Targeting TFAM K76 acetylation attenuates mitochondrial dysfunction and kidney injury in diabetic kidney disease

Abstract

Background

Mitochondrial dysfunction is a hallmark of diabetic kidney disease (DKD), yet its regulatory mechanisms remain poorly defined. Mitochondrial transcription factor A (TFAM), a central regulator of mitochondrial homeostasis, undergoes lysine 76 (K76) acetylation, but the functional significance of this modification in DKD has not been established.

Methods

We collected kidney tissues from DKD patients and DKD mice, and assessed TFAM acetylation in HK-2 cells and primary renal tubular cells under high-glucose conditions. In addition, to investigate the potential mechanism of TFAM acetylation in mitochondrial damage within the kidney, we explored relevant pathways using proteomics and utilized streptozotocin (STZ)-induced DKD mouse models with tubular-specific expression of TFAM wild-type and mutant forms to examine kidney injury. Moreover, we identified TFAM K76 acetylation-specific inhibitors through high-throughput virtual screening and thoroughly validated them in HK-2 cells, primary cells, and DKD mice, confirming the critical role of TFAM acetylation in DKD-related kidney injury.

Results

Here, we identify TFAM K76 acetylation as a critical mediator of mitochondrial injury in DKD. TFAM K76 acetylation was markedly elevated in kidney tissues from DKD patients and diabetic mouse models, correlating with mitochondrial damage, inflammation, and fibrosis under hyperglycemic conditions. In vivo, overexpression of acetylation-mimetic TFAM K76Q in renal tubular epithelial cells aggravated renal injury and ultrastructural damage, whereas its deacetylation attenuated these effects. Mechanistically, TFAM K76 acetylation impaired oxidative phosphorylation and excessively activated autophagy, further exacerbating mitochondrial damage. We identified sirtuin 3 (SIRT3) as an upstream deacetylase that regulates this modification. Importantly, through high-throughput virtual screening, we discovered a novel small-molecule inhibitor (C14) that selectively reduces TFAM K76 acetylation and effectively alleviates hyperglycemia-induced mitochondrial dysfunction, inflammation, and fibrosis in both in vitro and in vivo models.

Conclusions

Collectively, our findings define TFAM K76 acetylation as a pathogenic driver of DKD and propose C14 as a promising therapeutic candidate targeting mitochondrial metabolism.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-026-03089-0.

Highlights

TFAM K76 acetylation is a hallmark of DKD and is involved in DKD progression in patients and in vitro and in vivo models.

Proteomics analysis indicates that TFAM K76 acetylation reduces mitochondrial oxidative phosphorylation, activates the FAK and JAK1 pathway, promotes kidney injury, and that the level of TFAM K76 acetylation is specifically regulated by SIRT3.

TFAM K76 acetylation further aggravates mitochondrial damage by increasing autophagy promotion.

High-throughput virtual screening of 200,000 small molecule compounds identified a specific TFAM K76 acetylation inhibitor that can reduce inflammation and fibrosis in both in vivo and in vitro models.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-026-03089-0.

Research insights

• What is currently known about this topic?

  Mitochondrial damage is a key event in DKD, and improving mitochondrial damage can alleviate DKD kidney injury.

• What is the key research question?

  How do key molecules involved in mitochondrial damage cause injury to renal tubular epithelial cells?

• What is new?

  This study identified new biomarkers for diabetes. We screened a novel small-molecule inhibitor that can significantly improve mitochondrial damage and alleviate kidney inflammation and fibrosis progression.

• How might this study influence clinical practice?

  The research findings may provide a promising candidate for targeted mitochondrial metabolism therapy for diabetic patients.

Introduction

Diabetic kidney disease (DKD) is a leading cause of chronic kidney disease (CKD) and end-stage renal disease (ESRD) [1, 2]. According to recent statistics from the World Health Organization, the number of individuals with diabetes is projected to reach 700 million by 2045[3]. DKD and other renal disorders are closely linked to inflammation and fibrosis driven by mitochondrial dysfunction [4–6]. However, the mechanisms underlying mitochondrial damage in DKD progression remain incompletely understood. While newer therapies, such as sodium-glucose cotransporter 2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists, effectively lower blood glucose, they do not fully halt the decline in renal function, leaving patients at high risk for progression to renal failure. Thus, further investigation into DKD pathophysiology and the identification of novel therapeutic strategies are urgently needed.

Mitochondrial homeostasis and energy metabolism in tubular epithelial cells (TECs) are critical determinants in the development of DKD, particularly in the progression of interstitial fibrosis. Emerging evidence suggests that diabetes directly damages renal tubules, resulting in mitochondrial dysfunction characterized by impaired bioenergetics, excessive mitochondrial reactive oxygen species (mtROS) production, dysregulated mitophagy, and altered mitochondrial dynamics. These mitochondrial perturbations collectively lead to severe metabolic abnormalities [7–9]. Therefore, metabolic reprogramming induced by mitochondrial injury has been increasingly recognized as a key driver of DKD progression [10]. Although numerous studies indicate that targeting mitochondrial damage could serve as a promising therapeutic strategy for DKD [11, 12], no mitochondria-targeted therapies have yet achieved clinical application.

Mitochondrial transcription factor A (TFAM) is a nuclear-encoded mitochondrial protein that directly binds mitochondrial DNA (mtDNA) during replication and transcription, playing a critical role in maintaining mtDNA stability, integrity, and metabolic function [13–15]. Dysregulation of TFAM has been implicated in a variety of metabolic disorders, including obesity, non-alcoholic fatty liver disease, and diabetes [16, 17]. TFAM is subject to multiple post-translational modifications (PTMs), such as phosphorylation, O-GlcNAcylation, acetylation, and ubiquitination [18–20]. Among these, acetylation has emerged as a key regulator of mitochondrial function and represents a potential therapeutic target for DKD. Nevertheless, the precise role of TFAM acetylation in the pathogenesis of DKD remains largely unexplored.

In this study, we identified a marked increase in lysine 76 (K76) acetylation of TFAM in kidney tissues from patients with DKD, as well as in cellular and murine models. Proteomic analyses revealed that TFAM K76 acetylation promotes renal fibrosis and inflammation. Mechanistically, this modification impaired oxidative phosphorylation, leading to mitochondrial biosynthetic dysfunction and activation of the phosphorylated FAK and JAK1 signaling pathway, which drives fibrosis progression. Moreover, TFAM K76 acetylation enhanced its interaction with LC3, thereby inducing excessive autophagy and further exacerbating mitochondrial damage. We identified sirtuin 3 (SIRT3) as an upstream deacetylase that regulates this modification. Notably, through high-throughput screening, we discovered a potent small-molecule inhibitor, C14, that selectively binds to TFAM and reduces K76 acetylation. C14 effectively mitigated hyperglycemia-induced mitochondrial dysfunction and attenuated inflammation and fibrosis in both in vitro and in vivo models. Collectively, our findings uncover a critical role of TFAM K76 acetylation in DKD pathogenesis and highlight C14 as a promising therapeutic candidate targeting mitochondrial dysfunction in DKD.

Materials and methods

Cell culture and treatments, and transfection

Immortalized human kidney tubular epithelial cells (HK-2) were cultured in DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS) at 37 °C in a humidified atmosphere containing 5% CO₂. Cells were exposed to either 5.5 mM glucose (normal glucose, NG) or 30 mM glucose (high glucose, HG) for 48 h. To avoid potential batch-to-batch variations in HK-2 cells, the HK-2 cells used in the experiment were between the 5th and 10th passages and were regularly tested for mycoplasma to ensure they were mycoplasma-free.

Primary renal tubular epithelial cells are isolated from the mouse kidney cortex, which was finely minced into ~ 1 mm³ fragments and digested with 1 mg/mL collagenase II at 37 °C for 30 min. The resulting suspension was passed through a 70 μm cell strainer, centrifuged at 150 × g for 3 min, resuspended in fresh culture medium, and incubated for 48 h before subsequent treatments.

Wild-type TFAM, TFAM mutants, SIRT3, and SIRT3 H248A overexpression plasmids were obtained from Beauchamp Biosciences. siRNAs targeting SIRT3, SIRT4, and SIRT5 were purchased from Scientific Research Cloud Technology Co., Ltd. The sequences were: SIRT3 siRNA: CATCGATGGGCTTGAGAGA; SIRT4 siRNA: GGAGAAGAAGCTCCCGATT; SIRT5 siRNA: TGTCCAGCTTTATCAGGAA. Transfection was performed using Lip3000 Transfection Agent. Plasmids were introduced at a dose of 2.5 μg, while siRNAs were transfected with RNA TransMate (Sangon Biotech, E607402) at a final concentration of 60 pM. Cells were harvested for downstream experiments 48 h after transfection.

The compound (Targetmol, Shanghai, China) was dissolved in DMSO and administered to HK-2 and primary tubular epithelial cells for 48 h. Each experimental condition was performed in three biological replicates.

Human samples

Kidney tissues from patients with diabetic nephropathy were obtained through percutaneous renal biopsy, while normal control kidney tissues were obtained from the histologically confirmed non-tumorous regions of nephrectomy specimens from patients undergoing surgery for localized renal cell carcinoma. The interval between sample collection and fixation for all samples was maintained within 30 min to ensure antigen integrity and preserve tissue morphology for immunohistochemistry and pathological analysis. Tissues were embedded in paraffin and cross-sectioned (4 μm) for histology examination. This study was approved by the Review Board of Jinan Central Hospital (Approval No. R202303060085). A total of ten patients were recruited for this study, and all participants signed informed consent forms prior to inclusion.

TFAM expression and purification

The human TFAM gene plasmid was constructed based on previously reported methods. TFAM mutants were generated by PCR using oligonucleotides encoding the desired mutations. The plasmids were transformed into BL21 (DE3) Escherichia coli (TRANSGEN, Cat#: CD601-02). A single BL21 (DE3) colony was cultured in Luria–Bertani broth containing 100 μg/mL kanamycin at 37 °C overnight until an OD₆₀₀ of 0.8 was reached. Protein expression was induced with 1 mM IPTG (Solarbio, Cat#: I807) at 16 °C for 20 h. Following induction, bacteria were collected by centrifugation, and proteins were extracted according to the manufacturer’s instructions (Beyotime, Cat#: P2226). Purified proteins were snap-frozen in liquid nitrogen and stored at −80 °C until further use.

Chemicals and reagents

Antibodies against acetylated-lysine (ab22550), AQP2 (ab199975), Phospho-JAK1 (ab138005), Collagen Type I (ab270993) and TFAM (ab119684) were obtained from Abcam (Cambridge, MA, USA). SIRT4 antibody (PA5-81,259) was obtained from Invitrogen. The following antibodies were purchased from Proteintech (Wuhan, China): AQP1 (20,333–1-AP), N-cadherin (22,018–1-AP), Parkin(66,674–1-Ig),Vimentin (10,366–1-AP), α-SMA (14,395–1-AP), TGF-β1 (21,898–1-AP), Flag (20,543–1-AP), Bax (50,599–2-Ig), COX2 (55,070–1-AP), ND1 (19,703–1-AP), CYTB (55,090–1-AP), KIM-1 (30,948–1-AP), Phospho-FAK (Tyr397) (83,933–1-RR), FAK (12,636–1-AP), JAK1 (66,466–1-IG), VTN (15,833–1-AP), DAG1 (11,017–1-AP), SIRT3 (10,099–1-AP), SIRT5 (15,122–1-AP), MYC (16,286–1-AP), LC3B (14,600–1-AP), Desmin (16,520–1-AP), GFP (66,002–1-Ig), and β-Actin (20,536–1-AP). Trichostatin A (TSA, S1045) and nicotinamide (NAM, S1899) were purchased from Selleck, and the SIRT3 inhibitor 3-TYP (HY-108331) was purchased from MedChemExpress.

Animal experiments

All animal experiments were conducted in accordance with the WMA Statement on Animal Use in Biomedical Research and approved by the Institutional Animal Care & Use Committee of Jinan Central Hospital (Approval No. JNCHIACUC2021-79). Male C57BL/6 J mice were purchased from Beijing HFK Bioscience Co., Ltd. and housed in the Laboratory Animal Center of Jinan Central Hospital under standard conditions with free access to food and water. After a one-week acclimatization period, 7-week-old mice underwent unilateral nephrectomy (left kidney removal) to accelerate diabetic nephropathy progression. Animals were then randomly assigned to either control or diabetic groups. The diabetic group received a high-fat diet (HFD, 60 kcal% fat) for 6 weeks, followed by intraperitoneal injection of streptozotocin (STZ; Solarbio, 100 mg/kg). Blood glucose levels (BGL) were monitored weekly using an electronic glucometer, and BGL > 16.7 mmol/L was considered indicative of successful diabetes induction. Mice were maintained on their respective diets for an additional 12 weeks and sacrificed when clear DKD pathological lesions were observed during examination.

1. Tubular Epithelial Cell-Specific TFAM Overexpression in Type 2 diabetic mice: recombinant adeno-associated virus serotype 9 (AAV9) carrying TFAM or empty vector (EV) cDNA under the control of the tubular epithelial cell-specific promoter Ksp1.3 (WZbioscience Inc., Jinan, China) was used for TFAM overexpression or controls. Type 2 diabetic mice were randomly divided into four groups: T2DM AAV-EV, AAV-TFAM-WT, AAV-TFAM-K76Q, and AAV-TFAM-K76R (n = 6 per group). AAV9 was administered via tail vein injection at a dose of 5 × 10¹¹ vg.

2. C14 Small Molecule Treatment: The C14 compound (Targetmol, AI-204/31719045, Shanghai, China) and dapagliflozin (AstraZeneca) were dissolved in 0.5% sodium carboxymethylcellulose (CMC-Na) and administered orally once daily at 20 mg/kg. The C14 dosage was calculated based on human equivalent dosing according to body surface area.

Western blot

Proteins were extracted using RIPA buffer (Solarbio, R0010) supplemented with 1% protease inhibitor cocktail. Protein concentrations were determined using a BCA protein assay kit (Beyotime, P0012). Equal amounts of protein were separated on 12% SDS–polyacrylamide gels and transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, USA). Membranes were blocked, incubated with specific primary antibodies, and subsequently exposed to appropriate secondary antibodies for immunoblotting. Protein bands were visualized using the Tanon chemiluminescence detection system (Shanghai, China) and analyzed semi-quantitatively using ImageJ software.

Pan-acetylation level detection

Total protein was extracted from mouse kidney tissues or HK-2 cells using RIPA lysis buffer, supplemented with 1% protease inhibitor cocktail, 1 µM Trichostatin A (TSA, a deacetylase inhibitor), and 5 mM nicotinamide (NAM, a sirtuin inhibitor) to preserve endogenous acetylation. Protein concentrations were determined by the BCA assay, and equal amounts of protein were used for Western blot detection. Primary antibodies used for incubation included rabbit anti-pan-acetyl-lysine antibody (Abcam, Cat#: ab22550, diluted 1:1000) and rabbit monoclonal β-actin antibody (Proteintech, Cat#: 20,536-1-AP, diluted 1:5000) to detect the target proteins.

Histological analysis

Fresh kidney biopsy tissue and mouse kidney tissue were rapidly transferred to 4% paraformaldehyde (PFA) and fixed overnight at 4 °C.Tissues were embedded in paraffin and cross-sectioned (4 μm) for histology examination. Kidney glomerular and tubular injury scores were assessed as previously described, with tissue injury analyzed using Hematoxylin and eosin (H&E; G1120, Solarbio, Beijing, China) and Periodic acid-Schiff (PAS; G1285, Solarbio) staining, and fibrosis evaluated by Masson’s trichrome staining (G1340, Solarbio). Images were acquired using a light microscope (3DHISTECH, Pannoramic SCAN II). Semi-quantitation of staining was performed in a blinded fashion. The scoring system used was as follows: 0 points, no injury; 1 point, ≤ 25% injury; 2 points, 1 26–50% injury; 3 points, 51–75% injury; 4 points, > 75% injury [21]. Masson’s trichrome-stained sections were scored based on the percentage of cortical area occupied by fibrotic tissue.

Transmission electron microscopy (TEM)

Mouse kidneys were perfused with 4% paraformaldehyde, sectioned into ~ 1 mm³ pieces, and fixed in 2.5% glutaraldehyde overnight at 4 °C. Samples were post-fixed with 1% osmium tetroxide for 2 h at room temperature in the dark, dehydrated in a graded ethanol series, and embedded in resin for 48 h polymerization. Ultrathin sections were prepared (leica, Germany), stained, and examined using a transmission electron microscope (Hitachi, Japan). Analyze TEM images using ImageJ. Calculate the thickness of the glomerular basement membrane (GBM) for each group and the number of foot processes per micrometer of GBM. For each mouse, randomly select 3 glomeruli for analysis, and take eight electron micrographs for each glomerulus. GBM thickness is determined by measuring 20 points per image and using the arithmetic mean as the average thickness for that glomerulus. Podocyte injury is quantified by counting the number of foot processes per micrometer of GBM. Specifically, trace GBM segments on the electron micrographs for measurement and count all foot processes that meet the criteria on both sides of the traced GBM segments. Each glomerulus is measured for GBM segment lengths greater than 50 μm. The average density per mouse is used as the statistical value for that sample.

For the analysis of autophagy in TEM images, ImageJ software was used to measure the total cross-sectional area occupied by autophagosomes, and this was divided by the total cytoplasmic area analyzed to obtain the proportion of cytoplasmic volume involved in autophagy.

ELISA measurement for urine albumin level

Urinary albumin concentrations were measured using an ELISA kit (JL20493-96 T, JONLNBIO, Shanghai, China) following the manufacturer’s instructions. Optical density was measured at 450 nm using a microplate reader. Urine albumin-to-creatinine ratio (ACR) was calculated from random spot urine samples to assess renal function.

Biochemical analysis of serum and urine samples

Serum creatinine, urine creatinine (C011-2-1), serum nitrogen (C013-2-1), serum aspartate aminotransferase (AST, C010-2-1), and alanine aminotransferase (ALT, C009-2-1) were analyzed using commercial assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). ALT and AST levels were also determined using colorimetric endpoint assay kits.

Generation of anti-acetylated (Ac)-K76 TFAM Antibody

Based on the TFAM protein sequence and modification type, two antigenic peptides encompassing lysine 76 (K76) and one unmodified control peptide were designed and synthesized. Six specific pathogen-free New Zealand rabbits were immunized with these peptides. Serum was collected and evaluated for specificity using ELISA and Western blot assays. Antibodies were purified via affinity chromatography and further assessed by ELISA, dot blot, and Western blot. One highly specific antibody was selected for use as the anti-TFAM (K76) antibody in this study.

Antibody specificity assay

Absorption/blocking experiments were used to verify the specificity of the antibody. Specifically, the antibody working solution was separately incubated overnight at 4 °C with rotation with an excess (molar ratio ≥ 10:1) of TFAM K76-acetylated peptide, TFAM non-acetylated peptide, or the untreated original antibody, allowing the antibody to fully bind to the free antigen. After centrifugation, the supernatant was collected and used as the primary antibody for IHC staining on kidney tissue samples from DKD patients.

Mito-tracker assay

Mitochondria were labeled using the Mito-tracker assay (Invitrogen, M7512). Cells were treated as indicated and incubated with 200 nM Mito-Tracker for 30 min at 37 °C. Fluorescence signals were detected using confocal microscopy (Leica Microsystems, Germany). Data represent results from five independent experiments.

Reverse transcription quantitative PCR (RT-qPCR)

Total RNA was extracted from cells and kidney tissues using TRIzol (Thermo Fisher Scientific), and RNA concentrations were measured with a Nanodrop spectrophotometer. First-strand cDNA synthesis was performed using the Evo M-MLV Mix Kit with gDNA Clean for qPCR (AG11728, Accurate Biotechnology). Quantitative PCR was conducted using the SYBR Green Premix Pro Taq HS qPCR Kit (AG11701, Accurate Biotechnology) on an ABI QuantStudio 1 system (Applied Biosystems). Primer sequences were designed based on PrimerBank (Table 1). Gene expression was normalized to the housekeeping gene (Actb or ACTB) and presented as fold change relative to the control group using the 2^-ΔΔCT method.

Table 1.

Primer sequences used for RT-qPCR analysis

Primers Name	Sequences (5’-3’)
homo-IL-6-146F	CCAGAGCTGTGCAGATGAGTA
homo-IL-6-146R	GTCCTGCAGCCACTGGTTC
homo-IL-1-129F	CCAGGGACAGGATATGGAGCA
homo-IL-1-129R	TTCAACACGCAGGACAGGTACAG
homo-CCL2-118f	TCAGCCAGATGCAATCAATGCC
homo-CCL2-118r	TTCTTTGGGACACTTGCTGCTG
homo-CCL5-111f	GCATCTGCCTCCCCATATTCCT
homo-CCL5-111r	GCACTTGCCACTGGTGTAGAA
homo-CCN2-117f	GAAGGGCAAAAAGTGCATCCG
homo-CCN2-117r	CCGTCGGTACATACTCCACAGA
mus-ccl2-163F	ACTCACCTGCTGCTACTCATTC
mus-ccl2-163R	TGTCTGGACCCATTCCTTCTTG
mus-ccl5-125F	CTGCTGCTTTGCCTACCTCTC
mus-ccl5-125R	CACACACTTGGCGGTTCCTTC
mus-CCN2-99f	CTAGCTGCCTACCGACTGGAAG
mus-CCN2-99r	CTTAGAACAGGCGCTCCACTCT
mus-Tgfb-126f	ACTGGAGTTGTACGGCAGTGG
mus-Tgfb-126r	TTTGGGGCTGATCCCGTTGAT
Mus-Actb-171F	CATCCGTAAAGACCTCTATGCCAAC
Mus-Actb-171R	ATGGAGCCACCGATCCACA
mus-IL6-149f	AGCCAGAGTCCTTCAGAGAGAT
mus-IL6-149r	TGGTCCTTAGCCACTCCTTCTG
mus-TFAM-140f	GCAAAGGATGATTCGGCTCAG
mus-TFAM-140r	TCCCAAGACTTCATTTCATTGTCG
mus-Il1b-127f	ACAAGGAGAACCAAGCAACGAC
mus-Il1b-127r	TTGCTTGGGATCCACACTCTCC
D-Loop 2-F	5’-GGCTCTCAACTCCAGCATGT-3’
D-Loop 2-R	5’-AGGACGAGGGAGGCTACAAT-3’
G6PC-F	5’-CTGTCTTTGATTCCTGCCTCAT-3’
G6PC-R	5’-GTGGCTGTGCAGACATTCAA-3’

Immunohistochemistry (IHC)

Paraffin-embedded tissue sections were used for IHC analysis. Sections were 4 μm thick and underwent antigen retrieval in Tris–EDTA buffer. This procedure was performed according to the instructions of the IHC kit (Zhongshan Jinqiao Biotechnology, catalog number: PV-9000). Sections were blocked with 5% goat serum (Solarbio, catalog number: SL038) at room temperature for 1 h, followed by overnight incubation with the primary antibody at 4 °C. On the following day, sections were incubated with the secondary antibody at 37 °C for 1 h. Sections were scanned using the 3DHISTECH Pannoramic SCAN II system. IHC staining results were semi-quantitatively analyzed using QuPath software (version 0.5.1). For each kidney sample, 10–15 non-overlapping high-power fields (40 ×) in the cortical region were randomly selected and analyzed. Scoring is performed only in pathologically relevant areas. Positive staining was categorized into three intensity levels (1: weak, 2: moderate, 3: strong), and the percentage of positively stained cells at each intensity was calculated. Then, the H-score was automatically calculated using the standard formula: H-score = (percentage of cells with intensity 1 × 1) + (percentage of cells with intensity 2 × 2) + (percentage of cells with intensity 3 × 3). The final H-score for each sample was obtained from the analysis of multiple non-overlapping fields.

Immunofluorescence (IF)

Treated cells were fixed in 4% formalin and blocked with 3% BSA, followed by incubation with the appropriate primary antibody overnight at 4 °C. Cells were then incubated with corresponding secondary antibodies for 1 h at room temperature. Nuclei were counterstained with DAPI after washing. Immunofluorescence staining of paraffin-embedded tissue sections was performed using the TSA Fluorescence Triple Staining Kit (RK05903, Abclonal, China) according to the manufacturer’s instructions. Fluorescence images of cells and tissues were captured and analyzed using confocal laser scanning microscopy.

Seahorse extracellular flux assays

Oxygen consumption rate (OCR) of HK-2 cells was measured using the Seahorse XFe 96 Extracellular Flux Analyzer. Cells were seeded at 1 × 10⁴ cells/well in Seahorse XF 96-well microplates and treated with C14 for 48 h prior to measurement. Oligomycin (1 μM), p-trifluoromethoxy carbonyl cyanide phenylhydrazone (FCCP; 1 μM), and rotenone plus antimycin A (0.5 μM) were sequentially injected according to the Seahorse XF Cell Mito Stress Test Kit protocol. Data were analyzed with Seahorse XF 96 Wave software and are presented as pmol/min.

Detection of mitochondrial permeability transition pore (mPTP) openings

mPTP opening in HK-2 cells was assessed using the Cellular mPTP Assay Kit (C2009S, Beyotime, Shanghai, China). Cells were washed with PBS and incubated with Calcein AM staining solution at 37 °C in the dark for 30 min. After incubation, the dye solution was replaced with pre-warmed culture medium and incubated for an additional 30 min at 37 °C in the dark. Cells were washed 2–3 times with PBS, and detection buffer was added for observation under a fluorescence microscope.

Immunoprecipitation (IP)

HK-2 and primary tubular epithelial cells were lysed in lysis buffer (Thermo, 87,788) containing a complete protease inhibitor cocktail. Lysates were incubated with the primary antibody against the target protein overnight at 4 °C with rotation. The next day, protein A/G magnetic beads (Selleck, B23202, USA) were added and incubated for 2 h at room temperature. Beads were washed three times with lysis buffer and bound proteins were eluted using 1 × loading buffer and boiled at 95 °C for 5 min. Immunoprecipitated proteins in the supernatant were analyzed by Western blotting.

Cellular thermal shift assay

HK-2 cells were incubated with C14 or dimethyl sulfoxide (DMSO) for 24 h. Proteins were extracted via liquid nitrogen freeze–thaw cycles, and concentrations were determined using a BCA assay. Extracted protein supernatants from each group were divided into seven aliquots and heated at the indicated temperatures for 3 min using a PCR instrument, then stored at −80 °C. TFAM protein levels were subsequently detected and quantified by Western blotting.

Biolayer Interferometry (BLI)

Protein-small molecule interaction kinetics were assessed using biolayer interferometry on the Octet R2 system. TFAM WT and TFAM K76R proteins were biotinylated prior to the assay. Experiments were performed at 25 °C in PBS (pH 7.4) containing 0.02% Tween 20. Streptavidin (SSA) sensors were primed with PBST, loaded with biotinylated proteins, and then exposed to small-molecule compounds. Controls were included to monitor background binding. Data analysis employed a dual reference subtraction method to correct for nonspecific binding and biosensor variability. Molecules showing a response exceeding 0.04 nm during the binding phase were considered potentially binding, this threshold was set at more than twice the baseline noise and negative control signal.

Statistical analysis

All data were analyzed using GraphPad Prism 8.0 (San Diego, CA, USA) and are presented as mean ± SD. For data with a normal distribution, comparisons between two groups are conducted using an unpaired t-test, while for data with a non-normal distribution, a non-parametric test is used. While multiple-group comparisons were conducted using one-way ANOVA with Bonferroni’s correction. p-values < 0.05 were considered statistically significant.

Results

TFAM K76 acetylation is elevated in kidney tissue of DKD patients and diabetic mice

To explore the role of protein acetylation in DKD, we first assessed global protein acetylation in kidney tissues from control and DKD mice using pan-acetylation antibodies. Total protein acetylation level was markedly increased in DKD kidneys (Fig. 1A, Fig. S1A). Immunohistochemistry results showed that the increased acetylation in DKD predominantly occurred in the renal tubules (Fig. S1B). To identify key acetylated proteins under hyperglycemic conditions, HK-2 cells were subjected to immunoprecipitation with pan-acetylation antibodies followed by mass spectrometry. Among 36 enriched proteins, TFAM was one of the most prominently hyperacetylated candidates (Fig. 1B). Given the critical role of mitochondrial protein acetylation in mitochondrial dysfunction in DKD, we focused on TFAM, a master regulator of mitochondrial DNA (mtDNA) packaging and transcription whose function is directly modulated by post-translational modifications. IP confirmed that TFAM undergoes acetylation in HK-2 cells (Fig. 1C–E), and the level of TFAM acetylation increases under high-glucose conditions (Fig. 1F). Moreover, we isolated primary renal tubular epithelial cells from mouse proximal tubules (Fig. 1J), and TFAM acetylation was also further enhanced under high-glucose conditions (Fig. 1G and Fig. S1C, D). Previous mass spectrometry studies have reported acetylation at four lysine residues on TFAM (K52, K76, K111, and K154) (Fig. 1H). To identify the dominant acetylation site in our model, we overexpressed wild-type TFAM and site-specific deacetylation mutants in high-glucose-treated HK-2 cells. Mutation at K76 markedly reduced overall TFAM acetylation (Fig. 1I). Based on this, we generated a site-specific antibody against TFAM K76 acetylation (Fig. 1K), the antibody specifically recognizes the K76-acetylated peptide, but not the non-acetylated TFAM peptide. In addition, pre-incubation with the acetylated K76 peptide almost completely eliminated the specific cytoplasmic staining signal in renal tubular epithelial cells, whereas pre-incubation with the non-acetylated TFAM peptide had no significant effect on staining intensity or pattern (Fig. S1E). IF staining revealed increased TFAM K76 acetylation in HK-2 cells under high glucose (Fig. 1L), and Western blot analysis confirmed its elevation in primary tubular epithelial cells and kidney tissues from type 2 diabetes mellitus (T2DM) mice (Fig. 1M, N). Kidney tissues from DKD patients and normal controls were compared using H&E, PAS, and Masson staining. Blind semi-quantitative analysis after staining showed that, compared with normal controls, DKD patient kidney tissues exhibited significantly increased tubular injury, glomerular mesangial matrix expansion, and interstitial fibrosis (Fig. 1O–Q). IHC showed that TFAM K76 acetylation signals were mainly localized in the tubular epithelial cells of the renal cortex in DKD patients, whereas staining in the glomeruli was minimal or absent. Compared with the control group, TFAM K76 acetylation was markedly increased in the kidneys of DKD patients and T2DM mice (Fig. 1R, S and Fig. S1F). Consistent with previous reports, TFAM expression was decreased in DKD, yet TFAM K76 acetylation was significantly upregulated, suggesting that chronic hyperglycemia induces a disproportionate increase in TFAM acetylation despite reduced total TFAM levels (Fig. S1G). Collectively, these findings identify TFAM K76 acetylation as a prominent molecular feature of DKD progression.

Fig. 1.

TFAM K76 acetylation levels are elevated in DKD patients and mice. A Western blot analysis of pan-acetylation levels in kidney tissues from normal and DKD mice. B Schematic of anti-acetylation co-immunoprecipitation followed by label-free quantitative proteomics to identify acetylated proteins in cytosolic lysates of high-glucose-stimulated HK-2 cells. IgG served as a negative control. C, D HK-2 cells were treated with 30 mM glucose for 48 h, and TFAM acetylation levels were assessed by immunoprecipitation. E HK-2 cells transfected with Flag-TFAM plasmid were exposed to 30 mM glucose for 48 h, followed by immunoprecipitation to evaluate TFAM acetylation. F, G Immunoprecipitation in HK-2 and primary renal tubular cells confirmed increased TFAM acetylation under high-glucose conditions. H schematic of TFAM domains highlighting prominent acetylation sites. I HK-2 cells transfected with wild-type or mutant TFAM plasmids were analyzed by immunoprecipitation to detect TFAM acetylation levels. J Illustration of the isolation process for primary renal tubular epithelial cells and representative immunofluorescence images of AQP1 and AQP2 in primary cells. K Antibody dot blot assay assessing TFAM K76ac recognition of modified and unmodified peptides at different doses. L Representative immunofluorescence images of TFAM K76 acetylation in HK-2 cells. Scale bar: 25 μm. M Western blot and quantitative analysis of TFAM K76 acetylation in primary mouse renal tubular epithelial cells (mean ± SD, n = 3). N Western blot and quantitative analysis of TFAM K76 acetylation in renal cortex from control and DKD mice (mean ± SD, n = 5). O Representative histological images of renal tissues from normal control and DKD patients stained with H&E, PAS, and Masson’s trichrome (n = 3 per group, scale bar 50 µm). P, Q Semi-quantitative analysis of PAS and Masson staining in normal control and DKD groups. R, S Representative immunohistochemical staining and quantitative analysis of TFAM K76ac in renal tissues from DKD patients and normal controls (n = 5 per group, scale bar 50 µm). Data are presented as mean ± SD. ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

Renal tubular-specific TFAM K76 deacetylation alleviates kidney injury in T2DM mice

To investigate the functional role of TFAM K76 acetylation in DKD, we generated AAV9 vectors with a renal tubular epithelial cell-specific promoter (Ksp1.3) encoding wild-type TFAM or site-specific mutants (AAV-TFAM-WT, AAV-TFAM-K76Q, AAV-TFAM-K76R) and delivered them via tail vein injection into T2DM mice (Fig. 3A). Western blotting detected GFP expression in mouse kidney tissues (Fig. S2A), and IF staining revealed strong GFP fluorescence in the renal tubules of AAV-injected mice. IHC results also showed that mice injected with AAV9 carrying TFAM-WT, TFAM-K76Q, or TFAM-K76R had increased TFAM expression in the renal tubules compared to the control group. This confirms the successful overexpression of wild-type TFAM and its mutants in mouse renal tubules (Fig. S2B). Overexpression of either wild-type or mutant TFAM did not alter blood glucose levels or body weight (Fig. 2B, C). Functionally, TFAM-K76Q overexpression significantly exacerbated kidney injury, as evidenced by elevated serum creatinine, blood urea nitrogen, and urine albumin-to-creatinine ratios compared to TFAM-WT, whereas TFAM-K76R expression markedly ameliorated these parameters (Fig. 2D–F). Analysis of the renal cortex by Western blotting and IF showed that the expression of the tubular injury marker KIM-1 was decreased in TFAM-K76R mice (Fig. 2G, H and Fig. S2C). Consistently, mRNA levels of pro-inflammatory (Il1b, Il6, Tnf) and pro-fibrotic (Tgf-β1, Ccl2, Ccl5, Ccn2) genes were significantly reduced in TFAM-K76R kidneys but elevated in TFAM-K76Q mice (Fig. 2I, J). Blinded semi-quantitative histopathological analysis confirmed that, compared with the TFAM-WT group, TFAM-K76R significantly alleviates high glucose-induced mesangial matrix expansion and interstitial fibrosis, whereas TFAM-K76Q markedly exacerbates these lesions (Fig. 2K–M). TEM images further revealed that TFAM-K76Q aggravated GBM thickening, podocyte foot process effacement, and mitochondrial structural damage, while TFAM-K76R mitigated these abnormalities (Fig. 2N–Q). Multiple early and late fibrosis markers were assessed by IHC and IF, and the results consistently showed that fibrosis was significantly elevated in the TFAMK76Q group, whereas it was prominently reduced in the TFAMK76R group (Fig. 2R–T and Fig. S2D, E). Collectively, these results demonstrate that TFAM K76 acetylation promotes kidney injury in DKD, whereas K76 deacetylation exerts a protective effect by attenuating inflammation, fibrosis, and mitochondrial damage.

Fig. 3.

TFAM K76 acetylation exacerbates fibrosis, inflammation, and impairs mitochondrial biogenesis in HK-2 and primary tubular epithelial cells. A Immunoprecipitation analysis of TFAM acetylation levels in HK-2 cells transfected with TFAM K76R (acetylation-deficient) or K76Q (acetylation-mimetic) mutant plasmids. B Schematic workflow of mass spectrometry analysis in HK-2 cells overexpressing wild-type (WT) or mutant TFAM (K76R, K76Q) under high-glucose conditions. C Venn diagram showing the overlap of differentially expressed proteins among TFAM WT, K76R, and K76Q groups. D Gene Ontology (GO) analysis of proteins differentially expressed between TFAM K76Q vs. WT and TFAM K76R vs. WT groups. E–G Western blot and quantitative analysis of fibrosis- and apoptosis-related markers (N-cadherin, vimentin, TGF-β1, α-SMA, BAX, and Bcl-2) in HK-2 and primary tubular epithelial cells overexpressing TFAM WT or mutant plasmids. H Representative immunofluorescence images showing vimentin (red) expression in primary tubular epithelial cells, scale bar 10 µm. I mRNA expression levels of proinflammatory cytokines (IL-1β, IL-6, TGF-β1) measured by RT-qPCR in HK-2 cells expressing TFAM WT or mutants. J mRNA expression levels of chemokines (Ccl2, Ccl5, Ccn2) assessed by RT-qPCR in HK-2 cells. K Western blot analysis of VTN, DAG1, p-JAK1, JAK1, p-FAK, and FAK in HK-2 cells overexpressing TFAM WT or mutant plasmids. L, M Oxygen consumption rate (OCR) analysis showing the effects of TFAM variants on ATP production and maximal respiratory capacity in HK-2 cells (mean ± SD, n = 5). N Mitochondrial DNA (mtDNA) copy number quantified by PCR in HK-2 cells expressing TFAM WT or mutants. O Representative Western blot and quantification of mitochondrial proteins COX2, CYTB, and ND1 in HK-2 cells (mean ± SD, n = 3). P mRNA expression levels of MT-ND1 and MT-CO2 assessed by RT-qPCR in HK-2 cells expressing TFAM WT or mutants. Q Mito-tracker staining to evaluate mitochondrial morphology in HK-2 cells expressing TFAM WT or mutants. Scale bars: 10 μm. R Assessment of mitochondrial permeability transition pore (mPTP) opening by co-loading calcein-AM and CoCl₂ in HK-2 cells. Data are presented as mean ± SD. ns, no significance; ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 2.

Renal tubular-specific TFAM K76 deacetylation alleviates kidney injury in T2DM mice. A Schematic illustrating the experimental workflow: tubular epithelial cell-specific TFAM overexpression in male T2DM mice was achieved via tail vein injection of Ksp1.3-promoter-targeted AAV9 carrying control, TFAM WT, TFAM K76R, or TFAM K76Q constructs. B, C Weekly monitoring of blood glucose levels and body weight in each group (mean ± SD). D-F Serum creatinine (SCr), blood urea nitrogen (BUN), and urine albumin-to-creatinine ratio (ACR) measured in each group (mean ± SD, n=6). G, H Western blot and quantitative analysis of kidney injury molecule-1 (KIM-1) expression in renal cortex tissues (mean ± SD, n=5). I mRNA expression levels of inflammatory cytokines (Il-1β, Il-6, Tnf-α, Tgf-β1) in renal cortex tissues measured by RT-qPCR (mean ± SD, n=3). J mRNA expression levels of fibrosis- and inflammation-associated genes (Ccl2, Ccl5, Ccn2) in renal cortex tissues (mean ± SD, n=3). K Representative histological images of kidney sections stained with H&E, PAS, and Masson’s trichrome, scale bar 50 µm. L, M Semi-quantitative analysis of PAS and Masson staining in each group (n=6 per group). N, O TEM images of glomerular ultrastructure and mitochondrial morphology in renal cortex of T2DM mice receiving different AAV constructs. Red arrowheads indicate thickened glomerular GBM; yellow arrows indicate effaced podocyte foot processes. P, Q Quantification of GBM thickness and podocyte foot process number in glomeruli. R, S Representative immunohistochemistry images and quantitative analysis of desmin, vimentin and α-SMA expression in renal tissues, scale bar 50 µm (mean ± SD, n=3). T Representative immunofluorescence images showing vimentin (green) and collagen I (red) expression in renal tissues, scale bar 50 µm. Data are presented as mean ± SD. ns, no significance; ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

TFAM K76 acetylation exacerbates inflammation and fibrosis in vitro

To delineate the molecular mechanisms by which TFAM K76 acetylation contributes to DKD progression, we first performed lentiviral-mediated knockdown of TFAM in HK-2 cells (Fig. S3A). Subsequently, HK-2 cells and primary tubular epithelial cells were transfected with plasmids encoding wild-type TFAM (TFAM-WT), an acetylation-mimetic mutant (TFAM-K76Q), or a deacetylation-mimetic mutant (TFAM-K76R) (Fig. S3B, C). Overexpression of TFAM-K76Q markedly increased TFAM acetylation relative to TFAM-WT (Fig. 3A). Notably, mutation of the K76 site did not alter TFAM ubiquitination levels (Fig. S3D), indicating that this modification does not impact its ubiquitin-mediated turnover. Proteomic analysis by LC–MS identified 26 upregulated and 34 downregulated proteins in TFAM-K76Q versus TFAM-WT, and 29 upregulated and 23 downregulated proteins in TFAM-K76R versus TFAM-WT (|log₂FC|> 1, p < 0.05) (Fig. 3B, C and Fig. S3E, F). Gene Ontology (GO) enrichment revealed that K76Q primarily activated extracellular matrix (ECM) binding pathways, while K76R enhanced ATP-dependent processes (Fig. 3D). Functionally, TFAM-K76Q overexpression significantly promoted fibrosis, apoptosis, and partial EMT-like phenotypic transition in HK-2 and primary tubular cells, whereas TFAM-K76R exerted opposite, protective effects (Fig. 3E–H and Fig. S3G). Consistent with these findings, mRNA levels of pro-inflammatory cytokines and chemokine were markedly increased in the TFAM-K76Q group and reduced in the K76R group (Fig. 3I, J and Fig. S3H). Mechanistically, we identified vitronectin (VTN) and dystroglycan (DAG1) as significantly upregulated components of the ECM binding pathway under TFAM-K76Q expression. Elevated VTN and DAG1 activated focal adhesion kinase (FAK) phosphorylation, which in turn stimulated JAK1 signaling (Fig. 3K), providing a mechanistic link between TFAM K76 acetylation and fibrosis progression. Given the central role of mitochondrial dysfunction in DKD, we next assessed oxidative phosphorylation using oxygen consumption rate (OCR) analysis. TFAM-K76Q markedly decreased ATP production and maximal respiration (Fig. 3L, M). Consistently, mitochondrial DNA copy number, electron transport chain (ETC) subunit protein levels, and corresponding mRNA expression were all reduced in the TFAM-K76Q group (Fig. 3N–P). High-resolution microscopy revealed pronounced mitochondrial fragmentation and mitochondrial permeability transition pore (mPTP) opening upon TFAM K76 acetylation (Fig. 3Q, R). Collectively, these results demonstrate that TFAM K76 acetylation promotes renal inflammation, fibrosis, and partial EMT-like phenotypic transition by activating the FAK-JAK1 axis, while concurrently impairing mitochondrial oxidative phosphorylation and structural integrity, thereby driving DKD progression.

TFAM K76 acetylation excessively activates autophagy and worsens mitochondrial damage

Mitochondrial damage in DKD often triggers autophagy as a compensatory response; however, persistent oxidative stress can lead to maladaptive autophagy, further exacerbating mitochondrial injury [22, 23]. TEM of kidney tissues from T2DM mice revealed markedly increased autophagic structures. Among the groups, TFAM-K76Q overexpression induced the highest level of autophagy, whereas TFAM-K76R significantly reduced autophagic vesicles (Fig. 4A). IF analysis demonstrated that TFAM-K76Q markedly increased LC3 puncta formation and fluorescence intensity while reducing P62 fluorescence intensity and punctate aggregates compared to the WT group, confirming its excessive activation of autophagy. Unexpectedly, although overexpression of wild-type TFAM also increased total TFAM levels, TFAM’s function only drives pathological overactivation of autophagy when K76 is acetylated (Fig. 4B). In HK-2 cells, inhibition of autophagy using Bafilomycin A1 (BafA1) reversed the TFAM K76 acetylation-induced elevation of LC3-II and reduction of P62 (Fig. 4C, D). Because TFAM has been proposed to act as an autophagy receptor via LC3 binding [24], we assessed autophagic flux using an mCherry-GFP-LC3B reporter. TFAM K76 acetylation enhanced autophagic flux, as indicated by increased red puncta, which was abolished upon BafA1 treatment (Fig. 4E). Immunoprecipitation revealed that TFAM-K76Q exhibited stronger binding to LC3 compared to WT and K76R (Fig. 4F, G). Structural modeling (AlphaFold3) predicted that K76 acetylation reinforces the TFAM-LC3 interaction (Fig. 4H), and PLA further validated this enhanced interaction in cells (Fig. 4I). To determine the structural basis of this interaction, TFAM and TFAM-K76Q plasmids lacking the LC3-interacting region (LIR) motif were generated (Fig. 4J). Deletion of the LIR motif in the K76Q background completely abolished LC3 binding (Fig. 4K), confirming that the LIR motif mediates TFAM-LC3 interaction. Finally, we assessed mitochondrial integrity by analyzing ETC proteins. Both autophagy inhibition and LIR motif deletion partially restored ETC protein expression (Fig. 4L, M), indicating that mitochondrial injury was alleviated. These findings reveal that TFAM K76 acetylation promotes excessive autophagy, which paradoxically fails to eliminate damaged mitochondria and instead exacerbates mitochondrial dysfunction and renal injury in DKD.

Fig. 4.

TFAM K76 acetylation excessively activates autophagy and exacerbates mitochondrial damage. A TEM and quantitative analysis of renal cortex from T2DM mice in different groups. Red arrows indicate autophagosomes and autolysosomes. B Representative immunofluorescence images of LC3B and P62 in kidney tissues from T2DM mice. White asterisks indicate LC3 accumulation; yellow asterisks indicate P62 accumulation, scale bar 5 µm. C, D Western blot and quantitative analysis of P62 and LC3-II levels in HK-2 cells. Autophagy inhibition was performed using Bafilomycin A1 (BafA1, 100 μM, 8 h). E Representative fluorescence images of HK-2 cells expressing mCherry-GFP-LC3B under high-glucose conditions in TFAM WT and K76Q backgrounds. White arrows indicate GFP-quenched mCherry-LC3 puncta. Scale bars: 10 μm (merged), 1 μm (enlarged). F, G Co-immunoprecipitation in HK-2 cells under high-glucose conditions to validate interactions between LC3B and TFAM WT or mutants using anti-FLAG (F) or anti-GFP (G). H AlphaFold3 structural prediction of TFAM WT (magenta), TFAM K76Q (magenta), and SIRT3 (gray) interactions. I Proximity Ligation Assay (PLA) showing TFAM WT or K76Q interaction with LC3B in HK-2 cells, scale bar 10 µm. J Schematic of TFAM WT and mutant constructs with deletion of the LC3-interacting region (LIR). K Co-immunoprecipitation using anti-FLAG/anti-MYC to assess LC3B binding to TFAM WT or K76Q with or without the LIR under high-glucose conditions. L Western blot and quantification of N-cadherin, Vimentin, CYTB, and COX2 in HK-2 cells treated with or without BafA1. M Western blot and quantitative analysis of N-cadherin, Vimentin, CYTB, and COX2 in HK-2 cells transfected with indicated plasmids. Data are presented as mean ± SD. ns, not significant; ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

SIRT3 modulates mitochondrial function by regulating TFAM K76 deacetylation

Lysine acetylation is dynamically regulated by histone deacetylases and the sirtuin family. Treatment with the histone deacetylase inhibitor Trichostatin A (TSA) and the sirtuin inhibitor nicotinamide (NAM) revealed that NAM, but not TSA, markedly increased TFAM K76 acetylation (Fig. 5A). Among mitochondrial sirtuins (SIRT3, SIRT4, and SIRT5)[25], only SIRT3 knockdown significantly elevated TFAM K76 acetylation in HK-2 cells (Fig. 5B). Co-IP confirmed a direct interaction between SIRT3 and TFAM (Fig. 5C–E), and SIRT3 knockdown increased TFAM acetylation in both HK-2 and primary tubular cells (Fig. 5F, G). AlphaFold3 structural modeling predicted stable SIRT3-TFAM binding (Fig. 5H). Immunofluorescence analysis revealed co-localization of SIRT3 and TFAM under normal conditions, which was markedly reduced under high-glucose conditions, paralleling the increase in TFAM acetylation (Fig. 5I). Functionally, SIRT3 silencing enhanced TFAM K76 acetylation, promoting inflammation and fibrosis under high glucose, whereas SIRT3 overexpression exerted protective effects (Fig. 5J, K). Pharmacological inhibition of SIRT3 using 3-TYP further aggravated mitochondrial damage and fibrosis in high-glucose-treated HK-2 cells (Fig. S4A). Previous studies have identified histidine 248 (H248) as the catalytic site essential for SIRT3 deacetylation activity [26]. We generated a deacetylation-deficient SIRT3 mutant (H248A), which abolished the protective effect of SIRT3 overexpression and reduced the expression of ETC subunits (Fig. 5L, M and Fig. S4B). Importantly, although SIRT3 overexpression reversed high-glucose-induced injury in wild-type cells, it failed to rescue fibrosis and mitochondrial dysfunction in the presence of the acetylation-mimetic mutant TFAM K76Q (Fig. 5N). Consistent with these findings, both IHC and Western blot analyses revealed reduced SIRT3 expression in kidney tissues from diabetic patients and mice (Fig. 5O–Q). These results identify SIRT3 as the primary deacetylase regulating TFAM K76 via its catalytic H248 site and highlight that SIRT3 downregulation contributes to DKD progression by permitting pathological TFAM acetylation.

Fig. 5.

SIRT3 regulates mitochondrial function and fibrosis by modulating TFAM K76 acetylation. A Western blot and quantitative analysis of TFAM K76 acetylation in HK-2 cells treated with TSA (200 nM) or NAM (10 mM). B Western blot and quantitative analysis of TFAM K76 acetylation in HK-2 cells following siRNA-mediated knockdown of SIRT3, SIRT4, or SIRT5. C-E Endogenous and exogenous co-immunoprecipitation assays demonstrating the physical interaction between TFAM and SIRT3 in HK-2 cells. F, G Immunoprecipitation analysis showing increased TFAM K76 acetylation in HK-2 and primary renal tubular epithelial cells under high-glucose conditions after SIRT3 knockdown. H Molecular docking model illustrating the predicted interaction interface between TFAM (magenta) and SIRT3 (cyan), with hydrogen bonds indicated in yellow. I Immunofluorescence staining showing colocalization of TFAM and SIRT3 in HK-2 cells, scale bar 10 µm. J, K Western blot and quantitative analysis of fibrosis- and apoptosis-related markers (N-cadherin, vimentin, TGF-β, α-SMA, and BAX) in HK-2 cells upon SIRT3 overexpression or knockdown. L Western blot and quantitative analysis of the same markers in HK-2 cells expressing wild-type SIRT3 or the catalytically inactive mutant SIRT3 H248A. M RT-qPCR analysis of TGF-β1 mRNA expression in HK-2 cells expressing SIRT3 or SIRT3 H248A (n=3). N Western blot and quantitative analysis of HK-2 cells expressing SIRT3, K76Q, or co-overexpressing SIRT3 and K76Q (n=3). O, P Representative immunohistochemistry images and quantification of SIRT3 expression in kidney tissues from DKD patients and mice, scale bar 50 µm. Q Western blot and quantification of SIRT3 protein expression in the renal cortex of control and DKD mice (n=5). Data are presented as mean ± SD. ns, no significance; ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

High-throughput screening identifies a small-molecule inhibitor that specifically reduces TFAM K76 acetylation

Given the pivotal role of TFAM K76 acetylation in DKD progression, we conducted high-throughput virtual screening to identify small molecules that specifically suppress this modification. A virtual screening model was established based on the TFAM crystal structure (PDB ID: 3TMM), with the docking site centered around K76. Grid files were generated using Maestro, and a library of 200,000 structurally diverse compounds was docked to this site. From the top 150 candidates, clustering yielded 22 representative compounds for functional screening (Fig. 6A). Binding affinities were further assessed via BLI using purified TFAM proteins, and the results showed that compounds 1, 4, 14, 19, and 20 have relatively high affinity for the TFAM protein. (Fig. S5). Cytotoxicity assessment by CCK-8 revealed that compounds 1 and 4 significantly reduced HK-2 cell viability (Fig. S6A). Western blot analysis identified compound 14 as a potent inhibitor of TFAM K76 acetylation without affecting total TFAM expression; this compound was designated C14 (Fig. 6B, C). Molecular docking showed that C14 occupies the TFAM binding pocket, forming hydrogen bonds with K76 and E88 (Fig. 6D–F). BLI confirmed strong binding between C14 and wild-type TFAM (Kd = 69.32 μM), whereas binding to the K76R mutant was markedly reduced (Fig. 6G). CETSA demonstrated that C14 enhanced TFAM thermal stability in HK-2 cells (Fig. 6H). Under high glucose conditions, C14 reduced TFAM K76 acetylation in HK-2 cells in a dose- and time-dependent manner (Fig. 6I, J). Furthermore, C14 also decreased TFAM K76 acetylation in primary renal tubular epithelial cells (Fig. S6B, C). Mechanistically, C14 did not alter SIRT3 expression (Fig. S6D) but enhanced the interaction between TFAM and SIRT3, as confirmed by molecular docking, Co-IP, and IF (Fig. 6K–M). Functionally, C14 improved mitochondrial bioenergetics by increasing OCR, ATP production, and maximal respiration (Fig. 6N, O). Consistent with these findings, mitochondrial DNA copy number, ETC protein levels, and mRNA expression were elevated upon C14 treatment (Fig. 6P–R). C14 also alleviated mitochondrial structural damage and prevented mPTP opening induced by high glucose (Fig. 6S, T). Together, these data establish C14 as a selective TFAM K76 acetylation inhibitor that restores mitochondrial function and protects against high-glucose-induced injury, highlighting its potential as a therapeutic candidate for DKD.

Fig. 6.

Identification and validation of a selective inhibitor targeting TFAM K76 acetylation. A Schematic workflow of the high-throughput screening process for small-molecule inhibitors of TFAM K76 acetylation. B Western blot and quantification of TFAM K76 acetylation levels in HK-2 cells treated with 50 μM of compounds 1, 4, 14, 19, and 20 for 48 h. C Chemical structure of compound 14 (C14). D Molecular docking illustration showing the binding interaction between TFAM and C14. E, F 3D and 2D interaction plots depicting hydrogen bonds and other interactions between C14 and TFAM. G Bio-layer interferometry (BLI) analysis of the binding kinetics between C14 and wild-type TFAM or the K76R mutant (residues 43-246). H Cellular thermal shift assay (CETSA) assessing TFAM stabilization with or without C14 pretreatment. I, J Western blot and quantification of TFAM K76 acetylation levels in HK-2 cells treated with varying concentrations of C14 or for different durations. K Molecular docking model illustrating the interaction among TFAM (magenta), SIRT3 (cyan), and C14 (yellow). L Co-immunoprecipitation assays showing interactions between TFAM and SIRT3 in HK-2 cells, with or without 100 μM C14 treatment under high-glucose conditions. M Representative immunofluorescence images of HK-2 cells treated with C14 (100 μM) under high-glucose conditions. Cells were stained with DAPI (blue), TFAM (green), and SIRT3 (red). Scale bars: 10 μm. N, O Oxygen consumption rate (OCR) analysis of HK-2 cells treated with C14 (100 μM) for 48 h, measuring ATP production and maximal respiration (n=5). P Quantification of mitochondrial DNA (mtDNA) copy number in HK-2 cells after 100 μM C14 treatment for 48 h. Q RT-qPCR analysis of MT-ND1 and MT-CO2 mRNA expression in HK-2 cells with or without 100 μM C14 under high-glucose conditions. R Western blot and quantification of mitochondrial proteins (COX2, CYTB, and ND1) in HK-2 cells treated with 100 μM C14 for 48 h (n=3). S Mito-tracker staining to assess mitochondrial morphology in C14-treated HK-2 cells. Scale bars: 10 μm. T Assessment of mitochondrial permeability transition pore (mPTP) opening in HK-2 cells co-loaded with calcein-AM and CoCl2. Data are presented as mean ± SD. ns, no significance; ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

C14 attenuates high-glucose-induced fibrosis and inflammation in vitro

To investigate the protective effects of C14, HK-2 cells were treated with C14 under HG conditions, followed by quantitative proteomic analysis (Fig. 7A). LC–MS/MS identified 121 upregulated and 64 downregulated proteins between HG and NG groups, and 76 upregulated and 301 downregulated proteins between HG + C14 and HG groups (|log₂FC|> 1, P < 0.05) (Fig. 7B, C). PCA demonstrated clear segregation between HG and NG groups, as well as HG and HG + C14 groups, with C14 partially reversing HG-induced proteomic alterations (Fig. 7D and Fig. S7A, B). GO analysis revealed that differentially expressed proteins were predominantly involved in ATP synthesis-coupled electron transport and the respiratory electron transport chain (Fig. 7E, F and Fig. S7C, D). Notably, proteins upregulated under HG, including fibrosis- and inflammation-associated factors such as FBN2, CKLF, FIBB, PTN14, DAF, and DPP4, were significantly downregulated by C14 treatment (Fig. 7G, H and Fig. S7E). IPA further indicated that HG exposure activated pro-inflammatory and pro-fibrotic networks, including IL1β, IL4, IFNG, AGT, TEAD1, and MRTFA (Fig. 7I). Treatment with C14 suppressed these pathways, reducing the expression of key mediators such as VEGFA, IFNG, and TGF-β1 (Fig. 7J and Fig. S7F, G). Western blot analyses confirmed that C14 significantly decreased inflammation- and fibrosis-related protein levels in both HK-2 and primary tubular epithelial cells under HG conditions (Fig. 7K–M and Fig. S7H). IF staining demonstrated reduced vimentin expression in primary tubular cells upon C14 treatment (Fig. 7N). Correspondingly, mRNA expression of pro-inflammatory cytokines (IL-1β, IL-6, TGF-β1) and chemokines was markedly decreased in both HK-2 and primary tubular cells (Fig. 7O, P and Fig. S7I). Overall, these data indicate that C14 protects renal tubular epithelial cells from hyperglycemia-induced injury by mitigating inflammation and fibrosis, highlighting its therapeutic potential in DKD.

Fig. 7.

The small-molecule compound C14 attenuates high glucose-induced fibrosis and inflammation in HK-2 and primary tubular epithelial cells. A Schematic workflow of the experimental design: HK-2 cells were treated with or without C14 (100 μM, 48 h) under high-glucose conditions, followed by proteomic analysis via mass spectrometry. B, C Volcano plots showing differentially expressed proteins in NG vs. HG, and HG vs. HG + C14 groups. D Principal component analysis (PCA) of proteomic profiles from NG, HG, and HG + C14 groups. E, F Gene Ontology (GO) enrichment analysis of differentially expressed proteins for NG vs. HG, and HG vs. HG + C14 comparisons. G Venn diagram illustrating the overlap between proteins upregulated by HG and those downregulated by C14 treatment. H Heatmap showing hierarchical clustering of differentially expressed proteins across NG, HG, and HG + C14 groups. I Ingenuity Pathway Analysis (IPA) of HG-upregulated proteins (fold change > 1.2). Orange indicates pathway activation. J IPA of proteins downregulated by C14 treatment under HG conditions (fold change < 0.8). Blue indicates pathway inhibition. K-M Western blot and quantification of fibrosis- and apoptosis-related markers (N-cadherin, Vimentin, TGF-β, α-SMA, BAX, and Bcl-2) in HK-2 and primary renal tubular epithelial cells treated with 100 μM C14 for 48 h. N Representative immunofluorescence images showing vimentin (red) expression in primary tubular epithelial cells treated with 100 μM C14, scale bar 10 µm. O RT-qPCR analysis of IL-1β, IL-6, and TGF-β1 mRNA levels in HK-2 and primary tubular epithelial cells following 100 μM C14 treatment for 48 h. P RT-qPCR analysis of chemokine genes Ccl2, Ccl5, and Ccn2 in HK-2 cells after C14 treatment. Data are presented as mean ± SD. ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

The compound C14 significantly attenuated kidney injury in vivo

To assess the therapeutic potential of C14 in DKD, T2DM mice were orally administered C14, with dapagliflozin (DAPA) serving as a positive control (Fig. 8A). As a SGLT2 inhibitor, dapagliflozin is known to lower blood glucose and mitigate renal inflammation, cortical hypoxia, and proteinuria [27, 28]. Treatment with C14 did not significantly affect blood glucose levels or body weight (Fig. 8B, C). Nevertheless, both C14 and dapagliflozin markedly reduced serum creatinine, blood urea nitrogen, and urine albumin-to-creatinine ratios compared to untreated T2DM mice (Fig. 8D–F). C14 substantially decreased TFAM K76 acetylation in renal tissue (Fig. 8G, H) and reduced the expression of the kidney injury marker KIM-1 (Fig. 8I, J and Fig. S8A). Serum transaminase analysis indicated that, despite high-fat diet-induced hepatic stress, C14 did not produce significant hepatotoxicity (Fig. 8K). The H&E staining results of multiple organs indicated that after C14 treatment, no obvious histopathological abnormalities or clear signs of toxicity were observed in these organs (Fig. S8B). At the transcriptional level, both C14 and dapagliflozin significantly downregulated mRNA expression of pro-inflammatory cytokines (Il1b, Il6, Tnf-α, Tgf-β1) and chemokines (Ccl2, Ccl5, Ccn2) in T2DM kidney tissue (Fig. 8L, M). Blinded semiquantitative histopathological analysis confirmed that C14 treatment significantly alleviated hyperglycemia-induced mesangial matrix expansion and interstitial fibrosis compared with the solvent-treated T2DM group (Fig. 8N–P). TEM confirmed that C14 alleviated T2DM-induced kidney damage, including GBM thickening, podocyte injury, and mitochondrial abnormalities (Fig. 8Q, R and Fig. S8C, D). Mechanistically, C14 reduced LC3B-II expression and increased P62 levels in HK-2 cells under high-glucose conditions (Fig. S8E), while TEM analysis demonstrated a reduction in excessive autophagy (Fig. 8S, T). IHC and IF further validated the attenuation of renal fibrosis in C14-treated mice (Fig. 8U and Fig. S8F–I). Taken together, these results demonstrate that C14 confers robust renal protection in T2DM mice, comparable to dapagliflozin, by reducing TFAM K76 acetylation and alleviating inflammation, fibrosis, and mitochondrial damage.

Fig. 8.

C14 alleviates renal injury in T2DM mice. A Schematic overview of the experimental design: T2DM-induced mice were orally administered vehicle, C14 (20 mg/kg), or dapagliflozin (10 mg/kg). B, C Blood glucose levels and body weight were monitored across all treatment groups. Data are presented as mean ± SD (n=6). D-F Serum creatinine (SCr), blood urea nitrogen (BUN), and albumin-to-creatinine ratio (ACR) were measured in each group. Data are presented as mean ± SD (n=6). G, H Western blot and quantification of TFAM K76 acetylation levels in renal cortex tissue. Data are presented as mean ± SD (n=5). I, J Western blot and quantification of KIM-1 expression levels in renal cortex tissue. Data are presented as mean ± SD (n=5). K Serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were measured to assess liver function. L RT-qPCR analysis of inflammatory cytokine mRNA levels (Il-1β, Il-6, Tnf-α, and Tgf-β1) in renal cortex tissues. M RT-qPCR analysis of fibrosis- and inflammation-associated genes (Ccl2, Ccl5, and Ccn2) in renal cortex tissues. Data are presented as mean ± SD (n=3). N Representative histological images of kidney sections stained with H&E, PAS, and Masson’s trichrome, scale bar 50 µm. O, P Semi-quantitative analysis of PAS and Masson staining in different treatment groups (n=6 per group). Q, R TEM images showing glomerular ultrastructure and mitochondrial morphology in the renal cortex of T2DM mice treated with vehicle, C14, or dapagliflozin. Red arrowheads indicate thickened GBM; yellow arrows indicate effaced podocyte foot processes. S, T TEM images and quantitative analysis of renal cortex ultrastructure in T2DM mice across different treatment groups. The red arrow points to the autophagosome and the autolysosome. U Representative immunofluorescence images showing vimentin (green) and collagen I (red) expression in renal tissues, scale bar 50 µm. Data are presented as mean ± SD. ns, no significance; *p < 0.05, **p < 0.01, ***p < 0.001

Discussion

Mitochondrial oxidative phosphorylation (OXPHOS) dysfunction is a central contributor to the pathogenesis of DKD, as hyperglycemia promotes excessive mitochondrial superoxide production. Dysregulation of mitochondrial dynamics, including fission, fusion, and autophagy, further exacerbates renal injury in DKD [29, 30], highlighting the importance of maintaining mitochondrial homeostasis. Targeting mitochondrial dysfunction may therefore offer innovative therapeutic strategies for DKD [31]. Previous studies have identified key regulatory mechanisms underlying mitochondrial damage in DKD. For example, modulation of the RBBP6-ERRα interaction mitigates mitochondrial injury, while elevated AKAP1 expression in DKD impairs mtDNA replication and mitochondrial function in podocytes [32, 33]. In these contexts, the inhibition of TFAM expression contributed to mitochondrial dysfunction. However, the role of TFAM acetylation in DKD remains unexplored. Our study revealed elevated TFAM K76 acetylation in kidney samples from DKD patients and murine models. Using comprehensive in vitro and in vivo analyses, we demonstrated that hyperacetylation at K76 exacerbates inflammation and fibrosis in renal tubular epithelial cells and impairs mitochondrial OXPHOS, TFAM K76 acetylation emerges as a critical regulator of TFAM function and mitochondrial homeostasis. To further clarify the role of TFAM K76 acetylation in mitochondrial injury, we plan to validate our findings in other models, including db/db mice, in the future.

Acetylation modifications are critical regulators of protein function, influencing essential biological processes such as glucose metabolism, lipid homeostasis, and the tricarboxylic acid (TCA) cycle [34, 35]. In DKD, acetylation is closely associated with metabolic dysregulation, inflammation, and fibrosis [29, 36]. TFAM acetylation, in particular, represents a key mechanism controlling its function. Elevated TFAM acetylation has been observed in kidney cancer, contributing to mitochondrial dysfunction [37], while in vascular dementia models, restoring TFAM deacetylation improves mitochondrial function and alleviates disease phenotypes [38]. Elucidating the mechanisms of TFAM acetylation and its impact on mitochondrial function may therefore reveal novel therapeutic targets for DKD. In our previous study, we found that TFAM acetylation impairs its mitochondrial translocation by reducing interaction with TOM70 in acute kidney injury [39]. Here, we show that elevated TFAM acetylation at lysine 76 (K76) is associated with increased fibrosis and inflammation in DKD. Proteomic analyses revealed that TFAM K76 acetylation activates the FAK and JAK1 pathway. Mechanistically, K76 acetylation hinders mitochondrial biogenesis and mtDNA replication, highlighting its critical role in mitochondrial homeostasis. Additionally, TFAM K76 acetylation significantly enhances autophagy in DKD mice. TFAM can act as an autophagy receptor by binding to LC3, thereby limiting mtDNA leakage under stress [24]. We discovered that K76 acetylation increased TFAM-LC3 binding and activated autophagy flux, and the dysregulation of autophagy intensified mitochondrial damage caused by TFAM K76 acetylation. These findings provide a new mechanism linking TFAM acetylation with mitochondrial dysfunction in DKD.

In advancing research on acetylation modifications in DKD, several compounds have been developed to modulate both histone and non-histone acetylation pathways. Resveratrol, for instance, reduces tubular damage by activating SIRT1 and inhibiting NF-κB p65 acetylation [40, 41]. Active ingredients in astragalus, such as puerarin, synergistically interact with SIRT1 to decrease proteinuria and renal fibrosis [42]. SIRT3, a key mitochondrial deacetylase, regulates antioxidant defense and energy metabolism [43, 44], and recent studies report that 2-APQC, a small molecule SIRT3 activator, enhances its deacetylation activity without altering SIRT3 expression [45]. In addition, other compounds can attenuate tubular injury through regulation of histone deacetylases, for example, CAY10603 inhibits HDAC6 to reduce renal epithelial cell damage [46]. However, most acetylation-modulating compounds target multiple SIRT or HDAC family members, often resulting in off-target effects and potential toxicity, limiting therapeutic efficacy. In this study, we identified a small molecule inhibitor, C14, that selectively targets TFAM K76 acetylation via high-throughput virtual screening. C14 significantly reduced TFAM K76 acetylation without altering TFAM or SIRT3 expression. Protein-small molecule affinity assays demonstrated that C14 specifically binds the K76 site, enhancing TFAM’s interaction with SIRT3, suggesting that C14 may expose additional TFAM-SIRT3 binding interfaces. Importantly, mutation of the K76 site markedly reduced C14’s binding affinity, confirming its high specificity and minimizing potential off-target effects.

Currently, strict glycemic and blood pressure control remain the cornerstone of clinical management for diabetic patients [47, 48]. In recent years, SGLT2 inhibitors, such as empagliflozin and dapagliflozin, have emerged as novel therapies that lower blood glucose while reducing proteinuria and slowing eGFR decline by inhibiting renal glucose reabsorption [49]. However, glycemic control carries a risk of hypoglycemia, and these strategies have limited efficacy in patients with advanced DKD [50]. In addition, the discovery of anti-inflammatory drugs (such as IL-6 inhibitors) and antifibrotic drugs (such as TGF-β antagonists), as well as some other inhibitors, is currently undergoing preclinical or early clinical trials [51–53], but their therapeutic effects remain modest. This may be because DKD progression involves multiple complex pathways, including autophagy dysregulation, oxidative stress, and inflammation, which making it difficult for single-target therapies to fully intervene, while combination therapies require further investigation. In our study, C14 significantly attenuated kidney injury in DKD models, lowering serum creatinine and urea levels and reducing proteinuria, with renal protective effects comparable to dapagliflozin. Importantly, C14 exhibited high specificity and safety, without increasing systemic burden, highlighting its potential as a promising therapeutic strategy for DKD. Future investigations involving pharmacokinetics and detailed biodistribution studies are necessary to comprehensively characterize C14’s tissue penetration and potential extrarenal effects.

Although our findings were validated in both cellular and animal models, this study has certain limitations. The limited sample size in our human cohort prevents us from establishing a robust correlation between TFAM K76 acetylation levels and clinical parameters such as serum creatinine and urea nitrogen. The STZ combined with unilateral nephrectomy-induced DKD model used in this study exhibits pathological features corresponding to the early stages of human DKD. Therefore, this study primarily reveals the role of TFAM K76 acetylation in the early pathogenesis of DKD. Future studies should further validate the generalizability of these findings in models of later-stage pathology or in late-stage human samples. Furthermore, while we demonstrated that TFAM K76 acetylation inhibits mitochondrial oxidative phosphorylation in tubular epithelial cells and primary mouse renal cells, technical constraints prevented validation of C14’s anti-fibrotic and anti-inflammatory effects using kidney tissue organoids. In future studies, using patient-derived renal tubular organoids is an important and promising direction, which is crucial for further validating the therapeutic potential of C14 and its translational significance for human diseases.

In conclusion, our study elucidates the role of TFAM K76 acetylation in the progression of DKD through its regulation of mitochondrial oxidative phosphorylation. To counteract the mitochondrial dysfunction induced by TFAM K76 acetylation, we identified C14 as a targeted small-molecule inhibitor. Our results demonstrate that C14 effectively ameliorates renal damage in DKD models both in vitro and in vivo. Collectively, these findings highlight C14 as a promising therapeutic agent capable of mitigating inflammation and fibrosis by restoring mitochondrial homeostasis, providing a potential avenue for future interventions in DKD.

Abbreviations

DKD

Diabetic kidney disease

CKD

Chronic kidney disease

ESRD

End-stage renal disease

T2DM

Type 2 diabetes mellitus

GLP-1

Glucagon-like peptide-1

SGLT-2

Sodium-glucose cotransporter 2

TECs

Tubular epithelial cells

HG

High glucose

NG

Normol glucose

TFAM

Mitochondrial transcription factor A

mtROS

Mitochondrial reactive oxygen species

SIRT3

Sirtuin 3

mtDNA

Mitochondrial DNA

IHC

Immunohistochemistry

IF

Immunofluorescence

TEM

Transmission electron microscopy

HE

Hematoxylin & eosin

PAS

Periodic acid-Schiff

VTN

Vitronectin

DAG1

Dystroglycan

OCR

Oxygen consumption rate

ETC

Electron transport chain

mPTP

Mitochondrial permeability transition pore

Author contributions

QW and TTF conceived the study, and TTF and SNS conducted the cell and animal experiments; FZ and JHZ participated in reviewing pathological images and performed blinded semi-quantitative histopathological analysis; ZYW and FYX participated in the animal experiments; XZJ provided important academic advice; QM and YW collected and organized clinical information; TTF drafted the manuscript, and QW, YL, and SNS critically revised it. All authors reviewed the manuscript draft and approved the final version. We declare that the manuscript has been approved for publication by all authors. On behalf of my co-authors, I declare that the work described is original research, has not been previously published, and has not been considered for publication elsewhere, in whole or in part. All listed authors have approved the accompanying manuscript.

Funding

This study was supported by the National Natural Science Foundation of China (82500905, 82370725, 82571828, 82471621, 82371604, 82200049, 82271622); Shandong Provincial Natural Science Foundation (ZR2023QH316, ZR2024MH321); Shandong Province Medical and Health Development Plan (202403031074); Jinan Science and Technology Plan Project (202328037, 202512007); Jinan Key Laboratory of Medical Sciences Foundation (20211201); Shandong Provincial Taishan Scholar Young Expert Program (tsqn202507354); Jinan High-level Talents in the Medical and Health Industry Special Fund (202412) and the Introduced Talents of Jinan Central Hospital (GD002, YJRC2021011).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Human DKD samples were obtained from kidney puncture tissue specimens from patients with chronic kidney disease, and human pericancerous kidney tissue was obtained from patients with kidney cancer who underwent surgical resection. The study was approved by the Medical Ethics Committee of the Affiliated Central Hospital of Shandong First Medical University (Approval No. R202303060085). All subjects signed an Institutional Review Board (IRB)-approved informed consent before enrollment in the study. The study was performed in accordance with the Declaration of Helsinki. All animal studies were conducted with approval from the Animal Research Ethics Committee of Jinan Central Hospital (Approval No. JNCHIACUC2021-79) of China and performed in accordance with established guidelines.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.