THP-1 cells were cultured in RPMI 1640 medium (Beijing Solarbio Science & Technology Co., Ltd.) supplemented with 10% fetal bovine serum (FBS) and 2% penicillin-streptomycin, and maintained at 37°C in a humidified incubator with 5% CO₂. To induce differentiation from monocytes to macrophages, the cells were treated with 160 nM phorbol 12-myristate 13-acetate for 24 h. Following differentiation, the macrophages were exposed to 50 µg/ml oxidized low-density lipoprotein (ox-LDL) for 48 h to facilitate foam cell formation. THP-1-derived foam cells were seeded into 24-well plates at a density of 1–3×105 cells per well and cultured overnight to achieve 60–80% confluence prior to transduction. The cells were transduced with either an empty vector control (LV-Mock) or LV-FTO, both obtained from Genechem Co., Ltd., Shanghai, China, at a multiplicity of infection (MOI) of 100, in the presence of 8 µg/ml polybrene (cat. no. PP101; Genechem). The viral transduction mixture was prepared using serum-free RPMI 1640 medium and subsequently added to the cells, which were then incubated at 37°C in a 5% CO₂ humidified incubator for 24 h. Following the transduction process, the medium was replaced with fresh RPMI 1640 supplemented with 10% FBS, and the cells were cultured for an additional 48 h. The efficiency of transduction was verified through quantitative PCR analysis.

Cholesterol quantification was conducted utilizing a reversed-phase HPLC system. This system comprised a PerkinElmer Flexar^™ (PerkinElmer, Inc.) liquid chromatography apparatus, which was equipped with a UV–Vis detector and employed TotalChrom software (version 6.2; PerkinElmer, Inc.) for data acquisition and processing. Chromatographic separation was achieved using a Zorbax ODS column (dimensions: 4.6×250 mm, particle size: 5 µm; Agilent Technologies, Inc.). The column was maintained at ambient temperature (~25°C). The mobile phase was composed of acetonitrile and isopropanol in a 50:50 volume ratio, delivered at an isocratic flow rate of 1.0 ml/min. An injection volume of 20 µl was utilized. The eluate was monitored at a UV wavelength of 210 nm. Cholesteryl heptadecanoate (17:0 cholesteryl ester) served as the internal standard. For the quantification of free cholesterol (FC), 0.4 units of cholesterol oxidase were added to each sample. Total cholesterol (TC) was measured by incubating samples with 0.4 units of cholesterol oxidase in conjunction with 0.4 units of cholesterol esterase. The content of cholesteryl esters (CE) was calculated as the difference between TC and FC (CE=TC-FC).

Recombinant adenoviruses expressing mCherry-EGFP-LC3 (Ad-mCherry-EGFP-LC3) were obtained from Hanheng Biotechnology (Shanghai) Co., Ltd. For the study of adenoviral infections, cells were cultured in 6-well plates and exposed to the respective viruses for a duration of 48 h. In a similar experimental setup, short hairpin RNA (shRNA) transfections were performed utilizing the Lipofectamine^® 3000 reagent (Thermo Fisher Scientific, Inc.), followed by an incubation period of 48 h. The shRNAs targeting FTO (shFTO), ATG5 (shATG5), YTHDF1 (shYTHDF1), YTHDF2 (shYTHDF2), and YTHDF3 (shYTHDF3), as well as a scrambled negative control shRNA (Scr-shRNA), were procured from Guangzhou RiboBio Co., Ltd. in the form of shRNA-expressing plasmid vectors. For the transfection process, cells were seeded in 6-well plates and transfected with 500 ng of shRNA plasmid per well, employing the Lipofectamine^® 3000 reagent in accordance with the manufacturer's instructions. The transfection complexes were subsequently introduced to the cells and maintained at 37°C for 48 h prior to further analysis. All procedures were carried out in strict accordance with the manufacturer's instructions.

The following primes (human) were used: Scr-shRNA: 5′-TTCTCCGAACGTGTCACGTTT-3′; shFTO: 5′-GGACTTAAGGAATCCAGAATT-3′; shATG5: 5′-GCTTCGAGATGTGTGGTTTTT-3′; shYTHDF1: 5′-GGACATTGGTACTTGGGATT-3′; shYTHDF2: 5′-GGGATTGACTTCTCAGCATT-3′; shYTHDF3: 5′-TAAGTCAAAGAAGACGTATTATT-3′.

Cell and tissue proteins were extracted utilizing RIPA lysis buffer (P0013B; Beyotime Biotechnology), augmented with a protease and phosphatase inhibitor cocktail (cP1045; Beyotime Biotechnology) while maintained on ice. Protein concentrations were quantified using a BCA Protein Assay Kit (P0012S; Beyotime Biotechnology). Total protein (~30 µg per sample) was subjected to separation via 10% SDS-PAGE, followed by transfer onto polyvinylidene difluoride membranes (MilliporeSigma; IPVH00010). After blocking with 5% bovine serum albumin (BSA; Beyotime Biotechnology) for 1 h at room temperature, the membranes were incubated overnight at 4°C with primary antibodies: anti-FTO (cat. no. ab124892; Abcam; 1:1,000), anti-ABCA1 (cat. no. ab18180; Abcam; 1:200), anti-ABCG1 (cat. no. ab52617; Abcam; 1:1,000), anti-CD36 (cat. no. ab133625; Abcam; 1:1,000), anti-scavenger receptor A (SR-A) (cat. no. ab217843; Abcam; 1:500), anti-LC3 (ab192890; Abcam; 1:2,000), anti-p62 (cat. no. ab109012; Abcam; 1:5,000), anti-ATG7 (cat. no. ab52472; Abcam; 1:1,000), and anti-ATG5 (cat. no. ab108327; Abcam; 1:1,000). After three washes with TBST, membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit secondary antibodies (cat. no. ab6721; Abcam; 1:10,000 dilution,) for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system (ChemiDoc; Bio-Rad Laboratories, Inc.). The protein expression levels were assessed in a semi-quantitative manner using ImageJ software (version 2.14.0/1.54f; National Institutes of Health), employing GAPDH as a reference control.

The sections underwent deparaffinization using xylene and a descending series of anhydrous ethanol, followed by rinsing with PBS. Antigen retrieval was performed using a sodium citrate buffer. Sections were incubated overnight at 4°C with primary antibodies: The following primary antibodies were utilized: anti-CD31 (rabbit; cat. no. ab9498; Abcam) at a dilution of 1:100, anti-CD68 (rabbit; cat. no. ab283654; Abcam) at a dilution of 1:200, α-SMA (mouse; cat. no. A5228, MilliporeSigma) at a dilution of 1:500, and anti-FTO (rabbit; cat. no. ab280081; Abcam) at a dilution of 1:1,000. Subsequently, the sections were incubated with Alexa Fluor^™ 488-conjugated goat anti-rabbit IgG (H+L) secondary antibody (A-11008, Thermo Fisher Scientific, Inc.) at a 1:500 dilution and Alexa Fluor^™ 594-conjugated goat anti-mouse IgG (H+L) secondary antibody (cat. no. A-11005, Thermo Fisher Scientific, Inc.) at a 1:500 dilution for 1 h at room temperature. The nuclei were counterstained with DAPI (cat. no. D9542; MilliporeSigma) at a concentration of 1 µg/ml for 5 min at room temperature. Imaging was conducted using a Nikon A1R confocal microscope (Nikon Corporation), and subsequent analysis was performed using ImageJ software (version 2.14.0/1.54f, National Institutes of Health).

RNA Immunoprecipitation (RIP) was conducted utilizing the PureBinding^® RNA Immunoprecipitation Kit (P0101; Guangzhou Geneseed Biotech. Co., Ltd.) following the manufacturer's instructions. In summary, approximately 2×10⁷ cells were lysed in 405 µl of RIP Lysis Buffer, which was supplemented with protease and RNase inhibitors. A fraction of the lysate (10%) was set aside as the input control and stored at −80°C. The remaining lysate was allocated into two groups: The immunoprecipitation (IP) group and the negative control IgG group. Protein A/G magnetic beads were incubated with 3–5 µg of anti-YTHDF1 antibody or normal rabbit IgG (as a control) in IP buffer at 4°C overnight with gentle rotation. Following incubation, the beads underwent three washes with RIP Wash Buffer. The immunoprecipitated RNA and Input RNA were both subsequently purified via a column-based extraction technique and subjected to analysis through reverse transcription-quantitative PCR (RT-qPCR). The RT-qPCR was performed using the PowerUp^™ SYBR^™ Green Master Mix (A25742, Applied Biosystems; Thermo Fisher Scientific, Inc.) on an Applied Biosystems^™ 7500 Real-Time PCR System (Thermo Fisher Scientific, Inc.). The thermocycling protocol included an initial denaturation step at 95°C for 2 min, followed by 40 cycles of denaturation at 95°C for 15 sec, annealing at 60°C for 30 sec and extension at 72°C for 30 sec. The relative enrichment of the target RNA was quantified using the 2^−ΔΔCq method (14), with normalization to the Input RNA.

Following a 36-h transfection period with either Scr-shRNA or shRNA targeting YTHDF1/FTO, cells were treated with 4 µM actinomycin D (MedChemExpress) to inhibit global transcription. After incubation for the specified durations, cells were harvested and total RNA was extracted for reverse transcription. The resulting cDNA was subjected to RT-qPCR to quantify target mRNA levels.

Paraffin-embedded heart valve tissues were sliced into 6–8 µm thick sections, stained with hematoxylin-eosin and Masson staining at room temperature for 5 min, and assessed for necrotic core areas and collagen content in atherosclerotic lesions using ImageJ software (version 2.14.0/1.54f; National Institutes of Health).

For the purpose of immunohistochemical staining, formalin-fixed, paraffin-embedded sections (4 µm thick) derived from mouse atherosclerotic plaques treated with AAV-sh-FTO or from cardiac arterial valve specimens transfected with a negative control were subjected to deparaffinization in xylene and subsequently rehydrated through a graded ethanol series. Antigen retrieval was conducted by heating the sections in citrate buffer (pH 6.0) at a temperature range of 95–100°C for 20 min. Following a cooling period to room temperature, the sections were treated with 3% H₂O₂ in methanol for 10 min at room temperature to inhibit endogenous peroxidase activity. Subsequently, the sections were permeabilized using 0.1% Triton X-100 in PBS for 10 min at room temperature. To prevent non-specific binding, the sections were incubated with 10% normal goat serum (cat. no. 16210064; Gibco; Thermo Fisher Scientific, Inc.) in PBS for 1 h at room temperature. The sections were incubated overnight at 4°C with the following primary antibodies: Anti-LC3 (rabbit polyclonal; cat. no. ab63817; Abcam; 1 µg/ml), anti-p62 (rabbit polyclonal; cat. no. ab155686; Abcam; 1:500), anti-ATG5 (rabbit recombinant monoclonal; cat. no. ab108327; Abcam; 1:100), anti-ABCA1 (mouse monoclonal; cat. no. ab18180; Abcam; 1:200), and anti-ABCG1 (rabbit polyclonal; cat. no. 13578-1-AP, Proteintech Group, Inc.; 1:200). Following incubation, the sections were washed with PBS and subsequently incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature. The specific secondary antibodies employed were HRP-conjugated goat anti-rabbit IgG (H+L) (for rabbit primary antibodies; cat. no. 31460; Invitrogen; Thermo Fisher Scientific, Inc.) at a dilution of 1:200, and HRP-conjugated goat anti-mouse IgG (H+L) (for the mouse primary antibody, cat. no. 31430, Invitrogen; Thermo Fisher Scientific, Inc.) also at a dilution of 1:200. Immunoreactivity was visualized employing a 3,3′-diaminobenzidine (DAB) chromogen detection kit (cat. no. GK500705; Gene Tech Biotechnology Co., Ltd.). Image acquisition was performed using an Olympus BX50 (Olympus Corporation) microscope equipped with a digital color camera, and quantitative analysis was conducted using ImageJ software (version 1.8.0; National Institutes of Health).

AAV production and tail vein injection. AAV serotype 9 (AAV9) vectors expressing shRNA targeting FTO (AAV-sh-FTO) or a negative control shRNA (AAV-sh-NC) were packaged in AAV-293 cells using the triple-plasmid transfection system, following the manufacturer's standard protocol (Shanghai GenePharma Co., Ltd.). Briefly, the AAV vector plasmid containing the shRNA sequence (targeting FTO or negative control) under an appropriate promoter, the helper plasmid encoding AAV9 Rep/Cap genes, and the adenovirus helper plasmid were co-transfected into AAV-293 cells using the proprietary transfection reagent RNAi-Mate (GenePharma). At 72 h following transfection, both the cells and supernatant were collected and subjected to three freeze-thaw cycles. Subsequently, the samples were centrifuged at 2,000 rpm (304 × g) for 5 min at 4°C to eliminate cellular debris. The clarified lysate was then purified and concentrated to obtain AAV vectors. For in vivo delivery, mice received a single dose of 3×10¹¹ vector genomes of the respective AAV9 vectors, diluted in sterile PBS to a final volume of 200 µl, via lateral tail vein injection. The FTO shRNA sequence (5′-3′) was: 5′-AAATAGCCGCTGCTTGTGAGATT-3′.

The extraction of total RNA and subsequent RT-qPCR were conducted as follows: Total RNA was isolated from samples utilizing the TRIzol^® reagent (cat. no. 15596026; Invitrogen; Thermo Fisher Scientific, Inc.) in accordance with the manufacturer's protocol. The concentration and purity of the extracted RNA were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc.). Subsequently, 1 µg of total RNA was reverse-transcribed into complementary DNA (cDNA) employing the PrimeScript RT Master Mix Kit (cat. no. RR036A; Takara Biotechnology Co., Ltd.) as per the manufacturer's instructions. qPCR was executed using the SYBR^® Green Supermix Kit (cat. no. 1708882, Bio-Rad Laboratories, Inc.) on an Applied Biosystems 7500 Real-Time PCR System (Thermo Fisher Scientific, Inc.). The thermal cycling conditions were as follows: Initial denaturation at 95°C for 3 min, followed by 40 cycles of 95°C for 10 sec, 60°C for 30 sec, and 72°C for 30 sec. A melting curve analysis was conducted to verify amplification specificity. Relative gene expression levels were determined using the 2^−ΔΔCq method (14). RNA extraction, cDNA synthesis, and qPCR were all conducted following the manufacturers' protocols for the TRIzol^® Reagent, PrimeScript RT Master Mix Kit, and SYBR^® Green Supermix Kit, respectively. The following primers (human) were used: FTO Forward: 5′-CTTCACCAAGGAGACTGCTATTTC-3′; FTO Reverse: 5′-CAAGGTTCCTGTTGAGCACTCTG-3′; GAPDH Forward: 5′-GGTGGTCTCCTCTGACTTCAA-3′; GAPDH Reverse: 5′-GTTGCTGTAGCCAAATTCGTTGT-3′.

Bafilomycin A1 (cat. no. B1793; MilliporeSigma) and 3-methyladenine (cat. no. 3-MA; M9281; MilliporeSigma) were employed to regulate autophagy in THP-1-derived foam cells. For the Bafilomycin A1 intervention, cells were exposed to 100 nM Bafilomycin A1 for a duration of 3 h at 37°C. In the case of 3-MA treatment, cells underwent pre-incubation with 1 mM 3-MA for 45 min prior to subsequent stimulation.

Serum concentrations of MMP-9, myeloperoxidase (MPO), sCD40L, IL-18, and monocyte chemoattractant protein-1 (MCP-1) were quantified utilizing specific ELISA kits: MMP9 ELISA Kit (cat. nos. ab253227 ab246539; Abcam), MPO ELISA Kit (cat. nos. ab155458 and ab119605; Abcam), CD40L ELISA Kit (cat. nos. ab275105 and ab196268; Abcam), IL-18 ELISA Kit (cat. nos. ab216165 and ab215539; Abcam), and MCP-1 ELISA Kit (cat. nos. ab208979 and ab179886; Abcam), in accordance with the manufacturers' protocols. The activity of m6A demethylase was evaluated using the Epigenase m6A Demethylase Activity/Inhibition Assay Kit (Colorimetric; cat. no. P-9013-96; EpigenTek Group Inc.), following the manufacturer's instructions.

THP-1 cells were plated in 24-well plates at a density of 0.5–2×105 cells per well, 24 h prior to transfection. The cells were co-transfected with the pGL3-ATG5 plasmid, which encodes the firefly luciferase reporter, and the pRL-TK plasmid, serving as the Renilla luciferase control vector (E2241; Promega Corporation). Additionally, shRNAs targeting either FTO (5′-GGACTTAAGGAATCCAGAATT-3′) or YTHDF1 (5′-GGACATTGGTACTTGGGATT-3′), as well as negative control shRNAs (5′-TTCTCCGAACGTGTCACGTTT-3′), were employed. These shRNAs were procured from Guangzhou RiboBio Co., Ltd. The transfection process was facilitated using Lipofectamine^® 2000 (cat. no. 11668019; Invitrogen; Thermo Fisher Scientific, Inc.). Each well received a total of 500 ng of plasmid DNA and 1.0–2.5 µl of Lipofectamine^® 2000, diluted in Opti-MEM^® reduced serum medium (Gibco; Thermo Fisher Scientific, Inc.), following the manufacturer's instructions. At 48 h post-transfection, cells were lysed using passive lysis buffer (Promega Corporation). Firefly and Renilla luciferase activities were subsequently measured using the Dual-Luciferase Reporter Assay System (cat. no. E1910; Promega Corporation) on a luminometer. The relative luciferase activity was determined by calculating the ratio of firefly to Renilla luciferase activity, thereby normalizing for transfection efficiency.

Statistical analysis was performed using GraphPad Prism (Version 9; Dotmatics). The results were obtained from a minimum of five independent experiments and are expressed as the mean ± standard deviation (SD). The normality of the data distribution was assessed using the Shapiro-Wilk test. For parametric data, analyses were conducted using Student's t-test or one-way analysis of variance (ANOVA), followed by Bonferroni's post hoc test for multiple comparisons. whereas nonparametric data were evaluated using the Mann-Whitney U test or the Kruskal-Wallis test. The quantification of atherosclerotic lesion areas and the percentage of Masson-positive areas within these lesions was performed using ImageJ software (version 2.14.0/1.54f; National Institutes of Health). P<0.05 was considered to indicate a statistically significant difference.

Given that macrophage lipid accumulation critically promotes atherosclerosis, the potential involvement of FTO in this process was investigated. Initial assessments in THP-1-derived macrophages indicated that ox-LDL stimulation markedly reduced both mRNA and protein expression levels of FTO compared to untreated controls (Fig. 1A), implicating FTO in lipid metabolism. Following this association, its functional role was further defined. In THP-1-derived foam cells, FTO expression was manipulated through either overexpression using LV-FTO or knockdown with shRNA-FTO, with LV-Mock and Scr-shRNA serving as controls. Transfection efficiency was confirmed by RT-qPCR analysis (Fig. S1). HPLC analysis demonstrated that FTO overexpression markedly decreased intracellular TC levels, accompanied by reductions in its components, CE and FC. Conversely, FTO knockdown resulted in increased levels of intracellular TC, FC and CE (Fig. 1B and C). These findings indicate that FTO plays a protective role against lipid accumulation in macrophage-derived foam cells.

Reduced cholesterol efflux via ABCA1 and ABCG1 markedly contributes to intracellular lipid accumulation (15). To elucidate the mechanism by which FTO exerts its protective effects, the impact of FTO on the expression of key cholesterol transporters and cholesterol efflux in THP-1 macrophage-derived foam cells was analyzed. FTO overexpression was found to upregulate both mRNA and protein levels of ABCA1 and ABCG1 (Fig. 1D and E). Given that enhanced cholesterol influx mediated by scavenger receptors, such as CD36 and SR-A, also promotes lipid accumulation (16,17), their expression levels were concurrently assessed. Notably, no significant differences in CD36 or SR-A levels were observed following FTO overexpression (Fig. 1F). These results indicate that FTO reduces lipid deposition by specifically promoting ABCA1- and ABCG1-mediated cholesterol export.

In addition to their roles in promoting cholesterol release, ABCA1 and ABCG1 are known to suppress the inflammatory response (18,19). Considering that atherosclerotic plaque stability is negatively associated with inflammatory cell infiltration and necrotic core size, while positively associated with fibrous cap thickness (20), it was hypothesized that FTO might influence plaque stability by modulating inflammation. Supporting this hypothesis, compounds such as the contaminant derivative OXA17 have been shown to stabilize plaques by upregulating ABCA1 (21). Consequently, the effect of FTO on the secretion of pro-inflammatory cytokines and enzymes in THP-1 macrophage-derived foam cells was assessed. FTO knockdown markedly increased the secretion of multiple pro-inflammatory mediators, including MMP-9, MPO, sCD40L), lipoprotein-Associated Phospholipase A2 (Lp-PLA2), IL-18, and MCP-1 (Fig. 1G). These data suggested that FTO mitigates the macrophage inflammatory response, potentially contributing to improved atherosclerotic plaque stability.

Autophagy is a multi-step biological process (22). To investigate its regulatory role, functional studies involving both overexpression and knockdown of FTO in THP-1 cells were conducted. Evaluation of key autophagic markers revealed that FTO overexpression elevated the LC3-II/I ratio while decreasing P62 levels (Fig. 2A and B). To directly monitor autophagic flux, the mCherry-EGFP-LC3 reporter system was employed. Consistent with these findings, FTO overexpression increased the number of LC3 puncta. This system distinguishes autophagosomes (yellow puncta, mCherry + EGFP+) from autolysosomes (red puncta, mCherry + EGFP-). Notably, FTO overexpression markedly increased the counts of both yellow and red puncta (Fig. 2C). Furthermore, the addition of BafA1 exacerbated the accumulation of LC3-II induced by FTO knockdown (shRNA-FTO; Fig. 2D). These results indicated that FTO induces autophagy.

To determine whether FTO regulates lipid metabolism and plaque stability through autophagy, autophagic flux in THP-1 macrophage-derived foam cells was modulated using Rapamycin (an agonist) and 3-MA (an inhibitor). Consistent with autophagy-mediated regulation, the upregulation of ABCA1 and ABCG1 expression driven by FTO overexpression was further enhanced by Rapamycin and suppressed by 3-MA (Fig. 2E and F). Similarly, HPLC analysis of intracellular cholesterol revealed that the reduction in TC, FC and CE levels induced by FTO overexpression was markedly amplified by Rapamycin. Conversely, treatment with 3-MA attenuated this effect, leading to increased intracellular accumulation of TC, FC, and CE (Fig. 2G).

These results indicated that FTO alleviates lipid accumulation in THP-1 macrophage-derived foam cells by promoting autophagic flux. Furthermore, the effect of autophagic modulation on the secretion of pro-inflammatory mediators (MMP-9, MPO, sCD40L, Lp-PLA2, IL-18, MCP-1) was assessed. Treatment with the autophagy inhibitor 3-MA increased the levels of these factors, whereas the autophagy inducer Rapamycin markedly reduced their secretion (Fig. 2H). These findings demonstrate that FTO enhances autophagy and promotes atherosclerotic plaque stability by upregulating the expression of ABCA1 and ABCG1.

To determine whether FTO regulates m⁶A modification in THP-1 macrophage-derived foam cells, global m⁶A levels were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS). FTO knockdown resulted in a significant decrease in m⁶A modification (Fig. 3A), confirming the role of FTO as an m⁶A demethylase in these cells. To identify the autophagy-related targets of FTO, several candidate genes were screened via RT-qPCR. The mRNA level of ATG5 was markedly downregulated following FTO knockdown and upregulated upon FTO overexpression (Fig. 3B). Consistent with these findings, FTO overexpression also increased ATG5 protein levels (Fig. 3C). To establish a functional link, a rescue experiment was conducted. Silencing ATG5 effectively reversed the increase in the LC3-II/I ratio and the decrease in SQSTM1/p62 levels induced by FTO overexpression (Fig. 3D), revealing that FTO promotes autophagy primarily by upregulating ATG5 expression.

To assess whether FTO influences lipid accumulation through ATG5, a rescue experiment was again performed. Silencing ATG5 reversed the upregulation of ABCA1 and ABCG1 induced by FTO overexpression (Fig. 3D and E), indicating that FTO regulates lipid metabolism via ATG5. HPLC analysis confirmed that ATG5 knockdown aggravated the reduction in intracellular TC, FC, and CE levels associated with FTO overexpression (Fig. 3F).

To investigate whether FTO regulates ATG5 through m⁶A methylation, bioinformatics was initially employed to predict potential m⁶A modification sites on ATG5 mRNA (Fig. 3G). FTO knockdown decreased the abundance of ATG5 mRNA containing m⁶A modifications at these sites (Fig. 3H). Methylated RIP-qPCR verified a direct interaction between FTO and ATG5 mRNA in foam cells (Fig. 3I). Furthermore, mRNA stability assays demonstrated that FTO depletion accelerated the decay of ATG5 mRNA transcripts (Fig. 3J). A luciferase reporter assay indicated that FTO knockdown suppressed the activity of a wild-type, but not mutant, ATG5 3′ UTR construct (Fig. 3K). Additionally, the data showed that shRNA-ATG5 further upregulated the secretion of pro-inflammatory mediators, including MMP-9, MPO, sCD40L, Lp-PLA2, IL-18, and MCP-1 (Fig. 3L). These data demonstrated that FTO directly targets ATG5 mRNA in an m⁶A-dependent manner to enhance its stability and expression, thereby promoting autophagy and plaque stability.

A previous study established that reader proteins of m⁶A modification can regulate various aspects of RNA metabolism, including translation, splicing, stability, nuclear export and degradation (23). Among these proteins, the YTH domain-containing family (YTHDF1-3) has been extensively studied. YTHDF2 has been shown to selectively recognize m⁶A-modified mRNAs and facilitate their decay, while YTHDF1 promotes the translation of m⁶A-marked transcripts (24,25). YTHDF3 appears to interact with both YTHDF1 and YTHDF2, potentially coordinating their functional roles (26). To determine which YTHDF protein primarily influences ATG5 transcript regulation, individual knockdowns of YTHDF1, YTHDF2 and YTHDF3 were performed using small interfering RNAs, with knockdown efficiency validated for each case (Fig. S2). Knockdown of YTHDF1, but not the other reader proteins, markedly increased ATG5 mRNA levels (Fig. 4A). As depicted in Fig. S3, the knockdown of YTHDF2 via shRNA did not result in a significant alteration of ATG5 or ATG7 protein expression when compared to the Scr-shRNA control. Conversely, shRNA-mediated knockdown of YTHDF1 led to a significant reduction in ATG5 protein levels, while ATG7 levels remained unaffected. Moreover, as illustrated in Fig. S4, the knockdown of FTO did not affect ATG7 at the protein level. This led to the hypothesis that YTHDF1 specifically recognizes FTO-mediated m⁶A modification on the ATG5 transcript to promote its decay. A rescue experiment supported this notion; YTHDF1 knockdown effectively counteracted the effects of FTO overexpression, restoring the LC3B-II/LC3B-I ratio and ABCA1/ABCG1 expression while reducing P62 accumulation (Fig. 4B and C), thus demonstrating that YTHDF1 is a key mediator in the m⁶A-dependent regulation of ATG5 by FTO. Furthermore, immunofluorescence analysis using the mCherry-EGFP-LC3 reporter in foam cells revealed that the increase in autophagic puncta (both red and yellow) induced by LV-FTO was further augmented upon treatment with shRNA-YTHDF1 (Fig. 4D).

Additionally, RIP-qPCR experiments confirmed a direct interaction between ATG5 mRNA and YTHDF1 (Fig. 4E). Subsequent mRNA stability assays demonstrated that YTHDF1 knockdown delayed the decay of ATG5 transcripts and enhanced their stability (Fig. 4F). In luciferase reporter assays, FTO knockdown suppressed the activity of wild-type constructs but had no effect on mutant constructs (Fig. 4G). HPLC analysis indicated that YTHDF1 knockdown reversed the cholesterol-lowering effect of FTO overexpression, as evidenced by restored intracellular levels of TC, FC and CE (Fig. 4H). Furthermore, in a rescue experiment, YTHDF1 knockdown counteracted the changes in the secretion of MMP-9, MPO, sCD40L, Lp-PLA2, IL-18 and MCP-1 induced by FTO overexpression (Fig. 4I). In summary, these results revealed that YTHDF1 regulates ATG5 mRNA expression via an m⁶A-dependent mechanism.

To further elucidate the relationship between FTO and atherosclerotic plaque composition, plaque size in aortic tissues was evaluated using hematoxylin-eosin staining, while collagen content was assessed via Masson's trichrome staining. Hematoxylin-eosin staining revealed that atherosclerotic lesions in AAV-sh-FTO-treated mice were larger, exhibited increased lipid accumulation, and demonstrated thicker fibrous caps compared to controls. Masson's staining further indicated a significant reduction in collagen fiber content within plaques from AAV-sh-FTO mice relative to the AAV-Mock group (Fig. 5A). Immunofluorescence co-localization analysis on arterial plaque sections was performed to determine the cellular localization of FTO in endothelial cells (CD31+), smooth muscle cells (αSMA+) and macrophages (CD68+). The results demonstrated that FTO predominantly localizes within macrophages (Fig. 5B). Immunohistochemical analysis of key autophagy and cholesterol efflux-related proteins showed that levels of LC3, ATG5, ABCA1 and ABCG1 were downregulated, while P62 expression was upregulated in the AAV-sh-FTO group (Fig. 5C). Moreover, the production of plaque destabilization markers-including MMP-9, MPO, sCD40L, Lp-PLA2, IL-18 and MCP-1 was markedly upregulated in AAV-sh-FTO mice (Fig. 5D). These findings indicate that FTO knockdown exacerbates atherosclerotic plaque development and instability, likely due to impaired autophagic flux and cholesterol homeostasis, accompanied by enhanced inflammation. In conclusion, these findings support a protective role for FTO in mitigating the progression of atherosclerosis.