Abstract Aspartame, a widely used artificial sweetener, remains controversial due to neurotoxic risks from its metabolites—phenylalanine, aspartic acid, and methanol. While epidemiological studies link artificial sweeteners to cerebrovascular disease, the molecular mechanisms connecting aspartame to ischemic stroke are unclear. This study integrates network toxicology and molecular docking to identify key targets and pathways. Potential aspartame targets were predicted using STITCH, SwissTargetPrediction, and SEA databases, while ischemic stroke-related genes were retrieved from GeneCards, OMIM, and TTD. Venn analysis identified 201 overlapping genes, with IL1B, MMP9, SRC, AGT, and TNF as core targets. GO/KEGG enrichment revealed their roles in the renin-angiotensin system, complement/coagulation cascades, and inflammatory pathways. Molecular docking demonstrated strong binding affinities between aspartame and these targets, suggesting direct modulation. Our integrated analysis suggests that aspartame may contribute to ischemic brain injury through multi-target interactions, potentially disrupting inflammatory responses and vascular homeostasis. This study provides preliminary systematic insights into the potential neurotoxicity mechanisms of aspartame, offering insights for food additive safety evaluation and stroke prevention. Further validation is required to clarify metabolite synergies and dose–response relationships. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-025-08898-z. Keywords: Aspartame, Ischemic stroke, Network toxicology, Molecular docking Subject terms: Computational biology and bioinformatics, Health care, Risk factors, Neurological disorders Introduction Aspartame, chemically known as L-α-aspartyl-L-phenylalanine methyl ester, is a high-intensity artificial sweetener that has been widely used in the global food industry since its discovery in 1965. With a sweetness approximately 200 times that of sucrose and minimal caloric contribution, aspartame quickly became one of the most extensively utilized sugar substitutes worldwide^[32]1. Currently, aspartame has been approved for use in over 90 countries and is present in thousands of products, including sugar-free beverages, low-calorie desserts, chewing gum, pharmaceuticals, and energy drinks^[33]2. Its primary advantage lies in its ability to provide the desired sweetness with minimal consumption, making it particularly favored by individuals with diabetes and those managing their weight. However, debates over its safety have persisted for decades. Among artificial sweeteners, aspartame is unique due to its metabolic breakdown into phenylalanine, aspartic acid, and methanol—a combination implicated in neurotoxicity and oxidative stress^[34]3. In contrast, other sweeteners such as sucralose and steviol glycosides lack direct neurotoxic metabolites^[35]4,[36]5, making aspartame a critical focus for cerebrovascular risk assessment. Although the U.S. Food and Drug Administration (FDA) has established an acceptable daily intake (ADI) of 40 mg/kg body weight and deemed aspartame safe at recommended doses^[37]6, the classification of aspartame as a Group 2B potential carcinogen by the International Agency for Research on Cancer (IARC) in 2023 has intensified scientific controversy. The core of this debate stems from the limitations of early animal studies, where high-dose exposure results are difficult to extrapolate to long-term, low-dose human consumption^[38]6. Moreover, aspartame undergoes complex metabolic pathways, breaking down into phenylalanine, aspartic acid, and methanol, which may interfere with neurotransmitter balance, induce oxidative stress, and trigger inflammatory responses—mechanisms potentially linked to cerebrovascular diseases, depression, and autism^[39]3. Notably, several studies suggest that aspartame may increase the risk of neurodegeneration by disrupting the blood–brain barrier (BBB) and causing neuronal damage through its metabolic byproducts^[40]7, a mechanism that may intersect with the pathological processes of ischemic stroke. Ischemic stroke, the second leading cause of death globally, accounts for approximately 87% of all stroke cases. Its pathophysiological hallmark is the obstruction of cerebral arteries, leading to local hypoxia and blood flow disruption, followed by BBB breakdown, cerebral edema, and neuronal death^[41]8. According to the Global Burden of Disease (GBD) study, 13.7 million new stroke cases were reported worldwide in 2016, with ischemic strokes comprising 84.4% of cases, and East Asia exhibiting the highest age-standardized incidence rates^[42]9. The loss of BBB integrity is a critical determinant of poor prognosis in ischemic stroke, with mechanisms involving tight junction protein degradation, endothelial cell injury, and inflammatory mediator release, ultimately resulting in irreversible neurological deficits. Although reperfusion therapies, such as thrombolysis or thrombectomy, can partially restore cerebral blood flow, reperfusion injury-induced oxidative stress and inflammatory cascades remain major challenges in clinical management^[43]10. In recent years, increasing attention has been directed toward the chronic impact of exogenous chemicals (e.g., environmental toxins and food additives) on BBB function, with aspartame metabolites emerging as potential modulators of stroke risk by disrupting cerebrovascular homeostasis. However, while existing research primarily focuses on the carcinogenicity of aspartame, its mechanistic link to cerebrovascular diseases remains largely unexplored. Recent epidemiological studies have yielded mixed results regarding the association between artificially sweetened beverage consumption and stroke risk. The Women’s Health Initiative Observational Study, involving 81,714 postmenopausal women, found that participants consuming two or more artificially sweetened beverages per day had a 23% increased risk of stroke (HR = 1.23, 95% CI 1.02–1.47) compared to those consuming less than one per week^[44]11. Similarly, the Framingham Heart Study Offspring cohort reported that individuals consuming at least one artificially sweetened soft drink daily had nearly three times the risk of ischemic stroke (HR = 2.96, 95% CI 1.26–6.97) compared to non-consumers^[45]12. Conversely, other studies have not observed a significant association after adjusting for confounding factors such as body mass index and pre-existing conditions^[46]13. Against this backdrop, integrating network toxicology and molecular docking approaches provides a novel paradigm for investigating the potential association between aspartame and ischemic stroke. Network toxicology, an interdisciplinary approach combining bioinformatics, systems biology, and chemoinformatics, offers a systematic framework for understanding how chemical compounds disrupt biomolecular networks and cellular functions, ultimately contributing to disease pathogenesis^[47]14. Molecular docking further enables atomic-level simulations of aspartame and its metabolites binding to target proteins to predict potential toxicity mechanisms^[48]15. By integrating these interdisciplinary methodologies, this study aims to elucidate how aspartame influences ischemic stroke pathogenesis, providing a theoretical basis for risk assessment and intervention strategies, while also establishing a methodological framework for the systematic study of food additive neurotoxicity. Materials and methods Toxicity identification of aspartame The chemical structure and standardized SMILES sequence of aspartame were retrieved from the PubChem database^[49]16 ([50]https://pubchem.ncbi.nlm.nih.gov/) using its unique identifier (CID: 134,601). To comprehensively assess its toxicity profile, a dual-platform cross-validation approach was employed, utilizing ProTox-3.0 ([51]https://tox.charite.de/protox3/) and ADMETLab 2.0 ([52]https://admetmesh.scbdd.com/). ProTox-3.0 employs machine learning models to predict compound toxicity categories and median lethal dose (LD50), whereas ADMETLab 2.0 evaluates its drug-likeness and potential risks based on ADMET (absorption, distribution, metabolism, excretion, and toxicity) parameters. This dual validation strategy was adopted to minimize biases associated with a single database and enhance the reliability of toxicity assessment. Collection of aspartame target genes Potential targets of aspartame were systematically predicted through an integrated bioinformatics approach. The chemical structure of aspartame (PubChem CID: 134,601) and its canonical SMILES string (COC(= O)[C@H](CC1 = CC = CC = C1)NC(= O)[C@H](CC(= O)O)N) were first retrieved from the PubChem database. Target prediction was then performed using three complementary databases: the STITCH^[53]17 ([54]http://stitch.embl.de/) database was queried using “Aspartame” with species restricted to Homo sapiens; SwissTargetPrediction^[55]18 ([56]http://www.swisstargetprediction.ch/) analysis was conducted by inputting the SMILES string of aspartame with Homo sapiens as the target organism, retaining predictions with probability scores ≥ 0.1; and the SEA^[57]19 ([58]https://sea.bkslab.org/) database was searched using the same SMILES string. The retrieved data were analyzed for structural consistency, and overlapping targets were integrated and de-duplicated using Venn diagrams to establish a comprehensive aspartame target library. Collection of ischemic stroke-related genes The disease-related target library was constructed based on data retrieved from the GeneCards ^[59]20 ([60]https://www.genecards.org/), OMIM ([61]https://www.omim.org/), and TTD ([62]https://db.idrblab.net/ttd/) databases using the keyword “Ischemic Stroke.” The selection of overlapping genes between aspartame and ischemic stroke was guided by GeneCards’ relevance score, which integrates multiple lines of evidence including genetic associations, functional annotations, and literature support. The threshold of 1.0 was selected to ensure a balance between including genes with established disease associations while minimizing false positives, a common practice in similar network toxicology studies^[63]21. This approach aligns with methodologies used in recent investigations of environmental toxins and neurological disorders^[64]22. While this scoring system provides a systematic framework for target identification, it is important to note that it may prioritize well-studied genes and pathways, potentially overlooking novel targets with limited existing annotation. The threshold also does not fully capture population-specific genetic variations that could influence stroke risk, as demonstrated by differences in genetic associations observed across diverse ethnic groups^[65]23. After merging and removing duplicates from the three databases using a Venn diagram, a final target library for ischemic stroke was established. The database search for this study was completed as of March 3, 2025. Intersection of aspartame and ischemic stroke targets The overlapping genes between aspartame targets and ischemic stroke-related genes were identified using the “ggvenn” package in R. The results were visualized using Cytoscape 3.10.3 ([66]https://cytoscape.org/). GO/KEGG enrichment analysis To elucidate the biological significance of the overlapping genes, Gene Ontology (GO) functional annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were conducted using the “ClusterProfiler,” “Enrichplot,” and “Org.Hs.eg.db” packages in R. This analysis aimed to reveal the potential roles of key genes in biological processes, molecular functions, and cellular components, as well as their involvement in signaling pathways. Screening of core targets and construction of PPI network Protein–protein interaction (PPI) analysis was performed on the overlapping genes using the STRING database ^[67]21 ([68]https://cn.string-db.org/), with a confidence score threshold set at ≥ 0.7 and species limited to Homo sapiens. Isolated nodes were excluded to ensure biologically meaningful interactions. The PPI network was visualized using Cytoscape 3.10.3 ^[69]24 ([70]https://cytoscape.org/), and key hub genes were identified based on degree centrality using the cytoHubba plugin. Molecular docking analysis To evaluate the binding affinities and interaction patterns between aspartame and its core target genes, we employed AutoDock Vina 1.2.2, a widely used computational protein–ligand docking software.The X-ray crystal structures of AGT (PDB ID: 1T38), IL1B (PDB ID: 1T4Q), MMP9 (PDB ID: 1ITV), and SRC (PDB ID: 8BQ3) were retrieved from the RCSB Protein Data Bank^[71]25 ([72]https://www.rcsb.org/). The 3D structures of Aspartame (CAS: 22839–47-0), Captopril (CAS: 62571–86-2), and Thalidomide (CAS: 50–35-1) were obtained in SDF format from the PubChem database. Protein and ligand files were preprocessed by converting them into PDBQT format and adding polar hydrogen atoms. The docking grid box was defined as a cubic region with dimensions of 40 Å × 40 Å × 40 Å and a grid spacing of 0.05 nm. To assess the binding affinity of aspartame to its targets, we conducted two sets of docking analyses using positive control compounds—Captopril with AGT and Thalidomide with TNF. Docking poses and binding energies for all protein–ligand complexes were generated using AutoDock Vina 1.2.2, with the lowest binding energy conformation selected for further analysis. Protein–ligand binding interfaces were systematically analyzed using PLIP and LigPlus, and interaction details were further visualized and annotated with PyMOL 2.5. Binding energy values were used to evaluate ligand-receptor interactions: a binding energy of less than 0 kcal/mol indicated spontaneous binding, while a value below − 5 kcal/mol suggested a relatively stable interaction. The workflow of this study is illustrated in Fig. [73]1. Fig. 1. [74]Fig. 1 [75]Open in a new tab Flowchart of the study design. Results Collection of aspartame target genes Potential targets of aspartame were predicted using three independent databases: STITCH, SwissTargetPrediction, and SEA. After removing duplicates, we constructed an aspartame target gene database comprising 352 targets (Supplementary Table S1). Collection of ischemic stroke-associated genes We searched the GeneCards ([76]https://www.genecards.org/), OMIM ([77]https://www.omim.org/), and TTD ([78]https://db.idrblab.net/ttd/) databases using the keyword “Ischemic Stroke.” To ensure the relevance of the retrieved genes to both ischemic stroke and aspartame, we set the “Relevance Score” threshold to 1.0 in the GeneCards database and selected genes with scores above this threshold. After removing duplicate targets from the three databases, we established an ischemic stroke target gene database containing 6,833 targets (Supplementary Table S2). Identification of core targets and interaction network between aspartame and ischemic stroke Using the “ggvenn” package in R, we identified 201 overlapping target genes between aspartame and ischemic stroke. These were then visualized using Cytoscape 3.10.3 ([79]https://cytoscape.org/) (Fig. [80]2). We performed a protein–protein interaction (PPI) analysis on the overlapping genes using the STRING database ([81]https://cn.string-db.org/), setting the confidence score threshold to ≥ 0.7, restricting the species to Homo sapiens, and excluding isolated nodes. This process yielded 200 biologically significant interaction targets (Fig. [82]3A). The PPI network was visualized in Cytoscape 3.10.3, and core hub genes were identified using the cytoHubba plugin based on degree centrality. The five key hub genes—IL1B, MMP9, SRC, AGT, and TNF—were selected, with darker colors and larger circles indicating stronger interactions with other proteins (Fig. [83]3B). This visualization provides a comprehensive overview of the interactions among key targets. Fig. 2. [84]Fig. 2 [85]Open in a new tab (A) Venn diagram of targets associated with aspartame and ischemic stroke. (B) Network diagram depicting the potential targets shared between aspartame and ischemic stroke. Fig. 3. [86]Fig. 3 [87]Open in a new tab (A) Protein–protein interaction (PPI) network of potential targets, with a confidence score threshold set at ≥ 0.7; nodes represent proteins and edges indicate their interactions. (B) Visualization of the PPI network using Cytoscape, with node color and size adjusted based on degree values—darker colors and larger circles represent stronger interactions. (C) Gene Ontology (GO) enrichment analysis of aspartame–ischemic stroke targets. (D) Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of aspartame–ischemic stroke targets^[88]26–[89]28. GO and KEGG enrichment analysis By analyzing the intersection of aspartame-related genes and ischemic stroke-associated genes, we identified 201 overlapping genes. Gene Ontology (GO) functional enrichment analysis revealed that these genes are primarily involved in biological processes such as blood pressure regulation, coagulation/fibrinolysis balance, extracellular matrix (ECM) remodeling, hypoxia adaptation, inflammation regulation, apoptosis, and neurotransmitter homeostasis (Fig. [90]3C). Additionally, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis demonstrated that these genes are significantly enriched in multiple critical pathways, including the renin-angiotensin system, neuroactive ligand-receptor interaction, apoptosis, proteasome, Alzheimer’s disease, complement and coagulation cascades, lipid metabolism and atherosclerosis, prostate cancer, proteoglycans in cancer, and fluid shear stress and atherosclerosis (Fig. [91]3D). These findings suggest that aspartame may influence the occurrence and progression of ischemic stroke through pathways related to neuroprotection, inflammatory response, apoptosis, vascular regulation, and tissue repair. Molecular docking of aspartame with core target genes Molecular docking analysis demonstrated that all five key target genes could spontaneously bind to aspartame (binding energy < 0 kcal/mol) (Table [92]1). Among them, IL1B, MMP9, AGT, and TNF exhibited binding energies lower than − 5 kcal/mol, indicating stable interactions with aspartame (Figs. [93]4, [94]5, [95]6, [96]7, [97]8)^[98]29–[99]31. To better evaluate the binding affinity of aspartame to key target genes, we additionally performed molecular docking analyses between AGT and Captopril (a known AGT modulator), as well as between TNF and Thalidomide (a TNF inhibitor) (Figs. [100]8, [101]9, [102]10). These results suggest that aspartame may exert its effects on ischemic stroke by directly interacting with these key target genes, thereby modulating relevant biological processes. Table 1. Binding energies between the core targets and aspartame. Target Degree PDB number Hydrone Binding energy (kcal/mol) TNF 39 4V46 Aspartame − 5.0 Thalidomide − 6.0 IL1B 30 1T4Q Aspartame − 5.8 SRC 28 8BQ3 Aspartame − 4.8 AGT 25 1T38 Aspartame − 6.7 Captopril − 4.6 MMP9 24 1ITV Aspartame − 6.2 [103]Open in a new tab Binding energies were computationally estimated via AutodockVina. Fig. 4. [104]Fig. 4 [105]Open in a new tab Molecular docking of Aspartame with its target AGT. (A) Cartoon representation of the 3D interactions between the small molecule compound and its target protein. (B) 2D interaction diagram illustrating the binding between the compound and its target. Fig. 5. [106]Fig. 5 [107]Open in a new tab Molecular docking of Aspartame with its target IL1B. (A) Cartoon representation of the three-dimensional interactions between the small molecule and its target protein. (B) Two-dimensional interaction map of the compound and its target. Fig. 6. [108]Fig. 6 [109]Open in a new tab Molecular docking of Aspartame with its target MMP9. (A) Cartoon representation of the three-dimensional interactions between the small molecule and its target protein. (B) Two-dimensional interaction map of the compound and its target. Fig. 7. [110]Fig. 7 [111]Open in a new tab Molecular docking of Aspartame with its target SRC. (A) Cartoon representation of the three-dimensional interactions between the small molecule and its target protein. (B) Two-dimensional interaction map of the compound and its target. Fig. 8. [112]Fig. 8 [113]Open in a new tab Molecular docking of Aspartame with its target TNF. (A) Cartoon representation of the three-dimensional interactions between the small molecule and its target protein. (B) Two-dimensional interaction map of the compound and its target. Fig. 9. [114]Fig. 9 [115]Open in a new tab Molecular docking of Captopril with its target AGT. (A) Cartoon representation of the three-dimensional interactions between the small molecule and its target protein. (B) Two-dimensional interaction map of the compound and its target. Fig. 10. [116]Fig. 10 [117]Open in a new tab Molecular docking of Thalidomide with its target TNF. (A) Cartoon representation of the three-dimensional interactions between the small molecule and its target protein. (B) Two-dimensional interaction map of the compound and its target. Discussion Aspartame, an artificially synthesized high-intensity sweetener, has been widely utilized in the food industry since the 1980s. Its metabolites—phenylalanine, aspartic acid, and methanol—have raised persistent scientific debates regarding their potential toxicity^[118]2. Although the U.S. Food and Drug Administration (FDA) has established an acceptable daily intake (ADI) of 40 mg/kg body weight^[119]3, the World Health Organization (WHO) has recently classified it as a potential carcinogen, further underscoring the complexity of its safety concerns. Existing studies have primarily focused on the carcinogenicity and metabolic disturbances associated with aspartame. Notably, recent work by the team led by Yihai Cao demonstrated that aspartame can accelerate the formation of atherosclerotic plaques through an insulin-mediated inflammatory response, suggesting a link with cardiovascular disease^[120]32. However, ischemic stroke, the world’s second leading cause of death, is closely associated with blood–brain barrier disruption, inflammatory cascades, and oxidative stress^[121]8, and the chronic impact of aspartame metabolites on cerebrovascular homeostasis remains underexplored. Consequently, the potential mechanisms by which aspartame may influence stroke risk—whether via modulation of inflammatory mediators, vascular tone, or coagulation balance—warrant further investigation. Traditional risk factors for ischemic stroke include atherosclerosis, hypertension, and atrial fibrillation^[122]33, yet the continuously rising incidence suggests that emerging environmental factors should not be overlooked. Recent studies have gradually elucidated the prothrombotic effects of exogenous chemicals, such as erythritol^[123]34, Given that aspartame is one of the most pervasive artificial sweeteners used in the global food industry, its cumulative effects from long-term, low-dose exposure may indirectly affect cerebrovascular health through metabolic interference. Although epidemiological evidence implies a potential association between artificial sweetener consumption and stroke risk, the underlying molecular mechanisms remain unclear. This study is the first to integrate network toxicology and molecular docking techniques to systematically explore the target interaction network between aspartame and ischemic stroke. The aim is to fill the gap in understanding the neurotoxic mechanisms of food additives and to provide a new paradigm for risk assessment. Through the intersection analysis of genes related to aspartame and ischemic stroke, we identified 201 overlapping genes and further screened five core targets: IL1B, MMP9, SRC, AGT, and TNF. These targets were significantly enriched in pathways including the renin–angiotensin system (RAS), the complement and coagulation cascades, and inflammatory regulatory pathways. These findings may reflect the combined effects of aspartame’s metabolites. Methanol, a primary breakdown product of aspartame, is metabolized to formate, which inhibits mitochondrial complex IV and induces oxidative stress—a process implicated in angiotensin II-mediated vascular dysfunction^[124]32. Such mitochondrial impairment has been shown to exacerbate RAS overactivation in preclinical models of hypertension^[125]35,[126]36. Additionally, phenylalanine, another metabolite, may interfere with large neutral amino acid transporters at the blood–brain barrier, potentially disrupting neurotransmitter balance and amplifying neuroinflammatory responses, as observed in studies linking phenylalanine accumulation to microglial activation^[127]37. Aspartic acid, through its role as a precursor to excitatory neurotransmitters, could further aggravate neuronal calcium overload during ischemia, a mechanism consistent with the observed apoptosis pathway enrichment^[128]38. Overactivation of the RAS can induce vasoconstriction and oxidative stress, and AGT, as a key component of this system, may exacerbate post-ischemic microcirculatory dysfunction by promoting angiotensin II production. Moreover, the enrichment of the complement and coagulation cascades suggests that aspartame may interfere with the coagulation–fibrinolysis balance, thereby increasing the propensity for thrombosis—a mechanism analogous to the recently reported prothrombotic effect of erythritol. These findings indicate that aspartame may compromise cerebrovascular homeostasis and exacerbate ischemic injury through multi-target interactions, providing a theoretical basis for the neurotoxicity evaluation of artificial sweeteners. TNF-α, a central proinflammatory cytokine^[129]39, exhibits a complex dual role in ischemic stroke. On one hand, TNF-α promotes neutrophil infiltration and blood–brain barrier permeability via activation of the NF-κB signaling pathway, thereby aggravating cerebral edema and neuronal death^[130]40; on the other hand, animal studies suggest that moderate expression of TNF-α can induce ischemic tolerance by modulating apoptosis-related proteins, such as members of the Bcl-2 family, thereby mitigating reperfusion injury^[131]41. Our molecular docking results confirmed a stable binding between aspartame and TNF (binding energy − 5.0 kcal/mol), suggesting that aspartame may directly interfere with the conformation or signal transduction of TNF-α. However, the team led by Yihai Cao found that aspartame mediates inflammatory responses via the chemokine CX3CL1, with TNF-α potentially acting as a downstream effector. This hierarchical regulatory network implies that the intervention of aspartame on TNF may be dose-dependent, necessitating further in vivo studies to determine its critical threshold for either proinflammatory or neuroprotective effects. IL1B is a key driver of the inflammatory cascade in ischemic stroke, promoting the maturation and release of IL-18 and IL-1β through the activation of the NLRP3 inflammasome^[132]42, thereby amplifying neuroinflammation following cerebral ischemia^[133]43. The high affinity binding between IL1B and aspartame observed in this study suggests that aspartame may directly activate Toll-like receptor (TLR) signaling by mimicking pathogen-associated molecular patterns (PAMPs), leading to the overexpression of IL1B^[134]44. Preclinical studies have demonstrated that IL1B inhibitors, such as anakinra, can significantly reduce infarct volume and improve neurological deficits^[135]45. Nevertheless, complete inhibition of IL1B may impair host defense mechanisms and elevate infection risk. Therefore, targeting upstream regulators of IL1B (such as NLRP3 or ASC)^[136]46 or developing small-molecule allosteric inhibitors may offer safer therapeutic strategies. The interaction between aspartame and IL1B provides a potential target for the development of novel anti-inflammatory agents, although its clinical translation requires careful evaluation of the risk–benefit ratio. MMP9, a member of the matrix metalloproteinase family, plays a pivotal role in ischemia–reperfusion injury by degrading type IV collagen and laminin, thereby disrupting the blood–brain barrier and promoting inflammatory cell infiltration and cerebral edema^[137]47. Our study found that the binding energy between aspartame and MMP9 is as low as − 6.2 kcal/mol, suggesting that aspartame may competitively inhibit MMP9 activity or interfere with its substrate binding, thereby mitigating blood–brain barrier leakage. This hypothesis contrasts with findings from the team led by Yihai Cao, who reported that MMP9 plays a key role in plaque instability during atherosclerosis^[138]32. The apparent discrepancy may arise from tissue-specific differences: in brain vascular endothelial cells, aspartame’s inhibitory effect on MMP9 might be protective, whereas in atherosclerotic plaques, it may promote MMP9 expression through alternative pathways such as CX3CL1. AGT, the rate-limiting component of the renin–angiotensin system (RAS), is significantly associated with stroke risk due to its genetic polymorphisms^[139]48. This study is the first to propose that aspartame may upregulate AGT expression, thereby promoting the production of angiotensin II, which induces vasoconstriction, oxidative stress, and endothelial dysfunction. This hypothesis is supported indirectly by experimental models. In atherosclerosis-prone mice, chronic aspartame exposure (600 mg/day for 12 weeks) was shown to increase vascular inflammation and plaque formation, accompanied by elevated levels of IL-6 and TNF-α—cytokines downstream of angiotensin II signaling^[140]32. These effects align with the observed high-affinity binding between aspartame and AGT (− 6.7 kcal/mol) in our molecular docking analysis. Furthermore, methanol, a key aspartame metabolite, has been demonstrated to enhance reactive oxygen species (ROS) production in endothelial cells, a known driver of RAS pathway activation^[141]49–[142]51. However, direct evidence linking aspartame consumption to RAS modulation in humans remains sparse. A recent systematic review concluded that the effects of non-nutritive sweeteners on RAS components are inconclusive, largely due to heterogeneity in study designs and exposure assessment methods^[143]52. Similarly, This mechanism is consistent with recent epidemiological evidence suggesting a positive correlation between artificial sweetener consumption and the incidence of hypertension^[144]53, a core risk factor for ischemic stroke. Furthermore, angiotensin II can exacerbate free radical generation via NADPH oxidase activation, synergistically contributing to the oxidative stress effects of methanol, a metabolite of aspartame. Interventions targeting AGT—such as ACE inhibitors or angiotensin receptor blockers—have been widely applied in stroke prevention. The interaction between aspartame and AGT suggests that reducing the intake of artificial sweeteners may enhance the efficacy of existing antihypertensive therapies. SRC, a member of the non-receptor tyrosine kinase family, influences angiogenesis and neuronal survival in ischemic stroke through the regulation of VEGF and the PI3K/Akt signaling pathways^[145]54. Although the binding energy between aspartame and SRC (− 4.8 kcal/mol) is lower than that of the other core targets, the pivotal role of SRC in cellular migration and inflammatory responses cannot be overlooked. Overactivation of SRC may increase blood–brain barrier permeability and activate microglia, whereas its specific inhibition might interfere with physiological repair processes^[146]55. For instance, dasatinib, an SRC inhibitor, has been shown to alleviate cerebral ischemic injury, yet its long-term use may suppress neural stem cell proliferation^[147]54. Further in vitro studies are required to elucidate whether aspartame regulates SRC activity through direct binding or via epigenetic modifications. Molecular docking analysis revealed that aspartame binds to IL1B, MMP9, TNF, and AGT with binding energies lower than − 5 kcal/mol, with the interactions being most stable for AGT (− 6.7 kcal/mol) and TNF (− 5.0 kcal/mol). Although static molecular docking results may be affected by solvent effects and conformational dynamics, a growing body of evidence supports the predictive utility of binding energy thresholds for specific biological targets. For example, Kitchen et al. reported that compounds with binding energies below − 5 kcal/mol tend to demonstrate higher rates of experimental validation in virtual screening studies, particularly in the context of transmembrane receptors and inflammation-related proteins^[148]29. Furthermore, Wang and Zhu, in their review, highlighted that despite the inherent limitations of molecular docking, binding energy thresholds—such as − 5 kcal/mol—serve as a meaningful reference for assessing ligand–protein affinity, especially in scenarios where experimental data are lacking^[149]30. The binding energy analysis indicates that aspartame may interact with the active pockets of these targets via hydrogen bonds and hydrophobic interactions. These binding energies align with computational models of neurotoxic compounds that disrupt mitochondrial function by inducing reactive oxygen species (ROS) generation—a mechanism observed in hypoxia-related cellular damage^[150]56. For instance, mitochondrial complex III-derived ROS have been shown to stabilize hypoxia-inducible factor-1α (HIF-1α) during oxygen sensing, a pathway potentially modulated by aspartame metabolites such as methanol^[151]56. However, the biological relevance of these interactions requires cautious interpretation. Previous studies on ischemia-related biomarkers (e.g., D-lactate and intestinal fatty acid-binding protein) highlight the necessity of combining computational predictions with functional assays to validate pathological impacts^[152]57. Although our study focuses on computational analyses, the observed affinity of aspartame for inflammatory targets (e.g., TNF) is supported by experimental evidence linking mitochondrial dysfunction to blood–brain barrier permeability in aged skeletal muscle models^[153]58. Future validation through in vitro models—such as measuring IL1B secretion in aspartame-exposed microglia or assessing mitochondrial ROS production in endothelial cells—would strengthen these findings. Notably, the docking results for the aspartame prototype suggest that it may directly interfere with target protein functions; however, the indirect effects of metabolites such as methanol require further elucidation through dynamic simulations and in vitro experiments. It is noteworthy that the complexity of stroke risk is not limited to the influence of individual food additives; other dietary components and lifestyle-related confounding variables may exert synergistic or antagonistic effects. A prospective cohort study published in The BMJ demonstrated a significant association between excessive intake of food emulsifiers and an increased risk of cardiovascular disease^[154]59. This association may be mechanistically mediated through disruption of the intestinal barrier and induction of chronic inflammation, which in turn may promote atherosclerosis and elevate the risk of ischemic stroke. Moreover, the consumption of trans fatty acids has been confirmed to be associated with non-alcoholic fatty liver disease and metabolic syndrome, both of which are recognized as independent risk factors for stroke^[155]60. Beyond specific additives, the overall impact of dietary patterns should not be overlooked. Existing studies have shown that individuals with low intake of fruits, nuts, and whole grains face a significantly increased risk of stroke, whereas high consumption of sugar-sweetened beverages and red meat may aggravate vascular injury via mechanisms involving insulin resistance and oxidative stress^[156]61. The contribution of confounding variables such as obesity and hypertension should also be taken into account. Evidence indicates that overweight status and elevated blood pressure are primary risk factors for stroke onset^[157]62,[158]63, and these metabolic abnormalities often coexist with high consumption of processed foods rich in additives. Consequently, the potential neurotoxicity of aspartame may exacerbate these effects by intensifying metabolic dysregulation and thereby interacting synergistically with the aforementioned factors. Future research should employ Mendelian randomization analysis or prospective cohort studies to further clarify the interactions between food additives and lifestyle variables, and to evaluate the potential regulatory effects of various dietary patterns on stroke risk. To further evaluate the binding affinity of aspartame, we compared its interaction characteristics with those of known TNF inhibitors and AGT modulators. For instance, thalidomide, a TNF-α inhibitor^[159]64, exhibits a binding energy of approximately − 6.0 kcal/mol with TNF and exerts its anti-inflammatory effect by blocking the interaction between TNF-α and its receptor. Although aspartame shows a slightly lower binding energy with TNF (− 5.0 kcal/mol) than thalidomide, the result still indicates considerable binding potential, suggesting that aspartame may influence the inflammatory cascade by altering the conformation or signaling of TNF-α. Similarly, captopril, an angiotensin-converting enzyme (ACE) inhibitor, reduces the conversion of AGT to Ang II by inhibiting ACE^[160]65, with a reported binding energy of − 4.6 kcal/mol. In contrast, aspartame demonstrates a stronger binding affinity with AGT (− 6.7 kcal/mol), implying a potential role in upregulating AGT expression and indirectly enhancing Ang II production. This could exacerbate vasoconstriction and oxidative stress, aligning with clinical observations of a positive association between aspartame consumption and the risk of hypertension. This study innovatively integrates network toxicology with molecular docking techniques to systematically elucidate how aspartame may influence the progression of ischemic stroke through a multi-target interaction network, thereby addressing a significant gap in the research on the neurotoxic mechanisms of food additives. Nonetheless, certain limitations must be acknowledged. First, target prediction is dependent on algorithmic databases, and the in vivo relevance of these targets requires validation in animal models. Second, the molecular docking analysis did not account for the effects of metabolites or the cumulative impact of long-term exposure. Finally, the absence of clinical cohort data limits the ability to support the predicted pathways with epidemiological correlations. Future research should focus on developing transgenic animal models for chronic aspartame exposure to dynamically monitor blood–brain barrier permeability and inflammatory cytokine expression, as well as conducting large-scale human studies to analyze the relationship between aspartame intake and stroke incidence, and to explore the mediating effects of biomarkers such as CX3CL1. Such efforts will deepen our understanding of the neurotoxicity of artificial sweeteners and provide a scientific basis for the formulation of precise food safety policies. Conclusion This study systematically elucidated the potential molecular associations between the artificial sweetener aspartame and ischemic stroke by integrating network toxicology with molecular docking techniques. Based on cross-validation across multiple databases, five core targets—IL1B, MMP9, SRC, AGT, and TNF—were identified, with functional enrichment in the renin–angiotensin system (RAS), complement and coagulation cascades, and inflammatory regulatory pathways. These findings suggest that aspartame may exacerbate ischemic brain injury by modulating vascular tone, coagulation balance, and neuroinflammation. Molecular docking analyses further confirmed a high-affinity binding between aspartame and the aforementioned targets, indicating its potential to directly interfere with target protein conformation or signal transduction. Notably, this study is the first to propose that aspartame may activate the RAS by upregulating AGT expression, thereby inducing oxidative stress and endothelial dysfunction. This mechanism aligns with epidemiological evidence linking artificial sweetener consumption to hypertension, offering a novel perspective on its cerebrovascular toxicity. The results provide a theoretical basis for assessing the neurotoxicity of aspartame and serve as a methodological reference for the re-evaluation of food additive safety. Future research should validate target functions using animal models, delineate the synergistic effects of metabolites, and combine epidemiological cohort studies to clarify the dose–response relationship, thereby advancing the refinement of food safety guidelines and the optimization of stroke prevention strategies. Electronic supplementary material Below is the link to the electronic supplementary material. [161]Supplementary Material 1^ (13KB, xlsx) [162]Supplementary Material 2^ (93KB, xlsx) Acknowledgements