Human serum and aortic tissue samples were collected at the General Hospital of Northern Theater Command (Shenyang, China) between January 20, 2021, and January 19, 2023. Serum samples were obtained from 40 participants, including 20 healthy individuals (aged 40±10 years) and 20 patients diagnosed with AAA (aged 50±10 years). During the same period, aortic tissue samples were collected from patients diagnosed with aortic dissection (AD) who underwent open surgical repair at the same hospital. Patients with valvular heart disease, chronic kidney disease, autoimmune disease, or other cardiopulmonary organic diseases were excluded. All participants were men, and there were no significant differences in age, diabetes mellitus, hypertension, or smoking between the control and AAA or AD groups. Blood samples were obtained from all participants; serum was obtained and stored at −80°C. The present study was conducted in accordance with the World Medical Association Code of Ethics (Declaration of Helsinki) [approval no. Y(2021)002] and was approved by the Ethics Committee of General Hospital of Northern Theater Command. Written informed consent was obtained from all individual participants included in the study. All participants consented to the use of their surgically resected tissue samples and anonymized clinical data for scientific research.

A total of 198 male 8-weeks-old ApoE^−/− mice, with body weights ranging from 20- 25 g, were purchased from GemPharmatech Co., Ltd. The experiment started after 1 week of acclimatization. All mice were housed under temperature-controlled (22±1-2°C) and specific pathogen-free conditions on a 12/12-h day/night cycle with free access to food and water.

All experiments were approved by the Subcommittee on Animal Medical Research Ethics, General Hospital of Northern Theater Command (approval no. 2022-20; Shenyang, China) and conducted in accordance with the existing guidelines on the care and use of laboratory animals. All animal care and experimental protocols complied with the National Institute of Health Guide for the Care and Use of Laboratory Animals.

Animal experiments in the present study were divided into four parts. In the first part of the in vivo experiments, a total of 30 ApoE^−/− mice were randomly divided into four groups: i) Angomir-negative control (NC) + saline (n=5); ii) Angomir-NC + Ang II (n=10); iii) Angomir-378a-5p + saline (n=5); iv) Angomir-378a-5p + Ang II (n=10). In the second part of the in vivo experiments, a total of 40 ApoE^−/− mice were randomly divided into four groups: i) Antagomir-NC + saline (n=5); ii) Antagomir-NC + Ang II (n=15); iii) Antagomir-378a-5p + saline (n=5); iv) Antagomir-378a-5p + Ang II (n=15). In the third part of the in vivo experiments, a total of 60 ApoE^−/− mice were randomly divided into four groups: i) AAV-SM22-shNC + saline (n=5); ii) AAV-SM22-short hairpin (sh)NC + Ang II (n=25); iii) AAV-SM22-sh Ablim1 + saline (n=5); iv) AAV-SM22-shAblim1 + Ang II (n=25). In the fourth part of the in vivo experiments, a total of 68 ApoE^−/− mice were randomly divided into four groups: i) AAV-SM22-NC + saline (n=9); ii) AAV-SM22-NC + Ang II (n=25); iii) AAV-SM22 + Ablim1-saline (n=9); iv) AAV-SM22-Ablim1 + Ang II (n=25).

To minimize animal suffering, the following predefined humane endpoints were strictly observed, and any animal meeting one or more criteria was euthanized immediately: i) sustained weight loss exceeding 20% of baseline body weight within 72 h; ii) Impaired mobility or inability to access food or water autonomously; iii) Signs of severe distress or pain unrelieved by analgesia; iv) Ulceration, necrosis, or exceeding a tumor volume of 1.5 cm³ in any dimension in tumor-bearing models and v) Clinical signs indicating severe systemic illness. Animals were monitored daily, and any mouse reaching a predefined humane endpoint was euthanized humanely by CO₂ inhalation. No animals in the present study reached the predefined humane endpoints prior to the scheduled experimental endpoint. All animals were euthanized at the planned conclusion of the study for tissue collection, in accordance with the approved protocol.

ApoE^−/− mice (8-week-old) were infused with saline (0.9% sodium chloride) or Ang II (1,000 ng/kg/min) using an osmotic pump (Alzet model 1004; AlzaCorp; https://alzet.com/products/alzet_pumps/) for 28 days. Ang II (A1042) was purchased from APeXBIO Technology LLC. Briefly, mice were anesthetized with 2-3% isoflurane (RWD Life Science) in oxygen. During surgery, anesthesia was maintained with 1.5-2% isoflurane. Minipumps were implanted into the subcutaneous space of the mice at the back of the neck. All surgeries were performed under aseptic conditions. Animals were monitored daily for signs of distress.

At the experimental endpoint (Day 28), mice were euthanized by CO₂ inhalation at a flow rate displacing 30-70% of the chamber volume per min, followed by secondary cervical dislocation. Death was confirmed by the absence of heartbeat (via palpation), cessation of breathing, and fixed, dilated pupils. Aortic tissues were then harvested for subsequent analysis. The aortic diameters were measured using small animal ultrasound (Visual Sonics) at 28 days after Ang II infusion. Aortic tissues were harvested for RNA, protein, morphological and histological analyses. AAA incidence was defined as an increase in the external aortic diameter by 50% or greater than that in the aortas from saline-infused mice.

Blood pressure was measured in conscious mice using a non-invasive tail-sleeve system (CODA-6; Kent Scientific). The mice were placed in a holding tube within a heating chamber set at 37°C (Model LE5510; Panlab). All animals were acclimatized to the instrument for at least 1 week before baseline measurements and osmotic pump implantation. To avoid changes in blood pressure due to circadian cycles, all measurements were taken between 8 a.m. and 10 a.m. Each mouse received 10 initial pressure assessments so that they can adapt to the procedure, and then 10 additional readings were recorded to obtain average systolic and diastolic blood pressure. The acceptable standard was to consider at least 10 of the 20 acquired measurements and a standard deviation (SD) of <30 mmHg for each session (34).

As previously described (35), miRNA expression in mouse aortas was analyzed by Bio-Miao Biological Technology Co., Ltd. using the Agilent Mouse miRNA Microarray Kit, Release 21.0,8×60 K (Design ID: 070,155; Agilent Technologies, Inc.), which contained 1,902 probes for mature miRNA (35). Tukey's bi-weight average (log2) intensity was analyzed using the analysis of variance (ANOVA). Differentially expressed miRNAs were defined as those exhibiting an absolute average log2 fold change of ≥2.0 and an adjusted P<0.05.

Conditions of amplification reactions were as follows: 95°C for 15 min, followed by 40 cycles of 95°C for 30 sec, 55°C for 1 min, and 72°C for 30 sec. Peripheral venous blood was collected from patients into EDTA-containing tubes and centrifuged at 3,000 × g for 10 min at 4°C, and serum was stored at −80°C. Total RNA was extracted from 200 μl serum samples using the miRNeasy Serum/Plasma kit (Qiagen GmbH) and cDNA was synthesized using a reverse transcription kit (Guangzhou RiboBio Co., Ltd.) according to the manufacturer's instructions. After equal volume dilution of cDNA with DNase/RNase-free deionized water, the expression levels of miRNAs were evaluated by RT-qPCR using specific primers and the miRNA RT-qPCR kit (Guangzhou RiboBio Co., Ltd.) following the manufacturer's protocol. Reactions were performed on a CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.). Each reaction was performed in triplicate, and the relative expression level of miRNAs were calculated based on cycle threshold (Ct) values using the following formula: 2^-∆∆Cq (36). The expression levels of miRNAs in the tissues and cells were normalized to the expression levels of U6 snRNA, and those in the serum were normalized to the expression levels of the external reference cel-miR-39-3p.

Besides, total RNA from aortic tissues and VSMCs was extracted using TRIzol reagent (Qiagen). The isolated RNA was used to synthesize cDNA using PrimeScript RT with a gDNA Eraser kit (Takara), RT-qPCR was performed in triplicate using SYBR Premix Ex Taq II (Takara Bio, Inc.). The primer sequences are shown in Table SI.

The probe 5'-CUCCUGACUCCAGGUCCUGUGU'-3 was labeled by FAM and Cy3 and were synthesized from the sequence of hsa-miR-378a-5p. In situ hybridization was conducted according to the instructions of the FISH Detection Kit (Qiagen). In brief, 5-μm of human aorta tissue was digested with proteinase K and incubated in a blocking buffer for 30 min (37°C). The FAM and Cy3-labeled hsa-miR-378a-5p fluorescent probe working solutions were prepared at a volume ratio of 1:1. The aortic tissue slices were incubated for 14 h (37°C) using an in-situ hybridization instrument. The sections were then washed with deionized formamide at 43°C to denature the unhybridized probes. The sections were washed three times with sodium citrate buffer (60°C). FISH images were then captured by confocal microscopy.

Angomir-378a-5p and angomiR-NC were purchased from MedChemExpress. 8-week-old ApoE^−/− mice were injected with angomir-378a-5p or angomiR-NC via the tail vein 3 times per week for 4 weeks (20 nmol each time). Additionally, antagomir-378a-5p and antagomir-NC were purchased from Guangzhou RiboBio Co., Ltd. 6-week-old ApoE^−/− mice were injected with antagomir-378a-5p and antagomir-NC via the tail vein 3 times per week for 4 weeks (50 nmol each time).

Adeno-associated virus serotype 2/9 (AAV2/9) carrying the Ablim1 coding sequence (CDS) with a Sm22α promoter (AAV-SM22-Ablim1) or the control virus (AAV-SM22-NC) were constructed by OBiO Technology Corp., Ltd. 6-week-old ApoE^−/− mice were injected with AAV-SM22-Ablim1 or the control virus (AAV-SM22-NC) via the tail vein at a dosage of 5×10¹¹ vg per mouse. A total of 3 weeks later, aortas were collected to access overexpression performance.

To achieve VSMCs-specific knockdown of Ablim1, adeno-associated virus serotype 2/9 (AAV2/9) carrying mouse shAblim1 sequence with a Sm22α promoter (AAV-SM22-shAblim1) or the control virus (AAV-SM22-shNC) were constructed by OBiO Technology Corp., Ltd. 6-week-old ApoE^−/− mice were injected with AAV-SM22-shAblim1 or the control virus (AAV-SM22-NC) via the tail vein at a dosage of 5×10¹¹ vg per mouse. After 3 weeks of injection, aortas were harvested for evaluation of Ablim1 knockdown efficiency.

To check the incidence of AAA, three representative parameters including the diameters of the aorta, superior renal artery and inferior renal artery were examined using a small animal ultrasound system (Vevo 2100 apparatus; Visual Sonics) equipped with a 30-MHz probe. Each mouse was anesthetized with 2% isoflurane throughout the ultrasonographic procedure. Arterial diameters were assessed by a blinded researcher.

Mice were euthanized and the whole aortas were perfused with saline and fixed with 4% paraformaldehyde for 24 h at room temperature. The aortas were isolated from the ascending aorta to the entrances of both iliac arteries for macroscopic analysis. The aortas were then segmented to obtain suprarenal abdominal aortas. The aortic samples were harvested, fixed for 24 h, and embedded in paraffin. Histology was examined in cross sections (5-μm) that were taken from these aortic samples. Paraffin-embedded sections were used for staining. Hematoxylin and eosin (H&E) staining was used for morphological evaluation. Sirius Red staining (Beijing Solarbio Science & Technology Co., Ltd.) was used to evaluate collagen deposition, and Verhoeff Van Gieson Elastic staining (MilliporeSigma) was used to evaluate elastin. Elastin degradation was scored as follows: 1, no degradation and well-organized lamina; 2, mild degradation with some interruptions or breaks in the lamina; 3, moderate degradation with multiple interruptions or breaks in the lamina; and 4, severe fragmentation, loss, or aortic rupture.

Paraffin-embedded sections (5-μm) were deparaffinized in xylene and rehydrated through a graded ethanol series (100, 95, 70%) to distilled water and blocked with 5% goat serum (cat. no. X0907; Fuzhou Maixin Biotechnology Co., Ltd.) for 1 h at room temperature. The sections were incubated with anti-ABLIM1 (1:200; cat. no. PA5-70451; Thermo Fisher Scientific, Inc.), anti-α-SMA (1:200; cat. no. Ab7817; Abcam) primary antibodies overnight at 4°C. After washing with phosphate-buffered saline (PBS) 3 times, Alexa Fluor 488/594-conjugated secondary antibodies (1:200; cat. nos. A-11012 and A-11008; Thermo Fisher Scientific, Inc.) were applied for 1 h at 37°C in the dark. Sections were mounted using ProLong Gold anti-fade reagent with 4',6-diamidino-2-phenylindole (DAPI; cat. no. P36931; 1 μg/ml; Thermo Fisher Scientific, Inc.) for fluorescence microscopy (Carl Zeiss AG).

Primary mouse VSMCs were isolated from the freshly dissected aortas. Briefly, the tunica media were separated by peeling off the tunica intima from the aortic tissue in PBS. The tunica media were cut into ~2-mm pieces and digested with 0.15% type II collagenase (MilliporeSigma). VSMCs were seeded in 6-well plates and cultured in fresh Dulbecco's modified Eagle's medium (DMEM) containing 20% fetal bovine serum (FBS; Biochrom Ltd.). Positive immunofluorescence staining of α-SMA and SM22-α were used to confirm VSMCs. VSMCs were stimulated with different concentrations of tumor necrosis factor α (TNFα; MilliporeSigma) for 24 h.

miR-378a-5p mimics, miR-378a-5p inhibitor, and matched controls were purchased from Guangzhou RiboBio Co., Ltd. The concentration of miR-378a-5p mimics was 50 nM and miR-378a-5p inhibitor was 100 nM. Ablim1 and Mkl1 small interfering RNA (siRNA) and matched controls were purchased from Shanghai GenePharma Co., Ltd. The concentration of siRNA was 50 nM. Mouse Ablim1 overexpression plasmid (pcDNA3.1-Flag-Ablim1), cellular Myelocytomatosis oncogene (c-Myc) overexpression plasmid (pcDNA3.1-Flag-c-Myc), and Mkl1 overexpression plasmid (pcDNA3.1-His-Mkl1) were designed and constructed by the OBiO Technology Corp., Ltd. The quality of plasmids was 1 μg/ml. Lipofectamine 2000 (Thermo Fisher Scientific, Inc.) was mixed with 250 μl of serum- and antibiotic-free medium, and incubated for 5 min. The siRNA or plasmids and lipofectamine 2000 solutions were mixed and incubated at room temperature for 20 min. The transfection mixture was added to the cells in serum-free culture and incubated for 8 h and then replaced with normal serum and antibiotic-containing growth medium. The cells were incubated for 48 h before collection for testing. Sequences of miRNA mimic, miRNA inhibitor, and small interfering RNA are included in Table SII. It was first demonstrated that VSMCs could be successfully transfected with both siRNAs and overexpression plasmids, which resulted in effective gene silencing and overexpression of the targets (Fig. S1).

Cells and tissues were homogenized in ice-cold suspension buffer (RIPA Lysis Buffer) supplemented with a proteinase inhibitor cocktail (MilliporeSigma). Briefly, protein concentrations were determined using a BCA protein assay kit (Thermo Fisher Scientific, Inc.). Equal amounts (20 μg) of protein were fractionated on SDS polyacrylamide gels and transferred to a polyvinylidene fluoride (PVDF) membrane. The concentration of the acrylamide gel was chosen based on the size of the target proteins. Usually, proteins with a size range from 10 to 30, 30 to 100, and >100 kD are separated on 12, 10 and 8% gels, respectively. After blocking with 5 % non-fat milk at room temperature for 1 h, the membrane was incubated with primary antibodies at 4°C overnight. The membranes followed by immunoblotting with the primary antibodies: anti-α-SMA (1:1,000; cat. no. Ab7817; Abcam), anti-SM22-α (1:1,000; cat. no. Ab14016; Abcam), anti-CNN1 (1:1,000; cat. no. Ab46794; Abcam), anti-MMP2 (1:1,000; cat. no. Ab92536; Abcam), anti-ABLIM1 (1:1,000; cat. no. PA5-70451; Thermo Fisher Scientific, Inc.), anti-MKL1 (1:1,000; cat. no. Ab219981; Abcam), anti-Flag (1:1,000; cat. no. sc-166355; Santa Cruz Biotechnology, Inc.), anti-His (1:1,000; cat. no. 66005-1-Ig; Proteintech Group, Inc.), anti-c-MYC (1:1,000;. cat. no. 10828-1-AP; Proteintech Group, Inc.), anti-MYOD1 (1:1,000; cat. no. 18943-1-AP; Abcam) and anti-β-actin (1:100; cat. no. sc-517582; Santa Cruz Biotechnology, Inc.). The membranes were then incubated with mouse or rabbit appropriate peroxidase-conjugated secondary antibody (1:1,000; cat. no. 32460; Thermo Fisher Scientific, Inc.) for 1 h at room temperature. Specific protein bands were detected using ECL detection reagent (Beijing New Create Life Science Biotechnology Co., Ltd.) and specific protein bands were visualized by enhanced chemiluminescence using Amersham Imager 680 (Cytiva).

The cells in each group were digested using trypsin, and were resuspended using DMEM without serum to adjust the cell density to 1×10⁵ cells/ml, and 300 μl of cell suspension was added to the upper compartment of the 0.8-μm Transwell chamber (Corning, Inc.); meanwhile, 700 μl of DMEM containing 10% FBS was added to the bottom compartment and the cells were cultured at 37°C, 5% CO₂ for 24 h. After the chamber was removed, the cells on the bottom of the membrane were fixed with 4% paraformaldehyde for 15 min at room temperature, stained with 0.1% crystal violet solution for 15 min, and the remaining crystal violet solution was washed off using PBS; the cells in the upper chamber were cleaned using a cotton swab. A total of five fields of the membrane were used to count the number of cells using an optical microscope, and the average was calculated to indicate the migration ability of the VSMCs.

VSMCs were cultured in six-well plates. A scratch was made using 10-μl pipette tips. After washing to remove cell debris, the medium was replaced with FBS-free DMEM and incubated for 24 h at 37°C. Migratory cells were visualized using a phase-contrast microscope (Olympus Corporation).

The 3' untranslated region (3'UTR) of mouse Ablim1, Ddx5 (Dead-box helicase 5), Slc7a1(Cationic amino acid transporter 1) gene was amplified and cloned into the pMIR-REPORT Luciferase (OBiO Technology Company) to construct pMIR-REPORT Luciferase 3'UTR wild-type (WT) plasmids. A mutation was introduced at the seed sequence of miR-378a-5p to create pMIR-REPORT Luciferase-3'UTR mutated (MUT) plasmids. 293T cells were transfected with miR-378a-5p mimic or control (50 nM), together with pMIR-REPORT Luciferase-3'UTR (WT) (1,000 ng) or pMIR-REPORT Luciferase-3'UTR (MUT) (1,000 ng) and pRL (Renilla)-CMV (500 ng) for 48 h. Transfection was performed with Lipofectamine 2000. At 48 h after transfection, luciferase activity was measured using a dual-luciferase analysis system kit (Promega Corporation). The ratio of luciferase activity/Renilla Luciferase activity was normalized to the control and presented as the relative transcriptional activity.

JASPAR (https://jaspar.elixir.no/) and TRANSMIR (http://cmbi.bjmu.edu.cn/transmir) were used to predict the upstream transcription factors of miR-378a-5p. TargetScan7.0 (http://www.targetscan.org/) and miRDB (http://www.mirdb.org) databases were used to predict the target genes of miR-378a-5p. Finally, intersecting genes were obtained from the two databases for subsequent analysis.

A total of 3 datasets of AAA and control aortic samples were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). GSE183464 is an RNA sequencing analysis of abdominal aorta tissues from 14 participants, including 7 patients with AAA and 7 control individuals. GSE237229 is an RNA sequencing dataset of human aortic SMCs isolated from 5 patients with AAA and 3 non-AAA donors. The human thoracic aortic aneurysm (TAA) single-cell RNA sequence (scRNA-seq) dataset (GSE155468) was also downloaded from the GEO database.

Co-IP was performed to validate the interaction between ABLIM1 and MKL1. Briefly, 293T cells were transfected with pcDNA3.1-Flag-Ablim1 and pcDNA3.1-His-Mkl1 for 48 h and lysed in 1 ml IP lysis containing protease and phosphatase inhibitors. The cell lysate was collected and centrifuged at 12,000 × g for 15 min at 4°C, followed by incubation with 50 μl magnetic beads (MBL International Co.) as suggested at 4°C for 1 h. The tube was placed on a magnetic rack for a few sec, and the supernatant was removed. A total of 1 ml of cold wash buffer was added, and the magnetic beads were resuspended 3 times. The magnetic beads were resuspended in the loading buffer and heated for 5 min, and the tube was placed on a magnetic rack for a few sec. Total or separate cell contents and immunoprecipitants were separated using SDS-PAGE gels. Consistent with the aforementioned description, the concentration of the acrylamide gel was chosen based on the size of the target proteins. The proteins were transferred onto PVDF membranes and incubated with the corresponding primary antibodies at 4°C overnight. The primary antibodies were as follows: anti-FLAG (1:1,000; cat. no. sc-166355; Santa Cruz Biotechnology, Inc.), anti-HIS (1:1,000; cat. no. 66005-1-Ig; Proteintech Group, Inc.), anti-ADM (1:1,000; cat. no. Ab190819; Abcam), anti-LMOD1 (1:1,000; cat. no. 15117-1-AP; Proteintech Group, Inc.) and anti-PRKCD (1:100; cat. no. sc-365969; Santa Cruz Biotechnology, Inc.). HRP-conjugated secondary antibodies were added, and the proteins were examined using a chemiluminescence imaging system.

ABLIM1 antibody was added to 293T lysates for IP. Rabbit IgG was used as a negative control. Then, the LC-MS/MS analysis was carried out by PTM Bio Co., Ltd. Finally, the substrate proteins that could bind to ABLIM1 were screened out according to the score and the mass of detected proteins.

All data are presented as the mean ± standard error of the mean. P<0.05 was considered to indicate a statistically significant difference. Statistical analyses were performed using GraphPad Prism 9.3 (GraphPad Software Inc.; Dotmatics). For statistical comparisons, it was first evaluated whether the data were normally distributed using the Shapiro-Wilk normality test. Non-parametric tests were used when data were not normally distributed. Normality tests were performed using Shapiro-Wilk statistics. Differences between the two groups were compared using the unpaired Student's t-test. Differences between three or more groups were compared using one-way ANOVA, followed by Tukey's post-hoc test for multiple pairwise comparisons. The Kaplan-Meier survival curve was used to analyze the survival percentage of saline- or Ang II-infused mice.

To establish a mouse AAA model, 8-week-old ApoE^−/− mice were infused with Ang II via an osmotic pump for 28 days. Mice infused with saline served as the control group. To identify miRNAs associated with AAA, the miRNA array was performed to determine the differentially expressed miRNAs in the aortas of AAA and control mice. Compared with the control group, 64 differentially expressed miRNAs were found in the aortas of AAA group mice, of which 27 miRNAs were upregulated, and 37 miRNAs were downregulated. The heatmap illustrated the top 10 miRNAs that exhibited the most significant decrease, with those highlighted in red indicating a high degree of homology between human and mice (miR-200b-3p, miR-200c-3p, miR-342-5p, miR-149-5p, miR-150-3p and miR-378a-5p) (Figs. 1A and B and S2A). The results of RT-qPCR demonstrated that miR-200b-3p, miR-200c-3p, miR-342-5p, miR-150-3p, and miR-378a-5p were significantly decreased in the aortas of AAA mice compared with control mice (Fig. 1C). To further clarify whether these miRNAs played a role in AAA, serum samples were collected from 20 patients with AAA and 20 healthy individuals. Compared with normal group, the expression of miR-378a-5p in the serum levels of patients with AAA were reduced; however, the expression levels of miR-200b-3p, miR-200c-3p, miR-342-5p and miR-150-3p were unchanged between the two groups (Fig. 1D). Furthermore, compared with the saline group, circulating miR-378a-5p level was significantly reduced in the serum of AAA mice (Fig. 1E).

VSMCs were isolated and identified by immunofluorescence staining using SM22-α and α-SMA (Fig. S2B). TNFα is a well-established direct suppressor of VSMCs contractile phenotype, rapidly downregulating contractile genes and impairing function (37). To simulate the pathological process of AAA in vitro, TNFα was used to stimulate VSMCs. As shown in Fig. S2C and D, TNFα inhibited the protein expression levels of VSMCs differentiation markers (α-SMA, CNN1 and SM22-α) and increased the protein level of metal matrix degrading enzyme 2 (MMP2) in a dose dependent manner. TNFα also significantly decreased miR-378a-5p levels in VSMCs. However, the levels of miR-149-5P, miR-200b-3p, miR-200c-3p, miR-342-5p and miR-150-3p were unchanged in TNFα-treated VSMCs compared with control group (Fig. 1F). Besides, miR-378a-5p fluorescence was also decreased in human AD aortas by FISH analysis (Fig. 1G).

To identify the biological functions of miR-378a-5p and its target genes in AAA pathogenesis, miRNA-associated signaling pathways were analyzed using the DIANA TOOLS-miRPath algorithm (http://www.microrna.gr/miRPathv4) (38). For the target genes of miR-378-5p, most biological processes enriched in cell migration, differentiation, RNA synthesis and degradation (Fig. 1H-J). These findings suggested that miR-378a-5p may play a critical role in AAA pathogenesis.

To clarify why miR-378a-5p was downregulated in AAA formation, two transcription factor prediction websites, JASPAR and TRANSMIR, were used to predict the transcription factors of miR-378a-5p. Two possible transcription factors were identified: c-MYC and myogenic differentiation 1 (MYOD1) (Fig. S2E). Next, upregulation of c-MYC protein expression was detected in TNFα-treated VSMCs, but the expression levels of MYOD1 had no significant changes (Fig. S2F). Besides, the increase of c-MYC was also detected in human AD tissues (Fig. S2G). Luciferase assays revealed that c-MYC-dependent miR-378a-5p suppression was maintained upon transfection with a luciferase vector containing the miR-378a-5p promoter, and this inhibitory effect was reversed by mutations in the miR-378a-5p promoter region (Fig. S2H). C-MYC knockdown resulted in a significant increase in miR-378a-5p levels in VSMCs, whereas c-MYC overexpression decreased miR-378a-5p levels in VSMCs (Fig. S2I and J). These results indicated that c-MYC directly bound to the promoter of miR-378a-5p and negatively regulated miR-378a-5p expression in VSMCs.

ApoE^−/− mice were injected with angomir-NC or angomir-378a-5p for 4 weeks through tail vein before AAA modeling (Fig. S3A). No difference in body weight was observed between all the groups (Fig. S3B). Blood pressure increased similarly upon Ang II infusion in both angomir-NC and angomir-378a-5p-injected mice (Fig. S3C and D). Compared with the angomir-NC group, the expression of miR-378a-5p was significantly increased in the aortas of the angomir-378a-5p group (Fig. S3E).

In the presence of Ang II, mice in the angomir-NC and angomir-378a-5p groups developed aortic dilations and aneurysms, which were mitigated in the angomir-378a-5p group (Fig. 2A). miR-378a-5p overexpression blunted the AAA incidence induced by Ang II (Fig. 2B); however, there was no significant difference in the survival rates among all groups (Fig. S3F). All images of the abdominal aortic specimens were displayed (Fig. S3G). Compared with the angomir-NC group, miR-378a-5p overexpression inhibited the aortic enlargement at 28 days post-Ang II infusion (Fig. 2C and D). H&E staining indicated the alleviated aortic dilatation in angomir-378a-5p mice in response to Ang II. Concomitantly, the collagen deposition and media degeneration were significantly reduced in angomir-378a-5p mice in response to Ang II (Fig. 2E-G). The results of western blotting demonstrated that Ang II reduced the protein expression levels of contractile markers (α-SMA, CNN1 and SM22-α) and elevated protein expression of MMP2 in the aortas, indicating that VSMCs dedifferentiation occurred in AAA formation, and miR-378a-5p overexpression could inhibit the dedifferentiation of VSMCs (Fig. 2H and I). These results indicated that miR-378a-5p overexpression effectively alleviates Ang II-induced AAA formation.

Given the aforementioned data, male ApoE^−/− mice were injected with a locked nucleic acid-modified antagomir-378a-5p or a scrambled mir control (antagomir-NC). At 4 weeks after the injection, the mice were treated with Ang II or saline for an additional 4 weeks (Fig. S4A). Body weight remained consistent across all groups (Fig. S4B). Ang II treatment group had a significant increase in blood pressure, but there was no significant difference between the antagomir-378a-5p + Ang II and antagomir-NC + Ang II groups (Fig. S3C and D). Antagomir-378a-5p significantly reduced the expression of miR-378a-5p in the aortas (Fig. S4E). The incidence of AAA in the antagomir-NC + Ang II group was 60%, and the incidence of AAA increased by 20% in antagomir-378a-5p + Ang II group (Fig. S4F). There was no significant difference in the survival rate or incidence of AAA between antagomir-NC + Ang II and antagomir-378a-5p + Ang II groups (Fig. S4G).

All images of the abdominal aortic specimens were displayed in Fig. S4H. Mice with antagomir-378a-5p treatment were susceptible to aortic dilation and exhibited severe aneurysms after Ang II treatment (Fig. 3A). Ultrasound images revealed that the maximum abdominal aortic diameter was increased by antagomir-378a-5p injection (Fig. 3B and C). Compared with the antagomir-NC + Ang II group, increased collagen disruption and elastin degradation were observed in the abdominal aortas of the antagomir-378a-5p + Ang II group (Fig. 3D-F).

Western blot demonstrated that the protein expression levels of α-SMA, CNN1 and SM22-α were significnalty downregulated in the antagomir-378a-5p + Ang II group compared with the antagomir-NC + Ang II group, while the protein expression of MMP2 was significantly higher in the antagomir-378a-5p + Ang II group (Fig. 3G and H). These results revealed that the inhibition of miR-378a-5p aggravated Ang II-induced AAA.

The loss of VSMCs in the medial layer of the aortic wall is an early hallmark of AAA development (39). To further evaluate the cellular effects of miR-378a-5p on VSMCs function, miR-378a-5p mimics were transfected into mouse primary VSMCs. miR-378a-5p mimics significantly increased the expression of miR-378a-5p in VSMCs (Fig. S5A).

TNFα decreased the protein expression of the contractile genes including α-SMA, SM22-α and CNN1 in VSMCs, and increased the protein expression of MMP2. With or without TNFα stimulation, miR-378a-5p overexpression significantly increased the protein expression of VSMCs contractile markers, and decreased MMP2 protein expression (Fig. 4A). The results of RT-qPCR were consistent with those of western blotting (Fig. 4B).

The balance between contractile and synthetic VSMCs shifts toward synthetic VSMCs, facilitating the detachment of VSMCs from the ECM and promoting VSMCs' migration (40). Therefore, it was also examined whether miR-378a-5p influenced VSMCs' migration. miR-378a-5p overexpression could inhibit migration of VSMCs induced by TNFα stimulation (Figs. 4C and S5B).

To further evaluate the effects of miR-378a-5p inhibitor on VSMCs function, VSMCs were transfected with miR-378a-5p inhibitors. miR-378a-5p inhibitor significantly reduced the expression of miR-378a-5p in VSMCs (Fig. S5A).

With or without TNFα stimulation, the miR-378a-5p s inhibitor significantly reduced the mRNA and protein expression of VSMCs' contractile markers (α-SMA, SM22-α and CNN1) and increased the mRNA and protein expression of MMP2 (Fig. 4D and E). Transwell and wound healing assays confirmed that miR-378a-5p inhibitor promoted the migration of VSMCs induced by TNFα (Figs. 4F and S5C).

To elucidate the molecular mechanism by which miR-378a-5p regulated VSMCs' phenotypic transformation, the predicted target candidates of miR-378a-5p were screened in silico. The target genes of miR-378a-5p were predicted using miRDB and starBase. Moreover, two datasets of AAA and control aortic samples were obtained from the GEO database (GSE183464 and GSE237229). GEO2R was used to analyze differentially expressed genes (DEGs) in datasets, and 3,001 DEGs were revealed with the cut-off criterion of adjusted P≤0.05 and |log2 (fold change)≥1, containing 1,473 upregulated and 1,528 downregulated genes in GSE183464. There were 1,009 DEGs in GSE237229, which included 536 upregulated and 473 downregulated genes. A volcano plot was used to depict the expression patterns of DEGs in the dataset (Fig. 5A). The intersection of DEGs and predicted target genes are presented in a Venn diagram, which showed three intersecting genes: ABLIM1, DDX5 and SLC7A1 (Fig. 5A).

Furthermore, the mRNA levels of three intersecting genes were detected in TNFα-treated VSMCs. The results identified that the transcription level of Ablim1 was significantly increased in VSMCs after TNFα stimulation, and no change in the transcription levels of Ddx5 and Slc7a1 occurred in TNFα-treated VSMCs (Fig. 5B). Subsequently, the transcript levels of Ablim1, Ddx5 and Slc7a1 were measured in the aortic tissues of Ang II-treated and control mice. The results also revealed a significant increase in the transcript level of Ablim1 in the aortic tissues of Ang II-treated, however, no changes occurred in the transcript levels of Ddx5 and Slc7a1 (Fig. 5B). These results revealed that ABLIM1 may be a direct downstream of miR-378a-5p.

To test whether ABLIM1 was involved in the occurrence and development of AAA, the expression of ABLIM1 was first detected in the aortic tissues of patients with AD and normal individuals using immunofluorescence staining and western blotting. The expression of ABLIM1 were significantly increased in the aortic tissues of patients with AD (Fig. S6A and B). It was also found that the expression levels of Ablim1 were increased in the aortic tissue of AAA mice (Fig. S6C and D) and in TNFα-treated VSMCs (Fig. S6E and F).

To further validate the relationship between ABLIM1 and miR-378a-5p, the expression of Ablim1 was detected in the aortas of antagomir-378a-5p + Ang II and angomir-378a-5p + Ang II mice. Compared with the corresponding control group, Ablim1 expression was increased in the aortic tissue of the antagomir-378a-5p + Ang II group (Figs. 5C and S6G) and decreased in the aortic tissue of the angomir-378a-5p + Ang II group (Figs. 5D and S6H). It was also found that miR-378a-5p mimics significantly decreased Ablim1 expression levels (Fig. S6I and J), whereas the miR-378a-5p inhibitor increased Ablim1 expression levels (Fig. S6K and L).

Bioinformatic analysis suggested that the binding sites for miR-378a-5p and ABLIM1 were highly conserved in humans, mice and rats (Fig. 5E). The three predicted binding sites are shown in Fig. S7A. To determine whether miR-378a-5p could directly bind to the 3'UTR of ABLIM1, the WT ABLIM1 3'UTR and the MUT ABLIM1 3'UTR were reconstituted into the pMIR-REPORT Luciferase vector (Fig. S7B). Compared with the mimics-NC-ABLIM1-3'UTR-WT group, the luciferase activity in the mimics-miR-378a-5p-ABLIM1-3'UTR-WT group, mimics-miR-378a-5p-ABLIM11-3'UTR-MUT1 group, and mimics-miR-378a-5p-ABLIM1-3'UTR-MUT3 group was significantly decreased, but there was no difference in the mimics-miR-378a-5p-ABLIM1-3'UTR-MUT2 and mimics-m iR-378a-5p-ABLIM1-3'UTR-MUT1-3 groups (Fig. 5F). These results indicated the position 1403-1409 of ABLIM1-3' UTR was the binding site of miR-378a-5p.

To verify whether miR-378a-5p regulated the VSMCs biological effects by directly regulating the expression of ABLIM1, the role of ABLIM1 in the occurrence and development of AAA was firstly explored by re-analyzing the GSE155468 dataset, which contained high-quality single-cell transcriptome data from 8 patients with TAA and 3 healthy donor thoracic aortas. A total of 42,611 cells with 20,551 genes remained after the unqualified cells and genes were filtered. The unsupervised clustering algorithm clustered the 42,611 cells into 26 cell populations (Fig. S8A and B). These cells were divided into seven groups (Fig. S8C), and each group was identified and named according to the expression of biomarker genes, including B cells (CD69 and CCR7), blood cells (MZB1), endothelial cells (VWF and IFI27), macrophages (CD14, CD68 and S100A9), mast cells (CPA3 and HPGD), SMCs (MYL9, ACTA2, MYH11 and TAGLN) and T cells (NKG7 and ZGMA) (Fig. S8D). These cells exhibited consistently high biomarker expression, thereby validating the robustness of the categorization.

Given that VSMCs lesions were the main mechanism of AAA, VSMCs were categorized into eight subpopulations (Fig. S8E and F). Based on the expression of ABLIM1, VSMCs were divided into two subsets: ABLIM1 high expression (ABLIM1⁺) and low expression (ABLIM1^-) (Fig. S8G-I). Gene Ontology and Kyoto Encyclopedia of Gens and Genomes analyses of DEGs in the ABLIM1⁺ and ABLIM1^-groups showed that most of the DEGs were related to VSMCs' contraction (Fig. S8J and K). Subsequently, it was found that the expression of contractile marker ACTA2, CNN1 and TAGLN in ABLIM1⁺ subsets was reduced and MMP2 expression was increased (Fig. 6A), which supported the deleterious role for ABLIM1 in AAA pathophysiology.

The results showed that Ablim1 knockdown led to suppress the protein expression of MMP2 and increased contractile markers (α-SMA, SM22-α and CNN1) (Fig. 6B), and overexpression of Ablim1 significantly increased the protein expression of MMP2 and suppressed expression of contractile markers (α-SMA, SM22-α and CNN1) in VSMCs (Fig. 6C). Under TNFα stimulation, the effects of miR-378a-5p inhibitor on the expression of VSMCs' contractile markers and MMP2 as well as migration were reversed by ABLIM1 knockdown. In addition, the protective role of the miR-378a-5p overexpression against VSMCs' contractile markers and migration was abolished by ABLIM1 overexpression (Fig. 6D-G). These aforementioned results indicated that miR-378a-5p regulates the phenotypic switching and secretion of MMP2 by attenuating the expression of ABLIM1 in VSMCs.

The role of ABLIM1 in the development of AAA was investigated in vivo. ApoE^−/− mice were injected with AAV-SM22-shNC or AAV-SM22-shAblim1 for 3 weeks through the tail vein before AAA modeling (Fig. S9A). It was found that Ablim1 was specifically knocked down in the aortas (Fig. S9B-D). There was no difference in body weight between the groups (Fig. S10A). After infusion of Ang II into AAV-SM22-shNC or AAV-SM22-shAblim1 mice, blood pressure was increased (Fig. S10B). There was no significant difference in survival rate between AAV-SM22-shNC + Ang II group and AAV-SM22-shAblim1 + Ang II group (Fig. S10C).

Compared with AAV-SM22-shNC + Ang II group, ABLIM1 knockdown reduced the incidence of AAA (40 vs. 74%, Fig. 7A and B). In the presence of Ang II, mice in the AAV-SM22-shNC and AAV-SM22-shAblim1 groups showed aortic dilation and aneurysms, whereas the symptoms in the AAV-SM22-shAblim1 treatment group were relieved (Fig. 7C and D). Moreover, histological analysis revealed that aortic dilatation, collagen deposition and medium degeneration were alleviated in AAV-SM22-shAblim1 mice in response to Ang II (Figs. 7E and S10D and E). All images of the abdominal aortic specimens were displayed in Fig. S11. These results indicated that ABLIM1 knockdown alleviated Ang II-induced AAA formation.

ApoE^−/− mice were injected with AAV-SM22-NC or AAV-SM22-Ablim1 for 3 weeks through the tail vein before AAA modeling (Fig. S12A). There was no difference in body weight between all the groups (Fig. S12B). Blood pressure was similarly increased following Ang II infusion in both AAV-SM22-NC and AAV-SM22-Ablim1 mice (Fig. S12C). There was no significant difference in survival rate between AAV-SM22-NC + Ang II and AAV-SM22-Ablim1 + Ang II group (Fig. S12D). As expected, the incidence of aortic aneurysms was significantly higher in Ang II-infused AAV-SM22-Ablim1 + Ang II mice compared with AAV-SM22-NC + Ang II mice (Fig. 7F and G). In vivo ultrasound showed that overexpression of ABLIM1 mice exhibited larger maximal internal diameters than control mice in response to Ang II (Fig. 7H and I). Furthermore, histological analysis revealed that exacerbated aortic dilatation, collagen deposition, and disruption of the medial architecture were aggravated in AAV-SM22-Ablim1 + Ang II mice compared with AAV-SM22-NC + Ang II mice (Figs. 7J and S12E and F). All images of the abdominal aortic specimens were displayed in Fig. S13. These results indicated that ABLIM1 overexpression aggravated Ang II-induced AAA formation.

To further investigate the potential mechanisms underlying ABLIM1-related VSMCs phenotypic switching, a Co-IP assay and mass spectrometry were performed to identify potential downstream effectors (Fig. 8A). A total of four proteins associated with VSMCs' contraction were validated: Mitogen-activated protein kinase 1 (MAPK1), adrenomedullin (ADM), protein kinase C delta (PRKCD) and MKL1 (Fig. 8B). It was found that ABLIM1 could interact with MKL1 (Fig. 8C and D). In addition, molecular docking predictions suggested that ABLIM1 bound to MKL1 (Fig. 8E). A total of 3 other proteins (ADM, PRKCD and MAPK1) did not interact with ABLIM1 (Fig. S14).

MKL1 (also known as MRTF-A) was a member of the myocardin-related transcription family. MKL1 transduce extracellular signals through the cytoskeleton that promote SMC differentiation and modulate SMC phenotype (31). MKL1 functions as a transcription cofactor through the nuclear cytoplasmic shuttle. It was found that ABLIM1 overexpression reduced nuclear MKL1 expression and increased cytoplasmic MKL1 expression (Fig. 8F). To further verify that whether the adverse effect of ABLIM1 on VSMCs phenotypic switching was dependent on MKL1, MKL1 overexpression was forced and the consequences of ABLIM1 overexpression on phenotypic switching of VSMCs were assessed. The results showed that MKL1 overexpression reversed the ABLIM1 overexpression-mediated phenotypic switching of VSMCs (Fig. 8G). Furthermore, MKL1 knockdown weakened the role of ABLIM1 knockdown-mediated phenotypic switching of VSMCs (Fig. 8H). These aforementioned results indicated that ABLIM1 regulates the VSMCs contractile phenotype by interacting with MKL1 and inhibiting the expression of MKL1.