AARS1 promotes diabetic kidney disease through rewiring Akt and NF-κB signaling to suppress autophagy and sustain inflammation

Abstract

Background

Diabetic kidney disease (DKD) is characterized by metabolic reprogramming, autophagy impairment, and chronic inflammation, but the molecular mechanisms linking these processes are not fully understood. Lactylation has emerged as an important metabolic–epigenetic regulatory mechanism in diabetic tissues. Alanyl-tRNA synthetase 1 (AARS1) has recently been identified as a lactyltransferase, but whether and how AARS1-mediated lactylation contributes to tubular stress responses and DKD progression remains unclear.

Methods

Kidney-specific Aars1 knockout mice and β-alanine treatment were used in streptozotocin-induced and db/db diabetic mouse models. Human proximal tubular epithelial cells cultured under high-glucose conditions, including CRISPR/Cas9-mediated AARS1 knockout cells, were used for mechanistic studies. AARS1-dependent transcriptional programs were analyzed by CUT and Tag, ChIP assays, and luciferase reporter assays.

Results

AARS1 was upregulated in diabetic kidneys and directly lactylated Akt and the NF-κB subunit p65, enhancing their phosphorylation and activation. This modification promoted autophagy impairment, inflammatory cytokine expression, tubular injury, and macrophage accumulation. CUT and Tag analysis further revealed AARS1-dependent transcriptional control of HK2, PFKP, ZEB1, and PPP6C, linking AARS1 to glycolytic reprogramming and fibrotic signaling. Mechanistically, AARS1 operated within a glycolysis–lactate–NF-κB feedback circuit, in which glycolysis-driven lactate increased the lactylation and activation of NF-κB, promoting AARS1 transcription and reinforcing glycolytic reprogramming and chronic tubular stress. Genetic deletion of Aars1 or pharmacological inhibition with β-alanine reduced protein lactylation, restored autophagy, attenuated inflammation, and significantly slowed DKD progression in both diabetic mouse models.

Conclusions

These findings identify AARS1 as a metabolic–epigenetic amplifier that rewires Akt- and NF-κB-dependent signaling to sustain chronic tubular stress and fibrotic remodeling in DKD, highlighting the AARS1–lactylation axis as a potential therapeutic target.

Supplementary Information

The online version contains supplementary material available at 10.1186/s11658-026-00953-5.

Introduction

Diabetic kidney disease (DKD) is a major microvascular complication of both type 1 and type 2 diabetes and remains the leading cause of end-stage kidney disease (ESKD) worldwide [1]. Despite advances in glycemic and blood pressure management, the prevalence of DKD remains high, and effective therapies to halt or reverse disease progression are still lacking [2]. Accumulating evidence indicates that tubular injury often occurs earlier than glomerular damage, precedes overt proteinuria, and shows a stronger association with subsequent renal function decline [3, 4]. Indeed, patients with nonalbuminuric DKD or reduced kidney function prior to the onset of albuminuria frequently exhibit more severe tubulointerstitial fibrosis and tubular atrophy, underscoring tubular injury as a key driver of DKD progression [5]. Elucidating the molecular mechanisms underlying tubular injury and maladaptive stress responses is therefore critical for identifying effective therapeutic targets.

Metabolic reprogramming is a hallmark of diabetic kidneys, with tubular epithelial cells shifting from oxidative phosphorylation to glycolysis [6]. Growing evidence indicates that aberrant glycolytic activation and excessive lactate accumulation directly contribute to tubular injury and accelerate DKD progression [7]. Clinically, elevated urinary lactate is associated with faster kidney function decline and higher risk of progression to ESKD in DKD, underscoring the pathogenic role of altered lactate metabolism [8]. Beyond acting as a metabolic byproduct, lactate also functions as a signaling molecule and a substrate for lysine lactylation, an emerging post-translational modification that couples cellular metabolic state to gene regulation [9]. Recent work has implicated lactate-driven lactylation in transcriptional reprogramming, epithelial–mesenchymal transition (EMT), and fibrotic remodeling, emphasizing the importance of metabolic–epigenetic regulation in diabetic tubular injury [10]. In parallel with these metabolic disturbances, diabetic kidneys exhibit persistent inflammation, impaired autophagy, and progressive fibrotic remodeling, reflecting a state of sustained tubular stress that drives disease progression and adverse renal outcomes [11–13].

Recent studies have identified alanyl-tRNA synthetase 1 (AARS1), a canonical component of the protein synthesis machinery, as a bona fide lactate sensor and lysine lactyltransferase [14, 15]. AARS1 directly binds lactate and catalyzes its transfer onto lysine residues of both histone and nonhistone substrates, thereby coupling cellular metabolic status to epigenetic modification and signaling regulation. Notably, a recent study demonstrated that AARS1-mediated lactylation of histone H3 lysine 18 and STAT1 promotes ferroptotic cell death in diabetic nephropathy, establishing a role for AARS1 in lactylation-dependent ferroptosis pathways during diabetic kidney injury [16]. While this work highlights an important function of AARS1 in regulating tubular cell death, it remains unclear whether AARS1 also governs nonlethal tubular stress responses—including metabolic dysregulation, impaired autophagy, and inflammatory remodeling—that ultimately drive renal fibrotic remodeling and DKD progression.

In this study, we define a previously unrecognized role of AARS1 in orchestrating nonlethal tubular stress and fibrotic remodeling in DKD. Using Cleavage Under Targets and Tagmentation (CUT and Tag) analysis, we identify AARS1-dependent transcriptional programs linked to glycolytic reprogramming and fibrotic remodeling. In addition, we demonstrate that Akt and the NF-κB subunit p65 are novel nonhistone substrates of AARS1, and that AARS1-mediated lactylation enhances their phosphorylation and activation, leading to impaired autophagy and sustained inflammatory signaling in diabetic kidneys. There findings establish AARS1 as a central metabolic–epigenetic amplifier that drives chronic tubular stress and fibrotic remodeling and suggest that targeting the AARS1–lactylation axis may represent a promising therapeutic strategy for DKD.

Methods

Mice and in vivo interventions

C57BL/6J mice, Ksp-Cre transgenic mice, B6.BKS-Lepr^dbJ(db/db) mice, and nondiabetic db/m mice were obtained from the Jackson Laboratory. Aars1^fl/fl mice were generated by CRISPR/Cas9-mediated gene editing (Cyagen).

For streptozotocin (STZ)-induced diabetic model, 6-week-old male C57BL/6J mice received intraperitoneal injections of STZ (MedChemExpress) at 60 mg/kg dissolved in 0.05 mM sodium citrate buffer (pH 4.5) once daily for five consecutive days, while control mice received vehicle alone. Blood glucose and body weight were monitored before induction and at 2-week intervals thereafter, and mice with blood glucose levels exceeding 280 mg/dL were considered diabetic. Beginning at 12 weeks of age, STZ-diabetic mice were administered daily intraperitoneal injections of β-alanine (Sigma-Aldrich) at 120 mg/kg dissolved in phosphate-buffered saline (PBS; Sigma-Aldrich) or PBS alone until sacrifice at 24 weeks, when blood and kidney tissues were collected. For the type 2 diabetic model, db/db mice and their db/m littermate controls received daily intraperitoneal injections of β-alanine (120 mg/kg) or PBS beginning at 12 weeks of age and continuing until sacrifice at 24 weeks. Blood and kidney tissues were harvested for subsequent analyses.

Kidney-specific Aars1 knockout mice were used to evaluate the role of AARS1 in DKD. Aars1^fl/fl:Ksp-Cre mice were generated by crossing Ksp-Cre mice with Aars1^fl/fl mice to obtain Aars1^fl/+:Ksp-Cre offspring, which were then intercrossed to produce Aars1^fl/fl:Ksp-Cre mice. Similarly, db/db:Aars1^fl/fl:Ksp-Cre mice were generated by crossing db/m mice with Aars1^fl/+:Ksp-Cre mice, followed by intercrossing to obtain db/db:Aars1^fl/fl:Ksp-Cre offspring for experiments.

Cell culture and regents

Human renal proximal tubular epithelial cells (HK-2; CRL-2190, ATCC) were cultured in Dulbecco’s modified eagle medium (DMEM; Invitrogen) supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin at 37 °C under 5% CO₂ and 5% O₂. Cells were seeded at approximately 70% confluence, serum-starved for 12 h, and subsequently exposed to normal glucose (5 mM) or high glucose (30 mM) conditions for 48 h. β-alanine and l-alanine (MedChemExpress, HY-N0229) were dissolved in PBS to prepare stock solutions at concentrations of 100 mM and 500 mM, respectively. MK-2206 (MedChemExpress) and BAY-11-7085 (Cayman Chemical) were dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich) at stock concentrations of 1 mM and 40 mM, respectively. Sodium lactate and 2-deoxy-d-glucose (2-DG) (both from Sigma-Aldrich) were dissolved in PBS to prepare stock solutions at concentrations of 10 M and 1 M, respectively. All stock solutions were stored at −20 °C and diluted to working concentrations immediately before use.

The following primary antibodies were used for immunoblotting and immunostaining: anti-AARS1 (sc-165990) and anti-HA tag (sc-7392) from Santa Cruz Biotechnology; anti-p65 (10745-1-AP), anti-PPP6C (15851-1-AP), anti-HK2 (22029-1-AP), and anti-PFKP (13389-1-AP) from Proteintech; anti-AKT (no. 9297), anti-p65 (no. 8242), anti-Smad3 (no. 9513), anti-E-cadherin (no. 14472), anti-p62 (no. 39749), and anti-LC3B (no. 83506) from Cell Signaling Technology. Phospho-specific antibodies against AKT-S473 (no. 9271), p65-S536 (no. 3031), and Smad3-S423/425 (no. 9520) were also obtained from Cell Signaling Technology. Anti-l-lactyl lysine antibody (PTM-1401RM) was purchased from PTM BIO. Antibodies against ZEB1 (GTX105278), Kim-1 (NBP1-76701), α-SMA (ab7817), T7 tag (ab9138), fibronectin (MA5-11981), β-actin (A5316), and FLAG tag (F3165) were obtained from GeneTex, Novus Biologicals, Abcam, Invitrogen, or Sigma-Aldrich, as indicated. Horseradish peroxidase-conjugated secondary antibodies were purchased from Santa Cruz Biotechnology.

CRISPR-Cas9-based gene editing

The AARS1-knockout (KO) HK-2 cell lines were generated via CRISPR-Cas9 technology. In brief, the guide RNAs were cloned into a p×459/Puro vector, and then were transfected into HK-2 cells with Lipofectamine 3000 (Invitrogen) and selected as in the literature [17]. The guide RNAs sequences were listed in Supplementary Table S1.

Plasmids and transfection

Flag-tagged wild-type Akt1 and its lysine mutant constructs (K14R and K179R), as well as T7-tagged p65 (RelA) wild-type and lysine mutant constructs (K218R, K221R, and K310R), were obtained from Addgene. An HA-tagged AARS1 expression plasmid was purchased from Sino Biological. Flag-tagged Akt1 constructs were transfected into AARS1-knockout HK-2 cells using Lipofectamine (Invitrogen) in the presence or absence of HA-tagged AARS1. In parallel, T7-tagged p65 (RelA) constructs were transfected into AARS1-knockout HK-2 cells with or without coexpression of HA-tagged AARS1. Cells were harvested 48 h after transfection for downstream analyses.

Western blot analysis and immunoprecipitation

Cells were lysed in ice-cold lysis buffer (20 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100) supplemented with protease and phosphatase inhibitor cocktails (Roche and Sigma-Aldrich). Lysates were briefly vortexed and incubated on ice for 30 min to ensure complete lysis, followed by centrifugation at 20,000×g for 30 min at 4 °C. Clarified supernatants were collected and used for subsequent western blotting or immunoprecipitation assays. For immunoprecipitation, cell lysates were incubated with protein A agarose beads (Pierce) preconjugated with antibodies against AARS1, Akt, p65, or PPP6C, or with corresponding isotype control antibodies. Antibody–bead complexes were prepared in PBS containing 5 mg/mL bovine serum albumin (Sigma-Aldrich) and rotated at 4 °C for 6 h prior to incubation with lysates. Following overnight incubation at 4 °C, beads were washed extensively with lysis buffer supplemented with an additional 300 mM NaCl to reduce nonspecific binding. Bound protein complexes were released by boiling in SDS sample buffer for 7 min and analyzed by western blotting.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was isolated from cells using the RNeasy Plus Mini Kit (QIAGEN) according to the manufacturer’s instructions. For cDNA synthesis, 1 μg of total RNA was reverse-transcribed in a 20 μL reaction using the iScript cDNA Synthesis Kit (Bio-Rad). Quantitative PCR was performed using iTaq SYBR Green Supermix with ROX (Bio-Rad) on an iCycler iQ real-time PCR system. Amplification was carried out for 40 cycles with denaturation at 95 °C for 10 s and annealing/extension at 60 °C for 20 s, followed by melt-curve analysis to verify amplification specificity. Each sample was analyzed in technical triplicates, and experiments were independently repeated at least three times. Relative gene expression levels were normalized to actin, and primer sequences are provided in Supplementary Table S2.

Histology and immunohistochemistry

Paraffin-embedded kidney sections (5 μm) were subjected to Periodic Acid-Schiff (PAS), Masson’s trichrome, and Picrosirius Red staining using a PAS staining system (395B-1KT, Sigma-Aldrich), Trichrome Stain Kit (ab150686; Abcam), and Picrosirius Red Stain Kit (Polysciences, Inc.), respectively, according to the manufacturers’ instructions. For immunofluorescence staining, sections were incubated with primary antibodies against AARS1 (mouse), Kim-1 (rabbit), or F4/80 (rat; 14-4801-82, eBioscience), followed by species-appropriate Alexa Fluor-conjugated secondary antibodies. Alexa Fluor 555 anti-mouse IgG and Alexa Fluor 488 anti-rabbit IgG were used for detection of AARS1 and Kim-1, respectively, while Alexa Fluor 555 anti-rat IgG was used for F4/80 staining. Images were acquired using a Nikon Eclipse 80i microscope.

Transmission electron microscopy (TEM)

Kidney cortex tissues were cut into small pieces (1 mm³) and immediately fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) at 4 °C. Samples were postfixed with 1% osmium tetroxide, dehydrated through graded ethanol, and embedded in epoxy resin. Ultrathin sections (~70 nm) were prepared, stained with uranyl acetate and lead citrate, and examined by transmission electron microscopy. Representative images were acquired from cortical regions for ultrastructural analysis.

Chromatin immunoprecipitation (ChIP) assay

ChIP assays were conducted following established protocols [18]. Briefly, cross-linked chromatin was isolated and subjected to immunoprecipitation using antibodies against AARS1 or p65, with normal IgG serving as a negative control. After extensive washing, protein–DNA complexes were eluted and crosslinks were reversed. Purified DNA was subsequently analyzed by PCR to determine AARS1 occupancy at the promoters of HK2 (Hexokinase 2), PFKP (Phosphofructokinase, Platelet type), ZEB1 (Zinc Finger E-box-binding Homeobox 1), and PPP6C (Protein Phosphatase 6 Catalytic Subunit), as well as p65 binding at the AARS1 promoter.

CUT and tag analysis

CUT and Tag assays were performed using the CUT and Tag Assay Kit (77552; Cell Signaling Technology) following the manufacturer’s protocol. Cells were incubated with ConA-coated magnetic beads for 1 h at room temperature, permeabilized with digitonin, and then incubated overnight at 4 °C with anti-AARS1 antibody or normal IgG. After incubating with a secondary antibody for 1 h at room temperature, samples were treated with the pA/G-Tn5 transposase complex for 1 h, washed, and subjected to tagmentation for 1 h at 37 °C. A stop solution was added, followed by incubation at 58 °C for 1 h to release tagged chromatin fragments. Supernatants were collected, and CUT and Tag DNAs were used for library construction and sequencing on the Illumina NovaSeq 6000 platform.

siRNA-mediated gene silencing

Small interfering RNAs (siRNAs) targeting human PPP6C were purchased from Santa Cruz Biotechnology. Cells were transfected with siRNAs using the DharmaFECT transfection reagent (Dharmacon) according to the manufacturer’s instructions. At 48 h after transfection, cells were collected for protein analysis by western blotting.

Luciferase reporter assay

The proximal promoter regions of human HK2, PFKP, ZEB1, and PPP6C (−2000 bp to +200 bp relative to the transcription start site, TSS) were amplified from human genomic DNA and cloned into the pGL3‑Basic vector (Promega). All constructs were verified by Sanger sequencing. HEK293T cells were seeded in 96‑well plates at 70–90% confluence and cultured in DMEM supplemented with 10% fetal bovine serum. For each well, 100 ng of firefly luciferase reporter plasmid (pGL3‑promoter), 10 ng of Renilla luciferase internal control plasmid (pRL‑TK, Promega), and 50 ng of either empty vector or HA‑AARS1 were cotransfected using Lipofectamine 3000 (Invitrogen). The total amount of DNA was adjusted to 200 ng per well with empty vector. At 24 h post‑transfection, cells were treated with dimethyl sulfoxide (DMSO; vehicle control) or 10 μM BAY‑11‑7085 for an additional 24 h. At 48 h post-transfection, cells were lysed with 1× Passive Lysis Buffer (Promega), and firefly and Renilla luciferase activities were measured using the Dual‑Luciferase Reporter Assay System (Promega) on a GloMax plate reader (Promega). The relative luciferase activity was calculated as the ratio of firefly to Renilla luminescence and normalized to the empty vector control (set to 1). All experiments were performed in triplicate and repeated independently three times.

Quantitative blood urea nitrogen (BUN) determination

Serum samples were diluted fivefold with distilled water prior to analysis. Aliquots (5 μL) of distilled water (blank), urea nitrogen standard (50 mg/dL), or diluted serum samples were dispensed in triplicate into a clear 96-well plate. Subsequently, 200 μL of freshly prepared working reagent was added to each well, and the plate was gently mixed. After incubation at room temperature for 20 min, absorbance was measured at 520 nm using a microplate reader.

Measurement of intracellular lactate

Intracellular l-lactate levels were determined using an l-Lactate Assay Kit (Novus Biologicals, NBP3-25875) following the manufacturer’s protocol. Briefly, cells were washed with cold PBS, harvested, and lysed in assay buffer. Cell lysates were centrifuged, and the supernatants were used for lactate measurement. The reaction was performed in a 96-well plate by incubating samples with the reaction mix, and absorbance was measured using a microplate reader. Lactate concentrations were calculated on the basis of a standard curve and normalized to total protein concentration. Results are expressed as mmol lactate per g protein.

Statistical analysis

Data are presented as mean ± SEM. Statistical analyses were performed using SPSS Statistics version 22. p values were calculated by two-tailed unpaired Student’s t-test and one-way analysis of variance (ANOVA); and p < 0.05 was considered statistically significant.

Results

The expression of AARS1 was upregulated in DKD

With the analysis of the GSE178319 dataset [19], we found that the expression of Aars1, encoding AARS1, was increased in diabetic mouse kidneys compared with normal controls (Fig. 1A), which had also been validated in the datasets of GSE134327 and GSE199929 (Supplementary Fig. 1A, B) [20, 21]. To further define the cellular distribution of AARS1, we analyzed publicly available single-cell RNA-sequencing datasets from human DKD samples (GSE131882 and GSE151302) [22]. AARS1 was broadly expressed across multiple tubular epithelial segments, with predominant enrichment in proximal tubular cells (Supplementary Fig. 1C). Consistent with these transcriptomic data, immunohistochemical analysis of human kidney specimens showed increased AARS1 expression in tubular epithelial cells from patients with DKD compared with control kidneys (Supplementary Fig. 1D). We confirmed that the expression of AARS1 was increased in kidneys of two DKD mouse models, including STZ-induced type 1 diabetic mice and db/db mice, a spontaneous model of type 2 diabetes, compared with age-matched control mice as examined by western blot analysis (Fig. 1B, C) and immunofluorescence (IF) staining (Fig. 1D, E). Furthermore, we found that the expression of AARS1 was increased in HK-2 cells treated with high glucose (HG, 30 mM) (Fig. 1F). The mRNA of AARS1 was also increased in diabetic mouse kidneys and HG-induced HK-2 cells as examined by qRT-PCR analysis (Fig. 1G). These results suggest a potential role of AARS1 in DKD pathogenesis.

Fig. 1.

AARS1 was upregulated in diabetic kidneys and high glucose-induced HK-2 cells. A Transcriptomic analysis of database (GSE178319) showing an increase of AARS1 expression in diabetic mouse kidneys compared with controls. B, C Western blot analysis of the expression of AARS1 from kidneys in STZ-induced diabetic (B) and db/db (C) mice. D, E Immunofluorescent staining of AARS1 costained with Lotus tetragonolobus lectin (LTL) in kidneys from STZ-induced diabetic (D) and db/db (E) mice, compared with that in kidneys from age-matched control mice (n = 3). Scale bar: 50 μm. F Western blot analysis of the expression of AARS1 from whole cell lysates in HK-2 cells treated with normal glucose (NG, 5 mM) versus high glucose (HG, 30 mM). Relative AARS1 expression was quantified from three independent immunoblots and standardized to actin. G qRT-PCR analysis of the relative AARS1 mRNA expression in HK-2 cells treated with normal and high glucose, as well as in the kidneys from STZ-induced diabetic and db/db mice (n = 3)

Kidney-specific Aars1 deletion attenuated DKD progression in diabetic mice

To investigate the functional role of AARS1 in DKD in vivo, we generated kidney specific Aars1 knockout mice (Aars1^fl/fl:Ksp-Cre) and induced diabetes by the treatment with STZ (Fig. 2A). We found that knockout of Aars1 decreased kidney weight to body weight ratio (KW/BW) (Fig. 2B) and BUN levels (Fig. 2C) in STZ-induced mice compared with age-matched Aars1 wild-type (WT) mice induced with STZ. Knockout of Aars1 also ameliorated renal injury in STZ-induced diabetic mouse kidneys as seen by a decrease of (1) the expression of kidney injury molecule-1 (Kim-1), a marker of tubular injury (Fig. 2D), (2) tubular dilation, atrophy and loss of brush borders as examined by PAS staining (Fig. 2E), (3) glomerular basement membrane (GBM) thickening and podocyte foot process effacement as examine with electron microscopy (EM) (Fig. 2F), and (4) tubulointerstitial fibrosis as examined by Masson’s trichrome and Sirius red staining (Fig. 2G), compared with STZ-induced Aars1 WT controls. In addition, we found that the level of lactylation was increased in kidneys of STZ-induced diabetic mice, which could be decreased in kidneys of Aars1 knockout STZ-induced diabetic mice as examined with a pan-lysine lactylation (pan-Kla) antibody (Fig. 2H). Consistently, we found that kidney-specific knockout of Aars1 in db/db mice also mitigated DKD progression and reduced renal lactylation (Supplementary Fig. 2). There results demonstrate that kidney-specific deletion of Aars1 protects against DKD progression in both STZ-induced and db/db diabetic models by improving renal function, alleviating glomerular and tubular injury, decreasing renal fibrosis, and reducing global protein lactylation.

Fig. 2.

Kidney-specific knockout of Aars1 ameliorated DKD progression in a STZ-induced diabetic mouse model. A Experimental scheme of STZ-induced DKD model. Aars1^fl/fl:Ksp-Cre and wild type (WT) mice were age-matched, and persistent hyperglycemia was induced using STZ for 18 weeks. B, C KW/BW ratios (B) and BUN levels (C) were significantly decreased in STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice (n = 5 mice per group). D Immunofluorescent staining indicated that Kim-1 expression was decreased in kidneys from STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice. Scale bar: 50 μm (n = 5 mice per group; six independent fields per section were examined). E. Representative histological images of PAS staining in kidney sections from STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice. The bar graph indicated the tubular injury score. Scale bar: 50 μm (n = 5 mice per group; six independent fields per section were examined). F Representative electron microscopy (EM) images of podocytes from STZ-induced Aars1^fl/fl:Ksp-Cre mice and age-matched WT mice. Scale bar: 600 nm. G, H Representative images and quantification of Masson’s trichrome and Sirius red staining (G) and immunostaining with pan-Kla antibody (H) in kidney sections from STZ-induced Aars1^fl/fl:Ksp-Cre mice and age-matched WT mice. Scale bar: 50 μm

AARS1 regulated metabolic and fibrotic gene programs in DKD

Recent studies have shown that AARS1 can sense intracellular lactate and be translocated into the nucleus to modulate gene expression in cancer [23]. To investigate how AARS1 promotes DKD progression, we identified AARS1 target genes in HK-2 cells under HG or NG condition with CUT and Tag analysis (Fig. 3A). We found that the AARS1-binding peaks were predominantly enriched in promoter regions, and HG treatment further increased its promoter occupancy (74.15% versus 61.22%; Fig. 3B). These peaks displayed strong enrichment around transcription start sites (TSS) in both HK-2 cells treated with HG and NG, with the HG group showing a modest increase in signal intensity (Fig. 3C). We further identified 1915 specific AARS1-binding peaks in HG-induced HK-2 cells (Fig. 3D).

Fig. 3.

The identification of AARS1 target genes by CUT and Tag analysis in high glucose-treated HK-2 cells. A Experimental scheme of CUT and Tag analysis. B Genome-wide distribution of AARS1-binding peaks in HK-2 cells treated with normal or high glucose. C Metagene analysis of AARS1 peaks around TSS (± 3 kb) showed strong enrichment, with slightly higher signals in HG-treated than in NG-treated HK-2 cells. D Venn diagram showing the overlap of AARS1-occupied peaks in HK-2 cells treated with normal glucose or high glucose. E Flow diagram for identifying AARS1 target genes. HG-specific AARS1 peaks were linked to phosphatase, glycolytic, and TGF-β pathways, and Nephroseq data showed reduced PPP6C and increased HK2, PFKP, and ZEB1 expression in the tubulointerstitium of patients with DKD

Given the established roles of glycolysis, fibrosis, and the phosphorylation of key components of specific signaling pathways in DKD pathogenesis [24], we focused on AARS1 target genes that are involved in the regulation of these processes (Fig. 3E), which included 3 glycolysis-related genes, and 7 transforming growth factor-beta (TGF-β) signaling-associated genes, and 35 phosphatase-related genes (Supplementary Table S3). By comparing AARS1 target genes with transcriptomic data from the Nephroseq database (http://www.nephroseq.org), we found that the AARS1 target genes associated with glycolysis and fibrosis, including HK2, PFKP, and ZEB1, were transcriptionally upregulated (Supplementary Fig. 3A–C), whereas the genes of phosphatase, such as PPP6C, were downregulated in DKD kidneys (Supplementary Fig. 3D).

AARS1 transcriptionally regulated the expression of glycolysis associated genes in DKD

The CUT and Tag profiles revealed prominent AARS1 binding peaks at the promoters of HK2 and PFKP, two glycolysis associated genes (Fig. 4A, B). To determine whether AARS1 directly regulates their transcription, first, we confirmed the binding of AARS1 on the promoters of HK2 and PFKP in HK-2 cells by ChIP assay (Fig. 4C). To assess functional consequence of this binding, we performed luciferase reporter assays. The proximal promoter regions (−2000 to + 200 bp relative to the TSS) of HK2 and PFKP were cloned into the pGL3-Basic vector and cotransfected with HA-AARS1 or empty vector into HEK293T cells. Overexpression of AARS1 significantly increased the promoter activity of HK2 and PFKP (Fig. 4D). Consistently, both mRNA and protein levels of HK2 and PFKP were increased in HG-induced HK-2 cells compared with those treated with normal glucose (Fig. 4E–G). This HG-induced upregulation was markedly attenuated by sgRNA-mediated knockout of AARS1. In addition, knockout of Aars1 also decreased the expression of HK2 and PFKP proteins in kidneys from STZ-induced diabetic mice compared with STZ-induced WT mice (Fig. 4H), and consistently reduced HK2 and PFKP levels in kidneys from db/db mice (Supplementary Fig. 4A).

Fig. 4.

AARS1 transcriptionally regulated the expression of glycolysis associated genes in HG-treated HK-2 cells and diabetic kidneys. A, B The genome browser view of AARS1 at HK2 (A) and PFKP (B) locus. The y-axis represents CUT and Tag intensity. C The binding of AARS1 on the promoters of HK2 and PFKP as analyzed by ChIP assay. D Luciferase reporter assays of HK2 and PFKP promoter activity in HK-2 cells with or without AARS1 overexpression. E qRT-PCR analysis of the relative HK2 and PFKP mRNA expression in HK-2 cells treated with normal or high glucose. F Knockout of AARS1 with sgRNA decreased the expression of HK2 and PFKP mRNAs in HG-induced HK-2 cells. G Knockout of AARS1 with sgRNA decreased the expression of HK2 and PFKP proteins in HK-2 cells treated with high glucose. H Western blot analysis of the expression of HK2 and PFKP proteins in kidneys from STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice

AARS1 transcriptionally activated ZEB1 to promote EMT and fibrotic remodeling in DKD

EMT is a critical driver of tubulointerstitial fibrosis in DKD [25]. CUT and Tag analysis indicated that AARS1 bound to the promoter of ZEB1 (Supplementary Fig. 5A), a master regulator of EMT [26]. We confirmed the binding of AARS1 on the promoter of ZEB1 by ChIP assay in HK-2 cells (Supplementary Fig. 5B). Luciferase reporter assays demonstrated that AARS1 overexpression increased ZEB1 promoter activity (Supplementary Fig. 5C). Consistently, ZEB1 expression was upregulated in HG-induced HK-2 cells, whereas sgRNA-mediated knockout of AARS1 attenuated this induction (Supplementary Fig. 5D, E), accompanied by a decrease of fibrotic markers, fibronectin (FN) and α-smooth muscle actin (α-SMA), and an increase of E-cadherin (E-cad) in HG-treated HK-2 cells (Supplementary Fig. 5F). In addition, knockout of Aars1 decreased the expression of ZEB1, FN, and α-SMA, but increased the expression of E-cad in kidneys from STZ-induced diabetic mice (Supplementary Fig. 5G), and similarly altered the expression of these markers in kidneys from db/db mice (Supplementary Fig. 5H). These results suggest that AARS1 may promote EMT and fibrotic remodeling in DKD through targeting ZEB1.

AARS1 repressed the transcription of PPP6C to activate Smad3 signaling in DKD

The CUT and Tag analysis also revealed a prominent AARS1 binding peak at the promoter of PPP6C (Fig. 5A), the catalytic subunit of protein phosphatase 6. PPP6C is known to negatively regulate multiple signaling pathways, including ERK and cGAS [27, 28], both of which are involved in inflammation and fibrosis [29]. We confirmed the binding of AARS1 at the promoter of PPP6C by ChIP assay (Fig. 5B). Luciferase reporter assays further revealed that AARS1 overexpression suppressed PPP6C promoter activity (Fig. 5C). Consistent with this, we observed that mRNA level of PPP6C was reduced in HG-induced HK-2 cells (Fig. 5D). Conversely, knockout of AARS1 increased the expression of PPP6C (Fig. 5E), accompanied by a decrease of the phosphorylation of Smad3 in HG-treated HK-2 cells (Fig. 5F) and in kidneys from STZ-induced and db/db diabetic mice (Fig. 5G, H) compared with corresponding controls. Consistently, knockdown of PPP6C in HK-2 cells increased the phosphorylation of Smad3 (Fig. 5I). Furthermore, we found that PPP6C interacted with Smad3 in HK-2 cells as examined by immunoprecipitation assays (Fig. 5J), supporting that PPP6C as a phosphatase is responsible for the phosphorylation of Smad3 in DKD kidneys. These results suggest that AARS1 may regulate the phosphorylation and activation of DKD associated signaling pathways through the downregulation of the transcription of PPP6C in DKD.

Fig. 5.

AARS1 repressed the transcription of PPP6C to activate Smad3 signaling in HG-induced HK-2 cells and diabetic kidneys. A The genome browser view of AARS1 at PPP6C locus. The y-axis represents CUT and Tag intensity. B The binding of AARS1 on the promoter of PPP6C as analyzed by ChIP assay. C Luciferase reporter assays of PPP6C promoter activity in HK-2 cells with or without AARS1 overexpression. D qRT-PCR analysis of the relative PPP6C mRNA expression in HK-2 cells treated with normal glucose or high glucose. E Knockout of AARS1 with sgRNA increased the expression of PPP6C mRNA in HG-induced HK-2 cells. F Knockout of AARS1 with sgRNA increased the expression of PPP6C protein but decreased the phosphorylation of Smad3 (p-Smad3) in HG-treated HK-2 cells. G Western blot analysis of the expression of PPP6C and p-Smad3 in kidneys from STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice. H Kidney-specific knockout of Aars1 increased the expression of PPP6C protein but decreased the p-Smad3 in kidneys from db/db:Aars1^fl/fl:Ksp-Cre mice versus db/db:Aars1^+/+:Ksp-Cre mice. I Knockdown of PPP6C with siRNA increased the phosphorylation of Smad3 in HK-2 cells. J The interaction of PPP6C with Smad3 as analyzed by Co-IP in HK-2 cells

AARS1 interacted with and lactylated Akt and NF-κB p65 in HG-treated tubular cells

In addition, we found that the phosphorylation of Akt and the p65 subunit of NF-κB was also increased in HG-treated HK-2 cells (Fig. 6A), whereas sgRNA-mediated knockout of AARS1 attenuated this effect (Fig. 6B). Overexpression of HA-tagged AARS1 also increased the phosphorylation of Akt and p65 in HK-2 cells (Fig. 6C). Consistent with these in vitro findings, knockout of Aars1 decreased the phosphorylation of Akt and p65 in kidneys from Aars1-deficient STZ-induced diabetic mice compared with Aars1 WT STZ-induced diabetic mice (Fig. 6D), with similar reductions in db/db mice (Fig. 6E). Notably, the interaction between AARS1 and AKT1 was increased in HG‑treated cells, in parallel with the upregulation of both proteins (Fig. 6F). We also found that the lysine lactylation of Akt and p65 was increased in HG-treated HK-2 cells (Fig. 6G), and this modification was markedly diminished uponAARS1 knockout with sgRNA (Fig. 6H). These results suggest that Akt and p65 are novel substrates of AARS1.

Fig. 6.

AARS1 interacted with and lactylated Akt and NF-κB p65 in HG-treated tubular cells. A Western blot analysis of the phosphorylation of Akt and p65 from whole cell lysates of HK-2 cells treated with normal and high glucose. B Knockout of AARS1 with sgRNA decreased the phosphorylation of Akt and p65 in HG-treated HK-2 cells. C Overexpression of AARS1 increased the expression of the phosphorylation of Akt and p65 in HK-2 cells. D Western blot analysis of the phosphorylation of Akt and p65 in kidneys from STZ-induced Aars1^fl/fl:Ksp-Cre mice versus age-matched WT mice. E Kidney-specific knockout of Aars1 decreased the phosphorylation of Akt and p65 in kidneys from db/db:Aars1^fl/fl:Ksp-Cre mice versus db/db:Aars1^+/+:Ksp-Cre mice. F Interactions between AARS1 and Akt, as well as AARS1 and p65, in HK-2 cells treated with normal or high glucose were detected with anti-AARS1 antibody and then blotted with Akt and p65 antibody, respectively. IgG was used as a negative control. G Lactylation of Akt and p65 was increased in HG-treated HK-2 cells compared with NG-induced HK-2 cells. These cells were immunoprecipitated with anti-Akt (top panel) and anti-p65 antibody (bottom panel), and then blotted with Akt, p65, and pan–lactyl lysine antibody, respectively. IgG was used as a negative control. H Knockout of AARS1 with sgRNA decreased the level of lactylated Akt and p65 in HK-2 cells treated with high glucose. I Western blot analysis of the lactylation and phosphorylation of Akt in AARS1-knockout HK-2 cells cotransfected with HA-AARS1 and Flag-tagged WT or K → R mutant Akt1. J Western blot analysis of the lactylation and phosphorylation p65 in AARS1-knockout HK-2 cells cotransfected with HA-AARS1 and T7-tagged WT or K → R mutant p65 (RelA)

To identify the AARS1-mediated lactylation sites on Akt and p65, we expressed Flag-tagged WT Akt1 and its lysine-to-arginine mutants (K14R and K179R), as well as T7-tagged WT p65 and its mutants (K218R, K221R, and K310R), in AARS1-knockout HK-2 cells together with HA-tagged AARS1. We focused on the Akt1 isoform because Akt1 is a major Akt isoform expressed in tubular epithelial cells and its lysine residues are highly conserved across Akt isoforms [30]. Lactylation of Flag-Akt1 was assessed using a pan-lactyl lysine antibody following anti-Flag immunoprecipitation, and we found that lactylation was reduced in the Akt1 K14R mutant but not in K179R (Fig. 6I). Similarly, lactylation of T7-p65 was assessed following anti-T7 immunoprecipitation, and lactylation was decreased in the p65 K310R mutant but not in the K218R or K221R mutants (Fig. 6J). We further found that AARS1 overexpression enhanced phosphorylation of wild-type Akt1 and the Akt1 K179R mutant, but had minimal effects on phosphorylation of the Akt1 K14R mutant (Fig. 6I). Similarly, AARS1 overexpression increased phosphorylation of WT p65 as well as the K218R and K221R mutants, whereas phosphorylation of the K310R mutant was largely unaffected (Fig. 6J). These results suggest that lysine 14 of Akt1 and lysine 310 of p65 are critical sites for AARS1-mediated lactylation and the lactylation of these sites are required for their phosphorylation and activation.

AARS1-mediated lactylation of Akt and p65 promoted autophagy impairment and inflammatory remodeling in DKD

Because impaired tubular autophagy contributes to DKD progression and Akt signaling is a critical regulator of autophagy [31], we next investigated whether AARS1-mediated Akt lactylation and activation affects autophagy in DKD. First, we found that the level of LC3B was decreased, whereas the level of p62, a key substrate of autophagy, was increased in HG-induced HK-2 cells, and these changes were reversed when AARS1 was knocked out in HK-2 cells (Fig. 7A). Consistently, we also observed decreased LC3B and increased p62 levels in kidneys from STZ-induced and db/db diabetic mice, and these alterations were reversed in Aars1-knockout mice (Fig. 7B, C). Functionally, AARS1 overexpression decreased LC3B and increased p62 in HK-2 cells, whereas inhibition of Akt with MK-2206 abolished these effects (Fig. 7D), indicating that AARS1 impairs autophagy through Akt activation. Furthermore, EM revealed fewer autophagic vesicles in renal tubular cells of STZ-induced and db/db diabetic kidneys, whereas Aars1 deletion markedly increased the number of autophagic vesicles in both models (Fig. 7E, F). These results suggest that AARS1 may impair autophagy in diabetic tubular cells by promoting Akt lactylation and activation.

Fig. 7.

AARS1-mediated the lactylation of Akt and p65 promoted autophagy impairment and macrophage recruitment in DKD. A High glucose increased the expression of p62 but decreased the expression of LC3B, while knockout of AARS1 with sgRNA counteracted these changes in HK-2 cells. B Western blot analysis of the expression of p62 and LC3B proteins in kidneys from STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice. C Kidney-specific knockout of Aars1 increased the level of LC3B protein but decreased the level of p62 protein in kidneys from db/db:Aars1^fl/fl:Ksp-Cre mice versus db/db:Aars1^+/+:Ksp-Cre mice. D Western blot analysis showed that AARS1 overexpression decreased the level of LC3B and increased the level of p62 in HK-2 cells, whereas inhibition of Akt with MK-2206 abolished those effects. E Representative EM images showed that the number of autophagic vacuoles was decreased in STZ-induced WT diabetic mouse kidneys, while Aars1 knockout increased the number of autophagic vacuoles. Asterisks indicate autophagic vacuoles. Scale bar: 1 μm. F Kidney-specific knockout of Aars1 increased the number of autophagic vacuoles in the kidneys of db/db mice. G Immunofluorescent staining indicated that macrophage recruitment was decreased in kidneys from STZ-induced Aars1^fl/fl:Ksp-Cre mice compared with age-matched WT mice. Scale bar: 50 μm. H Immunofluorescent staining of F4/80 in kidneys from db/db:Aars1^fl/fl:Ksp-Cre mice compared with db/db:Aars1^+/+:Ksp-Cre mice. Scale bar: 50 μm (n = 5 mice per group; six independent fields per section were examined)

To determine whether AARS1-mediated p65 lactylation and activation regulates inflammatory responses in DKD, first, we found that the levels of NF-κB target genes, including IL-6 (interleukin-6), MCP-1 (monocyte chemoattractant protein-1), and TNF-α (tumor necrosis factor alpha), all of which have been reported to be involved in the pathogenesis of DKD [32], were upregulated in HG-induced HK-2 cells and STZ-induced mouse kidneys compared with proper controls (Supplementary Fig. 6A, B). We further found that AARS1 knockout decreased the mRNA levels of these genes in HG-treated HK-2 cells and in kidneys from STZ-induced diabetic mice (Supplementary Fig. 6C, D). In addition, Aars1 WT diabetic mice exhibited increased macrophage accumulation by F4/80 staining, which was reduced in Aars1-knockout STZ mice (Fig. 7G). Consistent with this, Aars1 knockout also reduced macrophage infiltration in db/db mouse kidneys (Fig. 7H). These results suggest that AARS1-mediated p65 lactylation and activation may enhance NF-κB-dependent inflammatory gene expression and contribute to the accumulation of macrophages in diabetic kidneys.

AARS1 and NF-κB formed a lactate-dependent positive feedback loop in DKD

Given that glycolysis is a major metabolic feature in DKD kidneys and lactate is the principal end product of glycolysis [33], we investigated whether glycolytic flux regulates AARS1 expression. We found that treatment with lactate increased the expression of AARS1 in HK-2 cells in a dose-dependent manner (Fig. 8A), whereas inhibition of glycolysis with 2-DG decreased the expression of AARS1 in HK-2 cells under high glucose conditions (Fig. 8B), supporting the existence of a positive feedback loop between AARS1 and glycolysis in DKD. In addition, we found that treatment with lactate increased the lactylation and phosphorylation of the p65 subunit of NF-κB in HK-2 cells (Fig. 8C), while treatment with 2-DG suppressed these modifications in HG-treated HK-2 cells (Fig. 8D). Furthermore, inhibition of NF-κB with BAY-11-7085 decreased AARS1 expression in HG-treated HK-2 cells in a dose-dependent manner (Fig. 8E). To determine whether NF-κB regulates AARS1 transcription, we performed ChIP assays and found that p65 directly bound to the promoter of AARS1 in HK-2 cells (Fig. 8F). Time-course experiments following acute lactate stimulation revealed a rapid and time-dependent increase in p65 lactylation, detectable as early as 1 h, followed by a progressive increase in p65 phosphorylation (Fig. 8G). Notably, AARS1 protein expression increased at later time points, consistent with a downstream transcriptional response. In line with this, ChIP-qPCR analysis showed a time-dependent increase in p65 occupancy at the promoter of AARS1 upon lactate stimulation (Fig. 8H), indicating enhanced NF-κB recruitment to the AARS1 promoter. These results delineate a temporal sequence in which lactate-induced p65 lactylation precedes NF-κB activation, increased promoter binding, and subsequent upregulation of AARS1 expression, supporting a lactate-driven feedback loop between AARS1 and NF-κB in DKD.

Fig. 8.

A positive feedback loop existed between AARS1 and NF-κB in high glucose-induced HK-2 cells. A Stimulation with lactate induced AARS1 expression in a dose-dependent manner in HK-2 cells. B Western blot analysis of AARS1 expression in HK-2 cells treated with the glycolysis inhibitor 2-DG under high-glucose conditions. C Stimulation with lactate increased the lactylation and phosphorylation of the p65 subunit of NF-κB in HK-2 cells. D Treatment with 2-DG decreased the lactylation and phosphorylation of the p65 subunit of NF-kB in HG-induced HK-2 cells. E Western blot analysis of the expression of AARS1 protein in the presence of the NF-κB inhibitor BAY-11-7085 in HG-treated HK-2 cells. F The binding of p65 on the promoter of AARS1 as analyzed by ChIP assay. G Time-course analysis of p65 lactylation, NF-κB activation, and AARS1 expression in HK-2 cells treated with lactate (20 mM). p65 was immunoprecipitated followed by immunoblotting with pan-Kla antibody. Input lysates were analyzed for AARS1, p-p65, and total p65. H ChIP–qPCR analysis showing time-dependent enrichment of p65 at the promoter of AARS1 in HK-2 cells following lactate treatment (20 mM)

Glycolysis-derived lactate promoted AARS1 binding to target gene promoters

To investigate whether high glucose-induced metabolic changes influence AARS1 association with its target promoters, we first measured intracellular lactate levels in HK-2 cells. High glucose significantly increased lactate production, which was effectively suppressed by 2-DG (Supplementary Fig. 7A). We next performed ChIP assays under normal glucose (NG), high glucose (HG), and HG+2-DG conditions. We found that HG markedly increased AARS1 enrichment at the promoters of HK2, PFKP, ZEB1, and PPP6C, whereas this effect was substantially attenuated by 2-DG (Supplementary Fig. 7B). These results indicate that glycolysis-derived lactate facilitates AARS1 recruitment to its target gene promoters under diabetic conditions.

β-alanine competitively interfered with AARS1-mediated lactylation to ameliorate the progression of DKD

To further evaluate the therapeutic relevance of AARS1 in DKD, we employed β-alanine, a structural analog of lactate that competitively interferes with AARS1 substrate utilization and suppresses lysine lactylation [14]. Although not a classical enzyme-specific inhibitor, β-alanine is widely used as a metabolic competitor to phenocopy AARS1 depletion in vitro and in vivo [34]. Consistent with this mechanism, we found that treatment with β-alanine suppressed HG-induced lactylation and phosphorylation of Akt and NF-κB subunit p65 in HK-2 cells (Supplementary Fig. 8A). In addition, β-alanine decreased the mRNA levels of HK2, PFKP, and ZEB1, while increasing PPP6C expression in HG-treated HK-2 cells (Supplementary Fig. 8B–E). Furthermore, ChIP assay revealed that β‑alanine treatment significantly decreased AARS1 enrichment at the promoters of HK2, PFKP, ZEB1, and PPP6C under HG conditions (Supplementary Fig. 8F).

To determine whether these effects are primarily mediated through the noncanonical lactyltransferase activity of AARS1, we treated HG‑induced HK‑2 cells with L‑alanine, which has been proposed to selectively compete with lactate for AARS1 binding. L‑alanine (5 mM, 24 h) significantly reduced lactylation and phosphorylation of Akt and NF‑κB subunit p65 in HK‑2 cells (Supplementary Fig. 9A). Similarly, l‑alanine decreased the mRNA levels of HK2, PFKP, and ZEB1, while increasing PPP6C mRNA expression in HG‑treated HK‑2 cells (Supplementary Fig. 9B–E). Moreover, ChIP assay demonstrated that l‑alanine also attenuated HG‑induced AARS1 binding to the promoters of these target genes (Supplementary Fig. 9F). These results support that the protective effects of β‑alanine are primarily mediated through inhibition of the lactyltransferase activity of AARS1.

In addition, we found that treatment with β-alanine decreased, (1) KW/BW ratios and BUN levels in STZ-induced diabetic mice (Fig. 9B, C), (2) the expression of Kim-1 (Fig. 9D), (3) glomerular mesangial expansion, basement membrane thickening and tubular injury as examined by PAS staining and EM (Fig. 9E, F), (3) interstitial fibrosis as examined by Masson’s trichrome and Sirius red staining (Fig. 9G), (4) level of renal lactylation as examined by Pan-Kla staining (Fig. 9H), and (5) the accumulation of macrophages in kidneys of STZ-induced diabetic mice (Fig. 9I). In addition, treatment with β-alanine decreased the protein level of Kim-1, as well as increased LC3 and decreased p62 levels, as examined by western blot analysis (Supplementary Fig. 10). Mechanistically, treatment with β-alanine normalized AARS1-regulated protein expression, decreasing HK2, PFKP, and ZEB1 while increasing PPP6C, and also reduced the phosphorylation of Akt, p65, and Smad3 in kidneys from STZ-induced diabetic mice compared with vehicle-treated controls (Supplementary Fig. 10). Similar protective effects of β-alanine were observed in db/db mice (Supplementary Figs. 11 and 12). These results suggest that treatment with β-alanine phenocopies the effects of Aars1 knockout in DKD kidneys, supporting that inhibition of AARS1 either alone or in combination with inhibition of its downstream targets/substrates are potential therapeutic strategies for DKD treatment.

Fig. 9.

Treatment with β-alanine ameliorated the progression of STZ-induced DKD. A Experimental scheme of β-alanine administration in STZ-induced diabetic mice. B, C KW/BW ratios (B) and BUN levels (C) were significantly decreased in STZ-induced diabetic mice treated with β-alanine compared with vehicle controls (n = 5 mice per group). D Treatment with β-alanine alleviated kidney injury in STZ-induced diabetic mouse kidneys, as detected with Kim-1 staining. Scale bars: 50 μm. E Representative histological images of PAS in kidney sections from STZ-induced mice treated with β-alanine or vehicle controls. The bar graph indicated the tubular injury score. Scale bar: 50 μm (n = 5 mice per group; six independent fields per section were examined). F Representative EM images of podocytes from STZ-induced mice treated with β-alanine or vehicle controls. Scale bar: 600 nm. G, H Representative images and quantification of Masson’s trichrome and Sirius red staining (G) and immunostaining with Pan-Kla antibody (H) in kidney sections from STZ-induced mice treated with β-alanine or vehicle controls. Scale bar: 50 μm. I Immunofluorescent staining indicated that macrophage recruitment was decreased in kidneys from STZ-induced mice treated with β-alanine compared with vehicle controls. Scale bar: 50 μm

Discussion

Lactylation, a post-translational modification in which lactate covalently modifies proteins, has emerged as an important mechanism linking cellular metabolic states to gene regulation and disease progression [35]. However, the role of lactylation—and the enzyme(s) regulating this process—in DKD remains poorly understood. Here, we demonstrate the unknown role and mechanisms of the lactyltransferase AARS1 in DKD progression. Here, we identify AARS1 as a lactate-responsive lactyltransferase that coordinates metabolic reprogramming, autophagy impairment, and inflammatory remodeling in DKD. Mechanistically, AARS1 integrates signaling and transcriptional regulation by directly lactylating Akt and the NF-κB subunit p65, thereby enhancing their phosphorylation and activation, while simultaneously controlling the expression of genes involved in glycolysis and fibrotic remodeling, including HK2, PFKP, ZEB1, and PPP6C. Through these coordinated actions, AARS1 suppresses autophagy, sustains inflammatory signaling, and promotes glycolytic reprogramming and profibrotic pathways—such as epithelial dedifferentiation and Smad3 activation—without inducing overt tubular cell death. Together, these findings indicate that AARS1 functions as a metabolic–epigenetic amplifier that couples high glucose-driven glycolysis and lactate accumulation to persistent nonlethal tubular stress, maladaptive repair, and progressive fibrosis during DKD progression (Fig. 10).

Fig. 10.

Working model of AARS1 in regulation of DKD progression. A schematic diagram depicting lactate-responsive AARS1-mediated pathways and processes in DKD. Diabetes and high glucose induce the upregulation of AARS1 in tubular epithelial cells, which is further enhanced by lactate accumulation resulting from glycolytic activation. Upregulated AARS1 in diabetic tubular cells: i) lactylates Akt, leading to its phosphorylation and activation, which impairs autophagy and contributes to autophagic vacuole depletion; ii) lactylates the p65 subunit of NF-κB, resulting in its phosphorylation and activation, which promotes inflammatory cytokine expression and macrophage accumulation; iii) regulates the transcription of novel AARS1 target genes involved in glycolysis, including HK2 and PFKP, two rate-limiting enzymes, thereby reinforcing glycolytic reprogramming and lactate production; iv) modulates fibrosis-related transcriptional programs, including the upregulation of ZEB1 and the repression of the phosphatase PPP6C, which enhances Smad3 phosphorylation and drives EMT and fibrosis. In addition, a metabolic-inflammatory positive feedback loop is formed: high glucose-driven glycolysis increases lactate production, lactate enhances NF-κB lactylation and activation, and NF-κB activation further promotes AARS1 expression, together amplifying AARS1-mediated signaling in diabetic kidneys. Treatment with β-alanine, a structural analog of lactate that competitively inhibits AARS1-mediated lactylation, attenuates DKD progression by restoring autophagy and suppressing glycolysis, inflammation, and fibrosis in diabetic mouse kidneys

Autophagy is a critical cellular process for removing damaged components, particularly in renal tubular epithelial cells and podocytes [36]. In DKD, sustained hyperglycemia is known to suppress autophagy, resulting in mitochondrial dysfunction, oxidative stress, and progressive tubular injury [37]. Hyperactivation of the PI3K/Akt/mTOR axis represents a central mechanism underlying autophagy inhibition in diabetic kidneys [31, 38]. Here, our findings uncover a dual-layered mechanism through which AARS1 enforces autophagy suppression under diabetic conditions. At the signaling level, AARS1 directly lactylates Akt at lysine 14 of the Akt1 isoform, enhancing Akt phosphorylation and downstream mTOR activation, thereby acutely inhibiting autophagy initiation. In parallel, CUT and Tag analysis reveals that AARS1 directly regulates a set of autophagy- and lysosome-associated genes, including ATG4A (autophagy-related 4A cysteine peptidase), PRKAA1 (protein kinase AMP-activated catalytic subunit alpha 1), LAMP1 (lysosomal-associated membrane protein 1), CTSD (cathepsin D), and CTSL (cathepsin L), which collectively govern autophagosome formation, lysosomal maturation, and autophagic flux (Supplementary Table S4) [39]. Together, these complementary signaling and transcriptional actions indicate that AARS1 not only rapidly suppresses autophagy but also chronically stabilizes autophagy impairment through transcriptional reprogramming, thereby locking tubular epithelial cells into a persistent nonlethal stress state that favors maladaptive repair and fibrotic remodeling.

Chronic inflammation represents another defining pathological feature of DKD and a major driver of tubulointerstitial fibrosis [40]. High glucose and oxidative stress activate innate immune pathways and induce inflammatory mediators such as tumor necrosis factor (TNF)-α and interleukin (IL)-1β, promoting recruitment and activation of macrophages that further drive fibrotic remodeling [41]. Activation of NF-κB is known to enhance transcription of its canonical target genes, including IL-6, MCP-1, and TNF-α, which promote tubulointerstitial inflammation and immune cell recruitment, thereby exacerbating chronic inflammatory injury in diabetic kidneys [32]. In this study, we identify NF-κB p65 as a novel nonhistone substrate of AARS1. AARS1-mediated lactylation of p65 enhances its phosphorylation and activation, leading to increased expression of IL-6, MCP-1, and TNF-α, as well as enhanced macrophage infiltration in diabetic kidneys. Importantly, p65 directly binds to the AARS1 promoter, establishing a positive feedback loop in which NF-κB activation upregulates AARS1 expression. Lactate exposure further enhances AARS1 expression and p65 lactylation, indicating that metabolic accumulation of lactate reinforces this AARS1-NF-κB inflammatory circuit in diabetic kidneys. Although our data establish a glycolysis–lactate–NF-κB–AARS1 positive feedback loop, the upstream signals initiating AARS1 upregulation under diabetic conditions remain to be fully defined. Our findings suggest that AARS1 does not initiate metabolic dysfunction but instead functions as a metabolic–epigenetic amplifier that sustains and exacerbates preexisting metabolic and inflammatory stress programs in DKD.

In addition to modulating Akt and NF-κB signaling through lactylation, AARS1 also functions as a transcriptional regulator of genes governing glycolysis, tubular stress response, and fibrosis. Using CUT and Tag and ChIP analyses, we found that AARS1 directly binds to the promoters of HK2, PFKP, ZEB1, and PPP6C in HK-2 cells. Although AARS1 lacks a canonical DNA-binding domain, coimmunoprecipitation assays revealed an interaction with NF‑κB p65, suggesting indirect recruitment to chromatin. Consistent with this model, bioinformatics analysis identified putative NF‑κB binding sites (κB motifs) within the proximal promoters of all four genes (Supplementary Table S5). Luciferase reporter assays further revealed that AARS1‑mediated activation of HK2, PFKP and ZEB1 promoters, as well as repression of the PPP6C promoter, were largely abolished by the NF‑κB inhibitor BAY‑11‑7085 (Supplementary Fig. 13A–D). Furthermore, ChIP assay demonstrated that lactate derived from high glucose is required for AARS1 occupancy at the promoters of its target genes (HK2, PFKP, ZEB1, and PPP6C), as inhibition of glycolysis or competition with β‑alanine or l‑alanine markedly attenuated AARS1 enrichment. Together, these data support a model in which AARS1 is recruited to promoters through interaction with p65, and this recruitment is enhanced by lactate‑driven metabolic cues. Functionally, AARS1 exerts coordinated transcriptional control over multiple pathogenic pathways in DKD. First, AARS1 positively regulated the transcription of HK2 and PFKP, two rate-limiting glycolytic enzymes [42], thereby enhancing glycolysis and lactate production under high glucose conditions. The resulting accumulation of lactate further increased AARS1 expression, establishing a metabolic positive feedback loop that amplifies tubular metabolic stress. Such glycolytic reprogramming has been linked to mitochondrial dysfunction, reactive oxygen species generation, maladaptive repair, and tubulointerstitial fibrosis in DKD [43]. Second, AARS1 promoted the transcription of ZEB1, a master regulator of EMT [26], thereby contributing to tubular dedifferentiation and fibrotic remodeling. Consistent with this role, analysis of public single-cell transcriptomic datasets from ischemia–reperfusion injury and unilateral ureteral obstruction models demonstrates that AARS1 expression is rapidly induced during acute tubular injury and remains elevated during maladaptive repair, supporting a role for AARS1 in the transition from injury to chronic fibrosis (Supplementary Fig. 14) [44]. Third, AARS1 negatively regulates the transcription of PPP6C, which encodes a catalytic subunit of the PP6 phosphatase complex [45]. Importantly, we demonstrated that downregulation of PPP6C under diabetic conditions reduces the availability of PPP6C to interact with and dephosphorylate Smad3. Consequently, AARS1-mediated suppression of PPP6C enhances Smad3 phosphorylation and activation, thereby promoting Smad3-driven fibrotic responses in DKD. Together, the selective upregulation of HK2, PFKP, and ZEB1, combined with repression of PPP6C, reveals a gene-specific transcriptional program orchestrated by AARS1 that drives glycolytic reprogramming and fibrotic signaling in diabetic kidneys.

Recent studies suggest that AARS1 functions as a noncanonical lactyltransferase that directly utilizes free lactate and ATP to catalyze protein lactylation without requiring high-energy acyl donors such as lactyl-CoA, which is rarely detected in mammalian cells [23, 46]. Consistent with this, we found that AARS1 overexpression markedly increased lactylation of Akt and p65, whereas AARS1 knockout reduced these modifications. However, residual lactylation persisted in AARS1‑knockout cells, suggesting that additional enzymes or nonenzymatic mechanisms may contribute to lysine lactylation. Indeed, acetyltransferases such as p300 (E1A binding protein p300), GCN5 (General Control Non-repressed protein 5), and HBO1 (histone acetyltransferase binding to ORC1) have been reported to exhibit lactyltransferase activity in the presence of lactyl‑CoA [9, 47], and elevated intracellular lactate may also promote nonenzymatic lysine lactylation. Nevertheless, the pronounced reduction observed upon AARS1 deletion indicates that AARS1 is a major lactyltransferase for Akt and p65 under diabetic conditions. In addition, β-alanine effectively inhibits AARS1-mediated lactylation in high glucose-treated tubular epithelial cells. This is consistent with previous reports demonstrating that β-alanine competitively binds the catalytic site of AARS1 and displaces lactate [14, 23]. We acknowledge that β-alanine may also affect the canonical aminoacyl-tRNA synthetase activity of AARS1. To address this, we performed complementary experiments using l-alanine, which has been proposed to compete with lactate for binding to AARS1 and may preferentially affect its noncanonical lactyltransferase activity. l-alanine treatment recapitulated the effects of β-alanine, reducing p65 lactylation, target gene expression, and AARS1 promoter occupancy. These data support the interpretation that the protective effects of β-alanine are largely mediated through inhibition of AARS1-associated lactylation signaling, although contributions from its canonical function cannot be completely excluded. We acknowledge that β-alanine is not a highly specific inhibitor of AARS1 and may exert additional metabolic effects. Importantly, the therapeutic relevance of AARS1 inhibition in DKD is primarily supported by genetic deletion of Aars1, with β-alanine serving as a pharmacological proof-of-concept for targeting AARS1-mediated lactylation. These findings provide a strong rationale for the future development of more selective and potent AARS1 inhibitors for DKD therapy.

Several limitations of this study should be acknowledged. Although immunofluorescence analyses indicate that AARS1 expression is enriched in Lotus tetragonolobus lectin (LTL)-positive proximal tubules, detectable AARS1 signals are also observed in non-LTL-positive tubular segments, indicating that AARS1 is not exclusively restricted to proximal tubules. This broader expression pattern provides a biological basis for the protective effects observed following Ksp-Cre-mediated Aars1 deletion, despite the enrichment of Ksp-Cre activity in distal tubular segments. Moreover, although we provide evidence that p65 contributes to AARS1 recruitment to chromatin, the precise molecular determinants of promoter specificity remain incompletely defined. In particular, whether additional transcription factors, coregulators, or chromatin-associated proteins participate in directing AARS1 to specific genomic loci requires further investigation. Future studies using approaches such as ChIP‑reChIP, proximity labeling, or DNA pull‑down mass spectrometry will be valuable for defining the AARS1-associated chromatin complex. Finally, we cannot completely exclude a contribution of the canonical aminoacylation function of AARS1 to DKD pathogenesis. However, under diabetic conditions characterized by enhanced glycolytic flux and elevated lactate levels, the noncanonical lactyltransferase activity of AARS1 appears to be the predominant driver of the observed phenotypes, as supported by genetic deletion and l‑alanine-based experiments.

Conclusions

This study establishes AARS1 as a central metabolic–epigenetic amplifier that integrates glycolytic reprogramming with autophagy suppression, inflammatory activation, and fibrotic remodeling in DKD. Genetic deletion of Aars1 or pharmacological inhibition of AARS1-mediated lactylation alleviates nonlethal tubular stress and slows DKD progression in both type 1 and type 2 diabetic mouse models. These findings highlight the AARS1-lactylation axis as a promising therapeutic target and provide a conceptual framework linking lactate metabolism to chronic tubular stress and fibrosis in DKD.

Abbreviations

AARS1

Alanyl-tRNA synthetase 1

DKD

Diabetic kidney disease

ESKD

End-stage kidney disease

EMT

Epithelial–mesenchymal transition

CUT and Tag

Cleavage Under Targets and Tagmentation

qRT-PCR

Quantitative real-time polymerase chain reaction

PAS

Periodic Acid-Schiff

TEM

Transmission electron microscopy

ChIP

Chromatin immunoprecipitation

HK2

Hexokinase 2

PFKP

Phosphofructokinase, platelet type

ZEB1

Zinc Finger E-box-binding Homeobox 1

PPP6C

Protein Phosphatase 6 Catalytic Subunit

BUN

Blood urea nitrogen

STZ

Streptozotocin

Kim-1

Kidney injury molecule-1

Kla

Lysine lactylation

TSS

Transcription start sites

FN

Fibronectin

α-SMA

α-Smooth muscle actin

E-cad

E-cadherin

IL-6

Interleukin-6

MCP-1

Monocyte chemoattractant protein-1

TNF-α

Tumor necrosis factor alpha

LTL

Lotus tetragonolobus lectin

Author contributions

L.T. performed most experiments and data analysis. Y.W., C. G. and E.A. performed some of the experiments and data analysis. J.X.Z. and S.M. assisted in data analysis and paper preparation. X.L. supervised the whole project, data analysis and paper writing.

Funding

This work was supported by NIH/NIDDK grants R01 DK129241 and R01 DK126662 (X. Li), and the Department of defense (DOD) Focused Program Award: PR221810 (X. Li).

Data availability

Raw CUT and Tag sequencing data have been deposited in the NCBI SRA under accession number PRJNA1372173.

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Institutional Animal Care and Use Committee of the Mayo Clinic (protocol A0007740-24, approved on 8 April 2024). This Ethics Committee operates under the governance of the International Council for Laboratory Animal Science (ICLAS)—Ethics and Animal Welfare Committee, Naoko Kagiyama, Hansjoachim Hackbarth, Karin Blumer, Marilyn Brown, Gilly Griffin, Byung-Hwa Hyun, Harry Rozmiarek†, John Schofield; Approved by the ICLAS Governing Board on 6.6.2013.

Consent for publication

No applicable.

Competing interests

The authors declare no competing interests.