Tenascin C promotes the formation of abdominal aortic aneurysm by regulating the homeostasis of vascular smooth muscle cells Authors: Xiaoyong Xiao, Jie Xiong, Yixuan Bao, Chixiong Huang, Guona Chen, HongYun Lu, CongMing Yang, Mengsha Li, Weiling Lu, Haoyu Song, Beiyou Lin, Yanxian Lai, Hongbin Liang, Lintao Zhong Published: 2026 ABSTRACT BackgroundAbdominal aortic aneurysm (AAA) is the most common type of arterial dilation disease. Once ruptured and bleeding, AAA can be life-threatening, with a mortality rate as high as 90%. Vascular smooth muscle cells(VSMCs) dyshomeostasis is the main pathological factor for AAA formation, but the specific regulatory process remains unclear.MethodsFirstly, we employed single-cell RNA sequencing (scRNA-seq) to reveal the heterogeneity and intercellular communication network of VSMCs in the aorta of AAA patients. We further integrated the data from the public database to clarify differences of the target signal in AAA, and also collected clinical data to analyze both differences and predictive performance for AAA. Subsequently, we generated VSMCs-specific knockout mice in the PPE-induced AAA mouse model to observe formation of AAA and VSMCs dyshomeostasis. Finally, we combined methods such as immunoprecipitation, bioinformatics, western blotting, qPCR, transwell co-culture and luciferase reporter assays to clarify the specific molecular mechanisms underlying VSMCs dyshomeostasis.ResultsThrough multi-level analysis and experiments, we confirmed that the Tenascin C (TNC) communication signal was significantly highly expressed in the arterial tissue and plasma of AAA, and elevated TNC levels in plasma could predict the occurrence of human AAA. In mouse models, we found that VSMCs-specific TNC knockout could inhibit PPE-induced AAA formation and VSMCs dyshomeostasis. Mechanistically, we revealed that TNC promoted nuclear translocation of NF-κB p50 and p65 through TLR4, thereby causing VSMCs dyshomeostasis, while the endoplasmic reticulum stress factor CHOP could induce TNC production in VSMCs.ConclusionOur research demonstrates the existence of a TNC-centered signaling axis mediating VSMCs dyshomeostasis in the formation of AAA.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s13062-026-00878-1. Introduction Abdominal aortic aneurysm (AAA) refers to the persistent tumor-like dilation of the abdominal aorta, with the tumor diameter increasing by more than 50% compared to the original vessel diameter [1, 2]. As the disease progresses, AAA can gradually increase in size, and the risk of rupture also rises accordingly. AAA rupture has an extremely high fatality rate, more than 90% of patients will lose their lives with in a short period of time once it suddenly ruptures, which is one of the most dangerous critical cardiovascular emergencies in clinical practice [3]. Therefore, further in-depth exploration of the endogenous pathogenesis of AAA and the search for appropriate therapeutic targets are of vital importance for the early prevention and intervention of AAA. Research reports, vascular smooth muscle cells (VSMCs) dyshomeostasis (such as phenotypic switching, oxidative stress, inflammation, etc.) is the main initiating factor for the formation of AAA [4–6]. However, VSMCs exhibit high heterogeneity in AAA, and the functional differentiation and regulatory networks of their specific subgroups remain unclear, which has become a key bottleneck in the research of limiting mechanisms. In recent years, the breakthrough of single-cell RNA sequencing (scRNA-seq) technology has provided a new tool for analyzing cellular heterogeneity in AAA [7–10]. The scRNA-seq research based on the aortic tissue of aortic disease patients showed that VSMCs in the lesion vessels presented significant heterogeneity and could be classified into multiple subgroups, including contractile, proliferative, inflammatory, fibroblast, and lipid type [11–14]. Given the significant role of scRNA-seq in revealing endogenous VSMCs lesions in AAA, we conducted scRNA-seq on human aortic samples. We analyzed the samples and found a group of VSMCs cell types with a significantly increased proportion in AAA. Through further in-depth analysis of their functional characteristics and differentiation features, we hypothesized that they might have a pathogenic effect on AAA. Subsequently, we conducted cell communication analysis on each VSMCs cell type and found that there was significant cell communication among the VSMCs cell types. We also discovered that TNC was the main interaction signal between the pathogenic VSMCs with an increased proportion in AAA and the contractile VSMCs that maintain homeostasis. Research has reported that tendon protein C (tenascin C, TNC) is composed of six polymer chain six polymers, mainly comes from the smooth muscle cell, fibroblast cell and macrophages. It participates in the early embryonic development of cardiac muscle and blood vessels and is hardly expressed in the heart and blood vessels of healthy adults [15–20]. Research has also found that the expression of TNC is significantly increased in cardiovascular diseases such as aortic dissection, myocardial infarction, hypertrophic cardiomyopathy, and coronary atherosclerosis, and it can also predict the final outcome [21–24]. Remarkably, existing clinical studies have found that the diameter of aortic aneurysms is positively correlated with the expression level of TNC [25, 26], suggesting that it may play an important role in the progression of AAA. We further collected plasma specimens from patients in cardiovascular research center and found that the level of TNC was significantly elevated in AAA patients, which had a significant diagnostic performance for AAA. Although TNC has been reported in AAA and aortic dissection, its cellular source, precise target cell type (contractile VSMCs), and downstream signaling cascade in the context of intercellular communication during AAA progression remain largely uncharacterized. The present study first delineates a complete cell communication-TNC-contractile VSMCs dysfunction axis that drives VSMCs homeostasis and AAA pathogenesis, thereby providing a novel mechanistic insight beyond previously reported correlations between TNC and AAA. Finally, we experimentally confirmed TNC in vivo and in vitro can promote VSMCs dyshomeostasis, which manifested as the expression of classic contraction genes decreases and the expression of inflammatory factors increases, and ultimately leading to the occurrence and progress of AAA. We also find the upper reaches of the endoplasmic reticulum stress factor CHOP and downstream TLR4/NF-κB inflammatory pathways play a crucial mediating role in the process of TNC-mediated maintenance of VSMCs homeostasis. This study reveals the core role of the “endoplasmic reticulum stress factor CHOP→ TNC transcription → TLR4/NF-κB inflammatory pathway” cascade regulatory network in VSMCs in AAA. It not only provides a new perspective for understanding the mechanism by which the heterogeneity of VSMCs drives AAA progression, but also lays a theoretical foundation for the precise treatment strategy of AAA targeting TNC. Materials and methods Detailed information can be found in the supplementary materials. Human specimens Three AAA and three healthy control (two donors and one recipient) samples used for scRNAseq analysis were obtained from surgical operations of patients with AAA in the Department of Surgery, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, China. The clinical information for scRNA-seq specimens is shown in Supplemental Table 1. The plasma specimens of the experimental group were derived from 40 consecutive AAA patients enrolled in the study. Forty healthy controls were recruited from the healthy population and were all matched with the experimental group in terms of age and gender. In accordance with the requirements of the Declaration of Helsinki, all specimens were collected under a protocol approved by the Ethics Committee of Guangdong Provincial People’s Hospital (Ethical Approval Number: KY-Z-20210219-02). All participants successfully and fully met the inclusion criteria assessment without triggering any exclusion criteria, including the presence of cardiovascular diseases (such as myocardial infarction, aortic trauma/dissection, pulmonary embolism, etc.), acute or chronic infectious diseases, history of immune system diseases, chronic kidney disease, active cancer, and previous aortic surgery, etc. A comprehensive dataset (Supplemental Table 2) variable containing demographic data(gender, age), cardiovascular risk factors (hypertension, diabetes), and BMI were systematically extracted from the electronic health records of the study population. Blood samples were collected with EDTA as an anticoagulant, followed by centrifugation for 10 min at 800 g. Plasma samples were cryopreserved in aliquots of 500 µL using cryovials (Thermo Scientific) at − 80 ℃(± 2 ℃ deviation) until biochemical analysis. Tissue dissociation and preparation of single-cell suspension The scRNAseq tissue was cut into vascular rings under a stereoscope. Then, the vascular rings were transferred into EP tubes and were washed with DPBS. For tissue dissociation, Pierce™ Primary Cardiomyocyte Isolation Kit (Thermo Fisher Scientifific™, Cat. No. 88281) was used following the manufacturer’s instructions [27]. Afterward, we performed blind shear and bathed it at 37 °C for 25 min. After the digestion process, 400–500 µl DPBS was added to the mixture, pipet-mixed gently, and then fifiltered with a 40-µm fifilter to collect the dissociated cells. Thereafter, the mixture was centrifuged for 5 min (4 °C, 400 × g). The supernatant was discarded and a 200 µl cold Sample Buffer (BD R added hapsody™, Cat. No. 650000062) was added to the cell mass. Finally, the mixture was pipe mixed gently to obtain the cell suspension for single-cell experiments. Cell capture, scRNA-seq library preparation, and sequencing For cell capture and library preparation for scRNA-seq, the BD Rhapsody™ system (BD Biosciences, USA) was applied based on the manufacturer’s protocols [28]. In brief, 1 µl Calcein AM (2 mM; Thermo Fisher Scientifific, USA, Cat. No. C1430) and 1 µL DRAQ7 (0.3 mM; Thermo Fisher Scientifific, USA, Cat. No. 564904) were added to 200 µL cell suspension (1:200 dilution). Then, the mixture was pipet-mixed gently and incubated at 37 °C in dark for 5 min. The cell viability and concentration of the suspension were measured by Hemocytometer Adapter (Cat. No. 633703). The single-cell suspension was then loaded onto a BD Rhapsody cartridge (Cat. No. 400000847) with > 200,000 microwells. Thereafter, cell capture beads were loaded into the microwells and washed to make sure that one magnetic bead binds to only one cell in a single microwell. Subsequently, the lysis mix was added and incubated at room temperature (15–25 °C) for 2 min. The cell capture beads were eventually retrieved for subsequent complementary DNA synthesis, exonuclease I digestion, and multiplex PCR-based library construction. Sequencing libraries were prepared using Whole Transcriptome Analysis Index PCR and the PCR product was purifified to enrich the 3′ end of the transcripts linked with the cell label and molecular indices. Finally, quality checks of the indexed libraries were performed with a Qubit Fluorometer using the Qubit dsDNA HS Assay. The sequencing of the libraries was conducted on an Illumina NovaSeq 6000 system. ScRNA-seq data preprocessing and analysing The raw sequencing reads were trimmed to keep the fifirst 75 bases. Then the trimmed reads were quality fifiltered using fastp. The BD Rhapsody whole transcriptome offificial analysis pipeline was applied under the default settings to obtain a cell-gene expression matrix and a quality control report for each sample [29]. The human reference genome was used for read alignment. The expression matrix was imported into Seurat for further preprocessing. We fifiltered out genes with counts in fewer than three cells to eliminate genes that were most likely discovered due to random noise. We fifiltered cell outliers (< fifirst quartile-1.5 interquartile range or > third quartile + 1.5 inter quartile range) based on the proportion of mitochondrial genes, the number of expressed genes, and the sum of unique molecular identififier (UMI) counts. To eliminate doublets, cells were fifiltered out using Scrublet. Furthermore, cells enriched in hemoglobin gene expression were also excluded. We normalized the UMI count sum of each cell to 10,000 and then log-transformed the data. To identify shared cellular states across samples, all the datasets were integrated via canonical correlation analysis (CCA) implemented in Seurat [30]. Principal component analysis (PCA) was used to reduce the dimensionality of scRNAseq expression data. A neighborhood graph of the cells was generated based on the fifirst 30 PCA components. Using UMAP or TSNE, we embedded the neighborhood graph in two-dimensional space. Using the likelihood-ratio test implemented in the function “FindMarkers” of Seurat, we detected differentially expressed genes between two groups of cells. The signifificance threshold was set to a log2-fold change > 0.25 and adjusted P value < 0.05. GO was used to perform functional enrichment analysis on a gene list with a Bonferroni-corrected P value < 0.05. In R software, monocle package was used for pseudo-temporal analysis, and cellchat was used for intercellular communication analysis. Experimental animals Healthy male C57BL/6J mice, aged 8 to 12 weeks, were sourced from Southern Medical University. All mice were housed under pathogen free conditions featuring a 12 h dark/light cycle with fuorescent lighting, at a humidity level around 55%, and a temperature range of 18–22 C. The mice had unrestricted access to food and purifed water. Conditional TNC knockout mice generation: gRNA to target the TNC gene, a donor vector containing loxP sites, and Cas9 mRNA were coinjected into fertilized mouse eggs to generate conditional TNC knockout offspring. F0 founder mice, which were bred to wild-type (WT) mice to test germline transmission and F1 animal generation, were identified via PCR followed by sequence analysis. An F1-targeted mouse was subsequently crossed with a tissue-specific Tagln-M-Cre deletion mouse to generate F2. Then, heterozygous Cre+ mice were bred with homozygous mice to generate F3. The genotypes of the knockout mice were as follows: primers 1, F2: 5’-GATGTTGAGAGTGGACCTGACTAC-3’, R2: 5’-CATTAAACCCACCAGGT CTTCTCA-3’; primers 2, Tagln-M-F, 5’-CAGACACCGAAGCTACTCTCCTT CC-3’; and Tagln-M-R, 5’-CGCATAACCAGTGAAACAGCATTGC-3’. The tissue-specific gene deletion can be confirmed by adding one additional primer to the PCR assay: primer 3: F3: 5’-GGCCTGTAAACTACAAGTCCAT-3’, R3: 5’-GATGTTGCAAGCTCCTATACGTTG-3’. The experiments conducted on animals were approved by the Chinese Academy of Basic Medical Sciences and the Animal Care and Use Committee of Southern Medical University and abided by the guidelines of the Animal Research Advisory Committee of the National Institutes of Health. Porcine pancreatic elastase (PPE)-induced AAA models The PPE-induced AAA models was established according to previous methods [31, 32]. Briefy, each mouse was anes-thetized through intraperitoneal injection of a mixture of xylazine (5 mg/kg) and ketamine (100 mg/kg). Adequate anesthesia was confrmed through the disappearance of the pedal withdrawal refex. A midline abdominal incision was made to fully expose the infrarenal abdominal aorta up to the bifurcation. The sponge soaked with PPE (0.2 U) was covered onto the abdominal aorta for 5 min. The control group mice received saline in sham operations. The residual PPE in the abdominal cavity was washed with physiological saline, the incision was sutured. The mice were euthanized after 7 days or 14 days of feeding with normal chow for the next experiment. Angiotensin II (Ang II) Infusion AAA Model To construct hyperlipidemic mice, eight-week-old male C57Bl/6J mice were intravenously injected via the tail vein with a single dose of 1 × 10¹¹ vg AAV8-PCSK9 particles (Weizhen Biotechnology) and fed a high-fat diet containing 0.15% cholesterol (HFD-C; 60% kcal fat, 26% kcal carbohydrate, Bio-Serv, F3282). After four weeks of hyperlipidemia induction, the Ang II-induced AAA model was established as previously described [14]. Briefly, the mice were anesthetized with 2% isoflurane. An osmotic minipump loaded with Ang II or normal saline was subcutaneously implanted through a small incision in the dorsum of the neck. Following 28 days of continuous infusion, the mice were euthanized, and aortic tissues were harvested for subsequent experimental analyses. Quantifcation of aneurysm formation Initially, the mice were perfused with 10% formaldehyde at physiological pressure via a pinhole on the left ventricle. Subsequently, the aorta was extracted and photographed using a digital camera to record changes in its diameter. The maximum diameter of the dilated segment of the infrarenal vessel was measured with Image-Pro Plus software (Media Cybernetics, USA). Based on the defnition of human aneurysm established in previous studies aneurysm forma-tion was defined by at least a 50% increase in infrarenal aortic diameter in the PPE-induced mice compared with the saline-treated mice. In cases where mice died during the experiment, necropsies were conducted. AAA were assessed by two investigators blinded to the experimental treatment, with at least thrice for accuracy. Elastin staining and degradation Aortic samples from different groups of mice were embedded in paraffin, dewaxed, hydrated, and then subjected to Elastica Van Gieson(EVG) staining using a commercial kit. Elastin degradation was assessed at 40x magnification. Severity was assessed according to the following previously established elastin degradation scoring criteria: (1) inelastic protein degradation and a good elastin layer; (2) mild degradation of elastin and some laminar disruption or rupture; (3) moderate elastin degradation and multiple interruptions or breaks in the lamina; (4) severe elastin breakdown or loss or aortic rupture. Cell culture Abdominal aortic aneurysm tissue was harvested from PPE-induced AAA mouse models. Samples of control aortic tissue was collected from healthy group of mouse. Mouse VSMCs used in this study were prepared as described in our previous study [33, 34]. Briefly, after cleaning adipose tissue and washing with PBS, the medial tissue was dissected from the adventitia and intima. Next, the media was cut into 1–2 mm3 pieces and carefully transferred to 25cm2 culture flask and incubated for adhesion at 37 °C for 1 h. After the tissue attaching to the culture flask, the medial tissues were gently cultured with Dulbecco’s modified Eagle medium (DMEM; Gibco) supplemented with 15% fetal bovine serum (FBS; Gibco) and 100 µg/mL penicillin and streptomycin (P/S, Thermo Fisher Scientific). The medium changed approximately every 3 to 4 days. VSMCs migrated out from the medial tissues within 1–2 weeks. The pieces were then removed and the cells were regularly collected and passaged. All VSMCs at passage 2–4 were used in this study. Western blotting assay The pretreated mouse aortic tissue was placed on ice, and then, the protein was extracted using RIPA bufer [consist-ing of Tris–HCl (25 mm, pH 7.6), NaCl (150 mm), NP-40 (1%), sodium deoxycholate (1%), and SDS (0.1%)]. The protein extracts were segregated on SDS‒PAGE gels and transferred to polyvinylidene fluoride (PVDF) blotting membranes. Initially, the membranes were incubated withprimary antibodies overnight at 4 C and secondary anti-bodies for 2 h at room temperature. Enhanced chemilumi-nescence reagent (RPN2235; GE-Healthcare Life Sciences, UK) was used for exposure imaging, which was repeated at least 5 times. Total protein was quantifed using ImageJ software (National Institutes of Health, USA), with β-actin or Tublin serving as the endogenous normalization controls. The antibody information is shown in Supplemental Table 3. RNA extraction and qPCR Total RNA from cells or tissues was extracted using TRIzol reagent (Invitrogen, Thermo Fisher Scientifc, USA) and its concentration was determined using a NanoDrop Lite spec-trophotometer (Thermo Fisher Scientifc, USA). With the SYBR Premix Ex Taq Kit (TaKaRa Biotechnology, China), quantitative polymerase chain reaction (qPCR) was carried out on a QuantStudio 5 real-time fuorescent qPCR system (Applied Biosystems, Thermo Fisher Scientifc, USA). The relative mRNA expression was measured by normalizing the internal reference (β-actin in this study). The primer sequences are listed in Supplemental Table 4. Histology and immunohistochemistry Murine aortas were fixed in 4% formalin and embedded in paraffin according to standard protocols. Hematoxylin and eosin (H&E) staining was performed for morphological assessment. For immunohistochemistry, anti-MCP1 antibody (1:1500; ab25124, UK), anti-IL6 antibody (1:50; 21865-1-AP, CHINA), and anti-MMP2 antibody (1:1000; ab97779, UK). Paraffin sections of murine aortas were dewaxed and hydrated, and endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 30 min. The sections were incubated with 5% bovine serum for 1 h at room temperature to block nonspecific binding sites followed by primary antibody at 4 °C overnight and secondary antibody (1:500; Cell Science, USA) for 60 min. All specimens were stained with DAB and hematoxylin staining solution. A minimum of three microscopic fields of stained slides were randomly observed by two independent researchers who were unaware of the group information. Information about antibodies is available in the data Supplemental Table 5. Immunohistochemical staining protein quantification Protein quantification was determined using IPP software (MEDIA CYBERNETICS, USA), based on the integral optical density (IOD) value, the sum of the chroma and the acreage of positive color (claybank). The mean IOD was calculated as the IOD value divided by the acreage of the samples. Each specimen had 3 lower power sections and 9 higher power sections. The mean IOD was the mean value of the IOD value in the 9 higher power sections. Molecular docking analysis First, the sequences of TNC_Mouse and TLR4_Mouse were submitted to the AlphaFold3 server. The sequence length of the TNC_Mouse protein is 2110, and that of the TLR4_Mouse protein is 835. Through the calculation of the AlphaFold3 server, we obtained the three-dimensional structures of the TNC_Mouse and TLR4_Mouse proteins [35]. We then used the hdock software to conduct binding prediction analysis on the TNC_Mouse and TLR4_Mouse proteins [36]. Finally, we used Pymol to analyze the binding sites of the TNC_Mouse and TLR4_Mouse proteins [37]. The molecular dynamics simulation of the protein-protein complex was carried out using GROMACS [38, 39]. The CHARMM36m force field was used for parameterization of the protein atoms. The complex was placed in a triclinic box (dodecahedron), with the minimum distance between the protein and the box boundary set to 1.0 nm, and the system was solvated with the TIP3P water model. Appropriate amounts of Na⁺/Cl⁻ ions were added to the system to neutralize the charge and simulate physiological salt concentration (0.15 M). The system was first subjected to energy minimization using the steepest descent method until the maximum force was less than 1000 kJ·mol⁻¹·nm⁻¹. Then, a 100 ps equilibration simulation was conducted under the NVT ensemble, maintaining the system temperature at 300 K through the V-rescale thermostat. Subsequently, a 100 ps equilibration simulation was performed under the NPT ensemble, maintaining the system pressure at 1 bar using the Parrinello–Rahman pressure coupling algorithm. Position restraints were applied to the heavy atoms of the protein during the equilibration process. The production simulation was carried out without position restraints for 10 ns with a time step of 2 fs. Long-range electrostatic interactions were treated using the Particle Mesh Ewald (PME) method, and the van der Waals interaction cutoff distance was set to 1.2 nm. All hydrogen bonds were constrained using the LINCS algorithm. The trajectory data were used for subsequent structural stability and interaction analysis. Co-immunoprecipitation and mass spectrometry analysis To identify TNC interacting proteins and verify their binding specificity, co-immunoprecipitation (Co-IP) combined with mass spectrometry (MS) was performed. Cells stably overexpressing Flag-tagged TNC (Flag-TNC) were cultured to 80%-90% confluence, lysed on ice for 30 min with RIPA lysis buffer (supplemented with protease and phosphatase inhibitors to prevent protein degradation/dephosphorylation), and centrifuged at 12,000 rpm for 15 min at 4 °C to collect the total protein supernatant.Next, 50 µL of Flag antibody-conjugated agarose beads were added to the supernatant and incubated overnight at 4 °C with gentle rotation. The beads were washed 5 times with cold PBS (pH 7.4, 0.1% Triton X-100) to remove non-specific binding. Immune complexes were eluted by heating with 2×SDS loading buffer at 95 °C for 10 min. Western blot was performed to verify the binding between TNC and TLR4, and silver staining was used for preliminary confirmation.Target silver-stained bands were excised, digested with trypsin, and analyzed by LC-MS/MS to further confirm TNC-TLR4 interaction. Luciferase reporter assays The “reporter plasmid” and “effector plasmid” were co-transfected into vascular smooth muscle cells, and the luciferase activity was detected. The reporter plasmids: pGL3-TNC-WT is a luciferase reporter plasmid containing the wild-type TNC gene promoter fragment. pGL3-TNC-MUT1, MUT2, and MUT12 are reporter plasmids that have undergone mutations at the predicted Chop binding site 1, site 2, or both sites simultaneously. The effector plasmid: pcDNA3.1-chop is a plasmid for overexpressing the Chop protein, and pcDNA3.1-NC is an empty negative control plasmid. The luciferase activity for each group was then evaluated using a dual-luciferase reporter assay system (Promega, USA). Statistical analysis All data are presented as the mean and standard error or median and interquartile range (the latter was used for the elastin degradation score). The data were analyzed using GraphPad Prism 7.0 and SPSS 27.0. Normality tests were assessed by Shapiro-Wilk statistics. When the data were normally distributed, Student’s t-tests were applied to determine statistical significance between two groups. Differences among groups were determined by one-way ANOVA, followed by Least significant difference post hoc test or Dunnett’s test. If a normal distribution could not be confirmed, nonparametric tests were applied using the Mann-Whitney U-test for two independent groups or Kruskal-Walli’s test with post hoc Dunn’s multiple comparisons test for ≥ 3 groups.The diagnostic performance of TNC in AAA was assessed using thereceiver operating characteristic (ROC) area under thecurve(AUC). Univariable and multivariable logistic regression analysis was used to evaluate the association of TNC and base line characteristics with AAA with results presented as the odds ratio with 95% CI. Fisher exact test was used to analyze the aneurysm incidence. p < 0.05 was considered statistically significant. Results ScRNA-seq reveals TNC as a key signaling molecule in VSMCs communication networks We collected aortic tissue samples from three healthy controls and three AAA patients for scRNA-seq analysis, with data uploaded to GEO database (GSE226492; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi). After rigorous quality control and batch effect correction, the sequencing library contained 32,487 high-quality cells. Unsupervised clustering and marker‑based annotation identified 10 distinct cell populations (Supplemental Fig. 1A–D). Notably, the proportion of VSMCs, a critical component of aortic tissue, showed the most significant variation in AAA patients (Supplemental Fig. 1, E.F). To dissect VSMC heterogeneity and function in AAA, we conducted detailed celltype analysis and identified 8 major celltypes of VSMCs (Fig. 1, A). Based on differential gene expression, these were classified into two fibroblast‑like subtypes (F‑SMC1/2, marked by DCN), two contractile subtypes (C‑SMC1/2, marked by ACTA2/α‑SMA), adipocyte‑like, macrophage‑like, and T‑cell‑like VSMCs, as well as a unique subtype. C‑SMC2 highly expressed contractile genes; adipocyte‑like VSMCs were enriched in APOD; macrophage‑like VSMCs expressed CD74; and T‑cell‑like VSMCs showed elevated CDH19 (Fig. 1B). The proportions and cell numbers of seven VSMC subtypes were reduced in AAA, whereas one subtype was markedly expanded in AAA tissues compared with controls. We collectively termed this expanded population “increased SMC subtype in AAA (I‑SMC)” (Fig. 1C–F). Fig. 1Single-cell RNA sequencing revealed a pathogenic VSMCs celltype in human AAA aortic tissue. (A) UMAP visualization demonstrated eight VSMCs celltype identified through scRNAseq in aortic tissue. (B-D) Comparative analysis of eight VSMCs subtype’ proportional distribution between AAA group and control group. (E) Average heatmap revealed differential genes across eight VSMCs celltype. (F) UMAP visualization demonstrates the distribution of eight celltypes in the control and AAA groups, clearly showing that a specific cell type is exclusively present in the AAA group while being virtually absent in the control group. (G) Dotplot illustrates expression patterns of classical VSMCs contractile factors, inflammatory factors, matrix metalloproteinases, and aging regulatory factors in both control and AAA groups. (H) GO enrichment analysis of the VSMCs subtype with increased proportion in AAA. (I) The distribution of time fulcrums of each celltype was analyzed by pseudotime analysis.(J) Pseudotime analysis of celltype temporal pivot distribution revealed seven distinct states of VSMCs subtype. (K) Violin graph shows the approximate temporal positions of the seven states of VSMCs. (L) The bar chart shows the distribution ratio of eight VSMCs celltypes in different pseudotime states. (M) Pseudotime analysis further analyzed density distribution of eight VSMCs celltypes. (N) Violin graph visualization presented temporal positions of eight VSMC celltypes Single-cell RNA sequencing revealed a pathogenic VSMCs celltype in human AAA aortic tissue. (A) UMAP visualization demonstrated eight VSMCs celltype identified through scRNAseq in aortic tissue. (B-D) Comparative analysis of eight VSMCs subtype’ proportional distribution between AAA group and control group. (E) Average heatmap revealed differential genes across eight VSMCs celltype. (F) UMAP visualization demonstrates the distribution of eight celltypes in the control and AAA groups, clearly showing that a specific cell type is exclusively present in the AAA group while being virtually absent in the control group. (G) Dotplot illustrates expression patterns of classical VSMCs contractile factors, inflammatory factors, matrix metalloproteinases, and aging regulatory factors in both control and AAA groups. (H) GO enrichment analysis of the VSMCs subtype with increased proportion in AAA. (I) The distribution of time fulcrums of each celltype was analyzed by pseudotime analysis.(J) Pseudotime analysis of celltype temporal pivot distribution revealed seven distinct states of VSMCs subtype. (K) Violin graph shows the approximate temporal positions of the seven states of VSMCs. (L) The bar chart shows the distribution ratio of eight VSMCs celltypes in different pseudotime states. (M) Pseudotime analysis further analyzed density distribution of eight VSMCs celltypes. (N) Violin graph visualization presented temporal positions of eight VSMC celltypes To further clarify the functional characteristics of this I-SMC subtype, we first performed differential gene analysis revealing that the I-SMC subtype highly expressed previously recognized key factors for SMC pathological malignancy, including aging-related factor TP53, inflammatory-related factor IL6, and matrix metalloproteinase MMP9. In contrast, the expression of contractile marker genes (ACTA2, CNN1, and MYL9) was significantly downregulated in this subtype. Subsequent GO enrichment analysis identified functional characteristics including chromatin partitioning, inflammatory regulation, and extracellular gene remodeling (Fig. 1, H). Additionally, pseudo-sequence trajectory analysis showed that most cells in the I-SMC subtype were positioned at the most upstream stage of differentiation, indicating a low-differentiated celltype (Fig. 1, I-N). Therefore, we propose that this I-SMC may play a significant pathogenic driving role in the formation of AAA. Finally, we use the software package of cellchat to investigate intercellular communication in VSMCs and observed distinct signaling interactions between the I-SMC celltype and other celltypes (Supplemental Fig. 2A). We identified a TNC signaling pathway, in which the I-SMC celltype exerts an effect on C-SMC2, which is a classical contractile celltype (Fig. 2, A-D and Supplemental Fig. 2, B.C). A comprehensive literature review indicated that TNC has been widely reported to play a pivotal role in the progression of various cardiovascular diseases. Based on this evidence, we hypothesize that TNC may act as a key pathogenic factor in AAA, exerting its pathogenic effects by disrupting the homeostasis of VSMCs. Fig. 2TNC signaling regulates VSMCs homeostasis and may serve as a plasma biomarker for predicting AAA development. (A, B) Dotplot demonstrated extensive intercellular communication networks among eight VSMCs celltypes.(C, D) String diagrams and classification heatmaps demonstrate that pathogenic VSMCs celltype emit TNC signals affecting contractile VSMCs celltype. (E) Violin plots also show TNC is predominantly enriched in VSMCs of aortic tissue. (F) Dotplot reveal significant high expression of TNC in AAA aortic tissue. (G) Violin plots indicate TNC is significantly overexpressed in VSMC of AAA aortic tissue compared to control groups. (H) (Transcriptomic sequencing data from GEO public database (GSE285959) shows TNC’s significant overexpression in mouse AAA models. (I) Transcriptomic sequencing data from GEO public database (GSE241265) demonstrates TNC’s marked overexpression in human diseased aorta. (J) Imaging clearly shows the formation of human AAA (red-marked position). (K) Elisa analysis reveals significant TNC overexpression in AAA patients ‘plasma. (L) ROC curve analysis demonstrates TNC’s strong predictive power for AAA (AUC = 0.877, P < 0.001). (M) Forest plots illustrate the correlation between clinical baseline data and AAA, revealing TNC plasma levels as a critical risk factor. (***P < 0.01) TNC signaling regulates VSMCs homeostasis and may serve as a plasma biomarker for predicting AAA development. (A, B) Dotplot demonstrated extensive intercellular communication networks among eight VSMCs celltypes.(C, D) String diagrams and classification heatmaps demonstrate that pathogenic VSMCs celltype emit TNC signals affecting contractile VSMCs celltype. (E) Violin plots also show TNC is predominantly enriched in VSMCs of aortic tissue. (F) Dotplot reveal significant high expression of TNC in AAA aortic tissue. (G) Violin plots indicate TNC is significantly overexpressed in VSMC of AAA aortic tissue compared to control groups. (H) (Transcriptomic sequencing data from GEO public database (GSE285959) shows TNC’s significant overexpression in mouse AAA models. (I) Transcriptomic sequencing data from GEO public database (GSE241265) demonstrates TNC’s marked overexpression in human diseased aorta. (J) Imaging clearly shows the formation of human AAA (red-marked position). (K) Elisa analysis reveals significant TNC overexpression in AAA patients ‘plasma. (L) ROC curve analysis demonstrates TNC’s strong predictive power for AAA (AUC = 0.877, P < 0.001). (M) Forest plots illustrate the correlation between clinical baseline data and AAA, revealing TNC plasma levels as a critical risk factor. (***P < 0.01) TNC exhibits significantly high expression and strong predictive performance in AAA To clarify the specific differences of TNC in AAA, we first conducted differential analysis on aortic tissues of scRNAseq, revealing that TNC is significantly overexpressed in AAA aortic tissues, predominantly enriched in VSMCs (Fig. 2, E-G). Subsequently, we downloaded RNA-seq data from two public databases (GSE241265 and GSE285959) containing human and mouse aortic tissues samples from the GEO database, which confirmed TNC’s significant overexpression in both human and mouse diseased aortic tissues (Fig. 2, H.I). Additionally, beyond bioinformatics analysis, we collected 40 plasma samples each from patients with AAA and a healthy control group for ELISA detection of TNC. All these patients underwent enhanced computed tomography (CT) scans of the abdominal aorta, which showed that the aortas of AAA patients exhibited obvious dilation (Fig. 2, J). Statistical analysis showed that TNC was significantly overexpressed in AAA patients’ plasma (Fig. 2, K). The ROC curve analysis indicated that the AUC of TNC for predicting AAA was 0.877. When the cutoff value was 33.13 ng/mL, the sensitivity was 72.5% and the specificity was 90% (Fig. 2, L). The forest plot drawn based on the clinical baseline data of human samples (Supplemental Table 6) shows that TNC, hypertension and BMC are significantly correlated with the onset of AAA under the condition of controlling for equal age and gender, among which the difference of TNC is the most obvious (Fig. 2, M). Through these multi-level analyses and experiments, we demonstrated that TNC’s differential overexpression in AAA carries significant statistical significance. Specific TNC knockout in VSMCs can delay the formation of AAA and inhibit the imbalance of VSMCs homeostasis To investigate the in vivo function of TNC expressed in VSMCs and determine whether TNC deficiency affects AAA formation, we generated VSMCs-specific TNC-deficient mice by creating hybrid knockout mice with Tagln/SM22α-Cre transgenic mice (Supplemental Table 7. Supplemental Figs. 3,4). Subsequently, we established PPE-induced AAA models in both TNC-deficient mice(CKO) and TNCflox/flox control mice(WT, Wild Type). In vivo imaging analysis confirmed PPE-induced AAA formation, which was inhibited after inactivating PPE function. Notably, TNC-specific knockout in VSMCs suppressed PPE-induced AAA formation (Fig. 3, A). Further analysis revealed that VSMCs-specific TNC knockout resulted in reduced maximum abdominal aortic diameter in PPE-induced mice (Fig. 3, B.C). Remarkably, Hematoxylin-eosin (H&E) staining showed that knockout of TNC on VSMCs could significantly alleviate vascular injury induced by PPE (Fig. 3, D). EVG staining showed that VSMCs-specific TNC knockout resulted in decreased elastic fiber degradation scores in PPE-induced mice (Fig. 3, E.F). In addition, we observed that the protein expression levels of classic contractile factors(α-SMA, CNN1 and SM22α) in VSMCs decreased after PPE intervention. However, the expression of these contractile factors was reversed and significantly upregulated after VSMCs-specific TNC knockout (Fig. 3, G.H). Fig. 3Knockout of TNC in VSMCs delays the formation of AAA and sustains the contractile phenotype homeostasis of VSMCs. (A) Vivo imaging confirmed PPE-induced AAA formation, TNC-specific knockout in VSMCs suppressed PPE-induced AAA formation (N = 6 in in-PPE group. N = 12 in PPE group). (B) VSMCs-specific TNC knockout resulted in reduced maximum abdominal aortic diameter in PPE-induced mice. (C) The intuitive images of mice aortic ultrasound. (D) H&E staining used to observe the histological panorama of the aorta which showed that knockout of TNC on VSMCs could significantly alleviate vascular injury induced by PPE (low power, 500 μm scale). (E) EVG staining used to observe the integrity of the aorta elastic fibers under high magnification which showed that VSMCs-specific TNC knockout resulted in decreased elastic fiber degradation scores in PPE-induced mice(50 μm scale). (F) Semi-quantitative scores (0–5 points, with higher scores indicating more severe degradation) were assigned to the pathological sections of elastin fibers in figure E for degradation, and the differences between groups were statistically analyzed(N = 6). (G, H) Western blot detection and quantitative analysis of VSMCs contractile markers(N = 4). (NS: no statistical significance. *P < 0.05) Knockout of TNC in VSMCs delays the formation of AAA and sustains the contractile phenotype homeostasis of VSMCs. (A) Vivo imaging confirmed PPE-induced AAA formation, TNC-specific knockout in VSMCs suppressed PPE-induced AAA formation (N = 6 in in-PPE group. N = 12 in PPE group). (B) VSMCs-specific TNC knockout resulted in reduced maximum abdominal aortic diameter in PPE-induced mice. (C) The intuitive images of mice aortic ultrasound. (D) H&E staining used to observe the histological panorama of the aorta which showed that knockout of TNC on VSMCs could significantly alleviate vascular injury induced by PPE (low power, 500 μm scale). (E) EVG staining used to observe the integrity of the aorta elastic fibers under high magnification which showed that VSMCs-specific TNC knockout resulted in decreased elastic fiber degradation scores in PPE-induced mice(50 μm scale). (F) Semi-quantitative scores (0–5 points, with higher scores indicating more severe degradation) were assigned to the pathological sections of elastin fibers in figure E for degradation, and the differences between groups were statistically analyzed(N = 6). (G, H) Western blot detection and quantitative analysis of VSMCs contractile markers(N = 4). (NS: no statistical significance. *P < 0.05) We simultaneously constructed Ang II-infused AAA models in WT and CKO mice, aiming to verify phenotype reliability and eliminate bias arising from the animal model itself. As expected, under AngII infusion, the incidence of AAA and maximal aortic diameter were markedly reduced in VSMCs-specific TNC CKO mice compared with WT controls (Supplemental Fig. 5, A–D). Histological staining showed alleviated elastic fiber degradation in the aortic wall of CKO mice (Supplemental Fig. 5, E–F). Western blot further demonstrated that the contractile phenotype markers of VSMCs (α-SMA, CNN1, SM22) were significantly upregulated in CKO mouse aortae (Supplemental Fig. 5, G–H). Collectively, VSMCs-specific TNC knockout mitigates AngII-induced AAA formation, attenuates elastin breakdown, and preserves the contractile phenotype of VSMCs. The consistent protective phenotypes across two independent AAA models strongly support the indispensable role of VSMCs-derived TNC in promoting AAA pathogenesis. To further clarify the impact of TNC on VSMCs homeostatic regulation, we also detected the expression of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2) in the aortic tissues of PPE-intervention and TNC knockout mouse models. We found that PPE stimulation enhanced the mRNA expression of inflammatory factors and MMP2 in VSMCs. Conversely, TNC knockout significantly suppressed the expression of these factors in VSMCs (Fig. 4, A-C). Furthermore, we confirmed the regulatory role of TNC in the protein expression of inflammatory factors and matrix metalloproteinases in VSMCs through immunohistochemical staining (Fig. 4, D-I) and Western blot analysis (Fig. 4, J.K). Fig. 4Knockout of TNC in VSMCs suppresses the expression of inflammatory factors and matrix metalloproteinases. (A-C) qPCR detected mRNA expression of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2) in the aortic tissues of PPE-intervention and TNC knockout mice models(N = 6). (D-F) Immunohistochemical staining detected expression of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2) in the aortic tissues of PPE-intervention and TNC knockout mice models. (G-I) Quantify the intensity of immunohistochemical staining by “optical density integral (IOD)/area“(N = 6). (J, K) Western blot detection and quantitative analysis of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2)(N = 4).(*P<0.05) Knockout of TNC in VSMCs suppresses the expression of inflammatory factors and matrix metalloproteinases. (A-C) qPCR detected mRNA expression of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2) in the aortic tissues of PPE-intervention and TNC knockout mice models(N = 6). (D-F) Immunohistochemical staining detected expression of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2) in the aortic tissues of PPE-intervention and TNC knockout mice models. (G-I) Quantify the intensity of immunohistochemical staining by “optical density integral (IOD)/area“(N = 6). (J, K) Western blot detection and quantitative analysis of inflammatory factors(including IL6 and MCP-1) and matrix metalloproteinases (MMP2)(N = 4).(*P<0.05) These results suggest that TNC knockout in VSMCs can alleviate the formation of AAA in mouse and inhibit the VSMCs dyshomeostasis. TLR4 is the downstream target protein of TNC in VSMCs To further investigate the specific pathological endogenous processes of VSMCs in TNC-deficient conditions, we conducted immunoprecipitation (IP) experiment analysis on VSMCs to identify TNC’s target proteins. The results revealed that TLR4(molecular weight: 95 kD) may be the downstream protein regulated by TNC in VSMCs (Fig. 5, A). Subsequently, we analyzed the transcriptome sequencing matrix data of mouse from the GEO public database (12 samples respectively in GSE285959), and found that the expressions of TNC and TLR4 were significantly correlated (R = 0.70, P = 0.01) in aorta tissues (Fig. 5, B). Furthermore, The MS/MS spectrum can clearly reflect the protein identification results (Fig. 5, C). Then, we used the hdock software to conduct a prediction analysis on the TNC_Mouse and TLR4_Mouse proteins. The results showed that the binding energy score of the two proteins was − 423.73, and the confidence score after calculation was approximately 0.98 (Fig. 5, D). Additionally, we use pymol to predict the binding sites of the two proteins revealed that they had multiple binding sites with each other (Fig. 5, E). Fig. 5The target proteins of TNC and the mechanism of its effect on the inflammatory response of VSMCs. (A) Western Blot was used to detect the capture of the target protein in the immunoprecipitation (IP) experiment. (B) The transcriptome sequencing matrix data of mice from the GEO public database (12 samples respectively in GSE285959) found that the expressions of TNC and TLR4 were significantly correlated (R = 0.70, P = 0.01) in aorta tissues. (C) The MS/MS spectrum clearly reflect the protein identification results. (D) We used the hdock software to conduct a prediction analysis on the TNC_Mouse and TLR4_Mouse proteins. (E) TNC_Mouse and TLR4_Mouse proteins had multiple binding sites with each other. (F, G) Western Blot was used to detect the subcellular localization and quantitative analysis of p50 and p65 in VSMCs when TNC is overexpressed(N = 3). (H, I) Western Blot was used to detect the subcellular localization and quantitative analysis of p50 and p65 in VSMCs when TNC is knockdown(N = 3).(*P<0.05) The target proteins of TNC and the mechanism of its effect on the inflammatory response of VSMCs. (A) Western Blot was used to detect the capture of the target protein in the immunoprecipitation (IP) experiment. (B) The transcriptome sequencing matrix data of mice from the GEO public database (12 samples respectively in GSE285959) found that the expressions of TNC and TLR4 were significantly correlated (R = 0.70, P = 0.01) in aorta tissues. (C) The MS/MS spectrum clearly reflect the protein identification results. (D) We used the hdock software to conduct a prediction analysis on the TNC_Mouse and TLR4_Mouse proteins. (E) TNC_Mouse and TLR4_Mouse proteins had multiple binding sites with each other. (F, G) Western Blot was used to detect the subcellular localization and quantitative analysis of p50 and p65 in VSMCs when TNC is overexpressed(N = 3). (H, I) Western Blot was used to detect the subcellular localization and quantitative analysis of p50 and p65 in VSMCs when TNC is knockdown(N = 3).(*P<0.05) Meanwhile, we also conducted molecular dynamics simulation analysis on the TNC_Mouse and TLR4_Mouse proteins, which showed that the stability and plasticity of the binding between them were very strong (Supplemental Fig. 5, A-C). The stability of the radius of gyration (Rg) also indicates that the overall conformation of the complex formed by the two is compact, with a uniform mass distribution and stable interactions (Supplemental Fig. 5, D-F). Furthermore, the SASA reflection system indicates that the interaction between TNC and TLR4 is persistent and relatively stable in an aqueous environment (Supplemental Fig. 5, G). Moreover, the 10,000 ps simulation process shows that there is a stable and specific recognition interface between TNC and TLR4, not just an accidental adhesion, but with a real and chemically credible binding driving force (Supplemental Fig. 5H). Finally, the 10 ns simulation process indicates that there is always a stable direct atomic contact between TNC and TLR4, with a high degree of interface adhesion and no sign of dissociation (Supplemental Fig. 5, I). Moreover, the interaction between the two groups is not a random pile-up, but is approaching a truly structurally reasonable binding mode (Supplemental Fig. 5, J). Finally, co-immunoprecipitation analysis was performed in VSMCs to verify the interaction between Flag-TNCand TLR4. TLR4 was co-precipitated together with Flag-TNC by anti-Flag antibody, while no specific binding was detected in the IgG control group, confirming the specific binding of Flag-TNC to TLR4 in VSMCs (Supplemental Fig. 7). Based on the above analysis, we can prove that the binding ability between the two proteins is very strong. TNC is likely to regulate the VSMCs dyshomeostasis through TLR4, thereby leading to the formation of AAA. However, the further endogenous process still needs to be deeply clarified. TNC promotes VSMCs dyshomeostasis through the TLR4-NF-κB pathway In recent years, numerous studies have reported that TLR4 mediates TNC’s regulatory effects on renal fibrosis and septic lung injury, with the NF-κB pathway playing a pivotal role in the TNC-TLR4 axis [20, 40]. To clarify whether TNC activates the NF-κB pathway via TLR4 in VSMCs to promote inflammatory responses, we cultured mouse aortic VSMCs to verify the role of a certain signal axis on VSMCs. We found that TNC stimulation significantly enhanced the nuclear translocation of p50 and p65, the core subunits of the NF-κB pathway, in VSMCs (Fig. 5, F.G). In contrast, silencing TNC expression in VSMCs reduced the nuclear translocation of p50 and p65, thereby inhibiting NF-κB pathway activation (Fig. 5, H.I). To further clarify the impact of TLR4-mediated TNC regulation on VSMCs homeostasis, we stimulated VSMCs with TNC while adding a TLR4 inhibitor (Fig. 6, A). We found that TNC stimulation significantly increased the protein expression levels of inflammatory factors and matrix metalloproteinases; however, the addition of the TLR4 inhibitor TAK-242 on this basis significantly suppressed the expression of these proteins (Fig. 6, B.C). Conversely, TNC stimulation significantly reduced the protein expression levels of classic VSMCs contraction factors, while the addition of TAK-242 abrogated this inhibitory effect (Fig. 6, D.E). Fig. 6TNC induces VSMCs homeostasis imbalance through TLR4. (A) Immunofluorescence staining (α-SMA+DAPI) showed that TNC stimulated VSMCs simultaneously and TLR4 inhibitors(TAK-242) were added. (B, C) Western Blot and quantitative analysis demonstrated that TNC could promote the expression of pro-inflammatory factors (MCP1, IL-6) and matrix degradation-related proteins (MMP2); TAK-242 could block this promoting effect of TNC on expression(N = 3). (D, E) Western Blot and quantitative analysis demonstrated that TNC can induce the up-regulation of contractile markers of VSMCs, TAK-242 can inhibit the differentiation of VSMCs induced by TNC(N = 3).(*P<0.05) TNC induces VSMCs homeostasis imbalance through TLR4. (A) Immunofluorescence staining (α-SMA+DAPI) showed that TNC stimulated VSMCs simultaneously and TLR4 inhibitors(TAK-242) were added. (B, C) Western Blot and quantitative analysis demonstrated that TNC could promote the expression of pro-inflammatory factors (MCP1, IL-6) and matrix degradation-related proteins (MMP2); TAK-242 could block this promoting effect of TNC on expression(N = 3). (D, E) Western Blot and quantitative analysis demonstrated that TNC can induce the up-regulation of contractile markers of VSMCs, TAK-242 can inhibit the differentiation of VSMCs induced by TNC(N = 3).(*P<0.05) Previous studies have repeatedly reported that angiotensin II (Ang II) is a classic pathway for inducing VSMCs dyshomeostasis and can be used to establish AAA cell models [41–43]. We utilized a Transwell co-culture system to investigate the regulatory effects of Ang II-induced stress on VSMCs dyshomeostasis and further determine whether pharmacological inhibition of TLR4 with TAK-242 could rescue such pathological changes (Fig. 7, A). We observed that the mRNA level of TNC was significantly upregulated in VSMCs co-cultured with AngII (Fig. 7, B). Consistent with above findings, VSMCs-specific TNC knockout markedly alleviated Ang II-induced AAA formation and VSMCs dyshomeostasis, demonstrating that TNC acts as a crucial effector mediating Ang II-regulated AAA progression and VSMCs homeostasis. Our experimental groups were precisely defined as control VSMCs (VSMC), Ang II-treated VSMCs with stress-induced dyshomeostasis (co-culture), and TAK-242-rescued VSMCs (co-culture + TAK-242). The core purpose of this grouping design was to verify two key scientific questions: (1) whether Ang II stimulation directly disrupts VSMCs homeostasis; and (2) whether blocking the TLR4/NF-κB signaling pathway via the specific inhibitor TAK-242 could effectively reverse Ang II-triggered VSMCs dysfunction. As shown in Fig. 7(C-F), western blotting results demonstrated that Ang II stimulation markedly disrupted VSMCs homeostasis, characterized by the downregulation of VSMCs contractile markers (α-SMA, SM22, CNN1) and the significant upregulation of pro-inflammatory and remodeling factors (MMP2, MCP1, IL-6). In contrast, TAK-242 intervention effectively reversed Ang II-induced VSMCs phenotypic switching and inflammatory activation, indicating that TLR4 signaling is required for Ang II-mediated VSMCs dyshomeostasis. Fig. 7Transwell co-culture system verifies that TLR4 inhibitor TAK-242 rescues Ang II-induced VSMC dysfunction and phenotypic transition. (A) Schematic diagram of the Transwell co-culture setup for VSMCs with Ang II treatment. (B) PCR quantification of TNC mRNA expression in monocultured VSMCs and co-cultured VSMCs stimulated with Ang II. (C) Representative western blot bands of remodeling factor MMP2 and inflammatory mediators MCP1, IL-6 across three groups. (D) Statistical quantification of relative protein levels for MMP2, MCP1 and IL-6. (E) Western blot detection of α-SMA, CNN1 and SM22; Tubulin was used as loading control. (F) Quantitative analysis of the relative expression of contractile markers. VSMC: monocultured control VSMCs; co-culture: AngII-treated co-cultured VSMCs; co-culture + TAK-242: AngII co-cultured VSMCs treated with TLR4 specific inhibitor TAK-242 *P<0.05 Transwell co-culture system verifies that TLR4 inhibitor TAK-242 rescues Ang II-induced VSMC dysfunction and phenotypic transition. (A) Schematic diagram of the Transwell co-culture setup for VSMCs with Ang II treatment. (B) PCR quantification of TNC mRNA expression in monocultured VSMCs and co-cultured VSMCs stimulated with Ang II. (C) Representative western blot bands of remodeling factor MMP2 and inflammatory mediators MCP1, IL-6 across three groups. (D) Statistical quantification of relative protein levels for MMP2, MCP1 and IL-6. (E) Western blot detection of α-SMA, CNN1 and SM22; Tubulin was used as loading control. (F) Quantitative analysis of the relative expression of contractile markers. VSMC: monocultured control VSMCs; co-culture: AngII-treated co-cultured VSMCs; co-culture + TAK-242: AngII co-cultured VSMCs treated with TLR4 specific inhibitor TAK-242 *P<0.05 These findings confirm that TNC activates the NF-κB pathway by inducing p50 and p65 nuclear translocation through paracrine act on TLR4, thereby causing VSMCs dyshomeostasis. The endoplasmic reticulum stress factor CHOP mediates the production of TNC, leading to VSMCs dyshomeostasis As a cytokine, the intracellular production of TNC necessarily involves upstream regulatory factors, but the mechanism by which TNC is produced in VSMCs remains unclear. Previous study has shown that the endoplasmic reticulum stress factor CHOP can mediate TNC transcription to activate inflammatory responses [20]. However, it remains unknown whether this process regulates the homeostasis of VSMCs in AAA. During our GO enrichment analysis of AAA’s TNC-positive VSMCs we identified significant endoplasmic reticulum stress signals (Fig. 8, A), indicating that TNC-positive VSMCs in AAA undergo endoplasmic reticulum stress, which may promote TNC production and secretion. Subsequently, our luciferase repoter assay demonstrated that in the pGL3-TNC-WT (wild-type promoter) group, compared with the negative control (NC), co-transfection with the CHOP overexpression plasmid significantly increased the luciferase activity (Fig. 8, B). This indicates that CHOP can strongly activate the TNC promoter. After any of the predicted CHOP binding sites (MUT1, MUT2) were mutated, the activating effect of Chop disappeared. In the double mutation (MUT12) group, CHOP was also unable to activate the promoter. These experimental results confirm that the activation of TNC transcription by CHOP depends on its binding to these two specific DNA sites on the TNC promoter. When the sites are mutated, CHOP cannot bind and thus cannot exert its transcriptional activation effect. Fig. 8The endoplasmic reticulum stress signaling factor CHOP mediates TNC transcription and thereby regulates the imbalance of VSMCs homeostasis. (A) GO enrichment analysis of AAA’s TNC-positive VSMCs we identified significant endoplasmic reticulum stress signals. (B) Luciferase repoter assay demonstrated that the activation of TNC transcription by Chop depends on its binding to these two specific DNA sites on the TNC promoter(N = 3). (C-E) The regulatory effects of CHOP and TNC on the mRNA expression of inflammatory factors and matrix metalloproteinases in VSMCs were investigated through different experimental groups (overexpression/knockdown of CHOP, overexpression/knockdown of TNC), and the relative mRNA levels of target genes were detected (N = 3). (F-H)The regulatory effects of CHOP and TNC on the mRNA expression of VSMCs contraction marker genes were detected by different group experiments (overexpression/knockdown of CHOP, overexpression/knockdown of TNC) to measure the relative mRNA levels of target genes (N = 3).(*P < 0.05) The endoplasmic reticulum stress signaling factor CHOP mediates TNC transcription and thereby regulates the imbalance of VSMCs homeostasis. (A) GO enrichment analysis of AAA’s TNC-positive VSMCs we identified significant endoplasmic reticulum stress signals. (B) Luciferase repoter assay demonstrated that the activation of TNC transcription by Chop depends on its binding to these two specific DNA sites on the TNC promoter(N = 3). (C-E) The regulatory effects of CHOP and TNC on the mRNA expression of inflammatory factors and matrix metalloproteinases in VSMCs were investigated through different experimental groups (overexpression/knockdown of CHOP, overexpression/knockdown of TNC), and the relative mRNA levels of target genes were detected (N = 3). (F-H)The regulatory effects of CHOP and TNC on the mRNA expression of VSMCs contraction marker genes were detected by different group experiments (overexpression/knockdown of CHOP, overexpression/knockdown of TNC) to measure the relative mRNA levels of target genes (N = 3).(*P < 0.05) To investigate the regulatory effects of CHOP and TNC on inflammatory factors (IL6, MCP-1), matrix metalloproteinase(MMP2) and contractile phenotype markers (α-SMA, CNN1, SM22) in VSMCs, we overexpressed or knocked down CHOP and simultaneously overexpressed or knocked down TNC, and observed the changes in mRNA expression of each gene (Fig. 8, C-H). We found that overexpression of CHOP promoted the expression of inflammatory factors, and this effect was dependent on TNC (knocking down TNC could weaken the promoting effect of CHOP). Additionally, CHOP deficiency (siCHOP) inhibited the expression of inflammatory factors; however, overexpression of TNC in the context of CHOP deficiency instead induced an increase in inflammatory factors (suggesting that TNC may replace the pro-inflammatory function of CHOP in the absence of CHOP). Meanwhile, overexpression of CHOP inhibited the expression of contractile phenotype markers (indicating that CHOP inhibits the contractile phenotype), and this inhibitory effect could be enhanced by TNC (knocking down TNC made the inhibitory effect of CHOP stronger). CHOP deficiency (siCHOP) promoted the expression of contractile phenotype markers; however, overexpression of TNC in the context of CHOP deficiency still inhibited the contractile phenotype markers (indicating that the inhibitory effect of TNC on the contractile phenotype of VSMCs is independent of CHOP). These findings demonstrate that CHOP in VSMCs regulates TNC transcription by binding to its promoter, thereby promoting TNC production and secretion, which ultimately leading to dyshomeostasis of VSMCs. Disscusion In this study, we conducted in-depth analysis of intercellular communication networks in VSMCs celltypes identified through scRNAseq of human aortic tissue. Our findings revealed that TNC signaling may influence the homeostasis of VSMCs contraction and inflammatory responses in human aortic tissue. Notably, significantly increased TNC levels were detected in plasma samples from AAA patients, demonstrating promising diagnostic potential. Subsequently, we established VSMCs specific TNC-knockout mouse model to elucidate the critical processes mediating AAA formation. Mechanistically, we discovered that CHOP, a key transcription factor in endoplasmic reticulum stress, regulates TNC transcription. TNC secretion induces VSMCs contraction homeostasis imbalance and inflammatory responses via the TLR4/NF-κB pathway. Collectively, our findings provide the first evidence of a TNC-centered signaling axis mediating VSMCs pathology in AAA development and progression. Early studies suggested that different celltypes of VSMCs play distinct roles in aortic aneurysm formation [44–46]. For instance, the reduction of contractile VSMCs and the increase of secretory/inflammatory VSMCs are closely associated with aneurysm development. Recent scRNAseq studies have provided critical insights into VSMCs subtype dynamics and functional changes during aneurysm progression, shedding light on its pathogenesis [12, 14]. However, since the application of scRNAseq in aneurysm research, multiple VSMCs classification definitions have emerged. As early as 2020, scholars had already identified several human aortic VSMCs celltypes, including contractile, stress, proliferative, and fibroblast-like celltypes [13]. Subsequent research expanded this classification to include macrophage-like, T-cell-like, and adipocyte-like celltypes [12, 47]. In angiotomy-induced mouse aneurysm models, scRNAseq also revealed multiple VSMCs clusters such as contractile, inflammatory, stress, and proliferative celltypes [8]. Nevertheless, there remains no unified standard for VSMCs classification at scRNAseq level. Our integrated scRNAseq dataset initially identified five recognized celltypes: contractile, fibroblast-like, macrophage-like, T-cell-like, and adipocyte-like. Notably, both contractile and fibroblast-like celltypes were further divided into distinct clusters, a pattern consistent with previous findings. However, our study did not define the uncertain VSMCs celltype with significantly increased proportions in AAA, but merely assigned a general name based on its distribution ratio before exploring its functional characteristics. We observed that this celltype highly expressed marker genes such as inflammation, aging, and matrix metalloproteinase, all of which have been confirmed to participate in AAA progression. Moreover, we also found this celltype is at the starting point of differentiation stage, indicates that it is a subtype of poorly differentiated cells. Previous studies have reported that low differentiation of VSMCs can lead to vascular wall dilation and extracellular matrix degradation, which led to the eventual formation of AAA [45, 48]. Therefore, we propose that this celltype may promote AAA progression and can be termed a pathogenic celltype. Cell-cell communication is a crucial process that profoundly influences the balance, development, and diseases within organisms [49, 50]. The dynamic communication network established through collaboration between cells plays a vital role in various biological processes [46, 51]. In recent years, multiple disease-related studies have reported the use of scRNAseq to analyze the dialogue between “pathogenic cell celltype” driving disease progression and “protective cell subpopulations” involving in tissue protection and repair—specifically, intercellular communication signal networks—which are key determinants of vascular disease progression. As early as 2021 [52], scholars conducted a study on the pathogenesis of hypertension and discovered that the cell communication between the aortic cells of mice with hypertension occurred frequently. Some signaling pathways were activated during hypertension, which might be the key signals causing hypertension. Another study revealed the pathogenesis of aortic dissection caused by hypertension and marfan syndrome [53]. CellChat identified the intercellular communication signals related to aortic dissection induced by hypertension and marfan syndrome, involving smooth muscle cell, fibroblasts, macrophages and T cell. A recent study has for the first time systematically revealed that the endothelial cell-derived signal BMP6 promotes osteoblastic differentiation of VSMCs through exosome form during carotid atherosclerosis, which provides an important reference for enriching the understanding of the VSMCs dyshomeostasis in vascular diseases [54]. Our scRNAseq communication analysis of VSMCs from human aortic revealed that TNC serves as a critical mediator of VSMCs homeostasis, undoubtedly providing richer insights into VSMCs heterogeneity changes during AAA progression. Previous studies have reported that TNC, as a biomarker, not only demonstrates excellent performance in diagnosing various diseases but can even accurately predict the occurrence of cardiovascular adverse events, making it clinically significant in cardiovascular disease research. Studies have shown that the level of TNC in the circulatory system is significantly associated with adverse prognosis and mortality in diseases such as myocardial infarction, dilated cardiomyopathy, pulmonary vascular disease and heart failure [55–58]. For instance, studies had long ago discovered that TNC expression shows a positive correlation with pulmonary artery pressure during the progression of pulmonary hypertension [59, 60]. Sato et al. found that serum TNC levels in patients with acute myocardial infarction (AMI) significantly increase after AMI [21]. Follow-up studies later revealed that patients with higher TNC levels exhibit more pronounced left ventricular remodeling [55]. Serum TNC levels have become a predictor of disability and mortality, and hold good predictive value for major adverse cardiovascular events (MACE) [15]. Additionally, In animals or humans with hypertensive pulmonary artery injury or reconstruction, TNC is expressed in vascular smooth muscle cell and promotes smooth muscle cell proliferation through paracrine mechanisms, thereby exacerbating pulmonary vascular remodeling [56, 59]. Recent studies have reported that the expression level of TNC is significantly positively correlated with the diameter of AAA [25, 26], suggesting that TNC may play a crucial role in the progression of AAA. Our research group has further explored this correlation from multiple levels and revealed at the cellular lineage level that the absence of TNC promotes the progression of AAA. This finding not only validates the conclusions of previous studies, but also provides more in-depth experimental support for them. Similar to pulmonary hypertension studies, we identified pathological characteristics of TNC in VSMCs function, enhancing our understanding of its crucial role in vascular diseases, particularly AAA progression. By analyzing AAA plasma samples, we predicted the diagnostic value of high TNC plasma concentrations in AAA diagnosis. These findings hold significant clinical implications for early prevention and treatment of AAA. Previous mechanistic studies have demonstrated that TNC is closely associated with inflammation, contributing to the development of inflammatory diseases through downstream inflammatory pathways [40, 61]. Our experiments revealed that TNC knockout in VSMCs of mice leads to decreased expression of pro-inflammatory factors and increased expression of VSMCs contractile factors. Research reports indicate that downregulation of contractile factors and infiltration of pro-inflammatory factors are critical components of VSMCs dyshomeostasis, which may also represent key pathological changes in TNC-mediated AAA progression. To further elucidate TNC’s pathogenic mechanisms, we need to investigate specific endogenous processes through which TNC regulates VSMCs dyshomeostasis. Although studies suggest TNC promotes vascular wall inflammation and participates in atherosclerosis progression, the exact pathological mechanisms remain unclear. Previous research showed that TNC activates NF-κB via the TLR4 pathway to induce inflammatory responses and cause lung tissue damage [20]. Through bioinformatics analysis and experiments, we confirmed that TNC binds to TLR4 to regulate the expression of VSMCs contractile factors and inflammatory factors, consistent with previous findings. The TLR4/NF-κB pathway is a classical regulatory pathway in vascular inflammation, which activates immune responses by recognizing endogenous danger signals, promotes inflammatory factor secretion and immune cell infiltration, and has been proven to regulate AAA progression through inflammatory factor expression [62–65]. Our subsequent experiments confirmed that TNC on VSMCs activates the downstream NF-κB pathway via TLR4, thereby regulating the expression of classical contractile genes and inflammatory factors. Our findings not only further confirm the role of the TLR4/NF-κB pathway in regulating VSMCs homeostasis and inflammatory responses during aortic aneurysm progression, but also for the first time clarify the specific role of TNC in this process, thereby significantly advancing our understanding of TNC/TLR4/NF-κB pathway in vascular diseases. As a cytokine, the intracellular production of TNC necessarily involves upstream regulatory factors. Current research on TNC, particularly its upstream regulation, remains limited. This study reveals for the first time that CHOP (C/EBP-associated transcription factor 3), an endoplasmic reticulum stress factor on VSMCs, directly binds to the TNC gene promoter region, activating transcription and promoting protein expression. CHOP, also known as DDIT3 or GADD153, located on human chromosome 12q13.3, belongs to the CCAAT/enhancer-binding protein (C/EBP) family [66]. With DNA-binding activity, it participates in adipogenesis, growth arrest induction, and endoplasmic reticulum stress response, making it a hallmark gene of ER stress [67]. Previous studies reported that ER stress plays a crucial regulatory role in AAA progression [68–70]. A recent study suggested that CHOP may influence VSMCs homeostasis and drive AAA progression through transcriptional regulation. However, the specific regulatory mechanisms of CHOP in AAA remain poorly understood. To investigate whether CHOP regulates TNC during AAA pathogenesis, we conducted functional GO enrichment analysis on TNC-highly-expressing celltype (I-SMC), and identified endoplasmic reticulum stress signaling within it. Subsequently, our experiments demonstrated that the activation of TNC transcription by CHOP depends on its binding to these two specific DNA sites on the TNC promoter. Furthermore, we demonstrated that TNC mediates CHOP’s regulation expression of downstream inflammatory factors. These results indicate that TNC regulates the TLR4/NF-κB pathway, which modulating VSMCs dyshomeostasis and AAA progression in upstream regulation mediated by CHOP during the endoplasmic reticulum stress. While this study has yielded a series of innovative findings, several limitations remain: (1). The small amount of human plasma samples we initially collected were used to verify the differences of TNC in AAA and its predictive performance for AAA diagnosis. However, there might be a problem of insufficient sample size. In the future, more plasma samples need to be included to validate its accuracy. (2) The investigation did not explore whether other endoplasmic reticulum stress factors (e.g., ATF4, PERK) regulate TNC expression. Future work could establish a double knockout mouse model to further elucidate the ER multifactorial synergistic network. (3) Previous studies have highlighted macrophage-mediated inflammatory regulation as a key driver of AAA progression. Moreover, research from other diseases has revealed that TNC can modulate macrophage-derived inflammatory signals through mechanisms beyond TLR4, such as ITGA7 receptors or ERK/CREB signaling pathways. Our scRNAseq also identified macrophages as a major cellular population in human aortas. The potential mechanisms by which TNC regulates macrophage-mediated inflammation to promote AAA progression require further investigation. In summary, this study identified a specific celltype of VSMCs in AAA through scRNAseq, which plays a key role in the regulation of the TNC signaling network. This discovery provides a new perspective for understanding the functional heterogeneity of VSMCs in AAA. Subsequently, we have also preliminarily confirmed that elevated TNC levels in plasma can effectively predict the occurrence of human abdominal aortic aneurysms. Finally, we systematically elucidates the complete pathway through which TNC drives VSMCs contraction homeostasis imbalance and inflammatory response via dual mechanisms: TLR4/NF-κB inflammatory pathway and CHOP transcriptional regulation, thereby promoting AAA development. This provides a novel perspective for AAA mechanism research and clinical intervention. Future work will focus on precise regulation of the TNC signaling network and explore its translational value in maintaining vascular homeostasis and disease treatment, we also need more clinical data to confirm the diagnostic performance of TNC in the treatment of AAA. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1