Abstract

   The transition of healthy contractile vascular smooth muscle cells to
   an inflammatory and senescent phenotype is a key driver of abdominal
   aortic aneurysm (AAA). Although CD147 is highly expressed in VSMCs and
   upregulated in aneurysmal tissue, the precise role of VSMC-derived
   CD147 in phenotypic switching and AAA pathogenesis remains elusive.
   Here, we identified a previously unrecognized nuclear localization of
   CD147 in VSMCs, and pathological stimuli upregulated the nuclear CD147
   expression through reactive oxygen species-dependent mechanisms.
   Multi-omics analysis integrating RNA sequencing, CUT&Tag, and protein
   interactome profiling revealed that nuclear CD147 directly interacts
   with the STAT1/STAT2 complex to activate the IRF7-IFNα/β axis under
   oxidative stress (H[2]O[2] exposure), thereby driving VSMC senescence
   and inflammatory reprogramming. Functionally, CD147 deletion in VSMCs
   significantly mitigated Angiotensin II- and CaPO[4]-induced AAA
   formation, accompanied by improved VSMC phenotype, reduced vascular
   inflammation and extracellular matrix degradation in vivo.
   Pharmacological inhibition of CD147 using Myricetin, a food-derived
   natural small-molecule compound, effectively discouraged oxidative
   stress-induced VSMC fate transition in vitro, and suppressed AAA
   progression and improved vascular integrity in two murine AAA models,
   underscoring its therapeutic potential. Collectively, these findings
   identify CD147 as a key driver of interferon-mediated VSMC fate
   transition, providing mechanistic insights into AAA progression and a
   promising therapeutic target for vascular diseases.

   Keywords: CD147, Type I interferon, Vascular smooth muscle cell,
   Abdominal aortic aneurysm, Myricetin, Oxidative stress

Graphical abstract

   [41]Image 1
   [42]Open in a new tab

Highlights

     * •
       Nuclear CD147 drives VSMC fate transition via activating type I
       interferon signaling axis under oxidative stress.
     * •
       VSMC-specific deletion of CD147 significantly attenuates AngII- and
       CaPO[4]-induced AAA formation in vivo.
     * •
       Myricetin, a natural CD147 inhibitor, alleviates oxidative
       stress-induced VSMC senescence and inflammatory reprogramming.
       Myricetin treatment reduces AAA progression in vivo.

1. Introduction

   Abdominal aortic aneurysm (AAA) is the second most prevalent aortic
   disease following atherosclerosis characterized by the abnormal and
   permanent dilation of the abdominal aorta. Although an abdominal aortic
   aneurysm can remain asymptomatic for a long time, sudden rupture leads
   to the life-threatening hemorrhage with a high mortality rate [[43]1].
   Despite advancements in surgical techniques, including open repair and
   endovascular aneurysm repair, the lack of effective pharmacological
   interventions to prevent AAA progression remains a significant
   challenge [[44]1]. Understanding the molecular mechanisms underlying
   AAA formation is crucial for identifying new therapeutic strategies.

   As the principal cellular component of healthy blood vessels, vascular
   smooth muscle cells (VSMCs) exhibit a quiescent, contractile phenotype
   under physiological conditions to maintain the structural integrity of
   the vasculature; however, pathological stimuli drive their transition
   to a senescent, inflammatory, and extracellular matrix-secreting
   phenotype, thus contributing to AAA development [[45]2,[46]3]. On one
   hand, as a contributor to vascular inflammation, inflammatory VSMCs
   exhibit high expression of innate immune pattern recognition receptors
   such as Toll-like receptors and Nod-like receptors [[47]4,[48]5]. Upon
   activation, VSMCs secrete a repertoire of pro-inflammatory cytokines,
   thereby recruiting macrophages, neutrophils, and other immune cells to
   the vascular wall, exacerbating inflammatory responses. On the other
   hand, inflammatory cytokines and reactive oxygen species (ROS) within
   the vascular microenvironment downregulate contractile gene expression
   in VSMCs and promote cellular dysfunction and senescence [[49][6],
   [50][7], [51][8]]. Moreover, the cGAS-STING signaling pathway, a
   critical component of innate immune cascades, is highly activated in
   damaged VSMCs. It senses cytosolic leakage of mitochondrial and nuclear
   DNA, triggering an autocrine/paracrine type I interferon (IFN–I)
   response, primarily involving IFNα and IFNβ, that drives VSMC
   senescence and phenotypic transition [[52]3,[53]9,[54]10]. Despite
   being the first line of defense against vascular injury and
   pathological stimuli, the precise mechanisms underlying VSMC-mediated
   vascular inflammation remain incompletely understood. Further
   refinement of these pathways may provide novel targets for blocking
   VSMC phenotypic alteration and preventing aortic aneurysm formation and
   progression.

   CD147 (cluster of differentiation 147, also known as EMMPRIN or
   basigin), encoded by the BSG gene of the immunoglobulin superfamily, is
   a multifaceted molecule with diverse regulatory roles due to its
   interactions with various molecular partners [[55][11], [56][12],
   [57][13]]. Targeting CD147 has a protective role in inflammation,
   various cancers, and COVID-19 [[58][13], [59][14], [60][15]]. Previous
   studies from our group and others have identified CD147 as a critical
   therapeutic target in cardiovascular diseases [[61][16], [62][17],
   [63][18]]. For instance, targeting pressure overload-induced CD147
   overexpression effectively attenuates heart failure and pathological
   cardiac hypertrophy [[64]18,[65]19]. Similarly, inhibiting
   myeloid-derived CD147 prevents atherosclerotic plaque progression by
   reducing inflammation and enhancing efferocytosis [[66]20,[67]21].
   Interestingly, CD147 upregulation has been observed in angiotensin II
   (AngII)-stimulated VSMCs [[68]22], elastase- or AngII-induced murine
   aneurysmal aortas [[69]23,[70]24], and human aneurysmal specimens
   [[71]22,[72]25,[73]26]. Moreover, systemic inhibition of CD147 has been
   shown to attenuate AAA formation in mice [[74]24]. However, the direct
   role of CD147 in VSMC biology and its mechanistic contribution to AAA
   pathogenesis remain largely unexplored. Moreover, exploration of
   therapeutic agents targeting CD147 remains urgent.

   In the present study, we found that CD147 is obviously expressed in the
   nucleus of VSMCs and promotes oxidative stress-induced VSMC
   inflammation and senescence by transcriptionally activating the
   IRF7-IFNα/β signaling pathway through binding to STAT1/2. VSMC-specific
   knockout of CD147 or treatment with its natural inhibitor Myricetin
   (3,3,4,5,5,7−hexahydroxyflavone, Myr) significantly alleviated AAA in
   vivo.

2. Material and methods

2.1. Animal

   The animal protocols for this study were approved by the Ethics
   Committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong
   University and all procedures were conducted in accordance with the
   National Institutes of Health Guidelines for the Care and Use of
   Laboratory Animals (8th Edition, 2011). VSMC-specific CD147-knockout
   mice (CD147^ΔSMC) were generated by crossing Tagln-Cre transgenic mice
   with CD147-floxed mice (on a C57BL/6J background), in which exons 2–8
   of the CD147 gene were flanked by two loxP target sites (Cyagen
   Biosciences, Guangzhou, China). Then, CD147^ΔSMC mice were bred on an
   ApoE^−/− background to establish ApoE^−/−/CD147^ΔSMC mice. Littermate
   controls were homozygous for the floxed CD147 allele
   (ApoE^−/−/CD147^flox/flox). All mice were maintained in a controlled
   environment at 24 ± 2 °C and 40–60 % humidity with a 12-h light/dark
   cycle and ad libitum access to food and water. Animal experiments were
   performed by two independent investigators in a blinded manner.
   Randomization was carried out using the RAND function in Excel,
   generating random group assignments based on body weight, with group
   allocations concealed from the experimenter. Sample sizes for the mouse
   aortic aneurysm studies were calculated using a priori power analysis
   via PASS software, based on data from similar AAA models in our
   research group [[75][27], [76][28], [77][29]]. A sample size of more
   than 8 animals per group was determined to achieve 90 % power (β = 0.1)
   to detect significant changes in maximal aortic diameters at an α value
   of 0.05. Each mouse served as an independent experimental unit.

   AngII- or calcium phosphate (CaPO[4])-induced AAA models were
   established in 8- to 10-week-old male CD147^flox/flox and CD147^ΔSMC
   mice (18–22g), with or without ApoE knockout background, as previously
   reported [[78][27], [79][28], [80][29]]. To evaluate the effects of
   Myricetin (Myr) (S2326, Selleck, TX, USA) on AAA, AngII-infused
   ApoE^−/− mice and CaPO[4]-treated AAA mice were treated with Myr
   (10 mg/kg/day) or vehicle via intraperitoneal injection for 28 days.
   Myricetin was dissolved in a solution containing 5 % DMSO, 40 % PEG300,
   5 % Tween 80, and 50 % sterile water at a concentration of 2.5 mg/ml.
   At the end of study, all mice were euthanized by cervical dislocation
   under deep anesthesia, and aortas were harvested at 28 days
   post-implantation. Aneurysmal aortas were defined as those exhibiting a
   diameter increase of 50 % or more. Mice that died before the study
   endpoint were subjected to autopsy, and aneurysm rupture was identified
   by the presence of blood clots outside the adventitia of the dilated
   aortic wall.

2.2. In vivo ultrasound imaging analysis

   Mice were positioned in a supine orientation on the operation platform
   and anesthetized with isoflurane. Abdominal hair was removed using
   depilatory cream. Aortic imaging was performed using a Vevo 2100
   high-resolution ultrasound system (Fujifilm VisualSonics, Toronto,
   Canada). Initially, the probe was applied along the long axis to
   visualize and capture morphological images of the abdominal aorta.
   Subsequently, the probe was rotated to the short axis for precise
   measurement of the maximal aortic diameter (B-mode).

2.3. Blood pressure measurements

   Noninvasive measurements of blood pressure were performed in conscious
   and trained mice using tail-cuff plethysmography (BP-2010A, Softron
   Biotechnology), as previously described [[81][17], [82][28], [83][29]].

2.4. Isolation, culture, and treatment of primary cells

   Primary VSMCs were isolated from the aortas of Sprague-Dawley rats
   (150–180 g), CD147^flox/flox or CD147^ΔSMC mice via collagenase
   digestion, following established protocols. Briefly, aortic tissues
   were dissected, rinsed in cold phosphate-buffered saline (PBS), and
   stripped of endothelium and adventitia. The tissues were then minced
   into approximately 1 mm^3 fragments and enzymatically digested with 1 %
   type II collagenase (B204; Worthington, USA) in a cell culture
   incubator for 6–8 h. Isolated VSMCs were cultured in Dulbecco’s
   Modified Eagle Medium/F12 (DMEM/F12) supplemented with 10 % fetal
   bovine serum and 1 % penicillin-streptomycin, and their identity was
   confirmed by immunofluorescence staining for α-smooth muscle actin
   (α-SMA). Cells from passages 2–4 were utilized for subsequent
   experiments unless otherwise specified. Human aortic smooth muscle
   cells (HASMCs) were provided by ScienCell (Cat. No. 6110, USA). All
   cells were maintained at 37 °C in a humidified atmosphere of 5 % CO[2].

   VSMCs at passages 2–4 and approximately 80 % confluence were used to
   establish in vitro models. For phenotypic switching, VSMCs were treated
   with 1 μM angiotensin II (AngII) for 24 h or 100 μM H[2]O[2] for 12 h.
   To induce cellular senescence, VSMCs were exposed to 100 μM H[2]O[2]
   for 12 h, followed by two washes with serum-free DMEM/F12 medium to
   remove residual H[2]O[2]. Cells were then cultured in fresh complete
   growth medium for an additional 72 h prior to analysis.

2.5. Histological and immunostaining analysis

   Aortic Sections and VSMC Crawls were used for immunofluorescent
   staining and immunohistochemical analyses. Double immunofluorescent
   staining of aortic tissue was performed to visualize protein expression
   and localization. Nuclei were visualized using
   4′,6-diamidino-2-phenylindol (DAPI, [84]P36931; Invitrogen, USA)
   staining. Fluorescence intensity was calculated by Image-Pro Plus
   software (Media Cybernetics, Rockville, MA, USA). Immunohistochemical
   staining was performed to visualize protein expression in the indicated
   groups. Images were captured and analyzed using PANNORAMIC MIDI Scanner
   (3DHISTECH, Budapest, Hungary) with CaseViewer software (3DHISTECH).
   The relative positive area was analyzed using Image-Pro Plus software.
   Hematoxylin and eosin (H&E), Masson’s trichrome, and elastic Van Gieson
   (EVG) staining were used to visualize morphological characteristics,
   collagen content, and elastic fiber, respectively. Qualitative
   evaluation of elastin integrity was performed by semiquantitative
   grading as described [[85]30,[86]31]. The severity of elastin
   fragmentation was graded as follows: grade 1, no degradation; grade 2,
   mild elastin degradation; grade 3, severe elastin degradation; and
   grade 4, aortic rupture.

2.6. RNA sequencing

   Total RNA was extracted from VSMCs using TRIzol reagent (Invitrogen,
   USA) and converted into a cDNA library. RNA sequencing was subsequently
   performed through the high-throughput commercial sequencing service
   provided by Metware Co., Ltd. ([87]www.metware.cn). P-value adjustments
   were made using the Benjamini-Hochberg method (R package DESeq2) to
   control for the false discovery rate. Genes with an adjusted P-value
   <0.05 and |log2(fold change)| ≥ 0.5 were considered differentially
   expressed. Heatmaps and volcano plots were generated using the pheatmap
   and ggplot2 R packages, respectively. Gene Ontology (GO) pathway
   enrichment analysis was carried out using DAVID Tools
   ([88]https://david.ncifcrf.gov/tools.jsp). To explore the pathway and
   biological changes induced by CD147 overexpression or knockout, Gene
   Set Enrichment Analysis (GSEA) was performed using the complete gene
   expression dataset, and normalized enrichment score for each gene set
   was calculated.

2.7. Analysis of microarray datasets

   The [89]GSE233625 single-cell RNA sequencing dataset was reanalyzed
   using the R programming language. The raw data were downloaded from the
   Gene Expression Omnibus (GEO) repository. Initial data processing
   involved quality control to exclude low-quality cells and genes with
   minimal expression. Subsequently, the dataset was normalized to correct
   for technical variations and batch effects. Principal component
   analysis (PCA) was applied for dimensionality reduction to identify
   major sources of variation. Cells were clustered based on their gene
   expression profiles using unsupervised clustering methods.
   Differentially expressed genes (DEGs) between different clusters were
   identified and GO enrichment analyses were conducted to explore the
   biological functions associated with the DEGs. Data visualization was
   performed using R packages, including Seurat for clustering and ggplot2
   for generating heatmaps, volcano plots, and other graphical
   representations. All statistical analyses were performed in R (version
   4.3.2), ensuring reproducibility and consistency throughout the study.

2.8. Cleavage under targets and tagmentation–sequencing (CUT&Tag) and
cleavage under targets & release using nuclease (CUT&Run)

   CUT&Tag and CUT&Run assays were performed using the Hyperactive
   Universal CUT&Tag Assay Kit for Illumina Pro (TD904, Vazyme Biotech)
   and Hyperactive pG-MNase CUT&Run Assay Kit for PCR/qPCR (HD101, Vazyme
   Biotech), respectively, following the manufacturer’s instructions.
   Briefly, cells were mixed with 10 μl Concanavalin A-coated magnetic
   beads and incubated at room temperature for 10 min. Bead-bound cells
   were resuspended in 100 μL of buffer containing primary antibody CD147
   (1:50, CY6711, Abways, Shanghai, China) or IgG Isotype Control (1:50,
   CY5125, Abways). The mixture was incubated overnight at 4 °C. After
   removing the primary antibody buffer using a magnetic rack, cells were
   incubated with the secondary antibody goat anti-rabbit IgG antibody
   (1:100, Ab207, Vazyme) for 60 min at room temperature. The Hyperactive
   pA-Tn5 transposase (CUT&Tag) or pG-MNase Enzyme (CUT&Run) was prepared
   and incubated with cells for 1h at room temperature. Next, the cells
   were resuspended in the Tagmentation buffer and incubated at 37 °C for
   1 h. DNA was purified using phenol-chloroform-isoamyl alcohol
   extraction and ethanol precipitation. The DNA product was used directly
   for qPCR in Cut&Run assay. For Cut&Tag, DNA library preparation and
   sequencing were performed by Romics Biotech (Shanghai, China). The
   libraries were amplified by PCR with a universal N5 and a uniquely
   barcoded N7 primer. Library size analysis was performed using the
   Agilent 4200 TapeStation system (Agilent Technologies, Santa Clara, CA,
   USA). Sequencing was carried out on the Illumina NovaSeq 6000 platform,
   generating 150 bp paired-end reads for subsequent analysis. Raw reads
   were processed using Trimmomatic to remove adapter sequences and
   low-quality bases. Clean reads were aligned to the rat reference genome
   (Rat6_ensembl_v112) using the BWA algorithm. MACS2 was used for peak
   calling (q-value <0.05), and peaks were annotated using the ChIPseeker
   R package. GO and KEGG enrichment analysis was conducted with
   clusterProfiler, and p-values were corrected using the false discovery
   rate (FDR).

2.9. Dual luciferase reporter assay

   Rat IRF7 promoter DNA (−2000 bp to +100 bp from the transcription start
   site) was synthesized by OBiO Technology (Shanghai, China) and cloned
   into the pGL3-Basic luciferase reporter vector. Rat VSMCs were
   co-transfected with the pGL3-IRF7-promoter firefly luciferase
   construct, a Renilla luciferase control vector, and either a CD147
   expression plasmid or empty vector using Lipofectamine 3000
   (Invitrogen). The pGL3-Basic vector (promoterless) and
   pGL3-SV40-promoter construct were used as negative and positive
   controls, respectively. Luciferase activity was measured using the
   Dual-Luciferase Reporter Assay System (Beyotime, Shanghai, China), and
   firefly signals were normalized to Renilla luciferase activity.

2.10. Alphafold3

   To predict the potential structure of the rat CD147-STAT1-STAT2
   complex, we utilized AlphaFold 3, a deep learning-based algorithm for
   protein structure prediction [[90]32]. The complex model was generated
   based on the known sequences of CD147, STAT1, and STAT2. After
   obtaining the predicted structure, visualization and further analysis
   were conducted using PyMOL software to examine the binding interactions
   and spatial configuration of the complex.

2.11. Molecular docking

   Autodock Vina (1.2.2) was employed to analyze the binding affinities
   and modes of interaction between Myricetin and CD147. The molecular
   structure of Myricetin was retrieved from PubChem Compound
   ([91]https://pubchem.ncbi.nlm.nih.gov/). The 3D coordinates of the
   monomer full-length CD147 were predicted by Alphafold3 [[92]32]. For
   docking analysis, all protein and molecular files were converted into
   PDBQT format with all water molecules excluded and polar hydrogen atoms
   were added. The grid box was centered to cover the domain of each
   protein and to accommodate free molecular movement. The grid box was
   set to 30 Å × 30 Å × 30 Å, and grid point distance was 0.05 nm. The
   docking scores were automatically generated by the software. The PyMOL
   software was used to analyze and visualize the binding patterns and
   hydrogen bonds.

2.12. RNA isolation and quantitative RT-PCR

   Total RNA was extracted using TRIzol reagent and subsequently used to
   synthesize cDNA with the HiScript ® III RT SuperMix for qPCR (Vazyme,
   China). mRNA levels were quantified through real-time PCR using ChamQ
   Universal SYBR qPCR Master Mix (Vazyme, China) on a LightCycler® 480
   Real-Time PCR System (Roche Applied Science). Gapdh served as the
   internal reference genes for normalization. Detailed primer sequences
   are listed in [93]Table S1.

2.13. Subcellular fractionation analyses

   VSMCs were collected and washed three times with ice-cold PBS. Nuclear
   and cytoplasmic fractions were isolated using the Nuclear Extract Kit
   (SB-PR013HZ, Share-bio, Shanghai, China) according to the
   manufacturer’s protocol. The nuclear pellets underwent three additional
   washes with a specialized nuclear washing buffer to thoroughly
   eliminate any cytoplasmic and mitochondrial contamination. The purity
   of the nuclear extracts was verified via Western blot analysis using
   anti-GAPDH (cytoplasmic marker) and anti-H3 (nuclear marker)
   antibodies.

2.14. Immunoelectron microscopy

   To investigate the subcellular localization of CD147 in VSMCs,
   immunoelectron microscopy was performed on both freshly isolated rat
   aortic tissue and primary cultured rat VSMCs. Small tissue segments
   (∼1 mm^3) were immediately fixed in immunoelectron microscopy fixation
   buffer (Servicebio, Wuhan, China). Cultured VSMCs were gently scraped,
   pelleted, and fixed using the same protocol. Samples were dehydrated
   and embedded in pure resin using a low-temperature UV polymerization
   method at −20 °C for at least 48 h. Ultrathin sections were obtained
   using an ultramicrotome and collected onto nickel grids. After three
   washes with TBS, the sections were blocked with 1 % BSA for 30 min at
   room temperature and then incubated overnight at 4 °C with a rabbit
   anti-CD147 primary antibody (1:200, Abways, Shanghai, China). Following
   three additional washes, sections were incubated with a 1:50 dilution
   of gold-conjugated anti-rabbit IgG secondary antibody (Servicebio) at
   room temperature for 20 min and then at 37 °C for 1 h. After thorough
   washing with TBS and ultrapure water, the grids were counterstained
   with 2 % uranyl acetate in ethanol for 15 min in the dark, air-dried,
   and examined using a transmission electron microscope (JEM-1400, JEOL,
   Japan).

2.15. Assay of MMP activity

   In situ zymography was performed on cryosections (8–10 μm thick) of
   freshly harvested abdominal aortas to assess matrix metalloproteinase
   (MMP) activity. Briefly, tissue sections were incubated with a
   fluorogenic gelatin substrate (DQ™ gelatin, E−12055; Molecular Probes,
   USA) according to the manufacturer’s instructions. MMP activity was
   visualized as green fluorescence, and cell nuclei were counterstained
   with DAPI.

   Gelatin zymography was also conducted to evaluate MMP activity in
   VSMCs. Protein samples were separated by SDS-PAGE on gels containing
   0.8 mg/mL gelatin. Following electrophoresis, gels were washed with
   2.5 % Triton X-100 to remove SDS and then incubated in zymography
   buffer [50 mmol/L Tris-HCl (pH 8.0), 10 mmol/L CaCl[2], and 0.05 %
   Brij-35] at 37 °C for 48 h. After incubation, gels were stained with
   Coomassie Brilliant Blue to detect zones of gelatinolytic activity.

2.16. Western blotting analysis and co-immunoprecipitation

   Proteins were extracted from abdominal aortas, whole cells, and nuclear
   fractions. Equal amounts of total protein from murine tissues or VSMCs
   were resolved by 7.5 %–12.5 % SDS‒PAGE and transferred onto
   polyvinylidene fluoride membranes, blocked with 5 % bovine serum
   albumin, and incubated with the respective primary antibodies. Protein
   bands were identified using enhanced chemiluminescence (Millipore)
   after incubation with the corresponding secondary antibodies. The
   quantification of individual Western blot bands was performed with
   ImageJ software (National Institutes of Health, Bethesda, MD, USA) and
   the relative expression levels of the target proteins were expressed as
   a fold-change against GAPDH expression.

   After the total VSMC proteins were extracted and quantified,
   500-1000 μg of protein was incubated with the indicated antibodies for
   co-immunoprecipitation assays. Following incubation at 4 °C overnight,
   the protein–antibody complexes were immunoprecipitated with 40 μL
   PureProteome™ Protein A/G Mix Magnetic Beads (LSKMAGAG02) for 10–30 min
   at room temperature with continuous mixing to capture the immune
   complex. Finally, the precipitated proteins were collected and
   separated by SDS-PAGE, and specific proteins were detected using
   western blotting.

2.17. Mass spectrometry analysis of CD147-interacted proteins

   Protein interactome analysis was conducted through the commercial
   services provided by Metware Co., Ltd. ([94]www.metware.cn). The
   samples underwent reductive alkylation, trypsin digestion, and peptide
   extraction. The resulting peptides were subsequently analyzed using a
   Vanquish Neo UHPLC system coupled with LC-MS/MS (Thermo Fisher
   Scientific, Waltham, MA).

2.18. RNA interference, plasmid construction, and adenovirus generation

   Small interfering RNAs (siRNAs) against rat STAT2 and a scrambled siRNA
   were designed and synthesized by OBiO Technology (Shanghai, China).
   Transfection of rat VSMCs with siRNAs (50 nM) in vitro was performed
   using Lipofectamine™ 3000 Reagent (L3000015; Thermo-Fisher, USA)
   according to the manufacturer’s protocol.

   To validate the predicted nuclear localization sequence (NLS) of CD147,
   we constructed two expression vectors. The full-length CD147 and a
   truncated version of CD147, where the predicted nuclear localization
   sequence (NLS, amino acids 350–353) was deleted, were both fused with
   FLAG and EGFP tags. These constructs were then cloned into the
   pcDNA3.1(+) vector. The plasmids were subsequently transfected into rat
   VSMCs using Lipofectamine™ 3000 Reagent, according to the
   manufacturer’s instructions. The transfected cells were cultured for
   further analysis to investigate the localization of CD147.

   To overexpress CD147, four recombinant replication-defective adenoviral
   vectors were constructed, each encoding FLAG-tagged rat CD147,
   FLAG-tagged mouse CD147, FLAG-tagged NLS-deleted mouse CD147, or a
   negative control, all under the control of the cytomegalovirus (CMV)
   promoter. The viruses were packaged and titrated by HanBio Technology
   (Shanghai, China).

2.19. CCK8

   To assess the cytotoxicity of Myr, VSMCs were treated with Myr at
   concentrations of 5, 10, and 20 μM for 48 h. After the treatment
   period, cell viability was evaluated using the CCK-8 assay according to
   the manufacturer’s instructions. The absorbance was measured at 450 nm
   to determine the extent of cell proliferation and cytotoxicity induced
   by Myr treatment at the specified concentrations.

2.20. Statistical analysis

   All statistical analyses and graphing were conducted using GraphPad
   Prism 8 (GraphPad Software, San Diego, CA, USA). Data normality was
   assessed using the Shapiro–Wilk test. For comparisons between two
   groups, the Levene test was applied to detect homogeneity of variances.
   Comparisons were performed by the Student’s t-test (for equal
   variances) or Welch’s t-test (for unequal variances) for normally
   distributed variables or the Mann-Whitney U test for non-normally
   distributed variables. For comparisons among three or more groups, the
   Brown-Forsythe test was used to evaluate the homogeneity of variance,
   followed by a one-way ANOVA and Tukey’s post-hoc test when variances
   were equal, or Welch’s ANOVA with Tamhane’s T2 post-hoc test when
   variances were unequal. Two-way ANOVA with Tukey’s post-hoc analysis
   was used when more than two groups and variables were compared. The
   Chi-squared test was employed for bivariate comparisons of categorical
   variables. Ordinal variables were expressed as median (interquartile
   range, IQR), and statistical comparisons between groups were performed
   using the Mann–Whitney U test. Kaplan–Meier survival curves were
   generated, and survival differences were evaluated using the log-rank
   test. Unless stated otherwise, parametric data are expressed as
   mean ± SE, and non-parametric data are presented as median (IQR). A
   P-value of less than 0.05 was considered statistically significant.

3. Results

3.1. CD147 is highly expressed in the nucleus of VSMCs

   To investigate the role of CD147 in VSMCs, we first examined its
   expression and subcellular localization in primary rat aortic VSMCs.
   Immunofluorescence co-localization staining revealed that CD147 is
   obviously expressed in the nucleus of primary VSMCs, in addition to its
   cytoplasmic localization ([95]Fig. 1A). This nuclear distribution is
   distinct from its traditionally cytoplasmic and membrane expression
   observed in the breast cancer 4T1 cell line, rat cardiomyocytes (H9C2),
   and neonatal rat cardiomyocytes ([96]Supplementary Fig. 1A–1C), as
   reported in previous studies [[97]33,[98]34]. Cellular Immunoelectron
   microscopy and nuclear-cytoplasmic fractionation experiments further
   confirmed the obvious nuclear expression of CD147 in primary VSMCs
   ([99]Fig. 1B and C). We obtained further, more persuasive evidence for
   the nuclear localization of CD147 by performing nuclear-cytoplasmic
   fractionation on primary VSMCs overexpressing FLAG-CD147 via adenoviral
   transduction. Western blot analysis showed that exogenously expressed
   FLAG-CD147 was also highly abundant in the nucleus ([100]Fig. 1D).
   Furthermore, both immunofluorescence and immunoelectron microscopy
   analyses of rat aortic tissue confirmed significant nuclear expression
   of CD147 in VSMCs in vivo ([101]Fig. 1E and F). Notably,
   pharmacological stimulation of VSMCs with PDGF-BB or AngII induced a
   significant nuclear translocation of CD147 ([102]Fig. 1G-J), suggesting
   that its nuclear localization may play a crucial role in the cellular
   response to pathological stimuli. Similar results were observed in
   primary human and mouse aortic VSMCs with another two commercially
   available specific antibodies from different manufacturers (240134,
   Zenbio; ab188190, Abcam; [103]Supplementary Fig. 2A–2G).

Fig. 1.

   [104]Fig. 1
   [105]Open in a new tab

   CD147 is obviously localized in the nucleus of primary rat aortic
   VSMCs. (A) Representative immunofluorescence images showing CD147
   expression and subcellular localization in rat VSMCs. The slides were
   co-stained with the SMC marker α-SMA. Nuclear localization of CD147 was
   determined by colocalization with DAPI. Different focal planes from
   Z-stack imaging are presented. Scale bar, 10 μm. (B) Representative
   immunogold electron microscopy images of CD147 protein expression in
   rat VSMCs. Scale bar, 200 nm. (C) Western blot analysis of nuclear and
   cytoplasmic protein fractions isolated from rat VSMCs. GAPDH was used
   as a cytoplasmic marker, and H3 (histone 3) was used as a nuclear
   marker. (D) Representative Western blot images of FLAG-tagged CD147 in
   nuclear and cytoplasmic protein fractions of rat VSMCs. (E)
   Representative immunofluorescence images of CD147 expression in rat
   aortic tissue. The slides were co-stained with α-SMA. CD147 nuclear
   localization was verified by colocalization with DAPI. Scale bar,
   10 μm. (F) Representative immunogold electron microscopy images of
   CD147 protein expression in rat aorta tissue. Scale bar, 200 nm. (G)
   Left: Western blot analysis of CD147 protein levels in nuclear and
   cytoplasmic fractions of rat VSMCs treated with PBS or PDGF-BB
   (10 ng/mL). Right: Quantification of nuclear CD147 protein expression
   (n = 4). ∗∗∗P < 0.001. (H) Representative immunofluorescence images
   showing CD147 expression and subcellular localization in rat VSMCs
   treated with PBS or PDGF-BB (10 ng/mL). The slides were co-stained with
   the SMC marker α-SMA. Scale bar, 10 μm. (I) Left: Western blot analysis
   of CD147 protein levels in nuclear and cytoplasmic fractions of rat
   VSMCs treated with PBS or AngII (1 μM). Right: Quantification of
   nuclear CD147 protein expression (n = 3). ∗∗∗∗P < 0.0001. (J)
   Representative immunofluorescence images showing CD147 expression and
   subcellular localization in rat VSMCs treated with PBS or AngII (1 μM).
   The slides were co-stained with the SMC marker α-SMA. Scale bar, 10 μm.

   Recent studies have suggested that the nuclear translocation of certain
   proteins can be regulated by redox-dependent mechanisms
   [[106]35,[107]36]. To determine whether ROS mediate AngII- or
   PDGF-BB-induced nuclear localization of CD147, VSMCs were pretreated
   with the ROS scavenger N-acetyl cysteine (NAC) for 1 h prior to
   stimulation. NAC treatment markedly inhibited AngII- or PDGF-BB-induced
   nuclear accumulation of CD147 ([108]Supplementary Fig. 3A–3D),
   indicating that ROS may play a critical role in regulating CD147
   nuclear expression. Furthermore, we identified a putative and highly
   conserved nuclear localization sequence (NLS) at positions 349 to 354
   of rat CD147 employing three bioinformatics online tools, matching the
   classical K (K/R) X (K/R) NLS motif [[109]37] ([110]Supplementary Fig.
   4A and 4B). To study the possible role of the NLS motif in regulating
   CD147 nuclear localization, VSMCs were transiently transfected with the
   pCD147-EGFP-FLAG or pCD147(NLS deletion)-EGFP-FLAG plasmid. Subsequent
   subcellular fractionation analysis showed that NLS deletion eliminated
   CD147 nuclear translocation ([111]Supplementary Fig. 4C), a finding
   that was also obtained by direct EGFP fluorescence ([112]Supplementary
   Fig. 4D). These results indicate that the nuclear localization of CD147
   in VSMCs depends on the conserved NLS sequence ‘KRRK’.

3.2. CD147 binds to the promoter region of Irf7 and transcriptionally
regulates its expression in VSMCs

   Considering the prominent nuclear expression of CD147 and its enhanced
   nuclear translocation in response to stressed stimuli, we hypothesize
   that CD147 may play a role in regulating the transcriptional functions
   of VSMCs. To investigate this further, we performed RNA sequencing of
   VSMCs transduced with either Vector control (Vector) or
   CD147-overexpressing (CD147-OE) adenovirus following PDGF-BB
   stimulation ([113]Supplementary Fig. 5A–C). Bioinformatics analysis
   revealed that DEGs were predominantly enriched in pathways related to
   the innate immune response and type I interferon (IFN–I) signaling
   ([114]Fig. 2A–B). To further assess whether CD147 interacts with
   specific DNA regions (such as promoters and enhancers) and regulates
   transcription, we performed a CUT&Tag assay using a recombinant
   monoclonal antibody against CD147 ([115]Supplementary Fig. 5D–E).
   Annotation of sequencing data showed that CD147 predominantly binds to
   transcriptionally relevant genomic regions ([116]Fig. 2C), such as
   promoters (31.37 %) and intergenic regions (32.7 %), further supporting
   the notion that CD147 may play a crucial role in transcriptional
   regulation. Subsequent KEGG functional enrichment analysis indicated
   that CD147 binding targets are primarily involved in regulating innate
   immune response-related pathways, including JAK-STAT, chemokine
   signaling, and Toll-like receptor pathways ([117]Fig. 2D).

Fig. 2.

   [118]Fig. 2
   [119]Open in a new tab

   CD147 binds to the Irf7 promoter region and regulates its transcription
   in VSMCs. (A) Top-ranked categories identified through Gene Ontology
   (GO) analysis of differentially expressed genes (DEGs) in
   CD147-overexpressing (CD147-OE) VSMCs compared to Vector controls
   treated with PDGF-BB (10 ng/mL). (B) Gene Set Enrichment Analysis
   (GSEA) of RNA-seq data highlighting the top-enriched gene sets based on
   normalized enrichment scores and p-values in CD147-OE VSMCs versus
   Vector controls. (C) Genome-wide distribution of CD147-binding sites in
   isolated VSMCs as determined by CUT&Tag analysis. (D) KEGG pathway
   enrichment analysis of target genes identified from CUT&Tag analysis.
   (E) Overlap of putative CD147 targets identified from CUT&Tag data with
   DEGs from RNA-seq analysis. Heatmap showing expression levels of
   overlapping genes across samples from the RNA-seq data. (F) IGV
   visualization of CUT&Tag analysis demonstrating CD147 enrichment at the
   Irf7 promoter. (G) CUT&Run was performed with a monoclonal anti-CD147
   antibody, and the target promoter region of Irf7 was quantified by
   qPCR. (n = 5). ∗∗P < 0.01. (H) VSMCs were co-transfected with either
   CD147 overexpression plasmid or control vector along with the
   pGL3-IRF7-promoter luciferase construct for 48 h. Luciferase activity
   was measured using a dual-luciferase reporter assay system (n = 6).
   ∗∗∗P < 0.001. (I) Quantitative PCR analysis of mRNA expression levels
   of key genes in the type I interferon signaling pathway in VSMCs
   treated with PDGF-BB (10 ng/mL) in the indicated groups. ∗P < 0.05;
   ∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. (J) Western blot analyses
   showing significant upregulation of IRF7 protein levels in CD147-OE
   VSMCs compared to Vector group following PDGF-BB treatment (10 ng/mL,
   n = 6). ∗∗∗∗P < 0.0001.

   To explore the relationship between CD147 occupancy and mRNA
   expression, we compared the CUT&Tag and RNA sequencing datasets. A Venn
   diagram revealed 20 overlapping genes identified from both CUT&Tag
   peaks and DEGs of RNA-seq ([120]Fig. 2E). Among these potential
   targets, IRF7 ([121]Fig. 2F), a member of the interferon regulatory
   factors family, along with its associated IFN-I response pathway
   [[122]38], was selected for further investigation based on three
   considerations: (1) The RNA-seq data revealed a significant
   differential expression of IRF7, and Gene Set Enrichment Analysis
   (GSEA) suggested that CD147 overexpression significantly enhanced the
   IFN-I response in VSMCs ([123]Fig. 2B); (2) VSMCs are both
   IFN-I-responsive and IFN-I-productive cells. IFN-I response induces
   VSMC phenotypic switching and innate immune responses in an
   autocrine/paracrine manner [[124]6,[125]9,[126]39]; (3) IRF7-mediated
   IFN-I signaling, including the upstream cGAS-STING-TBK1 axis and
   downstream JAK-STAT pathway, plays a pivotal role in vascular smooth
   muscle pathological responses [[127]10,[128]40,[129]41]. The putative
   binding site of CD147 in the promoter region of the Irf7 locus was
   confirmed by Cut&Run followed by qPCR analysis ([130]Fig. 2G).
   Moreover, dual luciferase reporter assays showed that CD147
   overexpression enhanced Irf7 promoter activity ([131]Fig. 2H;
   [132]Supplementary Fig. 5F). Consequently, CD147 overexpression
   significantly increased IRF7 mRNA and protein levels in PDGF-BB-treated
   VSMCs ([133]Fig. 2I and J). These results suggest that CD147 functions
   as a novel transcriptional regulator, involved in the transcriptional
   control of IRF7 in VSMCs.

3.3. CD147 promotes VSMC phenotypic transition, IFN-I response and cellular
senescence under H[2]O[2] exposure

   PDGF-BB primarily drives the phenotypic transition of VSMCs from a
   contractile to a proliferative and osteogenic state, its influence on
   IFN-I responses is limited ([134]Supplementary Fig. 6A and 6B). In
   multiple vascular pathologies, excessive ROS leads to mitochondrial and
   nuclear DNA leakage into the cytoplasm, significantly activating the
   cGAS-STING and JAK-STAT pathways to promote IFN-I responses and the
   phenotypic shift of contractile VSMCs to an inflammatory state
   [[135]9,[136]42,[137]43]. Thus, we used H[2]O[2], a potent inducer of
   ROS and cellular dysfunction, to trigger IFN-I signaling in VSMCs
   ([138]Supplementary Fig. 6C and 6D). Excessive ROS stimulation induced
   by H[2]O[2] significantly upregulates CD147 protein expression and
   promotes its translocation to the nucleus in VSMCs ([139]Fig. 3A-C). As
   anticipated, CD147 overexpression significantly enhanced the expression
   of key components in the IFN-I signaling pathway, including IRF7, IFNα,
   and IFNβ, in response to H[2]O[2] ([140]Fig. 3D and E;
   [141]Supplementary Fig. 7A–7C). Previous studies documented that
   enhanced IFN-I response induces cellular premature senescence and VSMC
   phenotypic switching in an autocrine/paracrine manner
   [[142]9,[143]44,[144]45]. Thus, we further investigated the role of
   CD147 in H[2]O[2]-induced VSMC senescence. Results demonstrated that
   CD147 overexpression markedly promoted H[2]O[2]-induced VSMC
   senescence, as demonstrated by significantly enhanced expression of
   senescence markers (P21, P16) and activated senescence-associated
   secretory phenotype (SASP) (IL-6, CXCL1) ([145]Fig. 3D-E;
   [146]Supplementary Fig. 7A–7C). Similarly, the area of
   senescence-associated β-galactosidase (SA-β-gal) positive cells per
   field was significantly increased in H[2]O[2]-induced VSMCs with CD147
   overexpression ([147]Fig. 3E). Additionally, H[2]O[2] induced a
   significant decrease in the expression of contractile phenotype-related
   proteins (CNN1, MYH11, α-SMA, and SM22), which were enhanced by CD147
   overexpression ([148]Fig. 3D; [149]Supplementary Fig. 7A–7C).

Fig. 3.

   [150]Fig. 3
   [151]Open in a new tab

   CD147 overexpression promotes H[2]O[2]-induced VSMC phenotypic
   transition, IFN-I response, and cellular senescence. (A) Left: Western
   blot analysis of CD147 protein levels in VSMCs treated with PBS or
   H[2]O[2] (100 μM). Right: Quantification of CD147 protein expression
   (n = 6). ∗∗∗P < 0.001. (B) Top: Western blot analysis of CD147 protein
   levels in nuclear and cytoplasmic fractions of rat VSMCs treated with
   PBS or H[2]O[2] (100 μM). Bottom: Quantification of nuclear CD147
   protein expression (n = 4). ∗∗P < 0.01. (C) Representative
   immunofluorescence images illustrating CD147 expression and subcellular
   localization in VSMCs treated with PBS or H[2]O[2] (100 μM). The slides
   were co-stained with SMC marker α-SMA and DAPI. (D) Western blot
   analysis of IRF7, IFNα, IFNβ, P21, α-SMA, SM22, CNN1 and MYH11 protein
   levels in VSMCs transfected with CD147-OE or Vector adenovirus,
   followed by exposure to PBS or H[2]O[2] (100 μM). (E) Left:
   Representative immunofluorescence images showing IRF7, P16, P21 protein
   expression, and SA-β-gal staining in VSMCs transfected with CD147-OE or
   Vector adenovirus, followed by exposure to PBS or H[2]O[2] (100 μM).
   The slides were co-stained with the SMC marker α-SMA and nuclear stain
   DAPI. Right: Quantification of IRF7, P16, P21 protein levels, and
   SA-β-gal positive areas in the indicated groups (n = 10–15).
   ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001.

   To further explore the effects of CD147 inhibition, we generated
   VSMC-specific CD147 deletion mice (CD147^ΔSMC) by intercrossing
   Tagln-Cre transgenic mice on a C57BL/6J background with CD147-floxed
   mice (CD147^flox/flox) ([152]Supplementary Fig. 8A). Genotyping was
   performed and successful ablation of CD147 was demonstrated by CD147
   detection via quantitative PCR assays and western blotting in isolated
   aortas ([153]Supplementary Fig. 8B–8E). Subsequently, mouse aortic
   VSMCs were isolated from CD147^ΔSMC and CD147^flox/flox mice and
   subjected to H[2]O[2] exposure. Remarkably, CD147 knockout effectively
   suppressed H[2]O[2]-induced IFN-I responses, cellular senescence, and
   VSMC phenotypic transition ([154]Supplementary Fig. 9A–9E). To
   investigate the role of nuclear CD147 in VSMC fate transition, we
   overexpressed FLAG-tagged wild-type CD147 (FLAG-CD147-WT) or
   FLAG-tagged NLS-deficient CD147 (FLAG-CD147-ΔNLS) in VSMCs isolated
   from CD147^ΔSMC mice via adenoviral vectors, followed by H[2]O[2]
   exposure. The results showed that overexpression of wild-type CD147
   significantly enhanced H[2]O[2]-induced IFN-I responses, cellular
   senescence, and VSMC phenotypic transition. In contrast, these effects
   were significantly abolished in cells overexpressing NLS-deficient
   CD147 ([155]Supplementary Fig. 9F–9J). These data hinted that nuclear
   CD147 may transcriptionally activate the IRF7-IFNα/IFNβ signaling
   pathway, thereby promoting H[2]O[2]-induced phenotypic transition and
   senescence in VSMCs.

3.4. CD147 regulates the IRF7-IFNα/IFNβ pathway via binding to STAT1/STAT2 in
VSMCs

   To further investigate the transcriptional regulatory mechanism of
   CD147 on IRF7, we performed immunoprecipitation using FLAG antibodies
   in lysates from senescent VSMCs overexpressing FLAG-CD147, followed by
   mass spectrometry to identify potential interacting proteins ([156]Fig.
   4A). Among the identified proteins, the STAT1/STAT2 complex, a key
   transcription factor for IRF7 and essential regulator for IFN-I
   induction, was of particular interest ([157]Fig. 4B) [[158]46,[159]47].
   Immunoprecipitation experiments confirmed that CD147 interacts with the
   STAT1/STAT2 complex ([160]Fig. 4C; [161]Supplementary Fig. 10A), and
   this interaction occurs within the nucleus ([162]Supplementary Fig.
   10B). Phosphorylation of STATs is critical for their nuclear
   translocation and transcriptional activity. Therefore, we performed
   co-localization staining of CD147 and phosphorylated STAT2 (p-STAT2),
   which revealed that CD147 co-localizes with p-STAT2 in the nucleus of
   VSMCs ([163]Supplementary Fig. 10C). Using AlphaFold 3 [[164]32], we
   predicted the structural features between CD147 and the STAT1/STAT2
   complex, revealing that CD147 primarily binds directly to STAT2
   ([165]Fig. 4D). Knockdown of STAT2 using small interfering RNA
   significantly disrupted the interaction between CD147 and the
   STAT1/STAT2 complex ([166]Fig. 4E). Further phenotypic analyses
   demonstrated that STAT2 knockdown effectively reversed the promoting
   effects of CD147 overexpression on the IRF7-IFNα/IFNβ signaling
   pathway, cellular senescence, and phenotypic transition in VSMCs
   ([167]Fig. 4F and G; [168]Supplementary Fig. 10D–10F). These results
   suggest that CD147 may transcriptionally regulate IRF7-IFNα/IFNβ
   signaling and promote VSMC fate transition by forming a complex with
   STAT1/STAT2.

Fig. 4.

   [169]Fig. 4
   [170]Open in a new tab

   CD147 interacts with STAT1/STAT2 complex to activate IFN-I signaling in
   VSMCs. (A) Schematic workflow of the CD147 protein interactome
   analysis. Senescent VSMCs were infected with adenovirus expressing
   FLAG-CD147. Cell lysates were immunoprecipitated using anti-FLAG beads
   and subjected to mass spectrometry for protein identification. (B) Five
   unique peptides from the STAT1/STAT2 complex were identified (left),
   and a representative LC-MS/MS spectrum is shown (right). (C) Western
   blot analysis showing the association of FLAG-CD147 with the
   STAT1/STAT2 complex in VSMCs. FLAG-CD147 was immunoprecipitated with
   anti-FLAG antibody from VSMC lysates, followed by Western blotting. (D)
   Predicted structural model of the CD147/STAT1/STAT2 complex generated
   by AlphaFold3. (E) VSMCs were infected with adenovirus expressing
   FLAG-CD147 and either STAT2-targeting siRNA or a negative control
   (si-NC, scrambled siRNA). Lysates were immunoprecipitated with
   anti-FLAG antibody, followed by Western blot analysis. (F) Western blot
   analysis of STAT2, IRF7, IFNα, IFNβ, P21, α-SMA, SM22, CNN1 and MYH11
   protein levels in VSMCs transfected with CD147-OE adenovirus and either
   STAT2-targeting siRNA or negative control (si-NC, scrambled siRNA),
   followed by exposure to PBS or H[2]O[2] (100 μM). (G) Left:
   Representative immunofluorescence images showing IRF7, P16, P21
   expression, and SA-β-gal staining in VSMCs transfected with CD147-OE
   adenovirus and either STAT2-targeting siRNA or negative control,
   followed by exposure to PBS or H[2]O[2](100 μM). The slides were
   co-stained with SMC marker α-SMA and DAPI. Right: Quantification of
   IRF7, P16, P21 protein expression levels, and SA-β-gal positive areas
   in the indicated groups (n = 10). ∗∗∗∗P < 0.0001.

3.5. Aneurysmal stimuli markedly upregulate CD147 expression in VSMCs and
significantly enhance its nuclear localization

   We next investigated the role of CD147-mediated VSMC phenotypic
   transition in aortic aneurysm (AA). We first analyzed single-cell RNA
   sequencing data of healthy aortic tissue (HC) and AngII-induced
   aneurysmal tissue (AA) obtained from the GEO database. Based on the
   expression patterns of cell type-specific marker genes, we identified
   four major vascular cell populations: VSMCs, fibroblasts,
   monocyte/macrophages, and endothelial cells (EC) ([171]Supplementary
   Fig. 11A and 11B). According to previous studies [[172]3] and gene
   ontology (GO) enrichment analysis of feature genes, VSMCs were
   classified into six principal functional subpopulations: contractile,
   extracellular matrix-producing, inflammatory,
   contractile/proliferative, stressed/adaptive, and EC
   function-associated SMCs ([173]Supplementary Fig. 11C–11E). Notably,
   cell distribution analysis revealed a significant expansion of
   inflammatory SMCs, alongside a marked reduction in proliferative and
   contractile SMCs in the aneurysm group compared to the control
   ([174]Fig. 5A). These findings suggest that aneurysmal stimuli induce
   VSMC phenotypic transition from a contractile state to inflammatory and
   senescent states. Furthermore, expression analysis revealed a
   pronounced upregulation of CD147 in VSMCs from the AngII-induced mouse
   aneurysmal tissues, while no significant changes were observed in
   endothelial cells or macrophages ([175]Fig. 5B). Moreover, the
   upregulation of CD147 in response to AngII stimulation was most evident
   in inflammatory SMCs ([176]Fig. 5B), reinforcing our in vitro findings
   that CD147 primarily drives VSMCs transition to the inflammatory
   phenotype. Further Western blot and immunofluorescence co-localization
   assays demonstrated that aneurysmal stimuli not only upregulated CD147
   expression in VSMCs but also facilitated its nuclear translocation
   ([177]Fig. 5C and D). Collectively, these findings suggest that
   CD147-driven VSMC phenotypic transition towards an inflammatory and
   senescent state may play a pivotal role in the pathogenesis and
   progression of AAA.

Fig. 5.

   [178]Fig. 5
   [179]Open in a new tab

   VSMC-specific knockout of CD147 significantly inhibits AngII-induced
   AAA formation in an ApoE^−/− background mouse model. (A) Bar chart
   showing the percentage of cluster populations in aortic smooth muscle
   cells (SMCs) from HC and AA groups of mice. Fisher’s exact test was
   performed to examine differences in the cluster proportions between
   groups. (B) Violin plots depicting the expression levels of CD147
   across different cell types in healthy control (HC) and aortic
   aneurysmal (AA) vascular tissues. Macro/Mono, macrophages/monocytes;
   EC, endothelial cells. (C) Western blot analysis indicating a
   significant upregulation of CD147 in mouse suprarenal abdominal aortic
   aneurysm tissue. ∗∗∗P < 0.001. (D) Immunofluorescence staining
   illustrating CD147 protein expression and nuclear localization in
   aneurysmal tissues of HC and AAA mice. The slides were co-stained with
   the smooth muscle cell marker α-SMA and DAPI. (E) Schematic protocol:
   ApoE^−/−/CD147^flox/flox and ApoE^−/−/CD147^ΔSMC mice were
   subcutaneously injected with saline or AngII by a mini osmotic pump for
   28 days (n = 21–22 per group). (F) Survival curves of the AngII-infused
   ApoE^−/−/CD147^flox/flox and ApoE^−/−/CD147^ΔSMC mice. Data were
   analyzed using the Kaplan-Meier method and compared by log-rank tests.
   (G) Incidence of AngII-induced aneurysm rupture in the indicated groups
   (n = 21–22). (H) Representative macroscopic images of AAA formation in
   the indicated groups. (I) Incidence of AngII-induced AAA in the
   indicated groups (n = 21–22). Data were analyzed using Fisher’s exact
   test. ∗P < 0.01; ∗∗∗P < 0.001. (J) Quantification of total aortic
   weight/body weight (BW) ratio in the indicated groups (n = 7–10).
   ∗∗∗∗P < 0.0001.

3.6. VSMC-specific CD147 deficiency represses AAA formation In vivo

   To elucidate the potential role of CD147 in the pathology of AAA, we
   generated VSMC-specific CD147-knockout mice and their littermate
   controls in an ApoE knockout background (ApoE^−/−/CD147^ΔSMC and
   ApoE^−/−/CD147^flox/flox) ([180]Supplementary Fig. 12A and 12B), which
   were subjected to AngII or saline and maintained on a high-fat diet for
   28 days ([181]Fig. 5E). Blood pressure measurements revealed no
   significant difference between ApoE^−/−/CD147^ΔSMC and
   ApoE^−/−/CD147^flox/flox control mice ([182]Table S3). However,
   ApoE^−/−/CD147^ΔSMC mice were less sensitive to AngII induction, as
   evidenced by significant reductions in mortality rate and aortic
   ruptures ([183]Fig. 5F and G). Moreover, compared with
   ApoE^−/−/CD147^flox/flox mice, decreased AAA incidence and the ratio of
   total aortic weight to body weight were observed in ApoE^−/−/CD147^ΔSMC
   mice under the condition of AngII infusion ([184]Fig. 5H-J). At the end
   of the induction period, micro-ultrasound imaging revealed that
   VSMC-specific CD147 knockout significantly attenuated the AngII-induced
   enlargement of maximal aortic diameter ([185]Supplementary Fig. 12C).
   Morphological and histological analyses further showed that CD147
   deletion in VSMCs mitigated elastic fiber degradation and decreased
   collagen deposition in the aortas of AngII-infused mice
   ([186]Supplementary Fig. 12D).

   Western blot analysis revealed that in response to AngII stimulation,
   the expression of contractile phenotype markers in abdominal aortas was
   significantly upregulated in ApoE^−/−/CD147^ΔSMC mice compared to
   ApoE^−/−/CD147^flox/flox controls ([187]Fig. 6A and B). In addition,
   CD147 deficiency suppressed the expression of inflammatory cytokines,
   including interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α), in
   the abdominal aorta ([188]Fig. 6C and D). Furthermore, CD147 deficiency
   markedly attenuated AngII-induced MMP activation, as evidenced by the
   significant reduction in MMP-2 expression in aortic tissue and MMP
   activity both in vivo and in cultured VSMCs ([189]Fig. 6C and D;
   [190]Fig. 6G and H). Immunostaining demonstrated lower levels of IFN-I
   components (IRF7, IFNβ) and senescence marker P21 in aneurysmal tissue,
   indicating improved cellular function following CD147 deletion
   ([191]Fig. 6E and F). In an AngII-induced in vitro VSMC model, we
   further confirmed that CD147 deficiency significantly suppressed
   IFN-I-related inflammatory signaling activation and cellular
   senescence, while promoting the transition toward a contractile VSMC
   phenotype through protein quantification ([192]Supplementary Fig.
   13A–13D) and RNA-Seq ([193]Supplementary Fig. 13E–13I).

Fig. 6.

   [194]Fig. 6
   [195]Open in a new tab

   CD147 knockout significantly alleviated AngII-induced VSMC phenotypic
   transition, IFN-I response, and MMP activation. (A–B) Representative
   Western blot and quantification of STAT2, IRF7, α-SMA, SM22, CNN1, and
   MYH11 protein levels in abdominal aortas in the indicated groups of
   mice on day 28 following AngII infusion. ∗∗P < 0.01; ∗∗∗P < 0.001;
   ∗∗∗∗P < 0.0001. (C–D) Representative immunohistochemistry images and
   quantification of typical matrix-degrading enzyme MMP2 and inflammatory
   markers (IL-6, TNF-α) protein expression levels in aneurysmal tissues
   from the indicated groups of mice on day 28 following AngII infusion
   (n = 5). ∗P < 0.05. (E–F) Representative immunofluorescence images and
   quantification of fluorescence intensity for protein expression of
   IFN-I components (IRF7, IFNβ) and senescence marker (P21) in aneurysmal
   tissues from the indicated groups of mice on day 28 following AngII
   infusion. ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. (G) Representative images and
   quantification of active MMP in aneurysmal tissues from the indicated
   groups of mice on day 28 following AngII infusion (n = 6).
   ∗∗∗P < 0.001. (H) MMP activity measured by gelatin zymography in
   isolated VSMCs treated with AngII (1 μM). ∗∗P < 0.01.

   To further validate the pathogenic role of VSMC-derived CD147 in AAA
   development, we employed an alternative model of AAA induced by CaPO[4]
   ([196]Supplementary Fig. 14A). Compared with their CD147^flox/flox
   littermates, CD147^ΔSMC mice exhibited markedly attenuated aneurysm
   formation and a reduced maximal diameter of the infrarenal abdominal
   aorta ([197]Supplementary Fig. 14B and 14C). Histopathological analysis
   revealed significantly less elastic fiber degradation and collagen
   accumulation in CD147^ΔSMC mice compared with littermate
   CD147^flox/flox controls ([198]Supplementary Fig. 14D–14F).
   Collectively, these findings underscore the critical role of
   VSMC-derived CD147 in driving AAA pathogenesis and highlight its
   therapeutic potential.

3.7. CD147 natural small molecule inhibitor Myr ameliorates H[2]O[2]-induced
VSMC phenotypic transition, IFN-I response and senescence

   We further assessed the effects of pharmacological inhibition of CD147
   in vitro. Myricetin (Myr) is a natural small-molecule inhibitor with
   the highest affinity for CD147 discovered to date and possesses
   significant medicinal value [[199]48,[200]49]. We next investigated the
   impact of Myr on CD147 expression in VSMCs and its role in VSMC
   phenotypic transition and senescence. Molecular docking was conducted
   to evaluate the binding affinity and target binding sites between Myr
   and CD147. The crystal structure of monomer full-length CD147 was
   predicted by Alphafold3 and that of Myr were obtained from PubChem
   chemistry database, respectively. As shown in [201]Supplementary Fig.
   15A and 15B, Myr bound to the N-terminal domain of CD147, forming five
   hydrogen bonds with amino acid residues GLU-53, TRP-56, LEU-70, ASP-72,
   and ALA-83. The docking score of −6.408 kcal/mol indicates a strong
   affinity between Myr and CD147. These data confirmed the stable
   interaction between Myr and CD147. Further Western blot analyses
   revealed that Myr downregulated CD147 protein expression in a
   dose-dependent manner in VSMCs ([202]Supplementary Fig. 15C–15E).
   Moreover, pre-treatment with the proteasome inhibitor MG-132
   significantly reversed Myr-induced degradation of CD147 protein
   ([203]Supplementary Fig. 15F and 15G). These results suggest that Myr
   exhibits a strong binding affinity for CD147 and downregulates its
   protein levels through proteasome-dependent degradation.

   We next evaluated the cytotoxicity of Myr in VSMCs. CCK-8 assays
   demonstrated that even at a higher concentration (20 μM), Myr did not
   significantly inhibit VSMC cellular viability ([204]Supplementary Fig.
   15H). We then investigated Myr’s potential protective effects on VSMCs
   under pathological stimuli. Results showed that Myr treatment
   significantly ameliorated the downregulation of contractile genes
   induced by H[2]O[2] ([205]Fig. 7A and B; [206]Supplementary Fig. 15I).
   Additionally, compared to the control group, Myr treatment markedly
   suppressed H[2]O[2]-induced upregulation of IRF7 and activation of the
   IFN-I signaling ([207]Fig. 7C and D; [208]Supplementary Fig. 15I).
   Quantitative analysis of senescence markers (P21 and P16) and SA-β-gal
   staining revealed that Myr significantly inhibited H[2]O[2]-induced
   cellular senescence ([209]Fig. 7E and F; [210]Supplementary Fig. 15I).
   These findings suggest that Myr treatment significantly attenuates
   H[2]O[2]-induced VSMC phenotypic transition, activation of the IFN-I
   signaling pathway, and cellular senescence.

Fig. 7.

   [211]Fig. 7
   [212]Open in a new tab

   Myricetin significantly attenuates H[2]O[2]-induced VSMC phenotypic
   transition, IFN-I activation, and senescence. (A–B) Western blot and
   quantification of α-SMA, SM22, CNN1, and MYH11 protein levels in VSMCs
   treated with myricetin (10 μM) or DMSO, followed by exposure to PBS or
   H[2]O[2] (100 μM) (n = 6). ∗∗P < 0.01; ∗∗∗∗P < 0.0001. (C–D) Western
   blot and quantification of IRF7, IFNα, IFNβ, and P21 protein levels in
   VSMCs treated with myricetin (10 μM) or DMSO, followed by exposure to
   PBS or H[2]O[2] (100 μM) (n = 6). ∗∗∗∗P < 0.0001. (E) Representative
   immunofluorescence images of IRF7, P16, and P21 expression, along with
   SA-β-gal staining, in VSMCs treated with myricetin (10 μM) or DMSO,
   followed by exposure to PBS or H[2]O[2] (100 μM). The slides were
   co-stained with the SMC marker α-SMA and nuclear stain DAPI. (F)
   Quantification of IRF7, P16, P21 fluorescence intensity, and
   SA-β-gal-positive areas in the indicated groups (n = 7–10).
   ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001.

3.8. Myr treatment significantly inhibits AAA formation In vivo

   To explore the potential therapeutic effects of Myr in AAA, we first
   administered a gradient of Myr concentrations via daily intraperitoneal
   injections to mice for 7 consecutive days ([213]Supplementary Fig.
   16A). Results showed that a dose of 10 mg/kg effectively inhibited
   CD147 protein expression in vascular tissues without causing any
   noticeable toxicity ([214]Supplementary Fig. 16B). Subsequently, we
   performed AngII infusion in ApoE^−/− mice and administered 10 mg/kg/day
   of Myr via intraperitoneal injection for 28 days ([215]Fig. 8A).
   Compared to the vehicle-treated group, Myr treatment improved the
   survival rate of aneurysmal mice ([216]Fig. 8B), and reduced
   AngII-induced aortic rupture, AAA formation, and the ratio of total
   aortic weight to body weight in ApoE^−/− mice ([217]Fig. 8C-F).
   Ultrasound analysis confirmed that the maximal aortic diameter was
   significantly lower in Myr-treated mice than that in Vehicle-treated
   littermates infused with AngII ([218]Fig. 8G). Morphologically,
   histological analysis revealed that Myr treatment obviously mitigated
   elastin degradation and pathological collagen deposition in
   AngII-administered mouse aortas ([219]Fig. 8G).

Fig. 8.

   [220]Fig. 8
   [221]Open in a new tab

   Administration of the CD147 natural inhibitor myricetin protects
   against AAA formation. (A) Schematic protocol: ApoE^−/− mice were
   infused with AngII or saline and intraperitoneally injected with
   vehicle or Myricetin (10 mg/kg/day) for 28 days. (B) Survival curves of
   the vehicle- or Myr-treated ApoE^−/− mice infused with AngII
   (n = 18–20). Data were analyzed using the Kaplan-Meier method and
   compared by log-rank tests. (C) Incidence of AngII-induced aneurysm
   rupture in the indicated groups (n = 18–20). (D) Representative
   macroscopic images of AAA formation in the indicated groups. (E)
   Incidence of AngII-induced AAA in the indicated groups (n = 18–20).
   Data were analyzed using Fisher’s exact test. ∗P < 0.05; ∗∗∗P < 0.001.
   (F) Quantification of total aortic weight/body weight (BW) ratio in the
   indicated groups (n = 6–9). ∗∗P < 0.01; ∗∗∗∗P < 0.0001. (G) Left:
   Representative ultrasound and histopathological images of abdominal
   aortas in mice after 28 days of saline or AngII infusion. Right:
   Quantification of the maximal diameter of abdominal aortas measured by
   B-mode ultrasound imaging (n = 6–11), elastin degradation, and collagen
   content based on histopathological staining (n = 6–7). ∗P < 0.05;
   ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001.

   Compared with the vehicle group, Myr treatment significantly
   upregulated the expression of VSMC contractile phenotype markers in
   abdominal aortas upon AngII stimulation ([222]Fig. 9A and B).
   Consistently, Myr administration markedly reduced the protein levels of
   pro-inflammatory cytokines (TNF-α and IL-6) in the vascular wall and
   suppressed the expression of type I interferons (IFN-α and IFN-β) as
   well as the senescence marker P21 in aneurysmal tissues ([223]Fig.
   9C-F). Furthermore, Myr treatment robustly attenuated AngII-induced MMP
   activation both in vivo and in cultured VSMCs ([224]Fig. 9G-H).

Fig. 9.

   [225]Fig. 9
   [226]Open in a new tab

   Administration of the CD147 natural inhibitor Myricetin attenuates
   AngII-induced VSMC phenotypic transition, IFN-I response, and MMP
   activation. (A–B) Western blot and quantification of STAT2, IRF7,
   α-SMA, SM22, CNN1, and MYH11 protein levels in abdominal aortas in the
   indicated groups of mice on day 28 following AngII infusion.
   ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. (C–D) Representative
   immunohistochemistry images and quantification of typical
   matrix-degrading enzyme MMP2 and inflammatory markers (IL-6, TNF-α)
   protein expression levels in aneurysmal tissues from the indicated
   groups of mice on day 28 following AngII infusion (n = 5). ∗P < 0.05;
   ∗∗∗P < 0.001. (E–F) Representative immunofluorescence images and
   quantification of fluorescence intensity for protein expression of
   IFN-I components (IFNα, IFNβ) and senescence marker (P21) in aneurysmal
   tissues from the indicated groups of mice on day 28 following AngII
   infusion (n = 5). ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. (G) Representative
   images and quantification of active MMP in aneurysmal tissues from the
   indicated groups of mice on day 28 following AngII infusion (n = 6).
   ∗P < 0.05. (H) MMP activity measured by gelatin zymography in isolated
   VSMCs treated with AngII (1 μM). ∗∗∗∗P < 0.0001.

   In the CaPO[4]-induced AAA model ([227]Supplementary Fig. 16C), Myr
   treated mice exhibited markedly attenuated aneurysm formation, reduced
   maximal diameter, less elastic fiber degradation and collagen
   accumulation of the infrarenal abdominal aorta compared with Vehicle
   group, ([228]Supplementary Fig. 16D–16H). These data indicate that Myr
   significantly mitigates the occurrence and progression of AngII- and
   CaPO[4]-induced AAA in vivo, highlighting its potential clinical
   translational value for the treatment of AAA.

4. Discussion

   This study identified a previously unrecognized role of VSMC-derived
   CD147 in the pathogenesis of AAA and VSMC phenotypic transition,
   providing several novel insights. First, CD147 was highly expressed in
   the nucleus of VSMCs and exhibited remarkable nuclear translocation
   upon pathological stimuli in a ROS-dependent manner, where it
   transcriptionally activated the IFNα/β signaling pathway through
   interaction with the STAT1/STAT2 transcriptional complex. This
   ROS-activated CD147-IFN-I axis promoted the phenotypic transition of
   healthy contractile VSMCs into pro-inflammatory and senescent states.
   Second, VSMC-specific knockout of CD147 protected mice against AAA
   formation in both AngII- and CaPO[4]-induced AAA models. Third,
   supplementation with myricetin, a natural CD147 inhibitor abundantly
   found in flavonoid-rich foods, effectively inhibited H[2]O[2]-induced
   VSMC phenotypic transition in vitro and AngII/CaPO[4]-induced AAA
   formation in vivo. Collectively, our study provides the first evidence
   identifying a novel ROS-activated CD147–STAT1/2–type I interferon
   signaling axis, which plays a pivotal role in driving VSMC fate
   transition and promoting AAA pathogenesis.

   Protein subcellular localization is dynamic, cell-specific, and
   functionally diverse, often influenced by tissue and cellular
   microenvironments to meet diverse biological demands [[229]50,[230]51].
   For instance, FGFR3, a transmembrane glycoprotein typically localized
   to the plasma membrane in normal cells, is aberrantly expressed in the
   nucleus of pancreatic cancer cells, where its nuclear presence has been
   associated with poor prognosis [[231]52]. Beyond its pyruvate kinase
   activity in the cytoplasm, PKM2 also translocates to the nucleus in
   response to various signals, functioning as a transcriptional regulator
   by acting as a protein kinase [[232]53,[233]54]. Similarly, as a highly
   glycosylated protein, CD147 has been reported predominantly localized
   to the cytoplasm and plasma membrane in most mammalian cells [[234]55].
   In this study, we provide the evidence that CD147 is highly expressed
   in the nucleus of VSMCs in an NLS ‘KRRK’-dependent manner.
   Interestingly, earlier study suggested that in hepatocellular carcinoma
   cells, a C-terminus domain of CD147 could be cleaved by γ-secretase and
   translocated into the nucleus [[235]56]. However, our data suggest that
   nuclear CD147 in VSMCs likely exhibits as the full-length protein, as
   its molecular weight is comparable between nuclear and cytoplasmic
   fractions. Moreover, external stimuli, such as PDGF-BB, H[2]O[2], and
   AngII, significantly enhance the nuclear expression of CD147 as a
   result of elevated ROS levels, suggesting a potential regulatory role
   in cellular redox processes. This finding expands our understanding of
   CD147’s cellular dynamics, and elucidating the mechanisms driving CD147
   nuclear translocation could provide the foundation for targeting its
   nuclear-specific functions as a strategy to mitigate VSMC phenotypic
   switching and AAA progression.

   Proper subcellular localization is fundamental to the functionality of
   biomacromolecules, including proteins and RNAs. CD147 is predominantly
   localized to the cytoplasm and plasma membrane in most mammalian cells,
   where it mediates cell-cell interactions [[236]13], lactate transport
   [[237]11], and signal transduction [[238]57]. In contrast, proteins
   highly expressed in the nucleus are typically involved in
   transcriptional regulation, DNA replication and repair, chromatin
   remodeling. In this study, we utilized multi-omics analyses (RNA-seq,
   CUT&Tag, and protein interactome) to uncover a novel signaling pathway
   in which nuclear CD147 interacts with the STAT1/2 complex to
   transcriptionally activate the IRF7-IFNα/β signaling axis under
   H[2]O[2]/AngII exposure in VSMCs. The IFN-I pathway was initially
   identified for its critical antiviral role; however, excessive IFN-I
   production has been implicated in autoinflammatory diseases [[239]58].
   In recent years, this pathway has attracted increasing attention for
   its involvement in a broad range of pathological conditions, including
   cardiovascular and oncological diseases [[240]59,[241]60]. For example,
   cardiomyocytes, rather than immune cells, have been identified as
   primary initiators of the pathological IFN-I response in the infarct
   border zone following myocardial infarction, positioning IFN-I as a
   potential therapeutic target in ischemic cardiomyopathy
   [[242]61,[243]62]. VSMCs not only express and secrete IFN-I in response
   to injury stimuli but also act as IFN-I-responsive cells. Through
   autocrine and paracrine signaling, IFN-I promotes VSMC phenotypic
   switching and senescence [[244]6,[245]9,[246]63]. Similarly, our study
   demonstrated that ROS-activated CD147-type I interferon signaling axis
   orchestrates the phenotypic transition of VSMCs toward a
   pro-inflammatory and senescent state, and VSMC-specific knockout of
   CD147 ameliorates AngII-induced AAA development and suppresses
   pathological VSMC fate transition. Although we did not directly
   manipulate nuclear CD147 in vivo, our in vitro rescue experiments, by
   overexpressing either CD147-WT or CD147-ΔNLS in CD147-deficient VSMCs,
   demonstrated that the pro-phenotypic switching and IFN-I signaling
   activation effects of CD147 largely depend on its nuclear localization
   sequence. These findings suggest that ROS-induced nuclear translocation
   of CD147 likely serves as the key mechanism by which CD147 promotes AAA
   pathogenesis. This discovery not only highlights CD147 as a versatile
   regulator with expanded functional repertoire, but also positions it as
   a mechanistic link between ROS and type I interferon signaling in VSMC,
   thereby bridging the redox-regulated immune-metabolic control of VSMC
   plasticity and AAA pathogenesis. Moreover, it offers a novel
   therapeutic avenue to inhibit VSMC plasticity and attenuate aneurysmal
   development.

   As a promising therapeutic target, CD147 is considered as a key driver
   in various diseases, including cancer [[247]11,[248]64], cardiovascular
   diseases [[249]16], and infectious diseases [[250]15,[251]65]. Despite
   the initial success of monoclonal antibodies and chemically synthesized
   small molecule inhibitors targeting CD147, their widespread clinical
   application remains limited due to concerns over biosafety,
   biocompatibility, high costs and low affinity [[252][66], [253][67],
   [254][68], [255][69]]. Natural products, however, offer advantages such
   as high bioavailability, low toxicity, and diverse pharmacological
   activities, making them an important resource for the development of
   targeted therapies for human diseases [[256]70]. Myricetin, a natural
   flavonoid abundant in various plants or foods and available in dietary
   supplement form, is currently the small molecule with the highest known
   binding affinity for CD147, and has been considered to possess
   considerable medicinal value [[257]49,[258]71]. It exhibits diverse
   pharmacological effects, including antioxidant, anti-inflammatory,
   antimicrobial, anticancer, antidiabetic, and hepatoprotective
   properties [[259]48]. Healthcare products containing myricetin have
   been developed and marketed for their antioxidant and anti-inflammation
   effects, although it is not yet approved as a new drug [[260]48].
   Accumulating evidence suggests a strong association between oxidative
   stress and both VSMC phenotypic modulation and AAA development,
   highlighting antioxidant intervention as a promising AAA therapeutic
   approach [[261]7,[262]72]. For example, dietary intake of antioxidant
   vitamins or fruits rich in antioxidative compounds has been shown to
   confer protective effects against AAA [[263]73,[264]74]. Consistently,
   our data show that antioxidant myricetin effectively inhibits oxidative
   stress-induced VSMC phenotypic transition and cellular senescence in
   vitro, as well as AngII- and CaPO[4]-induced AAA formation and
   progression in vivo, at least in part by targeting CD147-type I
   interferon signaling. These findings provide new mechanistic insights
   into antioxidant-based strategies for AAA prevention and treatment.

   Several limitations of this study should be acknowledged. First,
   although ROS may serve as a key initiator, the molecular mechanisms
   underlying CD147 nuclear translocation in VSMCs remain unclear. Further
   investigation into this process may provide novel opportunities to
   directly target nuclear CD147. Second, while our findings suggest
   myricetin as a potential therapeutic agent for AAA, additional clinical
   research is needed to confirm the efficacy and clinical benefits of
   myricetin in AAA treatment. Third, although our study confirmed the
   detrimental role of VSMC-derived CD147 in AAA development using two
   distinct murine models—subcutaneous AngII infusion and adventitial
   CaPO[4] application—both of which have inherent limitations in
   recapitulating the complex pathogenesis of human AAA. Moreover,
   although Tagln-Cre is widely used for VSMC-specific gene manipulation,
   its cellular specificity remains a matter of debate [[265]75]. Given
   that both our study and previous reports support CD147 as a
   therapeutically targetable molecule in AAA [[266]24], the potential
   contributions of CD147 derived from non-VSMC populations warrant
   further investigation.

5. Conclusion

   In summary, our study unveils a previously undescribed ROS-activated
   CD147-STAT1/2-IFNα/IFNβ axis in promoting VSMC inflammatory phenotype
   and senescence, and provides proof-of-concept evidence indicating the
   therapeutic value of targeting CD147 inhibition in AAA prevention and
   treatment.

CRediT authorship contribution statement

   Fangyuan Zhong: Writing – review & editing, Writing – original draft,
   Software, Methodology, Investigation, Data curation, Conceptualization.
   Hengyuan Zhang: Writing – original draft, Software, Methodology,
   Investigation. Xinning Guo: Writing – original draft, Visualization,
   Software, Methodology, Investigation. Yichao Zhao: Writing – review &
   editing, Conceptualization. Yufei Wang: Methodology, Investigation.
   Wenli Li: Methodology, Investigation. Yuyan Lyu: Writing – review &
   editing, Resources, Conceptualization. Heng Ge: Writing – review &
   editing, Visualization, Supervision, Resources, Funding acquisition,
   Conceptualization. Xiyuan Lu: Writing – review & editing, Writing –
   original draft, Validation, Supervision, Conceptualization. Jun Pu:
   Writing – review & editing, Writing – original draft, Supervision,
   Resources, Funding acquisition, Conceptualization.

Declaration of competing interest

   There are no conflicts of interest to declare.

Acknowledgments