Probucol inhibits palmitic acid-mediated renal calcium oxalate stone formation by targeting KLF5-PPARγ axis-mediated lipotoxicity

Abstract

The pathogenesis of renal calcium oxalate (CaOx) stone is multifaceted and closely associated with metabolic disturbances. Palmitic acid (PA), a major saturated fatty acid, has emerged as a key contributor to CaOx stone formation. In this study, high-throughput drug screening identified probucol as a potential therapeutic agent capable of alleviating PA-mediated renal CaOx crystal formation in vitro and in vivo. Mechanistically, PA upregulates Krüppel-like factor 5 (KLF5), which transcriptionally activates peroxisome proliferator activated receptor gamma (PPARγ) and thereby drives lipotoxicity. Probucol directly binds to arginine-440 (Arg440) within the zinc finger DNA-binding domain of KLF5, competitively inhibiting its DNA-binding ability. This inhibition suppresses PPARγ expression and mitigates PPARγ-mediated lipotoxicity. Collectively, our findings identify KLF5 as a novel intracellular target of probucol, and highlight its therapeutic potential in treating PA-mediated renal CaOx stone formation by mitigating lipotoxicity.

Graphical abstract

Highlights

• •High-throughput screening identifies probucol as a potent inhibitor of PA-mediated renal CaOx stone formation.
• •RNA-seq shows probucol suppresses PA-induced PPARγ, revealing its pathogenic role in benign kidney disease for the first time.
• •Probucol binds Arg440 in KLF5’s zinc finger, impairing its DNA binding and thereby reducing KLF5-driven PPARγ transcription.
• •Probucol alleviates PA-induced lipotoxicity by targeting the KLF5-PPARγ axis, thereby reducing renal CaOx stone formation.
• •Clinical validation confirms PA-associated upregulation of KLF5 and PPARγ in patients with renal CaOx stone.

1. Introduction

Urolithiasis is a prevalent urinary disorder with a substantial global impact [1]. Epidemiological studies indicate that the prevalence of urolithiasis is 8.8% in the United States, 5%–10% in Europe, 4% in South America, and 1%–19% across Asia, with an estimated 5.8% in China [2]. The incidence of urolithiasis has steadily increased in recent decades, largely driven by shifts in dietary habits and environmental factors. Notably, the recurrence rate remains high, with approximately 50% of patients experiencing relapse within 5–10 years and up to 75% within 20 years [3]. Despite the high burden, effective preventive and targeted therapeutic strategies remain limited in clinical practice, underscoring the urgent need to elucidate the underlying molecular mechanisms of stone formation.

Calcium oxalate (CaOx) stone constitutes nearly 80% of all urolithiasis cases [4]. The pathogenesis of renal CaOx stone is multifactorial, influenced by genetic predisposition, metabolic dysregulation, and environmental exposures [5]. Emerging evidence indicates that metabolic abnormalities, particularly dyslipidemia, are independent risk factors for CaOx stone formation [[6], [7], [8]]. To explore the mechanistic link between dyslipidemia and stone formation, we conducted untargeted urinary metabolomic profiling and observed a significant elevation of palmitic acid (PA), a major saturated fatty acid involved in lipid metabolism, in patients with renal CaOx stones compared to healthy controls. Consistently, functional experiments demonstrated that PA promotes CaOx stone formation by inducing injury in renal tubular epithelial cells (TECs) [9]. These findings warrant further investigation into how PA metabolic dysregulation contributes to CaOx stone formation. Targeting this pathogenic pathway may offer novel therapeutic strategies to prevent stone formation and recurrence.

Krüppel-like factor 5 (KLF5) is a key zinc finger transcription factor (TF) and plays a pivotal role in regulating various biological processes [10]. Previous studies have shown that KLF5 is closely associated with renal TECs injury, interstitial fibrosis, and inflammatory responses [[11], [12], [13]]. Additionally, peroxisome proliferator activated receptor gamma (PPARγ), a nuclear receptor TF, plays a central role in lipid synthesis and energy metabolism, with known anti-inflammatory and anti-fibrotic effects [14]. However, aberrant upregulation of PPARγ has also been implicated in several metabolic diseases, including non-alcoholic fatty liver disease and metabolic cardiomyopathy [[15], [16], [17]]. Whether KLF5 directly regulates PPARγ in the context of renal CaOx stone formation remains unknown. Given the dualistic roles of PPARγ in both protective and pathogenic processes, dissecting its regulation by upstream factors such as KLF5 is critical for understanding the complexity of stone pathogenesis.

Probucol, a lipophilic biphenolic compound traditionally used to treat hyperlipidemia, has attracted interest for its potential renoprotective properties owing to its lipid-lowering, anti-inflammatory, and antioxidant effects [18]. For instance, Yang et al. [19] demonstrated that probucol promotes AMP-activated protein kinase (AMPK) phosphorylation, enhancing the binding of sirtuin 1 (Sirt1) to the p66 Src homology 2 domain-containing protein (p66shc) promoter, ultimately slowing the progression of diabetic nephropathy. In polycystic kidney disease, Nagao et al. [20] reported that probucol effectively reduced kidney volume, lowered blood urea nitrogen (BUN) levels, and alleviated cyst expansion and interstitial fibrosis. Similarly, in a contrast-induced acute kidney injury, Ma et al. [21] found that probucol exerted renoprotective effects by modulating the extracellular signal-regulated kinase 1/2 (ERK1/2) and c-Jun N-terminal kinase (JNK) pathways and suppressing caspase 3 expression, thereby reducing TECs apoptosis. Furthermore, probucol has demonstrated clinical efficacy in treating atherosclerosis [22,23], a condition increasingly recognized to share pathogenic links with renal CaOx stone formation [24], further supporting its potential application in CaOx stone disease.

In this study, we employed a high-throughput screening approach using a library of 2080 approved drugs and identified probucol as a promising candidate. Through both in vitro and in vivo models, we demonstrated that probucol significantly inhibited renal TECs injury induced by abnormal PA metabolism, thereby reducing renal CaOx stone formation. Mechanistically, we discovered that PA stimulation upregulates KLF5, which transcriptionally activates PPARγ and drives lipotoxicity. Probucol exerts its protective effect by directly binding to the arginine-440 (Arg440) residue within the zinc finger DNA-binding domain of KLF5, thereby suppressing its transcriptional activation of PPARγ and the resulting lipotoxicity. Notably, this study is the first to reveal the pathogenic role of PPARγ in kidney diseases, providing novel insights into the pathophysiology of renal CaOx stone formation.

In conclusion, our findings indicate that probucol represents a potential therapeutic agent for PA-mediated renal CaOx stone formation, primarily by targeting the KLF5-PPARγ axis. These findings offer new mechanistic insights and identify a novel therapeutic strategy for precision management of renal CaOx stone formation.

2. Materials and methods

2.1. Human serum and kidney samples

Nontumor kidney tissue from patients diagnosed with renal cell carcinoma (RCC) who underwent nephrectomy was used as the normal group, while nontumor kidney tissue from patients with RCC concomitant with ipsilateral renal CaOx stone undergoing nephrectomy was designated as the positive group. Clinical characteristics are summarized in Tables S1 and S2. Paraffin-embedded kidney sections (5 μm) were prepared according to routine procedures. Serum PA concentrations were measured using an enzyme-linked immunosorbent assay (ELISA) kit (MSK Biotechnology Co., Ltd., Wuhan, China), following the manufacturer's protocol. This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Second Hospital of Tianjin Medical University (Approval No.: KY2022K060). Written informed consent was obtained from all participants for the collection and analysis of their serum and kidney samples.

2.2. Reagents

A detailed description of the materials and methods for reagents and antibodies is shown in Table S3.

2.3. High throughput drug screening

A library of 2080 approved clinical compounds—including drugs authorized by regulatory agencies (e.g., U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), China Food and Drug Administration (CFDA)) or listed in pharmacopoeial references (e.g., United States Pharmacopoeia (USP), European Pharmacopoeia (EP), British Pharmacopoeia (BP), Japanese Pharmacopoeia (JP), Chinese Pharmacopoeia (CP))—was provided by the High-Throughput Drug Screening Center of Tianjin Medical University.

For primary screening, HK-2 cells were seeded in 384-well plates (Explorer G3, PerkinElmer, Waltham, MA, USA) and incubated overnight for attachment. Cells were then divided into three groups: blank (no treatment), control (PA), and treatment (PA combined with test compounds). After 48 h incubation, cell counting kit-8 (CCK-8) reagent (Solarbio, Beijing, China) was added and incubated for 2 h. Cell viability was calculated as:

Viability improvement = (Treatment group−Control group)/(Blank group−Control group)

Compounds exhibiting ≥30% improvement in cell viability based on this ratio were considered hits.

For secondary screening, the 27 primary hits were re-tested in 96-well plates under identical conditions to confirm reproducibility. The top three most effective compounds were then subjected to phenotypic validation.

2.4. Animal models and treatment protocols

Male C57BL/6J mice (6 weeks old, 18–22 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) and housed under specific pathogen-free conditions (temperature: 22 ± 2 °C, humidity: 60% ± 5%) with a 12-h light/dark cycle. All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of the Institute of Radiation Medicine, Chinese Academy of Medical Sciences (Approval No.: IRM2-IACUC-2507-015) and conducted in accordance with institutional and national guidelines for the care and use of laboratory animals.

To investigate the effects of probucol on PA-mediated renal CaOx crystal deposition, mice were randomly divided into three groups (n = 12): chow diet, high-PA diet (7.5%), and high-PA diet (7.5%) + probucol (0.75%). The mice were orally fed the respective diets for 20 weeks. After 20 weeks, six mice from each group were euthanized, and blood samples were collected via cardiac puncture into 1.5 mL tubes with heparin. The left kidney was fixed for histological analysis, while the right kidney was frozen in liquid nitrogen and stored at −80 °C. Renal histological analysis and functional assessments confirmed that probucol alleviated PA-induced TECs injury.

For further modeling of renal CaOx crystal deposition, the remaining six mice in each group continued to receive daily intraperitoneal injections of glyoxylate (Gly; Sigma-Aldrich, Darmstadt, Germany) solution (80 mg/kg/day, dissolved in sterile saline) for 7 consecutive days [25].

To assess the role of KLF5 in PA-mediated renal CaOx crystals deposition, mice were randomly divided into four groups (n = 12): chow diet + vehicle, chow diet + ML264, high-PA diet (7.5%) + vehicle, and high-PA diet (7.5%) + ML264. Mice were orally fed the respective diets for 12 weeks. After 12 weeks, the normal diet and high-PA diet groups received intraperitoneal injections of either vehicle or ML264 (20 mg/kg/day), dissolved in a vehicle containing 5% dimethyl sulfoxide (DMSO), 40% polyethylene glycol 300 (PEG300), 5% Tween 80, and 50% sterile saline. All reagents, including ML264, PEG300, and Tween 80, were purchased from Selleck Chemicals (Houston, TX, USA) for 8 weeks. Six mice from each group were euthanized, and tissues were processed as described above. Renal histological analysis and functional assessments confirmed that KLF5 mediates PA-induced TECs injury. After one week rest, the remaining mice in each group were again intraperitoneally injected with Gly (80 mg/kg/day) for 7 consecutive days to continue the renal CaOx crystal model for subsequent experiments.

2.5. Cell culture and transfection

Human renal tubular epithelial cells (HK-2 cells) were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured in Dulbecco's modified Eagle medium/nutrient mixture F-12 (DMEM/F-12; Gibco, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Waltham, MA, USA) and 1% penicillin/streptomycin (Gibco, Waltham, MA, USA) at 37 °C in a 5% CO₂ atmosphere.

For in vitro experiments, cells were treated with 400 μM PA in the presence or absence of varying concentrations of probucol for 24 h.

Lentiviral vectors expressing human KLF5 or a scrambled control were constructed by Genechem Co., Ltd. (Shanghai, China). Small interfering RNAs (siRNAs) targeting KLF5, PPARγ, or a negative control were synthesized by RiboBio (Guangzhou, China). Transfections were performed using X-tremeGENE siRNA transfection reagent (Roche, Basel, Switzerland) according to the manufacturer's protocol, and cells were harvested 48 h after transfection for subsequent analysis.

2.6. Statistical analysis

All statistical analyses were performed using GraphPad Prism 8.0 (La Jolla, CA, USA) or SPSS 22.0 (IBM Corp., Armonk, NY, USA). For comparison between two sets of data, a two-tailed unpaired Student's t-test was used if the data followed a normal distribution; otherwise, a non-parametric Mann-Whitney U test was applied. For single-factor, multiple-group comparisons, one-way analysis of variance (ANOVA) was employed when the data followed a normal distribution (with homogeneity of variance tested using least significant difference (LSD) test for equal variances and Dunnett's T3 test for unequal variances), and a non-parametric test was used if the normality assumption was not met. A P-value of less than 0.05 was considered statistically significant. Data are presented as the mean ± standard error of the mean (SEM) from at least three independent experiments.

3. Results

3.1. Drug screening identifies probucol as an effective agent that alleviates PA-induced injury to renal TECs thereby reducing CaOx crystal adhesion

In our previous study, non-targeted urinary metabolomic profiling of patients with renal CaOx stone and healthy individuals identified PA, a predominant saturated fatty acid, as a key metabolite associated with CaOx stone formation. Further in vitro and in vivo experiments demonstrated that PA induces injury in renal TECs, leading to the upregulation of adhesion molecules such as osteopontin (OPN) and cluster of differentiation 44 (CD44). This upregulation further promotes the adhesion and deposition of CaOx crystal on TECs, contributing to the progression of renal CaOx stone formation [9].

Given that injury to renal TECs is a key factor underlying the adhesion and deposition of CaOx crystal [26], we aimed to identify potential therapeutic agents capable of alleviating PA-induced injury to renal TECs, and thereby reducing CaOx crystal adhesion and deposition. A high-throughput drug screening platform encompassing 2080 approved compounds was employed to identify candidate agents, and the detailed workflow is presented in Fig. 1A. Initially, 27 candidate compounds that alleviated PA-induced injury in HK-2 cells were identified (Fig. 1B). A secondary screen using the CCK-8 assay further narrowed the selection to the three most effective candidates (Fig. 1C). These candidates were evaluated using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and Western blot (WB) analyses to assess the expression levels of renal TEC-related adhesion molecules OPN, CD44 and injury markers neutrophil gelatinase-associated lipocalin (NGAL) [27], as well as a CaOx crystal adhesion assay (Figs. 1D–G). Ultimately, probucol was identified as the compound with the most significant therapeutic effect.

Fig. 1.

High-throughput drug screening identifies probucol as a potent inhibitor of palmitic acid (PA)-induced renal tubular epithelial cells (TECs) injury thereby reducing calcium oxalate (CaOx) crystal adhesion. (A) Workflow schematic of the drug screening strategy. (B) Heatmap showing 27 candidate compounds selected from the primary screen (see Section 2.3). (C) Cell counting kit-8 (CCK-8) validation of candidate compounds in secondary screening. (D) Representative images from the CaOx crystal adhesion assay in the indicated groups. (E) Quantitative analysis of fluorescein isothiocyanate (FITC)-fluorescence intensity from panel (D). (F, G) Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) (F) and Western blot (WB) analysis (G) showing the effects of selected compounds on PA-induced upregulatuon of adhesion and injury markers in renal TECs. Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎P < 0.05, ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001, compared with the PA group; ^####P < 0.0001, compared with the Control group. Pro: probucol; Laq: laquinimod; The: theophylline; OPN: osteopontin; CD44: cluster of differentiation 44; NGAL: neutrophil gelatinase-associated lipocalin; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole.

3.2. Probucol alleviates PA-mediated renal CaOx crystal adhesion and deposition in vitro and in vivo

The chemical structure of probucol is shown in Fig. S1A. To determine the optimal concentration range, a gradient concentration analysis was performed using the CCK-8 assay. Probucol effectively attenuated PA-induced injury in HK-2 cells at concentrations ranging from 10 to 100 μM, with maximal cell viability observed at 60–80 μM (Fig. S1B). RT-qPCR analysis demonstrated that co-treatment with 80 μM probucol significantly suppressed the PA-induced upregulation of renal TEC-related adhesion and injury markers. Based on these findings, 80 μM was selected for subsequent experiments (Fig. S1C). Further investigations using WB analysis and CaOx crystal adhesion assay demonstrated that probucol alleviated the PA-induced upregulation of renal TEC-related adhesion and injury markers, as well as CaOx crystal adhesion, in a dose-dependent manner (Figs. 2A and B).

Fig. 2.

Probucol alleviates palmitic acid (PA)-mediated renal calcium oxalate (CaOx) crystal adhesion and deposition in vitro and in vivo. (A) Representative images and quantification of the CaOx crystal adhesion assay showing that probucol dose-dependently alleviates PA-mediated CaOx crystal adhesion. (B) Western blot (WB) analysis showing that probucol dose-dependently inhibits PA-induced upregulation of adhesion and injury markers in renal tubular epithelial cells (TECs). (C) Schematic diagram illustrating the animal groups: chow diet, high PA diet (7.5%), and high PA (7.5%) + probucol (0.75%) diet. (D, E) Renal function assessment via serum creatinine (Scr) (D) and blood urea nitrogen (BUN) (E) levels in the indicated groups (n = 6). (F) Representative images of periodic acid–Schiff (PAS) (upper) and Masson's trichrome (lower) staining with corresponding quantification in kidney sections in the indicated groups (n = 6). (G) Representative Von Kossa staining and quantification of CaOx crystal deposition in kidney sections in the indicated groups (n = 6). (H) WB analysis of adhesion and injury markers expression in kidney tissues in the indicated groups (n = 6). Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001, compared with the PA group. Gly: glyoxylate; OPN: osteopontin; CD44: cluster of differentiation 44; NGAL: neutrophil gelatinase-associated lipocalin; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole.

To ensure the in vivo safety of probucol administration, we first performed a comprehensive toxicity assessment in C57BL/6J mice. Histological examination of the heart, liver, spleen, lung, and kidney revealed no significant morphological abnormalities in mice orally administered probucol compared with controls (Fig. S2A). Serum biochemical analysis showed that although high-density lipoprotein (HDL) levels were slightly decreased (Fig. S2B), there were no significant changes in serum alanine aminotransferase (ALT) or aspartate aminotransferase (AST) levels (Figs. S2C and D), indicating that probucol was well tolerated without inducing hepatotoxicity. The observed reduction in HDL levels is further discussed in the Discussion section in the context of probucol's pharmacological profile.

After confirming the in vivo safety of probucol, we proceeded to evaluate its therapeutic efficacy using a 20-week PA and PA + probucol mouse model established in C57BL/6J mice (Fig. 2C). Probucol treatment markedly suppressed PA-induced elevations in serum creatinine (Scr) and BUN (Figs. 2D and E). Histological analysis with periodic acid–Schiff (PAS) and Masson's trichrome staining indicated that probucol alleviated PA-induced TECs injury and interstitial fibrosis (Fig. 2F). Interestingly, Von Kossa staining revealed significant CaOx crystal deposition in both the renal cortex and medulla of mice fed PA for 20 weeks, followed by one week of intraperitoneal Gly injection, consistent with our previous findings [9]. However, probucol treatment effectively reduced CaOx crystal deposition in both the cortex and medulla (Fig. 2G). Furthermore, WB analysis showed that probucol significantly suppressed the PA-induced upregulation of adhesion and injury markers in renal TECs (Fig. 2H).

Taken together, these findings indicate that probucol effectively inhibits PA-induced injury in renal TECs, thereby reducing CaOx crystal adhesion and deposition in vitro and in vivo.

3.3. Probucol inhibits PA-induced PPARγ expression to reduce CaOx crystal adhesion in renal TECs

Given that injury to renal TECs underlies CaOx crystal adhesion and deposition, we performed RNA sequencing (RNA-seq) and bioinformatics analysis to elucidate the molecular mechanisms by which probucol attenuates PA-induced TECs injury. Gene set enrichment analysis (GSEA) demonstrated that PA treatment activated the lipid and atherosclerosis and interleukin-17 (IL-17) signaling pathways, both of which were substantially suppressed by probucol treatment (Figs. 3A and B). Notably, both pathways have been previously reported to contribute to the pathogenesis of renal CaOx stone formation [24,28]. Further analysis revealed that PPARγ not only regulates these two signaling pathways but also controls the expression of several key downstream genes within them, including perilipin 2 (PLIN2), matrix metalloproteinase-1 (MMP1), Jun proto-oncogene (JUN), and interleukin-6 (IL-6) (Fig. 3C) [15,[29], [30], [31]]. Given its well-established role in lipid metabolism, PPARγ was selected as a potential downstream target of probucol for further investigation.

Fig. 3.

Probucol inhibits palmitic acid (PA)-induced peroxisome proliferator activated receptor gamma (PPARγ) expression to reduce calcium oxalate (CaOx) crystal adhesion in renal tubular epithelial cells (TECs). (A) Heatmap of differentially expressed genes in the indicated groups. (B) Gene set enrichment analysis (GSEA) showing that probucol attenuates PA-induced enrichment of the lipid and atherosclerosis and interleukin-17 (IL-17) signaling pathways. (C) Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) validation of PPARγ and downstream targets in the indicated groups. (D) Western blot (WB) analysis of PPARγ and adhesion and injury markers expression in renal TECs in the indicated groups. (E) Representative images and quantification of CaOx crystal adhesion assay in the indicated groups. Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎P < 0.05, ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001. OPN: osteopontin; CD44: cluster of differentiation 44; NGAL: neutrophil gelatinase-associated lipocalin; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole; siCON: small interfering control.

First, we found that probucol significantly suppressed PA-induced PPARγ expression in vitro (Figs. S3A and B). To further investigate the role of PPARγ, we designed and validated PPARγ-specific siRNA constructs (Figs. S3C and D). Using the most effective siRNA, we conducted further experiments. RT-qPCR, WB analysis and CaOx crystal adhesion assay (Figs. S3E, 3D, and 3E) revealed that both PPARγ knockdown and probucol treatment significantly downregulated the expression of renal TEC-related adhesion and injury markers, as well as renal CaOx crystal adhesion compared to the PA groups.

These results identify PPARγ as a key pathogenic mediator of PA-mediated CaOx crystal adhesion in renal TECs and demonstrate that probucol attenuates crystal adhesion primarily by suppressing PPARγ expression.

3.4. Probucol mitigates PA-induced lipotoxicity in renal TECs via PPARγ modulation in vitro and in vivo

Previous studies have demonstrated that in non-adipose tissues, such as renal TECs, sustained lipid storage leads to mitochondrial dysfunction and apoptosis, a pathological process collectively referred to as lipotoxicity [27,[32], [33], [34]]. PPARγ, a pivotal nuclear receptor TF governing lipid metabolism, orchestrates the regulation of lipid uptake, synthesis, and storage.

In this study, we explored the transcriptional regulatory function of PPARγ by evaluating the expression of its downstream target genes involved in lipid metabolism. PPARγ knockdown significantly reduced the PA-induced mRNA expression of genes involved in lipid uptake cluster of differentiation 36 (CD36), fatty acid transport protein 2 (FATP2), lipid synthesis stearoyl-CoA desaturase (SCD), fatty acid synthase (FASN), solute carrier family 25 member 1 (SLC25A1), acetyl-CoA carboxylase (ACC), acyl-CoA synthetase short-chain family member 2 (ACSS2), lipid storage (PLIN2, MMP1), and inflammation/apoptosis (IL-6, JUN) [[15], [30], [35], [36]] (Fig. S3E).

To determine whether these transcriptional changes were accompanied by ultrastructural alterations, we performed transmission electron microscopy (TEM). PA-treated renal TECs displayed pronounced mitochondrial abnormalities, characterized by swelling, deformation, and cristae disruption, alongside substantial intracellular lipid droplet storage. In contrast, PPARγ knockdown or probucol treatment restored normal mitochondrial morphology and markedly reduced lipid droplet storage. Consistent with the ultrastructural findings, BODIPY 493/503 staining further confirmed a substantial increase lipid droplet storage in PA-treated cells, which was markedly attenuated following PPARγ knockdown or probucol treatment (Fig. 4A).

Fig. 4.

Probucol mitigates palmitic acid (PA)-induced lipotoxicity in renal tubular epithelial cells (TECs) via peroxisome proliferator activated receptor gamma (PPARγ) modulation in vitro and in vivo. (A) Transmission electron microscopy (TEM) images showing lipid droplet storage (indicated by red arrows) and mitochondrial morphology in the indicated groups. BODIPY 493/503 staining further showing lipid droplet distribution and quantification in the indicated groups. (B) Representative JC-1 and MitoSOX staining images with corresponding quantification in the indicated groups. (C) Western blot (WB) analysis of PPARγ, perilipin 2 (PLIN2), Bcl-2-associated X protein (BAX), B-cell lymphoma 2 (BCL2), caspase 3, and cleaved-caspase 3 expression in the indicated groups. (D) Representative terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining images and quantification in the indicated groups. (E) Representative flow cytometry plots and quantification of apoptosis by Annexin V–fluorescein isothiocyanate (FITC)/propidium iodide (PI) staining across groups. (F) Representative images and quantification of Oil Red O staining, TUNEL staining, and IHC staining for PPARγ, PLIN2, BAX, BCL2, and cleaved-caspase 3 in kidney tissues in the indicated groups (n = 6). Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001. GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole; siCON: small interfering control.

To further determine whether the observed mitochondrial abnormalities and lipid droplet storage were functionally associated with mitochondrial dysfunction and apoptosis, we performed a series of functional assays. WB analysis of apoptosis-related proteins including Bcl-2-associated X protein (BAX), B-cell lymphoma 2 (BCL2), and cleaved-caspase3/caspase3, JC-1 staining, MitoSOX staining, flow cytometry, and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays consistently demonstrated that PPARγ knockdown or probucol treatment attenuated PA-induced mitochondrial dysfunction and apoptosis, hallmark features of lipotoxicity in renal TECs (Figs. 4B–E).

In vivo, probucol significantly attenuated PA-induced metabolic disturbances, including elevated serum triglyceride (TG) and total cholesterol (TC) levels, as well as an increased kidney weight-to-body weight (KW/BW) ratio (Figs. S4A–C). Additionally, PA markedly increased malondialdehyde (MDA) levels while reducing superoxide dismutase (SOD) activity in kidney tissues, indicating enhanced oxidative stress and renal lipotoxicity. These alterations were significantly reversed by probucol treatment (Figs. S4D and E. Oil Red O staining and PLIN2 immunohistochemistry (IHC) demonstrated that probucol markedly reduced lipid droplet storage in renal TECs of PA-treated mice (Fig. 4F). Furthermore, IHC staining analyses of apoptotic markers (BAX, BCL2, and cleaved-caspase3) and TUNEL staining confirmed that probucol attenuated PA-induced mitochondrial dysfunction and apoptosis in renal TECs (Fig. 4F).

Together, these findings establish PPARγ-mediated lipotoxicity as a key pathogenic mechanism linking dysregulated lipid metabolism to CaOx crystal adhesion and deposition in renal TECs, and identify probucol as a promising therapeutic agent capable of interrupting this pathogenic cascade.

3.5. Probucol disrupts KLF5 DNA binding by targeting the R440 residue within its zinc finger domain

To explore the mechanism by which probucol affects PPARγ expression and its mediated lipotoxicity, we modified probucol by conjugating it with polyethylene glycol (PEG) and biotin to generate biotin-labeled probucol (Bio-Pro). This probe was then used in a pull-down assay coupled with mass spectrometry to identified proteins that directly interact with probucol, as detailed in Figs. 5A and B.

Fig. 5.

Probucol inhibits Krüppel-like factor 5 (KLF5) transcriptional activity by impairing its DNA-binding ability through interaction with the zinc finger domain. (A) Schematic representation of the pull-down assay combined with mass spectrometry to identify potential probucol-binding proteins. (B) Chemical structures of probucol and biotin-labeled probucol. (C) Venn diagram showing intersecting proteins identified from three independent screening strategies. (D, E) Biotin alone was used as a control. Biotin-labeled probucol was incubated with streptavidin-agarose beads. Lysates prepared from HK-2 cells (D) or mouse kidney tissues (E) were added to the biotin-labeled probucol-bound streptavidin-agarose beads, and eluted proteins were analyzed by Western blot (WB). Total lysates were used as input controls. (F) Residue arginine 440 (Arg440) of KLF5 was mutated to alanine (Ala). Pull-down assays were performed to assess the interaction between probucol and either wild-type (WT) or mutant KLF5, and compared to wild-type KLF5. The expression of Flag-KLF5 was detected by WB. (G) Cellular thermal shift assay (CETSA) was performed to evaluate the binding of probucol to KLF5, and to identify its binding site. Flag-KLF5 expression was detected by WB. (H) Representative immunohistochemistry (IHC) staining images and quantification of KLF5 in kidney sections in the indicated groups (n = 6). (I, J) Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) (I) and WB analyses (J) in HK-2 cells showing that probucol did not alter KLF5 expression in a dose-dependent manner, either with or without palmitic acid (PA) treatment. (K, L) WB (K), and representative immunofluorescence staining images and quantification (L) of KLF5 localization in HK-2 cells, showing no effect of probucol on KLF5 subcellular distribution. (M, N) Chromatin immunoprecipitation quantitative polymerase chain reaction (ChIP-qPCR) (M) and dual-luciferase reporter assays (N) were used to assess KLF5 enrichment on the PPARγ promoter under the indicated treatment conditions. Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001. SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis; LC-MS/MS: liquid chromatography-tandem mass spectrometry; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole; Bio-PEG-NH₂: biotin-polyethylene glycol-amine.

We first integrated the following two criteria (1) genes upregulated by PA treatment in RNA-Seq; (2) proteins directly binding to probucol identified by mass spectrometry. After integration, we found 15 related proteins, including ribosomal protein S13 (RPS13), FAU ubiquitin-like and ribosomal protein S30 fusion (FAU), RPS25, heat shock protein family A member 8 (HSPA8), KLF5, lactamase β (LACTB), ribosomal protein L13 (RPL13), mitotic spindle positioning (MISP), RPL36, RPL37A, microtubule-associated protein 1B (MAP1B), ribosomal protein S18 (RPS18), high mobility group AT-hook 1 (HMGA1), RPL8, and HSPA5. Among these 15 proteins, KLF5 caught our attention. Previous studies have reported that KLF5 binds to the PPARγ promoter region to promote its transcription [36,37]. Based on the above selection criteria, we further used the Animal TFDB V4.0 database to predict whether the remaining 14 proteins were associated with the transcriptional activation of PPARγ. The results showed that only KLF5 has the ability to promote PPARγ transcription. Based on these findings, we identified KLF5 as a direct binding protein of probucol for further investigation (Figs. 5C and S5A).

To validate the interaction between probucol and KLF5, surface plasmon resonance analysis was first performed, which confirmed the direct binding between probucol and KLF5 (Fig. S5B). To further substantiate this finding, a pull-down assay was carried out using Bio-Pro immobilized on streptavidin agarose beads, which were incubated with lysates from HK-2 cells or mouse kidney tissues. WB analyses of the eluent confirmed that Bio-Pro binds to KLF5 in both samples (Figs. 5D and E). Next, we investigated the specific binding site between probucol and KLF5 using molecular docking simulations (Figs. S5C and D). The simulations indicated that Arg440 in the zinc-finger domain of KLF5 is a potential interaction site. To confirm its role, we generated an R440A mutant and performed a pull-down assay. Results showed that the R440A mutation significantly reduced the interaction between KLF5 and probucol (Fig. 5F), identifying Arg440 as a critical residue for their interaction. This finding was further validated using cellular thermal shift assay (CETSA) experiments (Fig. 5G).

KLF5, a member of the KLF TFs family, contains a conserved zinc-finger domain at its C-terminus, and prior studies have shown that zinc-finger domains in KLF family members influence its nuclear translocation and DNA binding [10,38]. To determine whether probucol affects KLF5 abundance or localization in vivo, we first performed IHC staining of kidney tissues. The results showed that, compared to PA-treated mice, KLF5 expression levels and subcellular distribution in renal TECs remained unchanged following probucol administration (Fig. 5H). To further assess whether probucol modulates KLF5 expression at the molecular level, we examined KLF5 mRNA and protein levels under both physiological and pathological conditions. RT-qPCR and WB analyses revealed that probucol had no significant effect on KLF5 expression at either the transcriptional or protein level (Figs. 5I and J). Since Arg440 is located within the zinc finger DNA-binding domain of KLF5, we next investigated whether probucol affects its nuclear translocation. Nuclear-cytoplasmic fractionation and immunofluorescence staining further revealed that probucol had no effect on KLF5 nuclear localization compared to the PA-treated group (Figs. 5K and L).

We then hypothesized that probucol may counteract PA-induced activation of KLF5 and its binding to the promoter region of downstream target genes, thereby modulating PPARγ transcription. Chromatin immunoprecipitation quantitative polymerase chain reaction (ChIP-qPCR), dual-luciferase reporter, and electrophoretic mobility shift assays collectively demonstrated that PA stimulation enhanced KLF5 binding to the PPARγ promoter, whereas probucol markedly attenuated this PA-induced binding and consequently suppressed PPARγ transcription (Figs. 5M, N and S5E).

Mechanistically, probucol was shown to directly bind to Arg440, a residue located within the zinc finger domain of KLF5. This interaction is essential for the inhibitory effect of probucol, as R440 mutation abolished the binding between KLF5 and probucol in pull-down assays. We therefore propose that probucol impairs KLF5's DNA-binding activity by targeting the zinc finger domain, likely through interfering with its structural integrity or the DNA-contacting interface. This provides a structural basis for the observed inhibition of KLF5-driven transcription.

3.6. Probucol mitigates PA-induced lipotoxicity to reduce CaOx crystal adhesion in a KLF5-dependent manner in renal TECs

To validate the role of KLF5, we designed siRNA targeting KLF5 and confirmed their knockdown efficiencies (Fig. S6A). We then assessed the effects of KLF5 modulation on PA-mediated renal CaOx crystal adhesion. RT-qPCR, WB and CaOx crystal adhesion assay (Figs. S6B, 6A, and 6B) revealed that PA-induced upregulation of adhesion and injury markers in renal TECs, as well as the enhanced CaOx crystal adhesion, were markedly attenuated by KLF5 knockdown.

Furthermore, modulating KLF5 expression following PA treatment resulted in significant alterations in the mRNA levels of PPARγ-regulated genes (Fig. S6B). BODIPY 493/503 staining demonstrated that KLF5 knockdown significantly reduced lipid droplet storage compared to the PA group (Fig. 6C). In parallel, JC-1 and MitoSOX staining demonstrated restoration of mitochondrial membrane potential and a reduction in reactive oxygen species (ROS) production, while TUNEL staining and WB analysis of apoptotic markers (Figs. 6D–F) confirmed that KLF5 knockdown alleviated PA-induced mitochondrial dysfunction and apoptosis.

Fig. 6.

Probucol mitigates palmitic acid (PA)-induced lipotoxicity to reduce calcium oxalate (CaOx) crystal adhesion in a Krüppel-like factor 5 (KLF5)-dependent manner in renal tubular epithelial cells (TECs). (A) Western blot (WB) analysis of KLF5, peroxisome proliferator activated receptor gamma (PPARγ), and adhesion and injury markers in renal TECs in the indicated groups. (B) Representative images and quantification of CaOx crystal adhesion assay in the indicated groups. (C) Representative BODIPY 493/503 staining images and quantification of lipid droplet storage in the indicated groups. (D) Representative terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining images and corresponding quantification in the indicated groups. (E) Representative JC-1 and MitoSOX staining images with corresponding quantification in the indicated groups. (F) WB analysis of KLF5, PPARγ, perilipin 2 (PLIN2), Bcl-2-associated X protein (BAX), B-cell lymphoma 2 (BCL2), caspase 3, and cleaved-caspase 3 expression in the indicated groups. Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎P < 0.05, ^⁎⁎P < 0.01, ^⁎⁎⁎⁎P < 0.0001. OPN: osteopontin; CD44: cluster of differentiation 44; NGAL: neutrophil gelatinase-associated lipocalin; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole; siCON: small interfering control.

Importantly, across all assays—including downstream marker expression, lipid droplet storage, mitochondrial function, and apoptosis—no further improvement was observed in the PA + probucol + siKLF5 group compared to the PA + siKLF5 group, indicating that the protective effects of probucol are predominantly KLF5-dependent.

3.7. Lipotoxicity induced by KLF5-PPARγ axis contributes to PA-mediated renal CaOx crystal adhesion

To further investigate the functional relevance of the KLF5–PPARγ axis, we first established a stable KLF5-overexpressing HK-2 cell line via lentiviral transduction, and confirmed the overexpression efficiency by RT-qPCR and WB (Fig. S7A). Based on this model, we subsequently knocked down PPARγ and treated the cells with PA to assess downstream effects. RT-qPCR, WB and CaOx crystal adhesion assay (Figs. S7B, 7A, and 7B) revealed that PPARγ silencing in PA-treated, KLF5-overexpressing cells significantly reduced the expression of adhesion and injury markers in renal TECs and reduced CaOx crystal adhesion. Additionally, the mRNA levels of PPARγ target genes were also downregulated (Fig. S7B).

Consistently, BODIPY 493/503 staining demonstrated reduced lipid droplet storage (Fig. 7C), while JC-1 and MitoSOX staining revealed restored mitochondrial membrane potential and attenuated ROS production, respectively. Moreover, TUNEL staining and WB analysis of apoptotic markers further confirmed that PPARγ knockdown alleviated PA-induced mitochondrial dysfunction and apoptosis, even in the context of KLF5 overexpression (Figs. 7D–F).

Fig. 7.

Lipotoxicity via the Krüppel-like factor 5 (KLF5)-peroxisome proliferator activated receptor gamma (PPARγ) axis contributes to palmitic acid (PA)-mediated calcium oxalate (CaOx) crystal adhesion. (A) Western blot (WB) analysis of KLF5, PPARγ, and adhesion and injury markers in renal tubular epithelial cells (TECs) in the indicated groups. (B) Representative images and quantification of CaOx crystal adhesion assay in the indicated groups. (C) Representative BODIPY 493/503 staining images and quantification of lipid droplet storage in the indicated groups. (D) Representative terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining images and corresponding quantification in the indicated groups. (E) Representative JC-1 and MitoSOX staining images with corresponding quantification in the indicated groups. (F) WB analysis of KLF5, PPARγ, perilipin 2 (PLIN2), Bcl-2-associated X protein (BAX), B-cell lymphoma 2 (BCL2), caspase 3, and cleaved-caspase 3 expression in the indicated groups. Data are presented as mean ± standard error of the mean (SEM). ^⁎P < 0.05, ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001. OPN: osteopontin; CD44: cluster of differentiation 44; NGAL: neutrophil gelatinase-associated lipocalin; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole; LvCON: lentivirus vector control.

Together, these findings indicate that PPARγ silencing effectively counteracts the pro-lipotoxic effects of KLF5 overexpression under PA treatment, thereby attenuating PA-induced lipotoxicity and its downstream pathological consequences in renal TECs.

3.8. Pharmacological inhibition of KLF5 attenuates PA-mediated renal CaOx crystal deposition

To investigate the therapeutic potential of targeting KLF5 in PA-mediated CaOx crystal deposition, we performed in vivo experiments using ML264, a selective KLF5 inhibitor, as illustrated in Fig. 8A. ML264 treatment effectively improved the elevated levels of Scr, BUN, TG, and TC induced by PA, and also reduced the KW/BW ratio (Figs. 8B–F). In addition, ML264 treatment significantly reduced MDA levels and restored SOD activity in renal tissues, indicating an attenuation of PA-induced lipotoxicity (Figs. S8A and B). IHC staining revealed that ML264 treatment significantly reduced the expression levels of KLF5 and PPARγ in renal TECs of PA-treated mice (Fig. 8G). Concurrently, PAS and Masson's trichrome staining revealed that ML264 treatment attenuated PA-induced renal TECs injury and interstitial fibrosis. Moreover, Oil Red O staining and IHC staining of PLIN2 demonstrated that ML264 treatment reduced lipid droplet storage in renal TECs of PA-treated mice. Furthermore, TUNEL staining, along with IHC staining of BAX/BCL2, and cleaved-caspase 3 indicated that ML264 treatment inhibited PA-induced mitochondrial dysfunction and apoptosis in renal TECs (Fig. 8G). WB analysis further confirmed that ML264 treatment effectively suppressed the PA-induced upregulation of adhesion and injury markers in renal TECs (Fig. 8H). Finally, Von Kossa staining further demonstrated that ML264 treatment significantly reduced CaOx crystal deposition in both the cortex and medulla (Fig. 8I).

Fig. 8.

Pharmacological inhibition of Krüppel-like factor 5 (KLF5) mitigates palmitic acid (PA)-induced lipotoxicity to reduce renal calcium oxalate (CaOx) crystal deposition. (A) Schematic diagram illustrating the animal experimental groups: chow diet + vehicle, chow diet + ML264, 7.5% PA diet + vehicle, and 7.5% PA diet + ML264. (B, C) Renal function assessment via serum creatinine (Scr) (B) and blood urea nitrogen (BUN) (C) levels in the indicated groups (n = 6). (D, E) Serum triglyceride (TG) (D) and total cholesterol (TC) (E) levels were measured after 24 h fasting in the indicated groups (n = 6). (F) Kidney weight-to-body weight (KW/BW) ratio in the indicated groups (n = 6). (G) Representative images and quantifications of periodic acid–Schiff (PAS) staining, Masson's trichrome staining, Oil Red O staining, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining, and immunohistochemistry (IHC) staining for KLF5, peroxisome proliferator activated receptor gamma (PPARγ), perilipin 2 (PLIN2), Bcl-2-associated X protein (BAX), B-cell lymphoma 2 (BCL2) and cleaved-caspase 3 in kidney sections in the indicated groups (n = 6). (H) Western blot (WB) analysis of adhesion and injury marker expression in kidney tissues in the indicated groups (n = 6). (I) Representative Von Kossa staining and quantification of CaOx crystal deposition in kidney sections in the indicated groups. Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎P < 0.05, ^⁎⁎P < 0.01, ^⁎⁎⁎P < 0.001, ^⁎⁎⁎⁎P < 0.0001. OPN: osteopontin; CD44: cluster of differentiation 44; NGAL: neutrophil gelatinase-associated lipocalin; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; DAPI: 4′,6-diamidino-2-phenylindole.

Together, these findings highlight that pharmacological inhibition of KLF5 effectively disrupts the KLF5–PPARγ axis, thereby attenuating PA-induced lipotoxicity and consequently reducing renal CaOx crystal deposition, underscoring KLF5 as a promising therapeutic target for lipid metabolism-associated CaOx stone.

3.9. The expression levels of KLF5 and PPARγ correlated with the elevation of serum PA concentration in patients with renal CaOx stone

In our previous study, we found that the serum PA concentration in renal CaOx stone patients was higher than that in non-stone patients [9]. To further investigate this, we selected 19 clinical samples and measured the serum PA concentration (Tables S1 and S2). This observation was confirmed in our current study. Using the median serum PA concentration among 11 patients with CaOx stone as a cutoff, we defined those with concentrations above the median as the “high PA” group (Fig. 9A). Interestingly, IHC staining in the 11 CaOx stone patients revealed that increasing serum PA concentration was associated with elevated expression levels of KLF5, PPARγ, PLIN2, the BAX/BCL2 ratio, and cleaved-caspase 3 (Figs. 9B–D). This finding further underscores the importance of the PA-induced KLF5-PPARγ pathway in promoting renal CaOx stone formation (Fig. 10).

Fig. 9.

Elevated serum palmitic acid (PA) levels correlate with increased Krüppel-like factor 5 (KLF5) and peroxisome proliferator activated receptor gamma (PPARγ) expression in renal calcium oxalate (CaOx) stone patients. (A) Serum PA levels in the control, low PA, and high PA groups. (B, C) Representative immunohistochemistry (IHC) staining images (B) and quantification (C) of KLF5, PPARγ, perilipin 2 (PLIN2), Bcl-2-associated X protein (BAX), B-cell lymphoma 2 (BCL2), and cleaved-caspase 3 in kidney sections in the indicated groups (n = 19). (D) Correlation analysis between serum PA levels and the expression of KLF5, PPARγ, PLIN2, cleaved-caspase 3, and the BAX/BCL2 ratio in patients with renal CaOx stone. Data are presented as mean ± standard error of the mean (SEM). N.S: not significant; ^⁎P < 0.05, ^⁎⁎P < 0.01.

Fig. 10.

Schematic illustration of the mechanism by which probucol alleviates palmitic acid (PA)-mediated renal calcium oxalate (CaOx) stone formation by targeting Krüppel-like factor 5 (KLF5)-peroxisome proliferator activated receptor gamma (PPARγ) axis-mediated lipotoxicity. This schematic shows how probucol attenuates PA-mediated renal CaOx stone formation by targeting KLF5-PPARγ axis. PA upregulates KLF5, which binds to the PPARγ promoter region and enhances its transcription, thereby increasing the expression of downstream target genes, leading to lipid storage, mitochondrial dysfunction, and apoptosis, thereby inducing lipotoxicity, subsequently promoting CaOx crystal adhesion and deposition. Probucol binds to arginine-440 (Arg440) in KLF5, blocking its DNA-binding activity to the PPARγ promoter, suppressing PPARγ transcription, attenuating lipotoxicity, and ultimately reducing CaOx crystal adhesion and deposition. Findings are supported by high-throughput screening, transcriptomics, functional assays, and clinical data in patients with elevated serum PA levels. Pro: probucol; FATP2: fatty acid transport protein 2; CD36: cluster of differentiation 36. Created with BioRender.com.

4. Discussion

The present study demonstrates that probucol significantly mitigates PA-induced lipotoxicity, thereby reducing renal CaOx stone formation. This finding provides novel pharmacological insights into the pathogenesis of CaOx stone and highlights the potential of probucol as a therapeutic agent for lipotoxicity-associated renal disorders. Through high-throughput drug screening and multi-level mechanistic validation, we confirmed that probucol effectively attenuates lipid overload–induced injury in renal TECs, underscoring the pivotal role of lipotoxicity in CaOx stone development.

CaOx stone formation is closely linked to dyslipidemia, particularly disruptions in fatty acid metabolism that cause lipid storage and oxidative stress [6,9]. Our RNA-seq and bioinformatics analysis revealed that PA exposure activated multiple signaling pathways related to lipid metabolism and inflammation, notably the “Lipid and atherosclerosis” and “IL-17 signaling” pathways. Interestingly, probucol markedly inhibited the aberrant activation of these pathways, suggesting that its renoprotective effects are mediated in part through modulation of lipid metabolism-related TFs, such as PPARγ.

PPARγ, a central regulator of lipid metabolism, has been implicated in lipid storage, cellular dysfunction, and inflammatory responses when aberrantly activated [[15], [16], [17],29,30,39]. Here, we demonstrate for the first time in renal TECs that KLF5–PPARγ signaling mediates PA-induced lipotoxicity, leading to mitochondrial dysfunction, apoptosis, and enhanced CaOx crystal adhesion and deposition. We further identify probucol as a pharmacological inhibitor of this axis, blocking KLF5 binding to the PPARγ promoter, suppressing PPARγ-driven pathogenic lipid metabolism, preserving mitochondrial integrity, and ultimately reducing crystal burden. In other biological contexts, KLF5 has been shown to transcriptionally regulate PPARγ—for example, in adipogenesis, KLF5 binds the PPARγ promoter and cooperates with CCAAT/enhancer-binding protein beta/delta(C/EBPβ/δ) to drive adipocyte differentiation, and in esophageal adenocarcinoma, it binds the PPARγ promoter and a distal enhancer to promote fatty acid synthesis [36,37]. Collectively, our findings extend the pathological scope of the KLF5–PPARγ axis beyond adipose tissue and cancer into PA-induced lipotoxicity that drives stone pathogenesis, and reveal a previously unrecognized therapeutic strategy for dyslipidemia-associated nephrolithiasis.

While our data confirm that PA exposure elevates KLF5 expression in TECs, the upstream regulatory mechanisms remain to be fully defined. Evidence from studies in multiple metabolic diseases indicates that saturated fatty acids, including PA, can activate inflammatory and stress-responsive pathways—such as mitogen-activated protein kinase (MAPK) (p38/JNK) and endoplasmic reticulum (ER) stress–activating transcription factor 4 (ATF4)/C/EBP Homologous Protein (CHOP) cascades—that can transcriptionally induce KLF5 [40,41]. In addition, PA has been reported to stimulate specificity protein 1 (SP1) and early growth response 1 (EGR1), both of which enhance KLF5 promoter activity in certain cell types [[42], [43], [44]]. These observations suggest that PA-induced KLF5 upregulation may reflect the combined influence of multiple signaling pathways.

Beyond the mechanistic framework, the clinical implications of these findings warrant careful consideration, particularly regarding the potential of probucol as a therapeutic intervention. From a translational perspective, our findings are strongly supported by clinical data: serum PA levels were significantly elevated in patients with renal CaOx stone compared to healthy controls, and higher PA levels were accompanied by increased expression of both KLF5 and PPARγ. These observations reinforce the clinical relevance of the KLF5–PPARγ axis in stone pathogenesis and provide a solid rationale for exploring probucol as a therapeutic intervention. In its approved cardiovascular indications, probucol is typically administered at 250 mg twice daily, a regimen generally well tolerated but associated with a risk of QT interval prolongation, particularly in susceptible individuals. Therefore, any potential repurposing for nephrolithiasis would require careful baseline and follow-up electrocardiogram monitoring, along with avoidance of concomitant QT-prolonging medications [[45], [46], [47]]. Another consideration is its effect on plasma high-density lipoprotein cholesterol (HDL-C) via modulation of ATP-binding cassette sub-family A member 1 (ABCA1)-mediated cholesterol efflux. Although prior studies suggest this may not impair—and may even enhance—HDL's antioxidant function, its long-term impact on systemic lipid homeostasis in stone-forming patients warrants evaluation [46,48]. Finally, probucol's high lipophilicity and prolonged half-life could enable sustained renal protection but necessitate dosing adjustments and periodic monitoring of lipid profile, liver enzymes, and cardiac rhythm during extended therapy [49,50]. Collectively, these considerations underscore the need for well-designed clinical trials to balance efficacy and safety, thereby facilitating the bench-to-bedside translation of probucol for recurrent CaOx stone.

5. Conclusions

In summary, this study not only demonstrates the therapeutic potential of probucol in attenuating PA-induced lipotoxicity, thereby reducing renal CaOx crystal adhesion and deposition, but also elucidates the underlying molecular mechanism involving regulation of lipid metabolism via the KLF5–PPARγ axis. By restoring mitochondrial integrity, suppressing apoptosis, and reprogramming aberrant lipid metabolism, probucol emerges as a promising therapeutic strategy for renal diseases associated with lipotoxicity. Specifically, in the context of renal CaOx stone disease, probucol shows strong mechanistic support and translational potential.

However, although this study highlights the therapeutic potential of probucol in attenuating renal CaOx stone formation, further clinical trials and long-term follow-up studies are essential to validate its efficacy and safety. Additionally, the potential applicability of probucol in other lipotoxicity-related renal disorders warrants exploration. Future research should aim to determine the optimal dosage, treatment duration, as well as explore its potential synergistic effects with other therapeutic agents, to facilitate the development of more precise and personalized treatment strategies for renal CaOx stone.

CRediT authorship contribution statement

Haotian Ren: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Yutong Chen: Writing – original draft, Project administration, Methodology, Investigation, Formal analysis, Data curation. Yang Li: Writing – original draft, Methodology, Investigation, Formal analysis, Conceptualization. Jingdong Zhang: Validation, Investigation, Formal analysis. Yi Shao: Validation, Methodology, Investigation, Formal analysis, Data curation. Bangxian Yu: Validation, Investigation, Conceptualization. Linguo Xie: Validation, Investigation, Conceptualization. Shiyong Qi: Validation, Investigation, Conceptualization. Rui Wang: Writing – review & editing, Supervision, Formal analysis. Zhiqun Shang: Writing – review & editing, Writing – original draft, Validation, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Chunyu Liu: Writing – review & editing, Writing – original draft, Validation, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: