GO) enrichment analysis, as well as Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to elucidate the principal biological functions and identify the key toxicity-related pathways associated with these targets.

Molecular docking

Molecular docking simulations were utilized to forecast both the binding energy and three-dimensional binding conformation of ATBC with the core proteins discovered in this investigation, thereby facilitating further analysis of their intermolecular interaction mechanisms. Three-dimensional structural data for the principal targets were acquired from the RCSB Protein Data Bank (PDB), with subsequent refinement using PyMOL software to remove hydration molecules and endogenous ligands. The prepared protein structures underwent molecular optimization in AutoDockTools 1.5.7 through hydrogen addition, charge calculation, and non-polar hydrogen unification. The AutoGrid4 and AutoDock4 modules integrated in AutoDockTools facilitated successful docking of the ligand (ATBC) with respective receptor structures, enabling evaluation of binding stability between the toxicant and its protein target, while the outcomes of the molecular docking were graphically represented through PyMOL software.

Survival analysis

Using the KM Plotter pan-cancer RNA-seq module (https://kmplot.com/analysis), we performed exploratory survival analysis on the TCGA-based sarcoma cohort (n = 259). Patients were dichotomized using the platform’s auto-selected cutoff, and overall survival was evaluated by log-rank test. Because this web-based analysis used the pooled sarcoma dataset available in the public platform, subtype-stratified and multivariate Cox analyses were not performed in the present study.

Cell culture

SW872 and MG-63 cell lines (acquired from ATCC, Manassas, VA, United States) were maintained in high-glucose Dulbecco’s Modified Eagle Medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS; Gibco) and 1% penicillin (Gibco). The cells were grown in a humidified incubator at 37°C with 5% CO₂.

ATBC solution preparation

Acetyl tributyl citrate (ATBC, CAS No. 77–90–7, purity > 98%) was acquired from Sigma-Aldrich (St. Louis, MO, USA), while dimethyl sulfoxide (DMSO) was supplied by Solarbio (Beijing, China). To generate the stock solution, ATBC was solubilized in DMSO and subsequently subjected to dilution using cell culture medium to reach the necessary working concentrations, with the ultimate DMSO level consistently held under 0.1% (v/v) across all investigative protocols. The working concentrations of 25 and 50 µM were selected based on preliminary cytotoxicity assays and reference to previous in vitro toxicological studies of plasticizers [6].

CCK-8 assay

SW872 and MG-63 cell lines were seeded into 96-well plates and exposed to ATBC at 0, 25, and 50 µM concentrations. Following treatment periods of 24, 48, and 72 h, 10 µL of CCK-8 reagent (HY-K0301; MCE) was added to each well, with subsequent incubation for 2 h at 37°C in a 5% CO₂ atmosphere. Optical density measurements were then taken at 450 nm wavelength using a multimode microplate reader (Tecan Spark, Switzerland).

Wound healing assay

SW872 and MG-63 cells were plated in 6-well plates and cultured until reaching confluence, after which a standardized wound was introduced into the monolayer using a sterile pipette tip. Subsequently, the culture medium was exchanged for serum-free formulation supplemented with ATBC at 0, 25, or 50 µM. Wound images were acquired at 0, 36, and 48 h post-treatment using an inverted fluorescence microscope (Nikon Eclipse Ti2-U, Japan). The wound areas were quantified with ImageJ software, and healing rates were statistically analyzed to compare the effects of different treatments and time points.

RT-qPCR assay

Total RNA was extracted from cells using TRIzol reagent. Reverse transcription was performed using the PrimeScript RT Premix (Takara, Dalian, China) to synthesize cDNA from the extracted RNA. The cDNA was then amplified using Hieff qPCR SYBR Green Master Mix (High Rox Plus) (Yeasen, Shanghai, China), and quantitative analysis was carried out on the ABI StepOne Plus Detection System (Thermo Fisher Scientific, Massachusetts, USA). For RT-qPCR validation, cells were treated with vehicle control or 50 µM ATBC. The 50 µM concentration was selected for this preliminary transcriptional analysis because it produced more evident phenotypic changes in the proliferation and migration assays.

Primers for TLR4, ESR1, PPARG, SIRT1, NFKB1, PTGS2, and GAPDH were designed using NCBI Primer-BLAST based on the corresponding human gene sequences, and primer specificity was checked before use. GAPDH was used as the internal reference gene for normalization, and relative mRNA expression was calculated using the 2^−ΔΔCt method. The processed normalized RT-qPCR results are provided in Supplementary Table S1. The primer sequences used in this study are listed in Table 2. Melt-curve analysis was performed after amplification to verify the specificity of each primer pair and confirm the presence of a single amplification product.

Table 2.

Primer sequences used for RT-qPCR

Gene	Forward primer (5’-3’)	Reverse primer (5’-3’)
TLR4	TGGTGTCCCAGCACTTCATC	CTGTCCTCCCACTCCAGGTA
ESR1	GGTGCCCTACTACCTGGAGA	ACACTGCACAGTAGCGAGTC
PPARG	GCAATCAAAGTGGAGCCTGC	TCTCCGGAAGAAACCCTTGC
SIRT1	GGAGCAGATTAGTAGGCGGC	CCTCAGCGCCATGGAAAATG
NFKB1	AATGGGCTACACCGAAGCAA	TCTAGAGGTCCTTCCTGCCC
PTGS2	TTGCATTCTTTGCCCAGCAC	ACCGTAGATGCTCAGGGACT
GAPDH	GTCAAGGCTGAGAACGGGAA	AAATGAGCCCCAGCCTTCTC

Statistical analysis

Experimental results are presented as mean ± SD (n = 3). For comparisons involving two groups, the t-test was applied, whereas one-way ANOVA was utilized when comparing multiple groups. A probability threshold of P < 0.05 was defined as statistical significance, and all statistical evaluations were executed with GraphPad Prism software (version 9.0).

Results

ATBC toxicity analysis

According to the ProTox database prediction, the median lethal dose (LD50) of ATBC is 7000 mg/kg, corresponding to toxicity class 6. Furthermore, the database supplied critical molecular descriptors for this compound, comprising molecular weight, hydrogen bond acceptors and donors, atom and bond counts, rotatable bonds, molecular refractivity, and topological polar surface area (Fig. 2A). These parameters offer valuable insight into the physicochemical properties of ATBC and its potential carcinogenic effects.

Fig. 2.

Identification of common targets between ATBC and sarcoma. (A) Structural representation and toxicological profiling of ATBC obtained from ProTox; (B) Venn diagram illustrating the shared targets between ATBC and sarcoma; (C) Network visualization generated through Cytoscape software depicting the 102 overlapping targets identified in the Venn analysis

Identification of potential targets of ATBC in sarcoma

We identified potential target proteins of ATBC through searches in the Swiss Target Prediction, Super-PRED, TargetNet, and Comparative Toxicogenomics Database databases, yielding a total of 275 targets. Additionally, we retrieved 2,697 targets associated with osteosarcoma from the GeneCards, OMIM, PharmGKB, and Therapeutic Target Database databases. Integration and deduplication of these two datasets via a Venn diagram revealed 102 overlapping targets (Fig. 2B). These 102 overlapping targets were subsequently visualized using Cytoscape software (Fig. 2C).

PPI networks of potential targets and identification of core genes

A protein-protein interaction (PPI) network was constructed using the STRING database, consisting of 100 nodes and 734 interaction edges. Cytoscape software was employed to analyze the topological properties of the network nodes and generate an optimized visualization of the PPI network (Fig. 3A). Node size and color were scaled according to their degree values, where larger dimensions and darker hues represent higher degrees and greater functional significance within the network. Using the CytoHubba plugin, we applied a fusion strategy incorporating eight topological algorithms (Stress, MNC, MCC, EPC, Betweenness, Radiality, Closeness, and Degree). The integrated results from these algorithms revealed several densely interconnected clusters (Fig. 3B). The intersection of results from all eight algorithms identified the six core targets: TLR4, ESR1, PPARG, SIRT1, NFKB1, and PTGS2 (Fig. 3C).

Fig. 3.

Collection of core gene targets of ATBC and sarcoma. (A) PPI network optimized with Cytoscape; (B) Top 10 genes screened by eight topological algorithms, with composite scores shown in a gradient colormap; (C) Intersection analysis of algorithmic results identified six targets with the highest pathological relevance (TLR4, ESR1, PPARG, SIRT1, NFKB1, PTGS2)

GO and KEGG pathway enrichment analysis of potential targets

GO enrichment analysis was performed on 102 candidate targets through the Metascape platform under Homo sapiens species limitation, revealing 1173 statistically enriched GO terms distributed across biological process (BP, 970 terms), cellular component (CC, 55 terms), and molecular function (MF, 148 terms) categories. The terms were sorted by P-value, and the top 10 entries from BP, CC, and MF were selected for visualization in enrichment analysis plots (Fig. 4A&B). Enrichment results demonstrated principal involvement of core targets in BP including cellular response to lipid, cellular response to hormone stimulus, and cellular response to organic cyclic compound; CC were predominantly enriched in transcription regulator complex, RNA polymerase II transcription regulator complex, and external side of plasma membrane; and MF were characterized by nuclear receptor activity, ligand-activated transcription factor activity, and chromatin binding.

Fig. 4.

Gene Ontology enrichment assessment of 102 candidate targets highlighting the ten most significantly enriched terms. (A) The dimensions of the bubbles correspond to the quantity of genes linked to a particular pathway, whereas the saturation of the color reflects the statistical confidence of the enrichment evaluation. (B) From the GO enrichment analysis conducted on 102 potential targets, the bar graph illustrates the ten most substantially enriched terms spanning biological processes BP, CC, and MF. Evaluation of the enrichment outcomes relies on P-values, with reduced numerical magnitudes designating greater statistical importance. The vertical extent of each bar is directly proportional to the P-value, offering a graphical depiction of the enrichment intensity across diverse functional groupings

Additionally, KEGG pathway enrichment analysis on the same targets using Metascape identified 155 significantly enriched signaling pathways, with the top 20 most significant pathways displayed in statistical bubble charts and categorical histograms ranked by ascending P-value (Fig. 5A&B). These pathways included Thyroid hormone, Estrogen, Neurotrophin, AGE-RAGE in diabetic complications, Prolactin, Toll-like receptor, HIF-1, Sphingolipid, and PI3K-Akt signaling pathways, indicating potential molecular mechanisms through which ATBC may affect sarcoma, along with associated biological processes and signaling pathways.

Fig. 5.

Kyoto Encyclopedia of Genes and Genomes enrichment examination of candidate targets (top 20). (A) The graphical representation illustrates the twenty most substantially enriched KEGG signaling pathways arranged in increasing P-value sequence. Individual bubbles correspond to distinct pathways, where dimensional area corresponds to enriched gene quantity and chromatic saturation denotes statistical confidence in enrichment. (B) The histogram presents pathway enrichment frequency and statistical confidence levels, where bar length exhibits positive correlation with gene quantity enriched in each pathway while concurrently demonstrating enrichment magnitude and statistical relevance. Longer bars indicate a greater number of enriched genes and stronger enrichment significance

Molecular docking of ATBC and sarcoma core target proteins

Molecular docking simulations between ATBC and core target proteins were performed using AutoDock software. The results demonstrated spontaneous binding interactions between ATBC and six sarcoma-related core targets: TLR4 (PDB ID: 3ul9, -2.03 kcal/mol), ESR1 (PDB ID: 6psj, -2.78 kcal/mol), PPARG (PDB ID: 7e0a, -2.56 kcal/mol), SIRT1 (PDB ID: 4zzi, -2.99 kcal/mol), NFKB1 (PDB ID: 8tqd, -2.69 kcal/mol), and PTGS2 (PDB ID: 5f19, -1.15 kcal/mol). The binding energies ranged from -1.15 to-2.99 kcal/mol (Table 3), indicating thermodynamically favorable interactions. Although these values indicate relatively weak interactions, they suggest that ATBC may directly interact with these targets. The docking conformation with the lowest binding energy was visualized using PyMOL software (Fig. 6A − F), clearly illustrating the binding modes between ATBC and the core target proteins.

Table 3.

Binding energies of ligands and receptors

ligand	receptor	Bind. Energy (Kcal/mol)
ATBC	TLR4	-2.03
ATBC	ESR1	-2.78
ATBC	PPARG	-2.56
ATBC	SIRT1	-2.99
ATBC	NFκB1	-2.69
ATBC	PTGS2	-1.15

Fig. 6.

Molecular docking in each target protein with the ATBC. (A) ATBC and TLR4; (B) ATBC and ESR1; (C) ATBC and PPARG; (D) ATBC and SIRT1; (E) ATBC and NFKB1; (F) ATBC and PTGS2

Impact of core targets expression on overall survival

Exploratory Kaplan-Meier survival analysis was further conducted to evaluate the association between the expression levels of the six core genes and overall survival in the public sarcoma cohort. The survival curves demonstrated that subgroups with high expression of TLR4 or ESR1 exhibited significantly longer survival times (Fig. 7A&B). In contrast, high PPARG expression was significantly associated with poor survival and markedly reduced overall survival (Fig. 7C). Elevated expression of SIRT1, as well as reduced expression of NFKB1 or PTGS2, may also be associated with an increased mortality risk (Fig. 7D–F), although the Kaplan-Meier analysis did not reveal statistically significant associations between their expression levels and patient survival (P > 0.05). Because this analysis was based on a pooled public dataset and did not include multivariable adjustment or subtype-stratified analysis, these findings should be interpreted cautiously. This lack of significance may also be due to factors such as cohort heterogeneity or the relatively small sample size available in the online database.

Fig. 7.

Analysis of overall survival for core targets in sarcoma. (A–F) Association between TLR4, ESR1, PPARG, SIRT1, NFKB1, and PTGS2 expression and patient prognosis

ATBC enhances sarcoma cell proliferation and migration in vitro

To evaluate the direct impact of ATBC on sarcoma cells, two in vitro models—the human sarcoma SW872 and MG-63 cell lines—were employed to investigate its cancer-promoting effects across various concentrations. CCK-8 assays demonstrated that ATBC enhanced the proliferation of sarcoma cells in vitro in a dose-dependent manner (Fig. 8A&B). Moreover, wound healing assays indicated that ATBC stimulated the migration of both SW872 and MG-63 cells, with more pronounced effects at higher concentrations (Fig. 8C–F). These results suggest that ATBC enhances the proliferation and migration of sarcoma cells in vitro.

Fig. 8.

Results of the experimental study on the effects of ATBC on sarcoma cells. (A-B) CCK-8 assay assessing the effect of ATBC on SW872 and MG-63 cells proliferation; (C, E) Microscopic images of the scratch wound healing assay in SW872 and MG-63 cells; (D, F) Wound closure percentage of SW872 and MG-63 cells measured via the wound healing assay

ATBC regulates the mRNA expression of core sarcoma-related genes

To further explore the transcriptional changes associated with ATBC exposure, the mRNA expression levels of selected core target genes were assessed by RT-qPCR. The results showed that 50 µM ATBC treatment decreased the mRNA expression of TLR4 and ESR1 in both SW872 and MG-63 cells, while increasing PPARG mRNA expression (Fig. 9A&B). These changes were directionally consistent with the survival analysis results and may provide preliminary support for the potential involvement of these genes in the biological effects of ATBC in sarcoma cells.

Fig. 9.

RT-qPCR analysis of representative hub genes in sarcoma cells following treatment with 50 µM ATBC. TLR4, ESR1, and PPARG mRNA expression in (A) SW872 and (B) MG-63 cells. Statistical evaluations were performed using two-tailed Student’s t-tests between the control and 50 µM ATBC groups. The symbols ns, *, **, and *** denote non-significance and significance levels at P < 0.05, P < 0.01, and P < 0.001, respectively

Additionally, although the initial Kaplan-Meier analysis did not reveal statistically significant associations between the expression levels of SIRT1, NFKB1, or PTGS2 and patient survival, ATBC treatment induced changes in their mRNA levels that showed a concordant trend with clinical prognosis (Figure S1). This observation requires further validation at the protein and pathway levels.

Discussion

This research combines molecular docking, survival analysis, network toxicology, and in vitro experiments to investigate the potential role and underlying mechanisms of ATBC in the development and progression of sarcoma. As a widely used non-phthalate plasticizer, ATBC has been frequently detected in water bodies, indoor dust, and food packaging due to its high mobility and environmental persistence, making it an environmental contaminant of increasing concern [6, 29]. Although traditionally regarded as a relatively safe alternative, recent studies have revealed that ATBC exhibits endocrine-disrupting effects, which can lead to metabolic disorders and developmental toxicity in animals [9–11]. Environmental pollutants, particularly those with endocrine-disrupting properties, have been implicated in an elevated risk of tumorigenesis by multiple studies. For instance, plastic additives (such as DEP, DMP, and DOP) and PET microplastics have been shown to interfere with key signaling proteins such as MAPK1 and PIK3CA, thereby disrupting core pathways like PI3K-Akt-mTOR and facilitating the progression of multiple cancer types, such as breast cancer [30, 31]. However, the role and molecular mechanisms of ATBC in sarcoma remain unclear. Our research addresses this knowledge gap by suggesting, for the first time, that ATBC may exert tumor-promoting effects in sarcoma through modulation of multiple core targets, with several signaling pathways identified as candidate mechanisms requiring further validation.

Our study suggested that ATBC may interact with six core targets—TLR4, ESR1, PPARG, SIRT1, NFKB1, and PTGS2. Enrichment analyses indicated that these targets were associated with multiple oncogenic signaling pathways, including PI3K-Akt, HIF-1, AGE-RAGE, and sphingolipid signaling, which may be relevant to the observed promotion of sarcoma cell proliferation and migration in vitro. Through multi-database screening and protein-protein interaction network analysis, we first identified the aforementioned six core targets, which play critical roles in cell proliferation, apoptosis, immune response, and metabolic regulation. Molecular docking results suggested that ATBC could interact with all six core targets, with predicted binding energies ranging from -1.15 to -2.99 kcal/mol. Although these values indicate relatively weak interactions, they provide a preliminary theoretical basis for the hypothesis that ATBC may directly engage these proteins.

Functional analysis of the core targets revealed that ATBC may exert its tumor-promoting effects by regulating a complex molecular network. Among these, TLR4 and ESR1 were identified as potential tumor suppressors. As a key receptor in innate immunity, the activation of TLR4 is essential for the priming and activation of T cells [32, 33]. In osteosarcoma, stimulation of the TLR4 pathway has been shown to inhibit tumor progression through the activation of CD8 + T cells [34]. In this study, ATBC treatment significantly downregulated TLR4 expression, which is consistent with Kaplan–Meier analysis showing that low TLR4 expression is associated with poorer patient prognosis. This suggests that ATBC may create a favorable environment for tumor development by impairing TLR4-mediated anti-tumor immune surveillance. Similarly, ESR1 (estrogen receptor alpha) has been reported to suppress tumorigenesis in liver and breast cancer cells via regulation of CCN5 [35]. Our results demonstrated that ATBC downregulates ESR1 expression, and low ESR1 levels were correlated with adverse clinical outcomes, further supporting the hypothesis that ESR1 may act as a tumor suppressor in sarcoma.

In contrast, PPARG was identified in this study as a key pro-tumorigenic driver. Our survival analysis revealed that high PPARG expression significantly correlated with poorer sarcoma patient outcomes. This finding is highly consistent with previous studies: PPARG has been shown to promote angiogenesis by upregulating VEGF in cancers such as breast and prostate cancer [36, 37]. Moreover, recent research has also clearly demonstrated that it accelerates disease progression in osteosarcoma by regulating osteoclast proliferation [38]. In the present study, in vitro experiments confirmed that ATBC exposure significantly upregulates PPARG expression, while simultaneously enhancing the proliferation and migration abilities of sarcoma cells. Taken together, these findings support the potential involvement of PPARG in the effects of ATBC on sarcoma at both the molecular and cellular levels.

Other core targets, including SIRT1, NFKB1, and PTGS2, although not demonstrating significant prognostic value in our survival analysis, play important roles in tumor biology and are implicated in the potential carcinogenic mechanisms of ATBC. SIRT1, an NAD⁺-dependent deacetylase, participates in stress response and regulation of cell survival by deacetylating key proteins such as p53 and FOXO [39, 40]. In soft tissue sarcoma, high expression of SIRT1 is correlated with unfavorable clinical outcomes [41], and inhibition of SIRT1 has been shown to effectively suppress growth and induce cell death in pediatric soft tissue sarcoma [42]. In the present study, ATBC treatment resulted in a significant increase in SIRT1 expression in cells, suggesting that SIRT1 may also contribute to the pro-tumorigenic effects associated with ATBC exposure. NFKB1, a core component of the NF-κB signaling pathway, has been reported under certain contexts to exert anti-inflammatory and tumor-suppressive effects [43, 44]. Our experiments observed that ATBC downregulated NFKB1 expression, which may impair its potential tumor-suppressive function. Furthermore, PTGS2, a key enzyme involved in inflammation and shaping the tumor microenvironment, catalyzes the production of PGE₂, which can induce immunosuppression and promote tumor immune escape [45, 46]. In this study, ATBC downregulated PTGS2 expression, indicating its possible involvement in toxic effects via modulation of inflammatory signaling and the tumor microenvironment. This possibility is supported by recent evidence showing that alterations in the tumor immune microenvironment can critically influence tumor progression, immune cell infiltration, and therapeutic response [47–49]. Taken together, these findings suggest multiple hub genes. In particular, altered TLR4, NFKB1, and PTGS2 expression may contribute to changes in inflammatory signaling and tumor microenvironment regulation [34, 44–46], whereas dysregulation of ESR1, PPARG, and SIRT1 may further support a proliferative and stress-adaptive state [35, 38, 41]. These effects may converge on PI3K-Akt and HIF-1 signaling, thereby linking immune dysregulation, hypoxic adaptation, and tumor progression [50, 51]. From this perspective, ATBC exposure may contribute to sarcoma development through multi-target and microenvironment-related mechanisms, although the causal relationships still require further validation.

At the pathway level, KEGG enrichment analysis indicated that ATBC-related core targets were enriched in multiple cancer-associated pathways, providing a bioinformatic framework for interpreting its potential effects in sarcoma. Among these, aberrant activation of the PI3K-Akt pathway is highly prevalent in sarcomas and is linked to cell proliferation, survival, and chemotherapy resistance [50, 52]. Such activation is often linked to the inactivation of tumor suppressor genes such as PTEN [53–55]. Our enrichment results suggest that ATBC may interfere with the regulation of the PI3K-Akt pathway, although direct pathway activation remains to be experimentally validated. The HIF-1 signaling pathway plays a central role in the tumor hypoxic response, and its activation can promote angiogenesis [56–58]. In osteosarcoma, it drives cancer progression via activation of the AKT/Cyclin D1 axis [51]. Therefore, ATBC exposure may mimic or exacerbate hypoxic stress within tumors. Additionally, the sphingolipid signaling pathway warrants attention, as its key component SPK1 is critical for maintaining sphingolipid metabolic homeostasis and promoting cell survival [59]. Studies have confirmed that SPK1 is upregulated in osteosarcoma and contributes to doxorubicin resistance [60], suggesting a potential link between ATBC and sphingolipid signaling that warrants further investigation.

In terms of experimental validation, ATBC significantly enhanced the proliferation and migration capabilities of SW872 and MG-63 cells in vitro. In addition, RT-qPCR performed at the selected 50 µM concentration showed that ATBC treatment downregulated the expression of TLR4 and ESR1, while upregulating PPARG expression, which was consistent with the bioinformatic prediction at the target-gene level and with the observed clinical prognostic trends. Although the mRNA changes of SIRT1, NFKB1, and PTGS2 did not reach statistical significance in the survival analysis, their expression trends aligned with the direction of clinical outcomes, suggesting that they may play auxiliary roles in the ATBC toxicity network. Their specific mechanisms of action require further elucidation in more complex in vivo models. Overall, these experimental data support the involvement of selected core targets, but do not directly validate activation of the predicted signaling pathways.

Nevertheless, this research has a number of limitations. First, the cell line models used cannot fully recapitulate the complex heterogeneity of sarcomas in vivo or the tumor microenvironment, which includes immune cells and stromal components. Second, short-term drug interventions in vitro may not adequately reflect the pathophysiological processes associated with chronic, low-dose exposure to ATBC in humans. Accordingly, future studies should incorporate lower, more environmentally relevant exposure concentrations and, where possible, compare the in vitro concentrations used in the present study with available human exposure or tissue concentration data to better evaluate the real-world relevance of these findings. In addition, the survival analysis in this study was exploratory in nature, was based on a pooled public sarcoma cohort, and did not include multivariate adjustment or subtype-stratified analysis. Moreover, although RT-qPCR revealed transcriptional changes in several candidate genes, the present study did not include protein-level confirmation by Western blot, nor did it directly verify pathway activation. Furthermore, because RT-qPCR validation was performed at a single selected concentration, the dose dependence of ATBC-induced transcriptional changes could not be evaluated. Therefore, the mechanistic interpretation should be considered preliminary. Taken together, further studies are imperative to validate the tumor-promoting effects of ATBC using animal models, clinical models, and even epidemiological investigations, as well as to elucidate its direct regulatory impact on the tumor immune microenvironment.

It is important to note that the binding energies predicted by molecular docking in this study ranged from -1.15 to -2.99 kcal/mol, which are relatively modest compared to typical drug–target interactions. However, several considerations support the biological relevance of these findings. First, molecular docking serves primarily as a tool to predict potential binding modes and interaction residues, and its calculated binding energies are semi-quantitative; the presence of negative values, regardless of magnitude, indicates thermodynamically feasible interactions. Second, ATBC is an environmental plasticizer rather than an optimized pharmaceutical agent, and its interactions with protein targets are expected to be relatively weak, consistent with the concept that environmental chemicals often exert biological effects through low-affinity, multi-target mechanisms. Third, from a network toxicology perspective, ATBC is predicted to interact with multiple core targets. Even if individual binding affinities are modest, their combined perturbation of the molecular network can lead to significant phenotypic consequences, as supported by our in vitro observations of altered gene expression, enhanced proliferation, and increased migration in sarcoma cells. Moreover, the binding affinities predicted by molecular docking were relatively modest, and no positive control compounds were included for comparison. Future studies incorporating compounds with known binding affinities for these targets would help benchmark the docking results and more accurately assess the interaction strength between ATBC and its putative targets.

In summary, this study provides integrated evidence that ATBC may promote malignant phenotypes in sarcoma cells and may interact with several core targets, including TLR4, ESR1, PPARG, SIRT1, NFKB1, and PTGS2. Bioinformatic analyses further highlighted PI3K-Akt, HIF-1, and sphingolipid signaling as candidate pathways potentially involved in these effects, although direct mechanistic validation is still required. These findings not only provide a novel theoretical foundation for understanding the toxicological mechanisms of ATBC, but also suggest potential molecular targets for assessing its environmental health risks and developing preventive and therapeutic strategies for sarcoma.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Yue Wang: Conceptualization, Writing – original draft, Visualization, Methodology, Data curation, Funding acquisition. Xuan Lin: Writing – original draft. Yihuang Chen: Writing – original draft, Data curation. Yuanqun Zhang: Visualization, Software. Qiang Zhang: Writing – original draft. Miao Xu: Data curation, Xiaoning Lin: Data curation, Xin Wu: Data curation, Shifu Peng: Supervision, Conceptualization. Jianlin Shen: Writing – review & editing, Supervision, Conceptualization, Funding acquisition.

Funding

This study was supported by the Natural Science Foundation of Fujian Province (2025J08239), the Young and Middle-aged Teachers Educational Scientific Research Program of Fujian Province (JAT241104), the Startup Fund for Advanced Talents of Putian University (2024148), the Medical Research Project of the Jiangsu Commission of Health (grant M2022069), the Medical Research Foundation of Putian University (2024104) and the Science and Technology Bureau of Putian City (2025SZ3001PTXY08).

Data availability

Data will be made available on request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.