Regulation of renal ischemia-reperfusion injury and tubular epithelial cell ferroptosis by pparγ m6a methylation: mechanisms and therapeutic implications

Abstract

This study aimed to elucidate the role and underlying mechanisms of Peroxisome proliferator-activated receptor gamma (PPARγ) and its m6A methylation in renal ischemia-reperfusion (I/R) injury and ferroptosis of tubular epithelial cells (TECs). High-throughput transcriptome sequencing was performed on renal tissue samples from I/R injury models and sham-operated mice, complemented by in vivo and in vitro experiments focusing on the PPARγ activator Rosiglitazone and the manipulation of METTL14 and IGF2BP2 expression. Key evaluations included renal injury assessment, ferroptosis indicator measurement, and m6A methylation analysis of PPARγ. Our findings highlight the critical role of the PPARγ pathway and ferroptosis in renal I/R injury, with Rosiglitazone ameliorating renal damage and TEC ferroptosis. METTL14-mediated m6A methylation of PPARγ, dependent on IGF2BP2, emerged as a pivotal regulator of PPARγ expression, renal injury, and ferroptosis. This study reveals that PPARγ m6A methylation, orchestrated by METTL14 through an IGF2BP2-dependent mechanism, plays a crucial role in mitigating renal I/R injury and TEC ferroptosis. These insights offer promising avenues for therapeutic strategies targeting acute kidney injury.

Introduction

Ischemia/reperfusion (I/R) injury of the kidney is a major cause of acute kidney injury (AKI) and occurs when renal blood flow is temporarily interrupted and then restored [1, 2]. Renal I/R injury is prevalent in various clinical situations, such as renal transplantation, major surgery, and shock [3, 4]. The incidence of AKI varies significantly among different patient populations, but overall, its prognosis is closely related to long-term mortality, incomplete renal function recovery, and the development of chronic kidney disease [1, 5]. Therefore, renal I/R injury is not only a significant clinical problem but also a focus of medical research. Despite the progress in understanding renal I/R injury, the exact molecular mechanisms are still not fully understood, which limits the development of effective treatment methods.

Renal tubular epithelial cells (TECs) play a crucial role in the structure and function of the kidney [6,7,8]. During the process of renal I/R injury, TECs suffer significant damage, triggering a series of pathophysiological responses that lead to a significant decline in renal function [9]. In recent years, ferroptosis has been found to play a critical role in TEC injury [10]. Ferroptosis is a unique form of cell death driven by excessive intracellular iron ions and lipid peroxidation reactions [11,12,13]. In renal I/R injury, the activation of ferroptosis leads to the loss of a large number of TECs, exacerbating renal functional damage. Therefore, understanding and controlling ferroptosis of TECs is of significant importance for alleviating renal I/R injury and its long-term complications.

Peroxisome proliferator-activated receptor gamma (PPARγ), a type of nuclear receptor, is widely involved in cellular metabolism, inflammatory responses, cell growth, and death processes [14,15,16]. In the study of kidney diseases, PPARγ has gradually shown its importance [17,18,19]. Particularly, in the context of renal I/R injury, the activity and expression level of PPARγ significantly influences the fate of TECs [20]. On one hand, PPARγ activation has been found to alleviate renal injury and improve renal function. On the other hand, the dysregulation of PPARγ may make TECs more susceptible to ferroptosis, exacerbating renal damage [21,22,23]. Therefore, a thorough understanding of the role of PPARγ in renal I/R injury, especially its involvement in regulating ferroptosis of TECs, is of great significance for the development of novel therapeutic strategies.N6-methyladenosine (m6A) methylation is one of the most common RNA modifications and plays a vital role in RNA metabolism, stability, and function [24,25,26]. In gene regulation, m6A methylation regulates protein expression by affecting mRNA translation efficiency and stability [27,28,29]. In recent years, research has shown that m6A methylation plays a crucial role in various diseases, including kidney diseases [30,31,32]. In the context of renal I/R injury, the expression and function of PPARγ may be regulated by m6A methylation [20, 33]. In particular, two essential regulators of m6A methylation, METTL14 and IGF2BP2, play significant roles in the m6A methylation of PPARγ. METTL14 acts as the methyltransferase for m6A methylation, while IGF2BP2 is an m6A-modified RNA recognition protein. Together, they regulate the activity of PPARγ, thereby affecting the development of renal I/R injury and the process of ferroptosis of TECs [34].The purpose of this study is to explore the role of PPARγ in renal I/R injury and ferroptosis of TECs and its regulation mechanism by m6A methylation. We conducted in-depth research on the expression and functional changes of PPARγ in a renal I/R injury model and AKI patient samples through high-throughput transcriptome sequencing, transmission electron microscopy observation, and various molecular biology techniques. By experimentally manipulating the activity and expression of PPARγ, we investigate the role of PPARγ and its m6A methylation in the regulation of renal injury and ferroptosis of TECs. The research results can not only enhance our understanding of the molecular mechanisms of renal I/R injury but also provide a theoretical basis for the development of new strategies for treating AKI. Especially for clinical AKI patients, this study may bring new treatment ideas and methods with the potential to improve patients’ prognosis and quality of life. Therefore, this study has significant implications in both scientific and clinical fields.

Materials and methods

Ischemia reperfusion (I/R) injury mouse model construction

The experimental animals used in this study were 3-month-old male C57BL/6J mice (catalog number 219, Beijing Vetconlyhua, China). The mice were housed in a specific pathogen-free laboratory, and a total of 62 mice were used, with approval from the National Ethics Committee of The First Affiliated Hospital, Jiangxi Medical College, Nanchang University (No. CDYFY-IACUC-202303QR004). The mice were fasted for 12 h before surgery, and anesthesia was induced by intraperitoneal injection of a 1% sodium pentobarbital solution (40 mg/kg). After successful anesthesia, a 1 cm incision was made on both sides of the back to expose the kidneys. When the needle was felt to penetrate the renal capsule, a slow injection of lentivirus or drugs was started. To avoid bleeding, the needle should not be inserted too deeply. After completion of the injection, the needle was left in place for 1 to 2 s before removal to prevent lentivirus overflow. During the injection, the needle should be kept perpendicular to the kidney, and the injection speed should be fast to minimize kidney injury. Subsequently, the incision was sutured, and the mice underwent a renal ischemia/reperfusion (I/R) injury experiment after 10 days.

Model group (I/R group): Mice were anesthetized by intraperitoneal injection of a 1% sodium pentobarbital solution and placed on an operating table with their limbs fixed. The abdominal hairs of the mice were shaved, and the surgical area was disinfected with iodine and 75% ethanol. The abdominal cavity was then opened, and the left renal artery was carefully separated and clamped with a non-traumatic vascular clamp. After 0.5 min of clamping, the blocked area was visibly dark red, indicating successful clamping. After 30 min of continuous ischemia, the clamp was released, and the color of the left kidney changed from dark red to bright red, indicating successful reperfusion. After suturing, the mice were returned to the laboratory for regular observation of their condition and recording of any deaths, and follow-up experiments were conducted at 6, 12, and 24 h after reperfusion by anesthetizing the mice and collecting blood samples from the eye socket, as well as sampling the right renal vein and ligating the vein, followed by removal of the right kidney after 3 days. Kidney tissue samples of mice from each group were collected for subsequent testing [35].

Sham group: The left renal pedicle was left untreated, and only the right kidney was removed.

To study the effect of Rosiglitazone on I/R injury, the mice in the I/R group were randomly divided into the I/R + Vehicle group and the I/R + Rosiglitazone group, with 8 mice in each group. Rosiglitazone (10 mg/kg) or an equal volume of DMSO was intraperitoneally injected 30 min before ischemia and for 3 consecutive days after surgery. Rosiglitazone (catalog number R2408) was purchased from Sigma-Aldrich.

To study the effect of the METTL14-IGF2BP-PPARγ signaling axis on I/R injury, the mice in the I/R group were randomly divided into the oe-NC + sh-NC + oe-NC group (infected with lentivirus oe-NC, sh-NC, and oe-NC), the oe-METTL14 + sh-NC + oe-NC group (infected with lentivirus oe-METTL14, sh-NC, and oe-NC), the oe-METTL14 + sh-IGF2BP2 + oe-NC group (infected with lentivirus oe-METTL14, sh-IGF2BP2, and oe-NC), and the oe-METTL14 + sh-IGF2BP2 + oe-PPARγ group (infected with lentivirus oe-METTL14, sh-IGF2BP2, and oe-PPARγ), with 8 mice in each group. Each mouse was injected with 6.4 × 10⁶ TU of lentivirus, with an injection volume of 20 µL.

High-throughput transcriptome sequencing

We collected kidney tissue samples from the Sham group (n = 3) and I/R group (n = 3) for high-throughput transcriptome sequencing. The specific steps are as follows: Total RNA was extracted from each sample using Trizol reagent (T9424, Sigma-Aldrich, USA) according to the manufacturer’s instructions. The concentration, purity, and integrity of RNA were measured using Qubit^®2.0 Fluorometer^® (Life Technologies, USA) with the Qubit^® RNA analysis kit, Nanometer spectrophotometer (IMPLEN, USA), and Bioanalyzer 2100 system with the RNA Nano 6000 analysis kit (Agilent, USA), respectively. The measurement results for RNA concentration, purity, and integrity should ensure an A260/280 ratio between 1.8 and 2.0. The total RNA content of each sample was 3 µg, which was used as the input material for RNA sample preparation. Following the manufacturer’s instructions, cDNA libraries were generated using the NEBNext^® UltraTM RNA Directional Library Prep Kit for Illumina (E7760S, New England Biolabs, China), and their quality was evaluated on an Agilent Bioanalyzer 2100 system. According to the manufacturer’s guidelines, indexed samples were clustered using the Illumina TruSeq PE Cluster Kit v3 cBot HS on the cBot cluster generation system. After clustering, the libraries were sequenced on an Illumina Hiseq 550 platform to generate 125 bp/150 bp paired-end reads [36].

Quality control and differential analysis of sequencing data

The quality of paired-end reads in the raw sequencing data was examined using FastQC software (v0.11.8). The raw data was preprocessed using Cutadapt software (v1.18) to remove Illumina sequencing adapters and poly(A) tail sequences. A Perl script was used to discard reads with more than 5% N content. The FASTX Toolkit software (v0.0.13) was utilized to extract reads with a base quality of 20 or higher, representing 70% of total bases. Double-ended sequences were repaired using BBMap software (v39.01). Finally, the filtered high-quality reads were aligned to the human reference genome using hisat2 software (v0.7.12).

Based on the high-throughput transcriptome sequencing data, differential analysis was performed using the R package limma [37]. Genes were considered differentially expressed if |logFC| > 1 and P < 0.05. Differential expression genes (DEGs) [38] were identified by conducting separate analyses for S2_2 vs. S1_1, S2_2 vs. S1_1, and S2_2 vs. S1_1. The intersecting sets of upregulated and downregulated genes from the three comparisons were obtained.

Gene functional enrichment analysis

We performed gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis using the R software package clusterProfiler [39]. The intersection genes with both upregulated and downregulated expressions were analyzed with a significance threshold of P < 0.05. The GO terms comprised biological function (BP), cellular component (CC), and molecular function (MF).

Detection of renal biochemical indicators

Mouse blood samples (1 mL) were collected for the measurement of serum creatinine (Scr) and blood urea nitrogen (BUN) levels using the Mouse Serum Creatinine ELISA kit (catalog number: kt21413, MSK) and the Mouse Blood Urea Nitrogen ELISA kit (catalog number: kt21234, MSK). These measurements were further employed to assess renal function [40].

Periodic acid-Schiff (PAS) staining

Three days after ischemia/reperfusion (I/R), mice were euthanized, and renal tissues were collected. The samples were fixed, dehydrated, paraffin-embedded, and processed following routine procedures. PAS staining (ab150680, Abcam, UK) was then performed, and a histological examination was conducted under a light microscope. The degree of tubular pathology was determined based on the percentage of cell necrosis and brush border loss, as follows: 0: none; 1: 0–10%; 2: 11–25%; 3: 26–45%; 4: 46–75%; 5: >75% [41].

Immunofluorescence and immunohistochemistry

For cellular samples, the cells to be tested are seeded onto appropriate slides and cultured under suitable conditions until the desired cell density is reached. Cells are fixed with 4% formaldehyde for 20 min, followed by three washes with PBS. The cells were then treated with 0.1% Triton X-100 solution for 10 min and washed three times with PBS. Non-specific binding sites are blocked with 10% goat serum (catalog number: E510009, Shanghai Bioengineering Co., Ltd.) for 30 min. For tissue samples, paraffin-embedded kidney Sect. (4 μm) are deparaffinized in xylene, rehydrated in graded alcohol, and blocked with 10% goat serum for 30 min. Mouse monoclonal antibody anti-4-HNE (ab48506, 1:25, Abcam, UK), rabbit polyclonal antibody anti-CK18 (SAB4501663, 1:100, Sigma-Aldrich, USA), rabbit polyclonal antibody anti-Kim-1 (LAA785Mu81, 1:100, Wuhan Yunclone Technology Co., Ltd., China), rabbit polyclonal antibody anti-Ki67 (ab15580, 1:100, Abcam, UK), rabbit monoclonal antibody anti-Sox9 (ab185966, 1:1000, Abcam, UK), rabbit polyclonal antibody anti-PPARγ (ab59256, 1:50, Abcam, UK), rabbit monoclonal antibody anti-CD3 (ab135372, 1:10, Abcam, UK), rabbit polyclonal antibody anti-ly6G (ab238132, 1:2000, Abcam, UK), rabbit monoclonal antibody anti-F4/80 (ab300421, 1:5000, Abcam, UK) primary antibodies are incubated overnight at 4 °C, followed by visualization using goat anti-mouse IgG (ab6785, 1:1000, Abcam, UK) or goat anti-rabbit IgG (ab150083, 1:200, Abcam, UK) or goat anti-rabbit IgG (ab150077, 1:1000, Abcam, UK) or DAPI (ab104139, 1:1000, Abcam, UK). Fifteen randomly selected blind areas on each slide are carefully quantitatively stained, and the data are analyzed using Image Pro Plus software (Media Cybernetics, USA). The number of neutrophils, T cells, and macrophages infiltrating within 6–8 fields of view in the kidney is calculated [42].

In vivo pharmacokinetic study

We selected 6 healthy C57BL/6J mice (catalog number: 219, Beijing Vitonlihua, China), administered a dose of 3 mg/kg rosiglitazone, and euthanized them at 0.25, 0.5, 1, 2, 4, and 6 h after administration. LC-MS/MS method was used to measure the concentration of rosiglitazone in mouse plasma, kidney, liver, and brain tissues. The plasma sample processing method involved adding 160 µL of the sample to 8 µL of internal standard (IS) protein precipitation, vortex mixing, and centrifugation at 12,000 × g for 15 min at 4 °C. The supernatant (70 µL) was transferred to a 96-well plate, followed by centrifugation at 3,220 × g for 5 min at 4 °C. Finally, 6 µL of the supernatant was directly analyzed by LC-MS/MS. The tissue sample processing method involved homogenizing the tissues using a homogenization solution with a threefold volume of MeOH/15 mM PBS (1:2). The tissue sample processing steps included adding 400 µL of the sample to 20 µL of internal standard (IS) protein precipitation, vortex mixing, and centrifugation at 13,400 × g for 15 min at 4 °C. The supernatant (70 µL) was transferred to a 96-well plate, followed by centrifugation at 3,220 × g for 5 min at 4 °C. Finally, 6 µL of the supernatant was directly analyzed by LC-MS/MS [43].

Isolation, cultivation, identification, and cell model construction of primary tubular epithelial cells (TECs) in mice

TECs were isolated from C57BL/6J mice. In brief, after kidney removal, the renal tubular tissue was minced and digested with collagenase/dispase solution. The digested fragments were passed through 100 μm and 70 μm sieves. Subsequently, cells captured on a 40 μm sieve were isolated and grown in DMEM/HAM’s F12 (1:1) medium supplemented with GlutaMAX (31331-028, Gibco, USA), 10% fetal bovine serum (FBS) (10100147, Invitrogen, USA), and 1% penicillin/streptomycin (10378016, Invitrogen, USA) at 37 °C and 5% CO₂ in a humid environment. Cell morphology was observed, and the epithelial cell characteristics were confirmed using immunohistochemical staining with cytokeratin 18 (CK18) as a marker [44]. Construction of an in vitro hypoxia/reoxygenation (H/R) model involved culturing TECs under hypoxic conditions (1% O₂, 5% CO₂, and 94% N₂) for 24 h, followed by reoxygenation (21% O₂, 5% CO₂, and 74% N2) for 12 h [45].

Mitotracker

Red CM-H2XRos (MX4308-50UG, Mocan Biotechnology, China) is a non-fluorescent reduced form of MitoTracker Red CMXRos (MX4307), which fluoresces only when oxidized. This dye can also stain live cell mitochondria, with its accumulation dependent on membrane potential, and it retains good staining after aldehyde fixation. The general procedure involves placing the cells on a coverslip in a culture dish and adding the appropriate culture medium for cell growth. Once the cells reach an appropriate density, the culture medium is removed, and an appropriate preheated probe-containing medium is added at 37 °C and incubated for 45 min under specific growth conditions. The staining solution is replaced with a fresh culture medium, and the cells are observed under a fluorescent microscope (Olympus Corp, Japan). The cells are centrifuged, and the supernatant is removed. The cells are then resuspended in a preheated culture medium containing the MitoTracker probe at 37 °C and incubated for another 45 min under specific growth conditions. The centrifugation step is repeated, the staining solution is removed, and the cells are resuspended in a fresh culture medium before being observed under a fluorescent microscope. In this study, MitoTracker staining was used to label mitochondria in red, while Rosiglitazone was labeled in green to observe their co-localization [46].

CRISPR/Cas9 gene editing technology

The IGF2BP2-KO cells were constructed using CRISPR/Cas9 technology. The sgRNA sequences used were as follows: IGF2BP2-sgRNA: Forward primer: 5’-TCCGCTAGGCCTTCTCCTTG-3’ (PAM: AGG), Reverse primer: 5’-TGACATCAATAGTTAATACA-3’ (PAM: TGG). The sgRNA was inserted into the Lenti-CRISPR v2 vector (HanHeng Biotechnology, Shanghai, China) containing the Streptococcus pyogenes Cas9 nuclease gene. The cells were transduced with the lentiviral Lenti-CRISPR v2 vector to generate IGF2BP2-KO cells using the CRISPR/Cas9 editing system. The cells transfected with the sgRNA plasmid and the donor sequence were selected using 4 µg/mL puromycin (HY-K1057, MCE, USA). The selected IGF2BP2-KO cells were validated through limiting dilution cloning, and their knockout was confirmed by RT-qPCR and Western blot analysis [47, 48].

HEK293T cell culture

The human embryonic kidney cell line, HEK293T (CBP60661, Nanjing Kebei Biotechnology Co., Ltd., Jiangsu, China), was cultured in DMEM medium (Catalogue No: 11965092, Gibco, USA) supplemented with 10% FBS, 10 µg/mL of streptomycin, and 100 U/mL of penicillin. The cells were cultured in a humidified incubator (Heracell™ Vios 160i CR CO₂ Incubator, Catalogue No: 51033770, Thermo Fisher Scientific, USA) at 37℃ with 5% CO₂. Subculturing was performed when the cells reached 80%~90% confluence [49].

Construction of silence and overexpression lentiviral vectors

Potential short hairpin RNA (shRNA) target sequences were analyzed for the mouse cDNA sequences using GenBank as a reference. Three target sequences were designed for METTL14, IGF2BP2, and PPARγ, respectively. A control sequence without any interfering sequence (sh-NC) was also designed (for primer sequences, Table S1). These oligonucleotides were synthesized by GimaGen and used for the construction of lentiviral packaging systems in the lentiviral interference vector LV-1 (pGLVU6/GFP) (C06001, GimaGen, China). The packaging viruses and target vectors were co-transfected into HEK293T cells using lipofectamine 2000 when the cell confluence reached 80–90%. After 48 h of cell culture, the supernatant was collected and subjected to filtration and centrifugation to isolate viral particles. The viruses in the logarithmic growth phase were collected, and the viral titer was determined. Lentiviruses overexpressing METTL14 or PPARγ were constructed and packaged by GeneChem using the lentiviral gene overexpression vector LV-PDGFRA [50,51,52,53].

Cell transfection and grouping

When cells reached the logarithmic phase of growth, they were digested with trypsin and resuspended to prepare a cell suspension at a concentration of 5 × 10⁴ cells/mL. The cell suspension was then seeded into a 6-well plate with 2 mL per well. Prior to cell grouping, various lentiviruses (MOI = 10, viral titer of 1 × 10⁸ TU/mL) were added to the cell culture medium and incubated for 48 h. Stable cell strains were screened using 2 µg/mL puromycin for a continuous period of 2 weeks. The overall cell grouping was as follows: (1) sh-NC group, sh-METTL14 group, oe-NC group, oe-METTL14 group; (2) oe-METTL14 + sh-NC group, oe-METTL14 + sh-PPARγ group; (3) IGF2BP2-WT group, IGF2BP2-KO group; (4) IGF2BP2-WT (oe-NC) group, IGF2BP2-WT (oe-METTL14) group, IGF2BP2-KO (oe-NC) group, and IGF2BP2-KO (oe-METTL14) group. After 48 h of transfection, RNA and protein levels were examined to verify the knockdown or overexpression effects [54].

RT-qPCR

Total RNA was extracted using the Trizol reagent (T9424, Sigma-Aldrich, USA). The quality and concentration of RNA were measured using a UV-visible spectrophotometer (ND-1000, Nanodrop, Thermo Fisher, USA). Reverse transcription was performed using the PrimeScript™ RT-qPCR kit (RR086A, TaKaRa, Mountain View, CA, USA). Real-time quantitative reverse transcription polymerase chain reaction (RT-qPCR) was carried out using the SYBR Premix Ex TaqTM (DRR820A, TaKaRa, Japan) on a LightCycler 480 system (Roche Diagnostics, Pleasanton, CA, USA). GAPDH was used as the internal reference for mRNA. The primers used for amplification were designed by Shanghai Universal Biotechnology Co., Ltd. See Table S2 for primer sequences. The 2^−ΔΔCt method was used to determine the fold-change in target gene expression between the experimental and control groups, with the formula as follows: ΔΔCT = ΔCt experimental group - ΔCt control group, where ΔCt = target gene Ct - reference gene Ct [55].

Western blot

Cellular total protein was extracted using the RIPA lysis buffer (R0010, Solarbio, Beijing, China) according to the manufacturer’s instructions. After 15 min of lysis at 4 °C, the supernatant was collected by centrifugation at 12,000 × g for 15 min. The protein concentration of each sample was determined using the BCA assay kit (20201ES76, Yisheng Biotechnology Co., Ltd., Shanghai, China). Quantification was performed based on different concentrations, and the proteins were separated by SDS-PAGE and transferred to a PVDF membrane using a wet transfer method. The membrane was then blocked with 5% BSA at room temperature for 1 h. Primary antibodies were incubated overnight at 4 °C (see Table S3 for antibody information), followed by membrane washing with TBST for 5 min × 3 times. HRP-conjugated goat anti-rabbit IgG (ab205718, 1:20000, Abcam, UK) dilution solution was added at room temperature for 1 h. The membrane was washed with TBST for 5 min × 3 times and then visualized using a developing solution. Protein quantification analysis was performed using ImageJ (v1.48) software (National Institutes of Health) by calculating the ratio of the grayscale value of each protein to that of the internal control GAPDH [56]. The experiment was repeated three times.

Quantification of total m6A in RNA

The total m6A levels in RNA were measured using the EpiQuik m6A RNA Methylation Kit (Colorimetric) (P-9005-48, Epigentek, USA). Briefly, total RNA was extracted from the cells of interest, and its concentration was determined using a Nanodrop 2000 (Thermo Fisher Scientific, USA). In each well, 200 ng of total RNA and the binding solution were added, followed by incubation at 37 °C for 1.5 h. Then, the capture antibody, detection antibody, enhancer solution, colorimetric solution, and stop solution were sequentially added to each well and reacted for a specified time. Finally, the absorbance at 450 nm was measured using a microplate reader (Molecular Devices, USA) to compare the relative m6A levels in each group [57].

Methylated RNA immunoprecipitation (MeRIP)-PCR analysis

The Magna MeRIP™ m6A Kit (17-10499) from MERCK was used to detect m6A methylation as per the manufacturer’s instructions. In brief, total RNA (300 µg) was fragmented into approximately 100 nucleotide fragments. Then, an m6A antibody (10 µg; ab286164, Abcam, UK) was pre-coated onto protein A/G magnetic beads (88802, Thermo Fisher Scientific). Subsequently, the RNA fragments were thoroughly mixed with the m6A antibody-coated magnetic beads and incubated for 2 h under rotation. The immunoprecipitated methylated RNA fragments were eluted for PCR analysis [58]. The m6A levels of PPARγ mRNA were quantitatively analyzed using 2^−ΔΔCt with input RNA that did not undergo immunoprecipitation. All real-time PCR primers were designed to span at least one intron, and the amplification of a single product was confirmed by agarose gel visualization and/or melting curve analysis. See Table S4 for the MeRIP-PCR primers [59].

Dot blot

RNA was heat-denatured with an equal amount of 20× sodium citrate buffer (P4922, Sigma-Aldrich, USA) at 95℃ for 5 min. Then, 100, 200, or 400 ng of Poly(A) RNA was added onto a Hybond N membrane (RPN1520B, GE Healthcare, USA). After UV cross-linking for 30 min and blocking with 5% milk, the membrane was incubated overnight at 4℃ with an anti-m6A antibody (ab286164, Abcam, UK). Subsequently, the membrane was incubated with a secondary antibody (IgG, ab6721, 1:1000, Abcam, UK) and visualized using ECL [58].

RNA stability analysis

To measure the stability of PPARγ mRNA in each group of TECs cells, we treated the cells with Actinomycin D (SBR00013, Sigma-Aldrich, USA) at a concentration of 5 µg/ml. After a specific incubation period, we collected cell samples at 2, 4, 6, and 8 h, followed by RNA extraction from each sample for real-time quantitative polymerase chain reaction (qPCR) using GAPDH as the reference gene [60].

Dual-luciferase gene reporter assay

Based on data from the RMBase database (https://rna.sysu.edu.cn/rmbase/m6Amod.php), we screened for potential m6A methylation sites in the PPARγ mRNA and designed mutant sequences (Figure S5). These wild-type and mutant sequences were then inserted into the firefly luciferase reporter vector pGL3-basic (4351372, Thermo Fisher Scientific, USA) to construct the firefly luciferase reporter vectors encoding PPARγ mRNA sequences (pGL3-PPARγ-WT) and mutant sequences (pGL3-PPARγ-MUT1,2,3,4). TECs were seeded in a 24-well plate and cultured overnight. Lipofectamine 3000 reagent (L3000015, Invitrogen, USA) was used to co-transfect TECs with PPARγ pGL3-Empty and pGL3-5’UTR, pGL3-3’UTR, pGL3-CDS plasmids, along with oe-NC or oe-METTL14 and Renilla luciferase plasmids. Next, pGL3-Empty, pGL3-PPARγ-WT, pGL3-PPARγ-MUT1, pGL3-PPARγ-MUT2, pGL3-PPARγ-MUT3, and pGL3-PPARγ-MUT4 were co-transfected into TECs cells with oe-NC or oe-METTL14 and Renilla luciferase plasmids using Lipofectamine 3000 reagent. After 48 h, the cells were harvested for luciferase assay. Firefly luciferase activity was determined using the Dual-Luciferase Reporter Assay System E1910 (Promega, USA) and normalized to Renilla luciferase levels [61].

RNA immunoprecipitation (RIP)

The interaction between IGF2BP2 and PPARγ mRNA in TECs was examined using the EZ-Magna RIP RNA-Binding Protein Immunoprecipitation kit (Merck Millipore, USA). RIP was performed according to the manufacturer’s instructions, with the following general protocol: Firstly, the IGF2BP2 gene was fused with a Flag tag at the DNA level to construct an expression vector. The constructed Flag-IGF2BP2 expression vector was then transfected into TECs to induce overexpression within the cells. After harvesting the cells, they were lysed in a complete RIP lysis buffer. Next, the cell extracts were incubated overnight with magnetic beads coated with anti-mouse IgG (Merck Millipore, USA) or anti-Flag-IGF2BP2 (ab128175, Abcam, UK) in IP buffer. Subsequently, the RNA-protein complexes were washed and incubated with proteinase K (107393, Sigma-Aldrich, USA), and then total RNA was isolated from the extracts using TRIzol reagent for PCR-qPCR analysis [61].

Observation using transmission Electron microscope (TEM)

A transmission electron microscope was used to observe mitochondrial damage in TECs. Samples were fixed overnight in a 2.5% glutaraldehyde solution at 4 °C, then in a 1% osmium tetroxide solution for 1–2 h at room temperature. Dehydration was performed using a graded ethanol series (50%, 70%, 80%, 90%, and 95%), followed by treatment with pure acetone and overnight infiltration with pure embedding resin. Finally, the samples were embedded and heated overnight at 70 °C. This process resulted in properly embedded samples. Ultrathin sections of 70–90 nm were obtained using the UM10 ultramicrotome (Jiangsu Leibo Scientific Instruments Co., China). Subsequently, the sections were stained with lead citrate solution and saturated alcoholic uranyl acetate solution in 50% ethanol for 15 min each, enabling observation using the transmission electron microscope [62].

Reduced glutathione (GSH) and oxidized glutathione (GSSG)

GSH and GSSG levels in tissues or cells were measured using GSH and GSSG assay kits (S0053, Beyotime Biotechnology Co., Ltd, China) purchased from Beyotime. The absorbance of the samples was measured at 450 nm using a microplate reader (Infinite200, Tecan, Beijing), and quantification was done using a standard curve [63].

MDA

MDA levels in different cell groups or mouse kidney tissues were quantified using the malondialdehyde (MDA) assay kit (S0131S, Beyotime, China) [64].

Fe²⁺ content

Intracellular Fe²⁺ levels were evaluated using the FerroOrange probe (F374, Dojindo, Japan). Pre-treated TECs were seeded on confocal culture dishes, washed with Hank’s balanced salt solution (HBSS, 13150016, Gibco, USA), and incubated with 1 µM FerroOrange for 30 min. Fluorescence intensity was observed under a confocal laser scanning microscope (LSM780, Zeiss), and the average fluorescence intensity was measured to assess intracellular Fe²⁺ content [65]. Additionally, the iron assay kit (ab83366, Abcam, UK) was used following the manufacturer’s instructions to determine iron levels in cells and tissue lysates [62].

Reactive oxygen species (ROS)

Cellular ROS production was detected using the 2,7-dichlorofluorescein diacetate (DCFH-DA) fluorescent probe in the ROS assay kit (S30033S, Beyotime, China). To quantify ROS levels, cells were seeded in a 6-well plate and incubated with a 10 µM DCFH-DA solution in the dark for 20 min. After digestion, centrifugation, and resuspension, cells were analyzed using a flow cytometer (FC500ML, Beckman, USA). For visualization of ROS, cells were stained with DCFH-DA for 30 min and kept away from light. Fluorescence images were obtained using a confocal laser scanning microscope (LSM780, Zeiss) with an excitation wavelength of 488 nm and an emission wavelength of 525 nm. Average fluorescence intensity was measured for quantitative analysis of relative ROS abundance within the cells [66].

Lipid peroxidation

Cells grown on confocal culture dishes were stained with 5 µM C11-BODIPY581/591 in the dark for 15 min and observed under a fluorescence microscope. The fluorescence properties of C11-BODIPY581/591 shift from red to green when oxidized by free radicals. For MDA detection, the supernatant of cell homogenates and prepared working reagents were transferred to a 96-well plate. Absorbance measurements at 600 nm, 532 nm, and 450 nm were performed using an automated microplate reader (Infinite200, Tecan, China) for further calculation of non-oxidized and oxidized lipid levels, allowing for quantitative assessment of oxidized lipid levels [62, 65, 67].

Mitochondrial membrane potential (MMP)

To measure the mitochondrial membrane potential (MMP), cells were seeded into a 6-well culture plate and treated with the JC-1 mitochondrial membrane potential detection kit (40706ES60, Yisheng Biotech, Shanghai, China). 1 mL of JC-1 staining working solution was added to each well and thoroughly mixed. The cells were then incubated at 37 °C in a cell culture incubator for 20 min. To serve as a positive control, CCCP (50 mM), provided in the kit, was added to the cell culture medium at a 1:1000 ratio, resulting in a final dilution of 50 µM, and the cells were treated for an additional 20 min. After the 37 °C incubation, the supernatant was aspirated, and the cells were washed twice with 1x JC-1 staining buffer. Subsequently, 2 mL of cell culture medium containing serum and phenol red was added. Fluorescence microscopy was performed, with excitation and emission wavelengths set at 490 nm and 530 nm, respectively, to detect JC-1 monomers [68]. The fluorescence intensities of JC-1 monomers (green) and JC-1 aggregates (red) in the mitochondria were quantified using Image Lab 4.1 software (Bio-Rad Laboratories, Hercules, USA) to estimate the mitochondrial membrane potential. The JC-1 aggregates/monomers ratio was calculated for each group [69].

Masson’s trichrome staining

The extent of kidney fibrosis was assessed using the Masson’s trichrome staining kit (DC0032, Leagene Biotechnology, Beijing, China). Initially, 4 μm thick sections were deparaffinized and hydrated. Subsequently, staining with hematoxylin was performed for 5–10 min, differentiation in acid alcohol for 5–15 s, followed by rinsing and application of Masson’s blue solution for 3–5 min. After rebluing, the sections were incubated in 1% eosin for 5–10 min, washed and then treated with phosphomolybdic acid solution for 1–2 min. Subsequently, the sections were stained with aniline blue solution for 1–2 min, dehydrated in ethanol, cleared in xylene, and finally mounted for observation. Images were captured at 200x magnification using an Olympus BX51 microscope (Tokyo, Japan) and analyzed using ImagePro Plus 6.0 software. Three random sections per mouse were analyzed, with five fields per section observed under high power, following a single-blind protocol. The degree of fibrosis was quantified by measuring the ratio of blue-stained area to total kidney area (Collagen volume fraction, CVF).

TUNEL staining

Mouse kidney tissues were fixed in 4% paraformaldehyde for 15 min, washed thrice with PBS, permeabilized in 0.1% Triton-X 100 in PBS for 3 min. Subsequently, TUNEL staining kit (C1090, Beyotime, Shanghai, China) was used to stain the renal tissue cells. 50 µL biotin-labeled solution was added to the samples, followed by incubation at 37 °C in the dark for 60 min. After washing with PBS, 0.3 mL labeling reaction stop solution was added, followed by PBS washes. Then, 50 µL Streptavidin-HRP working solution was added and incubated at room temperature for 30 min, followed by PBS washes. DAB chromogen was added and incubated at room temperature for 5 min, followed by additional PBS washes. DAPI (10 µg/mL) was used for nuclear counterstaining for 10 min. The images were observed with a confocal microscope, and ImagePro Plus 6.0 software was employed to calculate the apoptotic cell ratio in each group.

Flow cytometry analysis of F4/80₊/CD11b₊/CD86₊ macrophage proportions

Following different treatments, kidney cells were stained to detect macrophages using PE anti-mouse CD86 (1 µg/test, 105007, BioLegend); APC anti-mouse CD206 (0.5 µg/test, 141707, BioLegend); PerCP-Cy5.5 anti-mouseCD11b (0.25 µg/test, 101227, BioLegend). Cells were incubated with antibodies, centrifuged, and resuspended, followed by incubation in the dark at room temperature for 15 min with binding buffer. Flow cytometry was utilized to detect cell apoptosis with an excitation wavelength of 488 nm. The experiment was repeated thrice. [70]

Statistical analysis

The R language version 4.2.1 was used for bioinformatics analysis in this study, compiled using the integrated development environment RStudio, version 2022.12.0-353. GraphPad Prism software, version 8.0, was used for the graphical representation of the statistical data. The results are presented as mean ± standard deviation for quantitative data. The independent samples t-test was used to compare two groups of data [71]. One-way analysis of variance (ANOVA) was employed to compare data among different groups, whereas two-way ANOVA was used to compare data among different time points. Post hoc tests were conducted using the Bonferroni method. The significance threshold was set at P < 0.05 [72].

Results

Identification of key pathways and factors involved in renal ischemia-reperfusion injury

The duration of renal ischemia determines the severity of acute kidney injury (AKI) induced by renal ischemia-reperfusion (I/R). In this study, we found that 15 min of ischemia in mice caused subclinical IRI, as serum creatinine (Scr) and blood urea nitrogen (BUN) levels did not increase. However, 30 min of ischemia led to significant changes in the clinical parameters of AKI (Figure S1A-B). Histological analysis using periodic acid-Schiff (PAS) staining revealed typical changes associated with severe AKI in the renal tissue of mice in the I/R group, including apoptosis and necrosis cells observed on the first day post-surgery and detectable even on the 7th day after I/R, with evident damage at 35 min. In contrast, the Sham group exhibited normal morphology (Figure S1C). These results demonstrate the successful establishment of a mouse model of renal I/R injury.

For high-throughput transcriptome sequencing, we collected kidney tissue samples from three mice in the I/R group (S2_2, S2_3, S2_6) and three mice in the Sham group (S1_1, S1_4, S1_6). Differential analysis was performed using |logFC| > 1 and adjP < 0.05 thresholds for S2_2 vs. S1_1, S2_2 vs. S1_1, and S2_2 vs. S1_1 comparisons. The comparison of S2_2 vs. S1_1 resulted in 1149 upregulated genes and 899 downregulated genes. Similarly, S2_2 vs. S1_1 yielded 1539 upregulated genes and 1638 downregulated genes. Lastly, S2_2 vs. S1_1 generated 347 upregulated genes and 1393 downregulated genes (Figure S2A-E). The intersection of upregulated and downregulated genes from the three comparisons produced 80 genes that were upregulated (Fig. 1A) and 111 genes that were downregulated (Fig. 1C).

GO and KEGG enrichment analyses revealed that the 80 upregulated intersected genes were primarily enriched in biological processes (BPs) such as long-chain fatty acid metabolic process, regulation of neurogenesis, and leukocyte proliferation. The cellular component (CC) enrichment included autolysosome, an intrinsic component of the presynaptic active zone membrane, and the apical part of the cell. Molecular function (MF) enrichment involved monocarboxylic acid binding, antioxidant activity, and peroxidase activity. Furthermore, the KEGG pathway enrichment analysis implicated ferroptosis, thyroid cancer, and retinal metabolism (Fig. 1B), among others. On the other hand, the 111 downregulated intersected genes were mainly enriched in BPs, including response to interferon-gamma, steroid metabolic process, and steroid biosynthetic process. CC enrichment included RNA polymerase II transcription regulator complex. MF enrichment comprised monooxygenase activity, oxidoreductase activity, acting on paired donors with incorporation or reduction of molecular oxygen, and N-acyltransferase activity. Moreover, the KEGG pathway enrichment analysis implicated the PPAR signaling pathway, Drug metabolism - cytochrome P450, and Biosynthesis of cofactors (Fig. 1D). These findings suggest that both the Ferroptosis-related pathway and PPAR signaling pathway play vital roles in the progression of renal I/R injury.

Based on existing literature evidence, cell death and injury are commonly associated with renal I/R injury. Iron-dependent lipid peroxidation-induced cell death, such as Ferroptosis, has been demonstrated to have significant harmful effects in the renal I/R injury model, making it a new type of cell death currently under investigation. Additionally, the PPAR pathway has also been shown to play an important role in renal I/R injury, with peroxisome proliferator-activated receptor-gamma (PPARγ) being a key factor. Therefore, we further examined the mRNA and protein expression of PPARγ through RT-qPCR and Western blot. We found that compared to the Sham group, the expression of PPARγ mRNA and protein in the renal tissue of mice in the I/R group was significantly decreased (Fig. 1E-F).

Fig. 1

Identification of key genes and pathways implicated in renal ischemia/reperfusion injury using high-throughput sequencing technology Note: (A) Venn diagram illustrating the intersection of upregulated genes in the comparison between three groups; (B) Bar graph showing the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for the 80 upregulated genes; (C) Venn diagram illustrating the intersection of downregulated genes in the comparison between three groups; (D) Bar graph showing the GO and KEGG enrichment analysis for the 80 downregulated genes; (E) mRNA expression levels of PPARγ in mouse renal tissue of each group measured by RT-qPCR; (F) Protein expression levels of PPARγ in mouse renal tissue of each group measured by Western blot; BP, CC, and MF in panels (B) and (D) represent Biological Processes, Cell Components, and Molecular Functions, respectively. Animal experiments consisted of 8 mice per group, * indicates p < 0.05 compared to the Sham group

Consequently, we believe that the Ferroptosis pathway and the PPARγ-mediated PPAR pathway play important roles in the progression of renal I/R injury. The level of Ferroptosis is significantly upregulated in renal I/R injury, while the expression of PPARγ and its mediated PPAR pathway is markedly downregulated.

Rosiglitazone alleviates kidney injury and promotes regeneration and repair in I/R injury mice

Rosiglitazone can be used as an activator of PPARγ [73, 74]. In this study, we subjected mice to renal ischemia/reperfusion (I/R) injury and treated them with the PPARγ activator Rosiglitazone for 3 consecutive days (Fig. 2A). We observed significant congestion in the cortical-medullary junction of the kidneys in the I/R group compared to the Sham group. There was no significant difference between the I/R group and the I/R + Vehicle group. However, the I/R + Rosiglitazone group showed reduced congestion compared to the I/R + Vehicle group (Fig. 2B). Furthermore, we calculated the ratio of injured kidney weight to body weight, and the results showed that the weight of the injured kidneys in the I/R group was higher than that in the Sham group, but the weight was reduced after Rosiglitazone treatment (Fig. 2C).

Subsequently, we performed pathological analysis using PAS staining to evaluate whether Rosiglitazone treatment improved kidney I/R injury in mice. We found that Rosiglitazone treatment significantly reduced serum creatinine and blood urea nitrogen levels (Fig. 2D-E) and alleviated tubular injury (Fig. 2F). The Masson staining revealed a marked decrease in renal fibrosis levels following the treatment with Rosiglitazone (Fig. 2G). Consistently, immunofluorescence staining showed reduced expression of Kim-1 (kidney injury molecule 1), a key mediator of AKI, after Rosiglitazone treatment (Fig. 2H). These results suggest that the PPARγ activator Rosiglitazone can significantly alleviate kidney injury in mice.

Since the regeneration and repair of renal tubular epithelial cells (TECs) are crucial for ischemic kidneys, Sox9 promotes the intrinsic repair process in injured kidney TECs. Similarly, the proliferative marker Ki67 is upregulated during the repair process after I/R [75, 76]. Therefore, we performed immunofluorescence staining for Ki67 and Sox9 to assess the proliferation and repair of injured kidney TECs. The results showed that the number of Ki67 and Sox9-positive cells in TECs of the I/R group was higher than that in the Sham group, and the number of Ki67 and Sox9-positive cells in TECs of the I/R + Rosiglitazone group was significantly higher than that in the I/R + Vehicle group (Fig. 2I-J). This indicates that the PPARγ activator Rosiglitazone significantly promotes the regeneration and repair of TECs after I/R injury.

Fig. 2

Effects of PPARγ agonist on mouse kidney injury and the repair of tubular epithelial cells (TECs) after ischemia/reperfusion injury (I/R) Note: (A) Experimental procedure for Rosiglitazone treatment in I/R mice; (B) Representative images of gross appearance of congested areas in the kidneys of different groups of mice; (C) Ratio of left injured kidney weight to body weight in different groups of mice; (D) Serum creatinine levels in different groups of mice; (E) Blood urea nitrogen levels in different groups of mice; (F) PAS staining of renal tissue in different groups of mice, Scale bars = 250 μm, with the right panel showing semi-quantitative scoring of tubular injury based on PAS staining; (G) Masson staining of fibrotic changes in the renal tissues of mice in each group; (H) Immunofluorescence staining of Kim-1 in kidney sections of different groups of mice, Scale bars = 100 μm, with the right panel showing quantitative analysis of Kim-1 + cells; (I) Immunofluorescence staining of Ki67 in kidney sections of different groups of mice, Scale bars = 50 μm, with the right panel showing quantitative analysis of Ki67 + cells; (J) Immunofluorescence staining of Sox9 in kidney sections of different groups of mice, Scale bars = 50 μm, with the right panel showing quantitative analysis of Sox9 + cells; Each group consisted of 8 mice, * indicates p < 0.05 compared to Sham group, # indicates p < 0.05 compared to I/R + Vehicle group

In conclusion, treatment with the PPARγ activator Rosiglitazone alleviated kidney injury and improved the regeneration and repair of renal TECs after I/R injury.

Rosiglitazone mitigates inflammation and reduces infiltrating immune cells in renal ischemia/reperfusion injury

Inflammation plays a crucial role in the progression of renal ischemia/reperfusion (I/R) injury [77]. Recent findings indicate that ferroptosis amplifies local inflammation and exacerbates kidney damage [78]. The recruitment of neutrophils, T cells, and macrophages to the site of injury contributes to the early inflammatory response in acute kidney injury (AKI) [79]. In this study, we utilized immunohistochemistry and immunofluorescence staining to label ly6G, CD3, and F4/80 positive cells in kidney tissue sections, enabling us to quantify the infiltrating neutrophils, T cells and macrophages within 6–8 visual fields per sample. Additionally, we measured the expression levels of pro-inflammatory cytokines RANTES, MCP-1, and IL-6 in mouse renal tissue using RT-qPCR.

The results revealed a significant increase in the number of neutrophils, T cells, and macrophages in the kidneys of the I/R group compared to the Sham group. No significant difference was observed between the I/R group and I/R + Vehicle group; however, the I/R + Rosiglitazone group exhibited a significant reduction in the infiltration of various inflammatory cells in the kidney (Fig. 3A-C). Flow cytometry analysis demonstrated an increased infiltration of M2 macrophages in the kidneys of mice from the early and late I/R + Rosiglitazone groups compared to the I/R + Vehicle group. Consistently, RT-qPCR analysis demonstrated a significant elevation in the expression levels of RANTES, MCP-1, and IL-6 in the kidneys of the I/R mice compared to the Sham group. No significant difference was observed between the I/R group and I/R + Vehicle group, while the expression levels of these pro-inflammatory factors were significantly suppressed in the I/R + Rosiglitazone group (Fig. 3E-G). In order to validate the transition from acute to chronic inflammation, we analyzed the long-term expression of chronic inflammation markers such as TGF-β and IL-6 in the PPARγ agonist treatment group. Additionally, we examined the expression levels of kidney injury markers Cr and BUN at 14 days. The results indicate that PPARγ activation not only alleviates acute inflammation by downregulating pro-inflammatory cytokines, but also prevents the occurrence of a cascade of chronic inflammation reactions. This suggests that PPARγ plays a crucial role in preventing the progression from acute kidney injury to chronic kidney disease by regulating the inflammatory response (Fig. 3H-I).

Fig. 3

Effects of PPARγ activator on the infiltration of inflammatory cells in mouse kidneys Note: (A) Neutrophils were labeled with anti-ly6G antibody, Scale bars = 50 μm, and the bar graph represents quantification of ly6G + cells; (B) T cells were labeled with anti-CD3 antibody, Scale bars = 50 μm, and the bar graph represents quantification of CD3 + cells; (C) Macrophages were labeled with anti-F4/80 antibody, red arrows indicate representative macrophages, Scale bars = 50 μm, and the bar graph represents quantification of F4/80 + cells; (D) Flow cytometry analysis of the percentage of F4/80+/CD11b+/CD86 + macrophages; (E-G) mRNA expression levels of RANTES, MCP-1, and IL-6 in mouse kidney tissues were detected by RT-qPCR; (H) Assessment of chronic kidney inflammation: Expression of 14-day creatinine (Cr) and blood urea nitrogen (BUN) in each group; (I) Western blot analysis of the levels of chronic inflammation-related proteins in the renal tissues of mice in each group. n = 8 mice per group; * indicates P < 0.05 compared to the Sham group, and # indicates P < 0.05 compared to the I/R + Vehicle group

Collectively, these findings suggest that Rosiglitazone, a PPARγ agonist, reduces the infiltration of inflammatory cells in the kidneys of I/R mice.

PPARγ activator rosiglitazone reduces ferroptosis in tubular epithelial cells following kidney ischemia/reperfusion injury

In this study, we further investigated the impact of the PPARγ activator Rosiglitazone on ferroptosis of tubular epithelial cells (TECs) following mouse kidney ischemia/reperfusion (I/R) injury. Using electron microscopy, we observed a decrease in mitochondrial damage in TECs after treatment with Rosiglitazone (Fig. 4A). Furthermore, we examined the levels of glutathione peroxidase (GSH) and malondialdehyde (MDA). As shown in Fig. 4B-C, compared to the Sham group, the I/R group exhibited a decrease in GSH levels and an increase in MDA levels. There were no significant differences between the I/R group and the I/R + Vehicle group. However, compared to the I/R + Vehicle group, the I/R + Rosiglitazone group showed an increase in GSH levels and a decrease in MDA levels. Additionally, we estimated the level of lipid peroxidation by staining for 4-hydroxynonenal (4-HNE), which indicated reduced lipid peroxidation after treatment with Rosiglitazone (Fig. 4D).

Previous studies have demonstrated that GPX4 and ACSL4 are markers of ferroptosis, with decreased GPX4 expression and increased ACSL4 expression indicating elevated levels of ferroptosis [80, 81]. Therefore, we performed Western blot analysis to examine the protein expression of GPX4 and ACSL4 in mouse kidney tissues from each group. As shown in Fig. 4E, compared to the Sham group, the I/R group exhibited a significant decrease in GPX4 expression and a significant increase in ACSL4 expression. There were no significant differences between the I/R group and the I/R + Vehicle group. However, compared to the I/R + Vehicle group, the I/R + Rosiglitazone group showed a significant increase in GPX4 expression and a significant decrease in ACSL4 expression. These results indicate that the PPARγ activator Rosiglitazone significantly reduces ferroptosis of TECs following mouse kidney I/R injury.

Given that ferroptosis and necroptosis are two primary factors contributing to renal I/R injury [82], we used TUNEL staining to assess the level of necroptosis in kidney tissues. Furthermore, we examined the protein expression of necroptosis-related factors RIPK3 and MLKL. Interestingly, in both the I/R + Vehicle and I/R + Rosiglitazone groups, there were no significant differences in the number of necrotic cells or the expression of p-RIPK3 or p-MLKL (Fig. 4F-G). In addition to ferroptosis and necroptosis, another form of cell death, pyroptosis, has been confirmed to play a role in renal I/R injury [83]. Consequently, we assessed the expression levels of pyroptosis-related factors. GSDMD-N is involved in the formation of membrane pores, thereby triggering pyroptosis [84]. As shown in Fig. 4H, there was no statistically significant difference in GSDMD-N expression between the I/R + Vehicle group and the I/R + Rosiglitazone group. This indicates that the PPARγ activator Rosiglitazone reduces ferroptosis of TECs following mouse kidney I/R injury but has no effect on necroptosis or pyroptosis.

Fig. 4

Effects of PPARγ activator Rosiglitazone on ferroptosis of TECs after renal I/R injury Note: (A) Transmission electron microscopy was used to observe the morphology of TECs in kidney tissues of each group; white arrows indicate mitochondrial atrophy and outer membrane rupture, Scale bars = 500 nm; (B) GSH levels in kidney tissues of each group; (C) MDA levels in kidney tissues of each group; (D) 4-HNE staining to estimate lipid peroxidation levels in kidney tissues of each group, Scale bars = 50 μm; (E) Protein expression levels of ferroptosis-related factors GPX4 and ACSL4 in mouse kidney tissues were detected by Western blot; (F) Detection of apoptosis in the kidneys of each group using the TUNEL staining method; (G) Protein expression and phosphorylation levels of necroptosis-related factors RIPK3 and MLKL in mouse kidney tissues were detected by Western blot; (H) Protein expression levels of pyroptosis-related factor GSDMD in mouse kidney tissues were detected by Western blot; n = 8 mice per group; * indicates P < 0.05 compared to the Sham group, and # indicates P < 0.05 compared to the I/R + Vehicle group

Pharmacokinetic analysis and tissue distribution of rosiglitazone as a potential therapeutic agent for kidney diseases

Due to the low plasma stability of the first-generation ferroptosis inhibitor, ferrostatin-1, we proceeded to investigate whether the PPARγ activator, Rosiglitazone, could be used as a therapeutic agent for kidney diseases. We employed an in vivo pharmacokinetic model and performed LC-MS/MS analysis of Rosiglitazone distribution in serum, kidney, brain, and liver extracts. As shown in Fig. 5A-C; Table 1, the calculated maximum concentration (Cmax) of Rosiglitazone in plasma, kidney, and brain occurred at 15 min after intravenous injection, with concentrations of 148.4 µg/ml, 231.1 µg/g, and 225.3 µg/g, respectively. The half-life (t1/2) was approximately 2 h, and the area under the concentration-time curve (AUC0-6 h) in plasma was 322.7 µg·h/ml, which was approximately four times higher in the kidney compared to the brain (234.7 µg·h/g and 4.1 µg·h/g, respectively). Overall, Rosiglitazone exhibited rapid plasma-tissue drug transfer. Furthermore, the concentrations in the kidney were almost equivalent to the plasma levels and had a longer duration, suggesting the therapeutic value of Rosiglitazone in kidney disease.

Next, we intravenously injected 3 mg/kg BODIPY-FL-Rosiglitazone into mice to explore its distribution in the kidney. We found higher radiolabeling efficiency in the kidney and liver compared to the low signal observed in the plasma and brain. Since Rosiglitazone was not detected in liver homogenates, the increased radiolabeling efficiency in the liver suggests instability of Rosiglitazone in liver homogenates (Fig. 5A-C; Table 1). Fluorescent microscopy observations demonstrated that most Rosiglitazone co-localized with tubular epithelial cells (TECs) (Fig. 5D), providing evidence for a high affinity between Rosiglitazone and TECs.

Table 1 PK parameters of rosiglitazone in plasma and tissues

TECs isolated from C57BL/6 mice displayed specific morphological features of renal tubular epithelial cells, which is an important indicator of successful isolation and culture of TECs (Figure S3A-B). Additionally, immunohistochemical staining using cytokeratin 18 (CK18) as a marker showed specific positive expression in the cells, further confirming the epithelial characteristics of the isolated cells (Figure S3C) and successfully identifying them as kidney TECs. Subsequently, we treated TECs with fluorescently labeled BODIPY-FL-Rosiglitazone to validate their localization in TECs. We found that Rosiglitazone accumulated in the cytoplasm and co-stained with mitochondria labeled by MitoTracker Red in TECs (Fig. 5E). These data suggest that Rosiglitazone exhibits excellent plasma stability, rapid plasma-kidney transfer, and high affinity for TECs, highlighting its potential as a novel ferroptosis inhibitor for kidney diseases.

Fig. 5

Pharmacokinetic study of Rosiglitazone in vivo and co-localization analysis with TECs mitochondria Note: (A) Concentration of Rosiglitazone in mouse plasma after I/R; (B) Concentration of Rosiglitazone in mouse kidney tissues after I/R; (C) Concentration of Rosiglitazone in mouse brain tissues after I/R; (D) Whole-body fluorescence imaging at 0 and 30 min after intravenous injection of fluorescently labeled Rosiglitazone, followed by measurement of fluorescence intensity in various organs (heart, liver, kidney, brain) after 30 min, the right panel shows frozen sections for observing the distribution of Rosiglitazone in the kidney, red arrows indicate representative TECs stained with Rosiglitazone, Scale bars = 50 μm; (E) Mitochondria were labeled with MitoTracker (red) and Rosiglitazone was labeled with green, white arrows indicate representative mitochondria in TECs stained with both Rosiglitazone and MitoTracker, Scale bars = 10 μm

In various forms of cell death induced by renal ischemia/reperfusion (I/R) injury, iron-dependent lipid peroxidation-regulated cell death in renal tubular epithelial cells (TECs) plays a critical role [43, 85]. PPARγ is a nuclear transcription factor that regulates lipid metabolism and suppresses neuronal inflammation. Previous studies have shown that PPARγ is downregulated in kidney I/R injury and may have a preventive and ameliorative role [86,87,88], but the exact mechanisms remain unclear. Therefore, this study aims to investigate the specific mechanism of PPARγ in TECs iron-dependent cell death after I/R.

METTL14 exerts a key role in the m6A methylation of PPARγ mRNA in renal ischemia/reperfusion injury

Research has shown that m6A methylation of PPARγ mRNA can affect its expression and, therefore, participate in disease progression [89, 90]. In order to identify key m6A regulatory factors in renal I/R injury, we examined the expression of common m6A regulatory factors (METTL3, METTL14, WTAP, FTO, ALKBH5) in the renal tissues of Sham and I/R mice using RT-qPCR and Western blot. The results indicated that compared to the Sham group, the mRNA expression of METTL3 was upregulated in the I/R group, but there was no significant change in protein expression. The mRNA and protein expression of METTL14 were significantly downregulated, while the mRNA and protein expression of ALKBH5 were both decreased. Additionally, there was no significant difference in the expression of WTAP and FTO, with METTL14 showing the most significant differential expression (Fig. 6A-B). These results suggest that METTL14 is significantly downregulated in renal I/R injury and may act as a key m6A regulatory factor in this process.

To further confirm the role of METTL14 in the m6A methylation regulation of PPARγ, we performed silencing and overexpression of METTL14. Three shRNA sequences targeting METTL14 were designed, and the knockdown effect of sh-METTL14 was validated. The sh-METTL14-1 sequence with the highest silencing efficiency was selected for subsequent experiments (Figure S4A-B). Additionally, the overexpression effect of METTL14 was verified (Figure S4C-D).

Dot blot experiments showed that silencing METTL14 led to a decrease in overall m6A methylation levels in TECs, while overexpression of METTL14 increased m6A methylation levels (Fig. 6C). MeRIP-qPCR experiments confirmed that silencing METTL14 significantly reduced the m6A levels of PPARγ mRNA in TECs, whereas overexpression of METTL14 had the opposite effect (Fig. 6D). RT-qPCR and Western blot results showed that silencing METTL14 significantly decreased the mRNA and protein expression levels of PPARγ, while overexpression of METTL14 significantly increased them (Fig. 6E-F). Furthermore, silencing METTL14 resulted in the shortened half-life of PPARγ mRNA, whereas overexpression of METTL14 had the opposite effect (Fig. 6G). These results indicate that m6A methylation is involved in METTL14-mediated regulation of PPARγ mRNA.

Subsequently, we used the RMBase database to screen for potential m6A methylation sites in PPARγ mRNA (Figure S5A) and constructed luciferase reporter vectors for the 5’UTR, CDS, and 3’UTR regions (Figure S5B). The results showed that overexpression of METTL14 increased the luciferase activity of pGL3-CDS but had no effect on the luciferase activity of pGL3-5’UTR and pGL3-3’UTR (Figure S5C), indicating that CDS is the main regulatory region for m6A methylation.

Then, based on the results from the RMBase database, we further constructed luciferase reporter vectors for both the wild-type (pGL3-PPARγ-WT) and mutant sequences (pGL3-PPARγ-MUT1,2,3,4) of PPARγ mRNA to identify specific m6A sites (Figure S5D). The results showed that overexpression of METTL14 reduced the luciferase activity of pGL3-PPARγ-WT and pGL3-PPARγ-MUT1,2,3 but had no effect on pGL3-PPARγ-MUT4 (Fig. 6H), indicating that the MUT4 site on PPARγ mRNA is the key m6A site mediated by METTL14 in the m6A methylation of PPARγ.

Fig. 6

Effects of METTL14 on m6A methylation and expression levels of PPARγ Note: (A-B) mRNA (A) and protein (B) expression levels of m6A regulatory factors in mouse kidney tissues were detected by RT-qPCR and Western blot, respectively; (C) Level of total m6A in TECs of each group; (D) Analysis of the effect of silencing or overexpression of METTL14 on m6A levels of PPARγ in TECs using MeRIP-qPCR; (E-F) mRNA (E) and protein (F) expression levels of PPARγ in TECs of each group were detected by RT-qPCR and Western blot, respectively; (G) Actinomycin D experiment was performed to evaluate the stability of PPARγ mRNA in TECs of each group; (H) Relative luciferase activity of pGL3-empty, pGL3-WT, pGL3-MUT1, pGL3-MUT2, pGL3-MUT3, and pGL3-MUT4 in oe-NC and oe-METTL14 groups; * indicates P < 0.05 compared to the sh-NC or Sham group, and # indicates P < 0.05 compared to the oe-NC group; 8 mice per group for animal experiments, and cell experiments were repeated 3 times

METTL14 regulates PPARγ expression to inhibit ferroptosis and enhance mitochondrial function in TECs

To investigate the role of METTL14 in regulating PPARγ expression and inhibiting ferroptosis in TECs, we designed three shRNA sequences targeting PPARγ. We confirmed the silencing effect of sh-PPARγ and selected the sh-PPARγ-3 sequence with the highest silencing efficiency for further experiments (Figure S6A-B). We examined specific indicators related to ferroptosis in different cell groups, including reactive oxygen species (ROS), malondialdehyde (MDA), reduced glutathione (GSH), oxidized glutathione (GSSG), and iron content (Fe²⁺) [91, 92]. Initially, we analyzed the expression of iron metabolism-related genes using the FerroOrange probe to investigate the regulatory role of METTL14 and PPARγ in iron homeostasis. As shown in Fig. 7A-B, overexpression of METTL14 resulted in a decrease in intracellular Fe²⁺ within 24 h, and compared to the oe-METTL14 + sh-NC group, the oe-METTL14 + sh-PPARγ group showed a significant increase in Fe²⁺. This suggests that silencing PPARγ can partially reverse the inhibitory effect of METTL14 overexpression on Fe²⁺ levels.

Next, we examined the impact of METTL14 and PPARγ on redox balance. Firstly, we compared oxidative stress markers and peroxides, which revealed that the GSH/GSSG ratio was significantly increased in the oe-METTL14 group compared to the oe-NC group, indicating alleviation of oxidative stress. Furthermore, compared to the oe-METTL14 + sh-NC group, the oe-METTL14 + sh-PPARγ group showed a significant decrease in the GSH/GSSG ratio (Fig. 7C). Confocal laser scanning microscopy and flow cytometry fluorescence quantification confirmed that the oe-METTL14 group exhibited lower levels of ROS production in TECs compared to the oe-NC group, while the oe-METTL14 + sh-PPARγ group showed a significant increase in ROS levels compared to the oe-METTL14 + sh-NC group (Fig. 7D-E).

In the cytoplasm, ROS can oxidize membrane polyunsaturated lipids, generating end products such as MDA [65]. We found that MDA levels significantly decreased after overexpression of METTL14, and in the oe-METTL14 + sh-PPARγ group, MDA levels were significantly higher compared to the oe-METTL14 + sh-NC group (Fig. 7F). Similarly, using C11-BODIPY581/591 to evaluate lipid peroxidation levels, we observed a significant increase in lipid peroxidation in the oe-METTL14 + sh-PPARγ group compared to the oe-METTL14 + sh-NC group (Fig. 7G-H). These results indicate that silencing PPARγ can partially reverse the improvement of oxidative-reductive balance caused by METTL14 overexpression.

To further validate the effects of METTL14 and PPARγ on ferroptosis of TECs, we examined the expression of ferroptosis markers GPX4 and ACSL4. RT-qPCR analysis showed that compared to the oe-NC group, GPX4 expression was significantly increased, and ACSL4 expression was significantly decreased in the oe-METTL14 group. Moreover, compared to the oe-METTL14 + sh-NC group, the oe-METTL14 + sh-PPARγ group exhibited significantly decreased GPX4 expression and significantly increased ACSL4 expression (Fig. 7I-J). This indicates that silencing PPARγ can partially reverse the inhibitory effect of METTL14 overexpression on ferroptosis of TECs.

Ferroptosis is accompanied by characteristic mitochondrial morphological changes [93, 94]. Transmission electron microscopy revealed that TECs in the oe-METTL14 group exhibited mitochondrial damage repair compared to the oe-NC group, while TECs in the oe-METTL14 + sh-PPARγ group showed mitochondrial shrinkage, reduced cristae, and increased membrane density (Fig. 7K). Evaluation of mitochondrial membrane potential using JC-1 staining showed a significant increase in mitochondrial membrane potential in TECs of the oe-METTL14 group compared to the oe-NC group, while the oe-METTL14 + sh-PPARγ group exhibited a significant decrease in mitochondrial membrane potential compared to the oe-METTL14 + sh-NC group (Fig. 7L). These findings suggest that METTL14 and PPARγ regulate ferroptosis of TECs, likely through their impact on mitochondrial function.

Fig. 7

The Effects of METTL14 and PPARγ on Ferroptosis in TECs.Note: (A) Representative fluorescence images of TECs from each group, as identified by FerroOrange. Scale bar = 100 μm. (B) Statistical analysis of iron content in TECs from each group using a detection kit. (C) GSH/GSSG ratio in TECs from each group. (D-E) Visualization of ROS production in TECs using DCFH-DA under confocal laser scanning microscopy (D). Scale bar = 200 μm. Statistical analysis of ROS production in TECs using flow cytometry (E). (F) Detection of MDA content in TECs from each group. (G-H) Representative confocal images of TECs stained with C11-BODIPY 581/591 (G). Red indicates non-oxidized lipids, and green represents oxidized lipids. Scale bar = 200 μm. Quantitative analysis of oxidized lipids in TECs (H). (I-J) mRNA expression levels of GPX4 (I) and ACSL4 (J) in TECs measured by RT-qPCR. (K) Transmission electron microscopy observation of mitochondrial morphology in TECs from each group. Scale bar = 5 μm. (L) Detection of mitochondrial membrane potential (MMP) in TECs using JC-1. * indicates a significant difference compared to the sh-NC group with a p-value less than 0.05, # indicates a significant difference compared to the oe-METTL14 + sh-NC group with a p-value less than 0.05. Cell experiments were repeated three times

In summary, METTL14 promotes PPARγ expression and inhibits ferroptosis of TECs.

IGF2BP2 regulates PPARγ mRNA stability in TECs through an m6A-dependent mechanism

IGF2BP2 is an m6A recognition protein that enhances mRNA stability [95]. Previous studies have indicated that IGF2BP2 can regulate macrophage phenotype activation and inflammatory diseases through the stabilization of PPARγ mRNA [96]. However, the role and mechanism of IGF2BP2 in TECs and renal I/R injury remain unclear.

To investigate whether IGF2BP2 is involved in the regulation of PPARγ mRNA stability, we first used CRISPR/Cas9 gene editing technology to knockout IGF2BP2 in TECs. We validated the knockout efficiency through RT-qPCR and Western blot analysis, which showed that IGF2BP2 was not expressed in IGF2BP2-KO TECs compared to IGF2BP2-WT TECs (Fig. 8A-B), confirming the successful knockout of IGF2BP2.

Furthermore, the RT-qPCR and Western blot results revealed a significant decrease in both mRNA and protein expression of PPARγ in IGF2BP2-KO TECs compared to IGF2BP2-WT TECs (Fig. 8C-D). We further confirmed the interaction between IGF2BP2 and PPARγ mRNA in TECs using RNA immunoprecipitation (RIP) analysis with an anti-IGF2BP2 antibody (Fig. 8E). Additionally, RNA stability analysis demonstrated that knocking out IGF2BP2 significantly reduced the stability of PPARγ mRNA (Fig. 8F). Therefore, we propose that IGF2BP2 is an m6A recognition protein that regulates the stability of PPARγ mRNA.

To determine whether the regulation of PPARγ m6A methylation by METTL14 depends on IGF2BP2, we overexpressed METTL14 in both IGF2BP2-WT and IGF2BP2-KO TECs. Changes in PPARγ mRNA and protein expression were detected by RT-qPCR and Western blot analysis in each group of cells. The results showed that in wild-type TECs, overexpression of METTL14 significantly increased PPARγ expression compared to the oe-NC group. However, in IGF2BP2 knockout TECs, overexpression of METTL14 had no significant effect on PPARγ expression compared to the oe-NC group (Fig. 8G-H). This indicates that the reversal of IGF2BP2 knockout can attenuate the impact of METTL14 overexpression on PPARγ expression, suggesting that METTL14 promotes PPARγ m6A methylation through an IGF2BP2-dependent mechanism.

Fig. 8

The Regulatory Role of IGF2BP2 in m6A Methylation and Expression Levels of PPARγ Note: (A-B) Verification of IGF2BP2 knockout efficiency by RT-qPCR (A) and Western blot (B). (C-D) Detection of mRNA (C) and protein (D) expression levels of PPARγ in TECs from each group by RT-qPCR and Western blot, respectively. (E) RIP analysis of the interaction between IGF2BP2 and PPARγ mRNA in TECs. (F) Actinomycin D experiment to test the stability of PPARγ mRNA in TECs from each group. (G-H) mRNA (G) and protein (H) expression levels of PPARγ in wild-type and IGF2BP2 gene knockout TECs by RT-qPCR and Western blot. * indicates a significant difference compared to the IGF2BP2-WT group, IgG group, or oe-NC group of wild-type TECs with a p-value less than 0.05. Cell experiments were repeated three times

Effect of METTL14 on ferroptosis of TECs via IGF2BP2-dependent mechanism

Next, we investigated the iron content, GSH/GSSG ratio, ROS, MDA, and other iron-specific markers of cell death in each group of cells after H/R treatment to explore whether METTL14 promotes PPARγ m6A methylation in TECs through an IGF2BP2-dependent mechanism and affects ferroptosis. We found that compared to the oe-NC group in wild-type TECs, the oe-METTL14 group showed a significant decrease in Fe²⁺ content, a significant increase in the GSH/GSSG ratio, a significant decrease in ROS levels and MDA and lipid peroxidation levels, along with a significant increase in GPX4 expression and a significant decrease in ACSL4 expression. Furthermore, noticeable mitochondrial damage repair and a significant decrease in mitochondrial membrane potential were observed (Fig. 9A-L). However, in IGF2BP2 knockout TECs, there were no significant changes in the above indicators between the oe-NC group and the oe-METTL14 group (Fig. 9A-L), indicating that the significant inhibition of ferroptosis of TECs by overexpressed METTL14 can be reversed by IGF2BP2 knockout.

Fig. 9

The Effects of IGF2BP2 Knockout on Ferroptosis in TECs Note: (A) Representative fluorescence images of TECs from each group, as identified by FerroOrange. Scale bar = 100 μm. (B) Statistical analysis of iron content in TECs from each group using a detection kit. (C) GSH/GSSG ratio in TECs from each group. (D-E) Visualization of ROS production in TECs using DCFH-DA under confocal laser scanning microscopy (D). Scale bar = 200 μm. Statistical analysis of ROS production in TECs using flow cytometry (E). (F) Detection of MDA content in TECs from each group. (G-H) Representative confocal images of TECs stained with C11-BODIPY 581/591 (G). Red indicates non-oxidized lipids, and green represents oxidized lipids. Scale bar = 200 μm. Quantitative analysis of oxidized lipids in TECs (H). (I-J) mRNA expression levels of GPX4 (I) and ACSL4 (J) in TECs measured by RT-qPCR. (K) Transmission electron microscopy observation of mitochondrial morphology in TECs from each group. Scale bar = 5 μm. (L) Detection of mitochondrial membrane potential (MMP) in TECs using JC-1. * indicates a significant difference compared to the oe-NC group of wild-type TECs with a p-value less than 0.05. Cell experiments were repeated three times

Enhanced METTL14 expression alleviates kidney injury through an IGF2BP2-dependent mechanism in mice subjected to renal ischemia/reperfusion

Finally, this study further validated the role of METTL14 in promoting m6A methylation of PPARγ in mouse kidney injury after I/R through in vivo animal experiments. Similarly, we designed three shRNA sequences for IGF2BP2 and confirmed the silencing effect of sh-IGF2BP2, selecting the sh-IGF2BP2-3 sequence with the highest silencing efficiency for subsequent experiments (Figure S7A-B).

RT-qPCR and Western blot analysis revealed that compared to the oe-NC + sh-NC + oe-NC group, the expression of METTL14 and PPARγ significantly increased in the oe-METTL14 + sh-NC + oe-NC group, while IGF2BP2 expression showed no significant change. Compared to the oe-METTL14 + sh-NC + oe-NC group, the oe-METTL14 + sh-IGF2BP2 + oe-NC group showed no significant change in METTL14 expression, but a significant reduction in the expression of IGF2BP2 and PPARγ. Furthermore, compared to the oe-METTL14 + sh-IGF2BP2 + oe-NC group, the oe-METTL14 + sh-IGF2BP2 + oe-PPARγ group showed no significant change in METTL14 and IGF2BP2 expression, but a significant increase in PPARγ expression (Fig. 10A-B). These results indicate that overexpression of METTL14 can promote PPARγ expression in renal tissue of I/R mice through an IGF2BP2-dependent mechanism.

We found that compared to the oe-NC + sh-NC + oe-NC group, the oe-METTL14 + sh-NC + oe-NC group exhibited significantly decreased levels of serum creatinine and blood urea nitrogen, reduced congestion area at the junction of the renal cortex and medulla, significantly decreased renal weight, lower tubular injury score, and decreased Kim-1 expression (Fig. 10C-J), indicating that overexpression of METTL14 can significantly alleviate kidney injury in mice after I/R. Compared to the oe-METTL14 + sh-NC + oe-NC group, the oe-METTL14 + sh-IGF2BP2 + oe-NC group showed significantly increased levels of serum creatinine and blood urea nitrogen, increased congestion area, significantly increased renal weight, higher tubular injury score, and increased Kim-1 expression (Fig. 10C-J), indicating that silencing IGF2BP2 can partially reverse the beneficial effects of METTL14 overexpression on renal injury in mice after I/R.

Furthermore, compared to the oe-METTL14 + sh-IGF2BP2 + oe-NC group, the oe-METTL14 + sh-IGF2BP2 + oe-PPARγ group exhibited significantly decreased levels of serum creatinine and blood urea nitrogen, reduced congestion area, significantly decreased renal weight, lower tubular injury score, and decreased Kim-1 expression (Fig. 10C-J), suggesting that overexpression of METTL14 can alleviate kidney injury in mice after I/R by promoting PPARγ expression through an IGF2BP2-dependent mechanism.

Fig. 10

The Effects of METTL14, IGF2BP2, and PPARγ on Mouse Kidney Injury Note: (A-B) mRNA and protein expression levels of METTL14, IGF2BP2, and PPARγ detected by RT-qPCR (A) and Western blot (B). (C) Serum creatinine (Scr) levels in each group of mice. (D) Blood urea nitrogen (BUN) levels in each group of mice. (E) Representative overall appearance of congested areas in the kidneys of each group of mice. (F) The ratio of weight of injured kidney to body weight in each group of mice. (G) PAS staining was used to detect pathological changes in the kidney tissues of each group of mice. Scale bars = 250 μm. (H) The bar graph shows the semiquantitative scoring of tubular injury based on PAS staining. (I) Immunofluorescence staining images of Kim-1 in kidney sections of each group of mice. Scale bars = 100 μm. (J) Bar graph showing quantitative analysis of Kim-1 + cells. Each group consisted of 8 mice. * indicates a significant difference compared to the oe-NC + sh-NC + oe-NC group with a p-value less than 0.05, # indicates a significant difference compared to the oe-METTL14 + sh-NC + oe-NC group with a p-value less than 0.05, & indicates significant difference compared to the oe-METTL14 + sh-IGF2BP2 + oe-NC group with a p-value less than 0.05

METTL14-mediated promotion of PPARγ expression inhibits ferroptosis and reduces renal inflammatory response in mouse kidney ischemia/reperfusion injury

In this study, we further investigated the impact of METTL14-mediated promotion of PPARγ expression through the IGF2BP2-dependent mechanism on ferroptosis in renal tubular epithelial cells (TECs) following mouse kidney ischemia/reperfusion (I/R) injury. Our observations using electron microscopy revealed mitochondrial damage in TECs, and we assessed the levels of glutathione peroxidase (GSH) and malondialdehyde (MDA) as indicators of oxidative stress. We also estimated lipid peroxidation levels through 4-HNE staining and determined the expression levels of GPX4 and ACSL4 using RT-qPCR and Western blot analysis. Our results demonstrated that compared to the oe-NC + sh-NC + oe-NC group, the oe-METTL14 + sh-NC + oe-NC group exhibited reduced mitochondrial damage, increased GSH levels, decreased MDA levels, lower accumulation of lipid peroxidation, significantly elevated GPX4 expression, and significantly reduced ACSL4 expression (Fig. 11A-F). These findings suggest that overexpression of METTL14 can decrease ferroptosis levels in mice after I/R injury. Furthermore, compared to the oe-METTL14 + sh-NC + oe-NC group, the introduction of sh-IGF2BP2 partially reversed the inhibitory effect of METTL14 overexpression on ferroptosis (Fig. 11A-F). This indicates that silencing IGF2BP2 can partially reverse the protective effect of METTL14 overexpression against ferroptosis after I/R injury. Moreover, compared to the oe-METTL14 + sh-IGF2BP2 + oe-NC group, the oe-METTL14 + sh-IGF2BP2 + oe-PPARγ group showed increased mitochondrial damage, decreased GSH levels, increased MDA levels, elevated accumulation of lipid peroxidation, significantly reduced GPX4 expression, and significantly elevated ACSL4 expression (Fig. 11A-F). These results suggest that METTL14 overexpression promotes PPARγ expression through the IGF2BP2-dependent mechanism and inhibits ferroptosis in mice after I/R injury.

In addition, to investigate the influence of METTL14 overexpression-mediated promotion of PPARγ expression through the IGF2BP2-dependent mechanism on the renal inflammatory response after mouse kidney I/R injury, we assessed the expression levels of the pro-inflammatory cytokines RANTES, MCP-1, and IL-6 in the renal tissue of the different groups using RT-qPCR. Our results demonstrated that compared to the oe-NC + sh-NC + oe-NC group, the oe-METTL14 + sh-NC + oe-NC group exhibited significantly reduced expression levels of RANTES, MCP-1, and IL-6. In comparison to the oe-METTL14 + sh-NC + oe-NC group, the introduction of sh-IGF2BP2 significantly increased the expression levels of RANTES, MCP-1, and IL-6 in the oe-METTL14 + sh-IGF2BP2 + oe-NC group. Furthermore, when compared to the oe-METTL14 + sh-IGF2BP2 + oe-NC group, the introduction of PPARγ in the oe-METTL14 + sh-IGF2BP2 + oe-PPARγ group significantly decreased the expression levels of RANTES, MCP-1, and IL-6 (Fig. 11G-I). Therefore, METTL14 overexpression promotes PPARγ expression through the IGF2BP2-dependent mechanism, thereby reducing the renal inflammatory response after mouse kidney I/R injury.

Fig. 11

Effect of METTL14, IGF2BP2, and PPARγ on ferroptosis and inflammatory response after renal ischemia/reperfusion (I/R) in mice. Note: (A) Morphology of tubular epithelial cells (TECs) in renal tissues of each group observed by transmission electron microscopy, white arrows indicate mitochondrial atrophy and outer membrane rupture, Scale bars = 500 nm; (B) Levels of glutathione (GSH) in renal tissues of each group of mice; (C) Levels of malondialdehyde (MDA) in renal tissues of each group of mice; (D) 4-HNE staining to estimate lipid peroxidation levels in renal tissues of each group of mice, Scale bars = 50 μm; (E-F) mRNA and protein expression of GPX4 and ACSL4 in renal tissues of each group of mice detected by RT-qPCR (E) and Western blot (F); (G-I) mRNA expression levels of RANTES, MCP-1, and IL-6 in renal tissues of each group of mice detected by RT-qPCR; 8 mice per group; * indicates P < 0.05 compared to the oe-NC + sh-NC + oe-NC group, # indicates P < 0.05 compared to the oe-METTL14 + sh-NC + oe-NC group, & indicates P < 0.05 compared to the oe-METTL14 + sh-IGF2BP2 + oe-NC group

Discussion

Renal ischemia-reperfusion (I/R) injury is a major cause of acute kidney injury (AKI), and its complex pathophysiological mechanisms have been a focal point in the study of renal diseases [3, 4, 97]. Previous research has primarily focused on cell metabolic disorders, oxidative stress responses, and inflammatory reactions following local renal ischemia [98]. However, these studies often overlook the diversity of renal cell death pathways, particularly the role of ferroptosis in renal I/R injury. Our findings shed new light on the role of PPARγ in the context of ferroptosis, enhancing our understanding of the mechanisms underlying renal I/R injury and providing a new direction for future therapeutic strategies. Tubular epithelial cells (TECs) play a crucial role in maintaining renal structure and function [6,7,8]. During renal I/R injury, TECs are particularly vulnerable, significantly influencing the severity of kidney damage and the possibility of recovery [8, 99, 100]. Previous studies have primarily focused on traditional cell death pathways such as necrosis and apoptosis, paying limited attention to the novel cell death mechanism of ferroptosis in TECs [8, 99, 101]. Our study reveals the significant role of ferroptosis in TECs, not only adding a new dimension to the pathological mechanism of renal I/R injury but also providing new insights into protective strategies targeting TECs.

PPARγ, as a nuclear receptor, plays a critical role in regulating cell metabolism, inflammatory responses, and cell growth and death processes [14,15,16]. In the context of renal I/R injury, our study discovers that PPARγ directly affects the survival status of TECs by regulating ferroptosis, which differs from previous understanding of PPARγ mainly through the regulation of inflammatory responses and cell metabolism pathways. Specifically, we find that the PPARγ agonist Rosiglitazone effectively alleviates renal I/R injury and TECs’ ferroptosis, providing new possibilities for clinical application. m6A methylation, as an important RNA modification, plays a crucial role in regulating gene expression and determining cell fate [89, 102]. However, the role of m6A methylation in renal diseases has not been sufficiently investigated [17, 18, 103]. Our study reveals the importance of m6A methylation in regulating PPARγ expression, particularly the roles of METTL14 and IGF2BP2 in the m6A methylation regulation of PPARγ. This finding not only fills the gap in understanding the role of m6A methylation in renal diseases but also suggests new molecular targets for treating renal I/R injury.

In this study, we employed multiple methodological approaches, including high-throughput transcriptome sequencing, transmission electron microscopy observations, and molecular biology techniques, allowing us to gain a comprehensive understanding of the complex mechanisms underlying renal I/R injury and TECs’ ferroptosis. Although these methods provide valuable information, they also have certain limitations. For example, while high-throughput transcriptome sequencing can reveal extensive gene expression changes, it cannot directly reflect protein-level changes. Similarly, transmission electron microscopy can provide detailed information about cellular ultrastructure, but its complex sample preparation process may impact the authenticity of observation results. However, the findings of this study are of great significance for understanding the mechanisms of renal I/R injury and TECs’ ferroptosis. We not only reveal the novel role of PPARγ in renal I/R injury but also elucidate the critical role of m6A methylation in regulating PPARγ expression. These findings provide a theoretical foundation for developing new strategies for treating AKI. In the future, applying the findings of this study in the clinical setting, especially using PPARγ agonists or m6A methylation regulators for treatment, may bring new hope to AKI patients.

In conclusion, we can preliminarily conclude that METTL14 promotes PPARγ m6A methylation through an IGF2BP2-dependent mechanism, leading to its enhanced expression. Overexpression or activation of PPARγ can alleviate mouse renal injury, inhibit TECs’ ferroptosis after I/R, and reduce renal inflammation (Fig. 12). This study has made progress in revealing the roles of PPARγ and its m6A methylation in renal ischemia-reperfusion injury and TECs’ ferroptosis. However, there are several limitations. Firstly, the study primarily relied on mouse models, which may have differences in physiological and pathological characteristics compared to humans, potentially limiting the direct applicability of conclusions. Secondly, the sample size was limited, which might affect the universality and reproducibility of the results. Additionally, the study focused on specific aspects of PPARγ and m6A methylation, possibly overlooking other key molecular mechanisms in renal I/R injury. Therefore, future research needs to be verified in broader biological models and larger sample sizes while exploring other potential molecular mechanisms.

Fig. 12

Hypothesis diagram of the molecular mechanism by which METTL14 promotes PPARγ m6A methylation through an IGF2BP2-dependent mechanism, thereby attenuating renal injury, inhibiting ferroptosis of TECs, and reducing intra-renal inflammatory response

Future research directions should focus on overcoming the existing limitations and expanding the depth and breadth of the study. The use of human kidney cell or tissue models could be considered to simulate the human disease state more accurately, thereby improving the clinical relevance of research findings. Additionally, interdisciplinary collaborations, such as integrating systems biology, computational biology, and clinical medicine, will help comprehensively unravel the complex mechanisms of renal I/R injury. Clinical trials of PPARγ agonists and m6A methylation regulators are also important components of future work, which will help evaluate the potential of these molecules for the clinical treatment of AKI. Finally, long-term follow-up studies will contribute to understanding the long-term impact of renal I/R injury on patient health and may reveal new treatment and prevention strategies.