m⁶A demethylase FTO promotes the progression of abdominal aortic aneurysm through PDK4-mediated apoptosis

Abstract

N⁶-methyladenosine (m⁶A) is one of the most abundant methylation modifications in mRNA, regulating different stages of mRNA metabolism, including folding, maturation, export, translation, and decay. However, the role of m⁶A modification and the m⁶A demethylase fat mass and obesity-associated (FTO) in abdominal aortic aneurysms (AAAs) remains unclear. We found that FTO expression levels were elevated in both in vitro and in vivo models of abdominal aortic aneurysms, and silencing FTO could partially reverse AngII-induced apoptosis of vascular smooth muscle cells (VSMCs). Integrated RNA-seq and MeRIP-seq analysis further identified pyruvate dehydrogenase kinase 4 (PDK4) as the target gene of FTO-mediated m⁶A modification. FTO mediates m⁶A demethylation in the 3’ untranslated region (3’ UTR) of PDK4 mRNA and induces its degradation through a YTH N⁶-methyladenosine RNA binding protein 2 (YTHDF2)-dependent mechanism. Overexpression of PDK4—a key enzyme in mitochondrial glucose metabolism and a novel regulator of mitochondria-associated endoplasmic reticulum integrity—reversed the inhibitory effect on apoptosis after FTO silencing. These results suggest that FTO regulates VSMC apoptosis by mediating m⁶A demethylation of PDK4 mRNA and promoting its degradation in a YTHDF2-dependent manner. Moreover, PDK4 overexpression reverses the apoptosis-inhibitory effect induced by FTO silencing.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-47997-3.

Introduction

Abdominal aortic aneurysm (AAA) is a type of aortic wall injury disease caused by multiple factors, which mainly manifests as permanent local bulging or diffuse dilatation of the aorta. If effective treatment is not administered in a timely manner, the mortality rate is as high as 65%-85% once rupture occurs¹. At present, AAA with large diameter, rapid progression, risk of rupture or rupture is mostly treated by open surgery or endovascular exclusion^2,3. However, for the small aneurysms that do not meet the surgical standards, only long-term monitoring and observation can be carried out, and there is a lack of effective intervention drugs^4,5. Therefore, exploring effective molecular targets that can delay the progression of AAA will be an important direction for clinical prevention and treatment of AAA.

N⁶-methyladenine (m⁶A) modification is a kind of epigenetic modification, which leads to the methylation of the sixth N position of adenosine^6,7. It is the most common modification of eukaryotic RNA molecules and affects mRNA processing and metabolism, including secondary structure, subcellular localization and degradation⁸. m⁶A modification is a dynamic and reversible regulatory process, and its modification level is regulated by ‘Writers’ and’ Erasers’. As the first m⁶A demethylase discovered, fat mass and obesity associated (FTO) has been extensively studied^9,10. At present, FTO has been found to be a biomarker for disease risk assessment and a therapeutic target in a variety of diseases such as malignant tumors and metabolic diseases^11–13, but its role in AAA is rarely reported.

Vascular smooth muscle cells (VSMCs), as the main cellular component of the aortic media, play an important role in maintaining the elasticity and compression resistance of the abdominal aorta and repairing the extracellular matrix of the aortic wall¹⁴. In this study, we silenced the expression of FTO in VSMCs and performed KEGG enrichment analysis, and found that the differential hypermethylation peaks were significantly enriched in the apoptosis pathway, while a large number of previous studies have shown that VSMCs apoptosis can lead to aortic wall weakness and dilation^15,16. We further demonstrated that silencing FTO expression in VSMCs attenuated AngII-induced apoptosis and delayed AAA progression in vitro and in vivo by inducing pyruvate dehydrogenase kinase 4 (PDK4) mRNA degradation in a YTH N⁶-Methyladenosine RNA Binding Protein F2 (YTHDF2) dependent manner.

Methods

Animal experiments

All animal studies adhered strictly to the guidelines outlined in the “Guide for the Care and Use of Laboratory Animals,” which is prescribed by the National Academy of Sciences and published by the National Institutes of Health. These studies were further approved by the Animal Care and Utilization Committee of Dalian Medical University in China (AEE23116). The ApoE^−/− mice were sourced from Beijing Vital River Laboratory Animal Technology Co., Ltd. For housing, the mice were kept in cages with a density of five mice per cage, all under stringent pathogen-free conditions. All experimental procedures were conducted in strict compliance with the ARRIVE guidelines, along with relevant regulatory standards and institutional protocols.

Construction of the AAA mouse model

Mouse models of Angiotensin II (Ang II)-induced AAA were generated as follows: prior to the subcutaneous implantation of micro-osmotic pumps (ALZET 2004), 12-week-old male mice were anesthetized using isoflurane inhalation at a concentration of 2–3% in 100% oxygen to ensure a painless and stress-free procedure. The micro-osmotic pumps contained either saline or Ang II (A9525, Sigma‒Aldrich) and were subcutaneously implanted, administering a dose of 1000 ng/(kg.min) for 28 days. FB23-2 (HY-127103, MedChemExpress) was administrated at 2 mg/kg/d, 6 mg/kg/d (i.p.) for 28 days, and the control group was administrated with vehicle (10% DMSO (HY-Y0320C, MedChemExpress) + 40% PEG300 (HY-Y0873, MedChemExpress) + 5% Tween-80 (HY-Y1891, MedChemExpress) + 45% normal saline). At the end of the treatment period, mice were euthanized via an overdose of pentobarbital sodium (150 mg/kg, i.p.) to ensure a humane and rapid death. Subsequently, the suprarenal abdominal aorta was dissected and collected for further analysis.

Cell culture and generation of stable cell lines

Immortalized Human Aortic Smooth Muscle Cells-SV40 were purchased from Zhejiang Meisen Cell Technology Co., LTD. Cells were cultured in DMEM/F12 medium (Meilunbio, China) containing 10% fetal bovine serum (Gibco, USA) and 1% dual antibiotics (Meilunbio, China). Cells were cultured in a 5% CO2 cell incubator at 37℃.

Plasmid transfection and lentiviral infection

The FTO plasmid, siRNA, and PDK4 overexpression plasmid were purchased from Gemma Genes Technology Co., LTD. (Shanghai, China). Plasmid DNA and siRNA were transfected using Lipofectamine 3000 (Life Technologies, USA) according to the manufacturer’s instructions. Detailed sequence information is provided in the Supplementary Material.

Western blotting

Equal amounts of protein extracts were separated by 12.5% SDS-PAGE and transferred to nitrocellulose membranes and incubated with the relevant antibodies overnight at 4 °C. The cells were incubated with fluorescent secondary antibodies for 1 h at room temperature. Proteins were visualized using the Odyssey^® CLX dual-color infrared laser imaging system. Antibodies were purchased from the following companies: anti-FTO (Proteintech, 27226-1-AP), anti-PDK4 (Proteintech, 12949-1-AP), anti-cleaved-caspase3 (Zenbio, 341034), and anti-GAPDH (Proteintech, 60004-1-Ig).

RNA isolation and RT-qPCR

Total RNA was isolated using RNA-Easy Isolation Reagent (Vazyme, R701-01) according to the manufacturer’s instructions. RNA quality was analyzed by NanoDrop2000c. Complementary DNA was synthesized using HiScript II Q RT SuperMix for qPCR (Vazyme, Q711), Real-Time reverse transcription PCR was performed using a ChamQ Universal SYBR qPCR Master Mix in a qTOWER³ Real-time PCR Thermal Cycler (Analytik Jena).Relative gene expression was normalized to β-actin using the standard 2−∆∆Ct quantification method. The primer sequences are listed in Supplementary Material.

RNA m⁶A quantification

The EpiQuik m⁶A RNA Methylation Quantification Kit (Colorimetric) (P-9005, Epigentek, USA) was utilized to assess the m⁶A content within total RNA samples. In brief, 200 nanograms of RNA were coated onto the assay wells. Subsequently, capture antibody solution and detection antibody solution, each diluted to an appropriate concentration as specified by the manufacturer’s instructions, were individually added to the assay wells. The m⁶A levels were then quantified through a colorimetric approach by measuring the absorbance of each well at a wavelength of 450 nanometers.

Transcriptome sequencing (RNA-seq)

Total RNA was extracted from both stable shFTO VSMCs and their respective control samples using Trizol reagent sourced from ThermoFisher (USA). Subsequently, rRNA was depleted from these RNA samples using the Ribo-Zero rRNA Removal Kits (Illumina, San Diego, CA, USA), adhering strictly to the manufacturer’s guidelines. rRNA-depleted RNA was then utilized to construct RNA libraries with the TruSeq Stranded Total RNA Library Prep Kit (Illumina, San Diego, CA, USA), following the manufacturer’s instructions meticulously.Quality control and quantification of the constructed libraries were conducted using the BioAnalyzer 2100 system (Agilent Technologies, Inc., USA). These libraries, at a concentration of ten picometers, were processed to convert them into single-stranded DNA molecules, which were then captured onto Illumina flow cells. In situ amplification was performed to form clusters, and finally, the libraries underwent 150 cycles of sequencing on an Illumina HiSeq Sequencer, all according to the manufacturer’s protocols. Novogene (Tianjin, China) provided the services for library preparation and high-throughput sequencing.

Methylated RNA immunoprecipitation sequencing (MeRIP-seq)

The RNA m⁶A was sequenced by MeRIP-seq at Novogene (Beijing, China). Briefly, atotal of 2 µg RNAs were extracted from the VSMCs. The integrity and concentration of extracted RNAs were detected using an Agilent 2100 bioanalyzer (Agilent) and simpliNano spectrophotometer (GE Healthcare), respectively. Fragmented RNA (~ 100 nt) was incubated for 2 h at 4℃ with anti-m⁶A polyclonal antibody (Merk Millipore) in the immunoprecipitation experiment. Then, immunoprecipitated RNAs or Input was used for library construction with Ovation SoLo RNA-Seq System Core Kit (NuGEN). The library preparations were sequenced on an Illumina Novaseq or Hiseq platformwith a paired-end read length of 150 bp according to the standard protocols.

RNA binding protein immunoprecipitation (RIP)-qPCR

The VSMCs were rinsed twice with ice-cold PBS and then lysed in 1 mL of RIP Lysis buffer (comprising 150 mM KCl, 25 mM Tris, 5 mM EDTA, 0.5% Triton X-100, 0.5 mM DTT, protease inhibitor at 1:100 dilution, and RNAase inhibitor at 1:1000 dilution) on ice for a duration of 30 min. Following lysis, the cell lysates were centrifuged at 12,000 × g for 15 min at 4 °C. A 10% aliquot of the supernatant was retained as the input sample, while the remaining supernatant was incubated with either anti-YTHDF2 antibody (or anti-YTHDF1/3 antibody from Proteintech 17479-1-AP, 24744-1-AP, 25537-1-AP) or control IgG antibody, along with protein A/G-conjugated magnetic beads (sourced from MCE, USA), in 900 µL of RIP Lysis buffer at 4 °C for 4 h.The RNA bound to the antibodies was immunoprecipitated using the beads and subsequently washed four times with RIP buffer (consisting of 150 mM KCl, 25 mM Tris, 5 mM EDTA, and 0.5% Triton X-100). The beads were then treated with 10 µL of 10% SDS, 10 µL of Proteinase K, and 130 µL of RIP buffer for 30 min at 55 °C. RNA was recovered from both the IP (immunoprecipitated) and input groups using Trizol reagent (Invitrogen, USA), as per the manufacturer’s instructions, and analyzed by RT-qPCR.

Detection of mRNA half-life

Cells that stably expressed shRNA targeting FTO or a non-targeting control shRNA (shNC) were plated in 60-mm dishes to achieve approximately 70% confluence. Subsequently, 24 h later, these cells were exposed to 5 µg/mL of actinomycin D to effectively inhibit RNA synthesis. At 0, 2, 4, and 6 h post-treatment, the cells were collected for total RNA extraction. The extracted RNA was then subjected to RT-qPCR analysis, following the previously described protocol.

Luciferase reporter assays and mutagenesis assays

Three potential m⁶A recognition sites within the 3′ Untranslated Region (UTR) of the PDK4 gene were identified in its 3′UTR sequence using the sequence-based N⁶-methyladenosine (m⁶A) modification site predictor, SRAMP. Site-directed mutagenesis, specifically converting adenine (A) to cytosine (C), was carried out utilizing the QuikChange II Site-Directed Mutagenesis Kit (Agilent, USA; catalog number 200,523), following the manufacturer’s protocols meticulously. To quantify luciferase activity, the Dual Luciferase Reporter Gene Assay Kit (YEASEN, China; catalog number 11402ES60) was employed in the GloMax luminometer (Promega). All experiments were conducted in triplicate to ensure reproducibility. The luciferase activity of the firefly luciferase reporter was normalized against the activity of the Renilla luciferase reporter, which served as an internal control to reflect expression efficiency.

Statistical analysis

Experiments were performed independently at least three times. Data were expressed as mean ± SD. Student’s t test or one-way analysis of variance (ANOVA) was used to assess statistical differences between two or multiple groups, respectively. The experimental data were analyzed using GraphPad Prism 9.0 software (GraphPad Inc., La Jolla, CA, USA). p value < 0.05 was considered statistically significant. p value < 0.01 was considered very significant.

Results

FTO, an N⁶-methyladenosine RNA demethylase is up-regulated in Abdominal aortic aneurysm

Bioinformatics analysis of Gene Expression Omnibus data base showed that demethylase FTO mRNA was significantly up-regulated in abdominal aortic aneurysm tissues compared with normal abdominal aortic tissues (GSE17901) (Fig. 1A), indicating that FTO plays an important role in m⁶A modification of AAA. To verify the expression of FTO in AAA, AAA was induced in ApoE-/- mice by subcutaneous infusion of Ang II (1 µg/min/kg) with a slow-release pump for 28 days, while normal saline was injected into the control group¹⁷(Fig. 1B,C). After 28 days, the mice were dissected and abdominal aortic tissue was evaluated. The results indicated that, in comparison to the control group, there was a notable elevation in the expression levels of both FTO mRNA and protein in AAA. (Fig. 1D,F). At the same time, in vitro injury model of human aortic smooth muscle cells induced by Ang II (1 µM) in VSMCs, Ang II also significantly increased the expression level of FTO protein (Fig. 1G,H).

Fig. 1.

Increased FTO expression in aortic tissue of AAA mice. (A) FTO expression levels in the GSE17901 GEO database, comparison between AAA tissues (n = 18) and NC tissues (n = 17). (B) Representative anatomical images of the abdominal aorta from ApoE⁻/⁻ mice at 28 days after continuous infusion of AngII (AAA group) or saline (NC group) via subcutaneously implanted micro-osmotic pumps. (C) Quantification of the maximum diameter of the suprarenal abdominal aorta measured by digital vernier calipers. (D–F) RT-qPCR and western blot were used to detect the expression of FTO in vivo. (G–H) The expression of FTO in vivo was detected by western blot. p < 0.05, ** p < 0.01, *** p < 0.001.

FTO promotes VSMCs apoptosis in vitro

VSMCs are the main cellular components of the media of abdominal aorta, and their apoptosis often leads to the decrease of the repair ability of the extracellular matrix of the aortic wall and the decrease of aortic elasticity and pressure resistance, thus promoting the formation of AAA^18,19. To determine whether FTO has a regulatory effect on the apoptosis of VSMCs, we constructed FTO stable silencing VSMCs cells using lentiviruses, and verified the knockdown efficiency of FTO at the RNA and protein levels (Fig. 2A–C). Quantitative analysis by Western blot revealed that FTO silencing significantly suppressed the expression level of Ang II-stimulated caspase3 compared with that of the control group (Fig. 2D–F), suggesting that FTO may affect apoptosis in vascular smooth muscle by regulating key proteases of apoptosis. Flow cytometry results showed that the late apoptosis rate of VSMCs in the Ang II-treated group was increased compared with that of the control group, whereas the rate was significantly decreased after FTO silencing (Fig. 2G–H). Notably, FTO knockdown not only reduced the expression of apoptotic proteins, but also significantly decreased the number of apoptotic cells. Together, these data suggest that FTO plays a critical role in Ang II-induced programmed death of VSMCs by regulating the caspase3-dependent apoptotic pathway. Given the critical role of VSMC survival in maintaining vascular wall integrity, these findings imply that FTO deficiency may be associated with the modulation of vascular wall homeostasis, which provides a potential mechanistic link to the inhibition of AAA pathological progression, a hypothesis that requires further in vivo validation.

Fig. 2.

FTO induced VSMCs apoptosis in vitro. (A–C) Specific lentivirus down-regulated the expression of FTO, and RT-qPCR and western blot were used to detect the knockdown efficiency. (D–F) Western blot analysis showed that the expression of cleaved-caspase3 was decreased after silencing FTO. (G,H) Annexin V/ propidium iodide (PI) staining and flow cytometry demonstrated that silencing FTO staining effectively reduced Ang II-induced apoptosis.*P < 0.05,**P < 0.01,***P < 0.001,****P < 0.0001.

FTO specific inhibitor FB23-2 attenuates Ang II-induced AAA formation

To further verify the effect of FTO on AAA formation in vivo, we selected FB23-2, a small-molecule inhibitor of FTO that directly binds to FTO and selectively inhibits its m⁶A demethylase activity²⁰. Previous studies have shown that FB23-2 can significantly inhibit the proliferation of human acute myeloid leukemia cell lines and clear cell renal cell carcinoma^21,22, but its role in AAA remains unclear. In this study, we administered FB23-2 (2 mg/kg/d, 6 mg/kg/d) intraperitoneally to an Ang II-induced AAA mouse model, while the control group received intraperitoneal injection of vehicle for 4 consecutive weeks. The results showed that FB23-2 treatment significantly reduced the diameter of AAA compared to the vehicle control. HE staining further revealed that, in comparison to the AAA model group, the aortic wall lumen of the FB23-2 treatment group exhibited a relatively preserved structure, with the medial layer arranged in an orderly fashion. Additionally, there was a notable reduction in inflammatory cell infiltration throughout all layers of the aortic wall (Fig. 3A,B). Furthermore, Western blot (WB) analysis showed that the FB23-2 (6 mg/kg) group had a significantly reduced activation level of caspase3 protein, suggesting that inhibition of FTO functional activity can suppress Ang II-induced AAA formation and apoptosis in ApoE⁻/⁻ mice (Fig. 3C,D) (Fig. 4A,B).

Fig. 3.

FB23-2, a specific inhibitor of FTO, inhibits Ang II-induced abdominal aortic aneurysm formation. (A) Representative anatomical images and HE staining of different groups of abdominal aortic aneurysm mice. (B) Quantification of the maximum diameter of the suprarenal abdominal aorta measured by digital vernier calipers. The results showed that the diameter of abdominal aortic aneurysm in the FB23-2 (6 mg/kg) dose group was significantly reduced. (C, D) Western blot showed that the expression of the apoptotic protein cleaved-caspase3 was decreased in the aortic tissue of mice in the FB23-2 (6 mg/kg) group. *P < 0.05,**P < 0.01,***P < 0.001.

Fig. 4.

FB23-2 inhibits Caspase-3 expression in abdominal aortic VSMCs in AngII-induced AAA mice.Immunohistochemical staining of Caspase-3 in the abdominal aortic wall of mice from the Control, AngII+Vehicle, and AngII+FB23-2 (6 mg/kg) groups (scale bar, 50 μm). Quantitative analysis of the integrated optical density (IOD)/area ratio for Caspase-3 staining showed that AngII treatment significantly upregulated Caspase-3 expression in aortic VSMCs, while administration of FB23-2 (6 mg/kg) markedly attenuated this effect. Data are presented as mean ± SEM (n = 3 per group). *P < 0.05,**P < 0.01.

RNA-seq and MeRIP-seq confirm that PDK4 is a downstream target of FTO-mediated m⁶A modification

To investigate the downstream targets of FTO-mediated m⁶A demethylation in AAA regulation, we performed MeRIP-seq and transcriptomic sequencing in VSMCs. RNA-seq analysis identified 1,125 overlapping genes between the shNC and shFTO groups (Fig. 5A). Compared with the control group, 225 genes were significantly upregulated and 333 genes were significantly downregulated in the shFTO group (Fig. 5B). MeRIP-seq analysis detected 4,421 overlapping methylation peaks between the shNC and shFTO groups (Fig. 5C), revealing that m⁶A methylation levels were increased at 1,831 mRNA sites in the shFTO group relative to controls (Fig. 5D). The m⁶A modification sites were primarily concentrated in the coding sequence (CDS), 3’ untranslated region (3’ UTR), and stop codon regions: the pie chart illustrates the proportional distribution of peaks across these genomic elements, while the bar chart shows the enrichment ratio of peaks on each corresponding element (Fig. 5E–G). Representative motif analysis was performed concurrently on both cell groups (Fig. 5H).

Fig. 5.

RNA-seq and MeRIP-seq confirm PDK4 as a downstream target of FTO mediated m6A modification. (A,B) Venn diagram of RNA-seq data showing common and unique genes between the shFTO group and the NC group. Bar graphs show differentially expressed genes between NC and shFTO groups. (C,D) Venn diagram of MeRIP-seq data showing the common and unique peaks of m6A RNA methylation in the control and shFTO groups. Bar graphs show the number of differentially methylated peaks between NC and shFTO groups. (E–G) Peak distribution of five mRNA transcription functional regions (CDS region, 5’utr region, 3’ UTR region, TSS region, and stop codon region) in the shFTO group. The pie chart shows the proportional distribution of peaks on each component, and the bar chart shows the enrichment ratio of peaks on the corresponding component. (H) Representative motif analysis of hypermethylated peaks. (I) FTO target gene screening strategy.

To identify the key downstream target through which FTO regulates VSMC apoptosis, we implemented an integrative, biologically driven screening strategy. We first intersected hypermethylated genes from MeRIP-seq with differentially expressed genes from RNA-seq to focus on candidates exhibiting concurrent alterations in both m⁶A methylation and transcriptional levels, a hallmark of FTO-mediated post-transcriptional regulation. We then cross-referenced these candidates with the POST-trAnscriptional Regulation database and prioritized those functionally linked to apoptosis—the core biological process underlying AAA progression in our study. Given that FTO knockdown elevates global m⁶A methylation levels, we further refined our selection to genes with increased m⁶A abundance in the shFTO group. PDK4 emerged as the primary target, as it displayed significant hypermethylation and transcriptional changes, and its well-documented role in regulating apoptosis is closely associated with FTO-dependent VSMC survival. Thus, PDK4 was identified as the downstream target gene through which FTO modulates VSMC apoptosis (Fig. 5I).

FTO-mediated m⁶A modification leads to degradation of PDK4 mRNA in a YTHDF2-dependent manner

A genomic browser visualization of the MeRIP-seq data showed a significant increase in the abundance of m⁶A in PDK4 mRNA 3’ UTR after FTO silencing (Fig. 6A). To verify the MeRIP-seq results, m⁶A modifications in the PDK4 mRNA 3’UTR region were identified using MeRIP-qPCR. The results showed that the m⁶A level of PDK4 mRNA was significantly increased after FTO silencing (Fig. 6B). In addition, EpiQuik m⁶A RNA methylation quantitative assay kit found that the overall m⁶A level of VSMCs was also increased after FTO knockout (Fig. 6C). To verify whether m⁶A modification affects PDK4 expression, we applied FB23-2 in ApoE-/- mouse AAA model and lentivirus silencing FTO in VSMCs. We found that interference with FTO can reduce the increase of PDK4 expression induced by AngII(Fig. 6D–G). To further validate the impact of m⁶A modification on PDK4 expression, we predicted sequence-based RNA adenosine methylation sites by utilizing the SRAMP tool. The mutant (MUT) PDK4 plasmid eliminated the m⁶A modification at the three predicted mRNA adenosine methylation sites in the 3’ UTR of PDK4 mRNA by replacing adenosine with thymidine at this site. Luciferase reporter gene experiment showed that the luciferase activity of wild-type PDK4 decreased significantly after FTO knockout. However, FTO knockout had no effect on MUT PDK4 luciferase activity. The findings indicate that FTO can modulate the expression of PDK4 by altering the m⁶A modification level of PDK4 mRNA. (Fig. 6H,I).

Fig. 6.

FTO-mediated m6A modification leads to the degradation of PDK4 mRNA in a YTHDF2-dependent manner. (A) MeRIP-seq data showed a significant increase in m⁶A abundance in the 3 ‘untranslated region (UTR) of PDK4 mRNA after FTO knockdown, according to the Integrated Genome Browser. (B,C) EpiQuik m⁶A RNA methylation quantification and MeRIP-qPCR showed increased levels of total m⁶A modification and PDK4 mRNA methylation in the shFTO group. (D,E) Western blot showed that the protein expression level of PDK4 was increased in Ang II-induced VSMCs. (F,G) Silencing FTO in VSMCs reduced PDK4 expression. (H) Dual luciferase reporter gene vector structure of wild-type or m⁶A site mutant pdk4-3’ UTR; (I) Measurement of the relative luciferase activity of PDK4-WT and PDK4-MUT after FTO knockdown in VSMCs. *P < 0.05,**P < 0.01, ***P < 0.001.

Next, we conducted a deeper investigation into the underlying mechanism through which m⁶A methylation modulates PDK4 expression. Since m⁶A modification at mRNA 3’ UTR can lead to mRNA degradation, RT-qPCR was used to detect the degradation of PDK4 mRNA in VSMCs after coculture with actinomycin D for 0, 2, 4, and 6 hours. Actinomycin D inhibits gene transcription and mRNA synthesis by binding to double-stranded DNA to form a stable complex that prevents DNA double helix disintegration²³. The results indicated that, in comparison to the control group, the silencing of FTO notably speeded up the degradation process of PDK4 mRNA, consequently reducing its half-life. (Fig. 7A). Although the modification of m⁶A is regulated by the “writer” and “eraser”, its function depends on the selective recognition of the m⁶A site by the “reader“⁸. As an important reading protein of m⁶A modified RNA, YTH domain family proteins can specifically recognize RNA bound to m⁶A modified RNA and mediate mRNA degradation. Distinct from the YTHDF family, which catalyzes mRNA degradation^24,25, the Insulin-like growth factor 2 mRNA-binding proteins family primarily functions to stabilize targeted mRNAs, thereby modulating their expression levels²⁶. This study explored which YTHDF protein played a major role in the degradation of PDK4 mRNA through RIP experiments. RIP-qPCR verification found that YTHDF2 could directly bind to PDK4 mRNA, but not YTHDF1 and YTHDF3 (Fig. 7B). DC-Y13-27 (10 µM) was used to bind YTHDF2 protein directly in vitro to inhibit its functional activity²⁷. The results showed that inhibiting the function of YTHDF2 could delay the degradation of PDK4 mRNA (Fig. 7C). These results suggest that FTO-mediated m⁶A modification leads to the degradation of PDK4 mRNA in a YTHDF2-dependent manner.

Fig. 7.

FTO regulates PDK4 expression levels in a YTHDF2-dependent manner. RT-qPCR was used to detect the stability and degradation of PDK4 mRNA in VSMCs after 0, 2, 4, and 6 h of actinomycin D co-culture. (A) FTO knockdown significantly accelerated PDK4 mRNA degradation compared with the control group. (B) RIP-qPCR confirmed the direct binding of YTHDF2 and PDK4. (C) Inhibition of YTHDF2 delayed PDK4 mRNA degradation. * * * * P < 0.0001.

Overexpression of PDK4 promoted apoptosis induced by FTO

To further confirm whether FTO-mediated apoptosis is dependent on PDK4, we used lentivirus overexpression of PDK4, RT-qPCR and WB to verify the overexpression of PDK4. Flow cytometry experiments showed that PDK4 overexpression partially offset the inhibitory effect of FTO silencing on VSMCs apoptosis. (Fig. 8A-E)Taken together, these results suggest that PDK4 is a key downstream target for FTO to exert its function by erasing m⁶A modifications.

Fig. 8.

PDK4 overexpression promotes FTO-induced apoptosis. (A–C) Western blot was used to verify the efficiency of lentivirus overexpression of PDK4 and cleaved-caspse3. (D,E) Flow cytometry was used to verify the effect of PDK4 expression level change on VSMCs apoptosis. **P < 0.01, ***P < 0.001.

Discussion

As the first RNA demethylation modification enzyme discovered, FTO mediated m⁶A modification disorder has been shown to affect the progression of many diseases, such as liver cancer^28,29, breast cancer³⁰ and so on. However, the precise role of FTO in abdominal aortic aneurysm (AAA) remains incompletely understood. Zhang Jian et al. have reported, for the first time, a significant elevation in the protein levels of FTO within AAA tissue. Although the underlying cause of the discrepancy between FTO mRNA and protein levels remains elusive, the cellular localization of FTO in VSMCs and CD3 + T cells implies that FTO may exert context-dependent effects on AAA progression via distinct cell types³¹. These results suggest that FTO may regulate AAA progression by influencing these two types of cells. In this study, we found for the first time that inhibition of FTO’s functional activity can inhibit Ang II-induced AAA formation and VSMCs apoptosis in mice in vivo and in vitro. From a mechanistic point of view, silencing FTO increases the m⁶A modification level of PDK4 mRNA in the 3’ UTR region; notably, the 3’ UTR region of PDK4 mRNA contains conserved m⁶A consensus motifs (RRACH), which serve as specific binding sites for YTHDF2. The m⁶A modification in the 3’ UTR region of PDK4 mRNA mediates the degradation of its mRNA by enhancing the affinity of YTHDF2 for PDK4 mRNA, thereby reducing the expression level of PDK4. PDK4 is a key negative regulator of mitochondrial pyruvate dehydrogenase complex (PDC), reduced PDK4 expression relieves the inhibitory effect on PDC, promotes the entry of pyruvate into the tricarboxylic acid cycle, enhances mitochondrial oxidative phosphorylation, and ultimately inhibits the apoptosis of VSMCs by maintaining the energy function metabolism of VSMCs mitochondria and enhancing the glucose oxidation rate³². Our results suggest that FTO plays an important role in regulating apoptosis of VSMCs and can be used as a potential target for drug therapy of AAA. Although our results suggest that increased FTO expression in VSMCs is associated with the progression of AAA, it is undeniable that our experiments need to introduce FTO gene knockout/knock-in mouse models to further verify the effect of FTO on the progression of AAA in vivo.

The role of m⁶A modification in the development of AAA is still controversial. The level of m⁶A modification is regulated by a variety of ‘writers’ and’ erasers’, and it is still uncertain which modifying enzyme plays the main role in AAA. Based on the analysis of public databases, Tan et al. found that the expression levels of Methyltransferase (METTL) 14 and Heterogeneous Nuclear Ribonucleoprotein C in AAA were lower than those in the normal group, while the expression levels of RBM15B and FTO were increased³³. However, this prediction result was not completely consistent with their RT-qPCR results. This inconsistency may stem from two aspects: first, the tissue location of AAA detected in this study was different from the proximal neck of the AAA selected in the sequencing data; second, the dynamic changes of m⁶A modification regulatory factors in different stages of AAA development may lead to conflicting results. n other studies, it was found that the expression level of METTL3-METTL14 complex in AAA was higher than that in normal group, and the formation of AAA was accelerated by promoting necrotic apoptosis and inflammation of vascular smooth muscle cells^31,34,35. For the above conflicting results, we hypothesized that three key factors may be responsible: (1) Cell-type specificity: m⁶A regulatory enzymes may exert distinct effects in VSMCs, endothelial cells, and immune cells; (2) Disease stage dependence: the expression of m⁶A writers/erasers may dynamically change during AAA initiation, progression, and rupture; (3) Model heterogeneity: the subcutaneous Ang II infusion model is mostly located at the aorta above the branch of renal artery, which is different from the infrarenal location of human AAA^36,37. Therefore, we may need to verify the expression level and localization of m⁶A modified regulatory factors in other types of animal models. To address the limitations of the single-factor AngII-induced model, future studies can employ elastase-perfusion AAA models or ApoE−/− mice fed a high-fat diet, which better simulate the pathological features of human AAA, thus avoiding the deficiency that a single model cannot fully simulate the pathological link of human AAA.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

m⁶A

N⁶-methyladenosine

FTO

fat mass and obesity associated

VSMCs

vascular smooth muscle cells

RNA-seq

transcriptome sequencing

MeRIP-seq

methylated RNA immunoprecipitation sequencing

PDK4

pyruvate dehydrogenase kinase 4

YTHDF2

YTH N⁶-Methyladenosine RNA Binding Protein F2

Ang II

Angiotensin II

RIP

RNA binding protein immunoprecipitation

UTR

Untranslated Regions

METTL

Methyltransferase

Author contributions

Lei Li and Yu Tian designed or/and supervised this project and revised the manuscript .Wei Gong and Xue Yang wrote the manuscript, performed the experiments and analyzed data. Han Wang and Shenjie Fu analyzed the data. All authors read and gave final approval of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China grant number: (82070485).

Data availability

The datasets generated and/or analysed during the current study are available in the Sequence ReadArchive repository, [PRJNA1345027]

Declarations

Competing interests

The authors declare no competing interests.

Consent for publication

All of the authors are aware of and agree to the content of the paper and their being listed as a co-author of the paper.