Degradable MMP9–responsive high-density lipoprotein nanoparticles enable efficient siRNA-PGRN and quercetin delivery to remodel the fibrotic microenvironment and attenuate renal fibrosis

Abstract

Renal fibrosis is the final common pathway of chronic kidney disease and remains a challenge due to the complex fibrotic and inflammatory microenvironment. This study aimed to develop an enzyme-responsive, fibrosis-targeted nanotherapeutic system capable of simultaneously delivering siRNA against progranulin (siPGRN) and quercetin (QT) to remodel the fibrotic microenvironment and effectively attenuate renal fibrosis. The TIMP-HDL-Nano@siPGRN&QT nanoparticles prepared in this study exhibit uniform nanoscale size (~ 100 nm), high encapsulation efficiency, excellent stability, and favorable biocompatibility. The MMP9-responsive modification significantly enhanced renal accumulation and fibrotic lesion targeting in vivo. Treatment with TIMP-HDL-Nano@siPGRN&QT markedly alleviated UUO-induced renal injury, apoptosis, and fibrosis, outperforming free drug. Transcriptomic and mechanistic analyses revealed that the therapeutic effects were mediated through disruption of the PGRN-PPARα interaction, restoration of PPARα signaling, and subsequent inhibition of NF-κB-driven inflammatory pathways. Furthermore, the nanoparticles suppressed pro-inflammatory M1 macrophage polarization and promoted M2 polarization in a PPARα-dependent manner. This study demonstrates that TIMP-HDL-Nano@siPGRN&QT enables efficient, targeted co-delivery of siRNA and small-molecule therapeutics to fibrotic kidneys. By remodeling the fibrotic immune microenvironment through the PGRN–PPARα–NF-κB axis, this nanoplatform provides a potent and mechanistically defined strategy for the treatment of renal fibrosis and offers translational potential for other inflammation-driven fibrotic diseases.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12951-026-04773-9.

Background

Chronic kidney disease (CKD) is a major global public health burden, affecting approximately 10–13% of the adult population worldwide and contributing to increasing morbidity, mortality, and healthcare costs [1, 2]. Irrespective of etiology, progressive CKD is invariably associated with renal fibrosis, a defining pathological feature characterized by excessive extracellular matrix deposition, tubular atrophy, and persistent inflammation, ultimately leading to irreversible renal failure and end-stage renal disease [3]. Although significant advances have been made in understanding the molecular mechanisms of renal fibrosis, effective disease-modifying therapies remain limited, and current treatments largely delay rather than reverse fibrotic progression, underscoring the urgent need for innovative strategies that target the fibrotic microenvironment and multiple pathogenic pathways.

Progranulin (PGRN), a multifunctional secreted glycoprotein, has emerged as an important regulator of inflammation, immunity, and tissue remodeling [4]. Recent studies suggest that PGRN is upregulated in renal injury and fibrosis and participates in inflammatory signaling and macrophage activation [5, 6]; however, its precise role in renal fibrotic progression remains controversial and context dependent. Targeted suppression of PGRN expression using small interfering RNA (siRNA) therefore represents a powerful strategy to elucidate its pathological function and achieve therapeutic benefit. Nevertheless, the clinical translation of siRNA-based therapies is severely hindered by poor stability, limited cellular uptake, and lack of tissue specificity [7, 8], underscoring the need for efficient and biocompatible delivery systems. In parallel with gene-silencing approaches, pharmacological modulation of inflammatory and fibrotic signaling using bioactive small molecules has also shown promise in experimental renal fibrosis [9, 10]. Quercetin, a naturally occurring flavonoid, has attracted considerable attention due to its anti-inflammatory, antioxidant, and anti-fibrotic properties [11, 12]. Experimental studies have demonstrated that quercetin attenuates renal fibrosis by inhibiting profibrotic pathways, including transforming growth factor-β (TGF-β) and nuclear factor-κB (NF-κB) [13, 14]. However, its poor aqueous solubility, rapid systemic clearance, and low renal accumulation substantially limit its therapeutic efficacy in vivo [15], highlighting the need for delivery platforms that enhance its bioavailability and enable targeted renal deposition.

Nanomedicine-based drug delivery systems provide unique opportunities to overcome these challenges by improving pharmacokinetics, protecting therapeutic cargos, and enabling tissue- or disease-specific targeting [16]. Among various nanocarriers, high-density lipoprotein (HDL)–based nanoparticles are particularly attractive due to their excellent biocompatibility, intrinsic kidney affinity, and ability to transport both hydrophobic small molecules and nucleic acids [17]. Reconstituted HDL nanoparticles mimic the structure and function of endogenous HDL, allowing efficient cellular uptake via scavenger receptors while minimizing systemic toxicity [17]. Importantly, HDL-based nanocarriers can be readily engineered to incorporate functional peptides or stimuli-responsive elements to achieve site-specific drug release [18]. Matrix metalloproteinase-9 (MMP9) is markedly upregulated in fibrotic kidneys and contributes to ECM remodeling, inflammatory cell recruitment, and disease progression [19]. Exploiting elevated MMP9 activity as an endogenous trigger for nanoparticle degradation and drug release represents a rational strategy to achieve fibrosis-specific targeting. Tissue inhibitor of metalloproteinases (TIMP)–derived peptides that are selectively cleaved by MMP9 offer a promising means to construct enzyme-responsive nanoplatforms capable of releasing therapeutic cargos specifically within the fibrotic microenvironment.

In this context, we developed a degradable, MMP9-responsive HDL-based nanotherapeutic system co-loading siRNA targeting PGRN (siPGRN) and QT, termed TIMP-HDL-Nano@siPGRN&QT, to overcome the inherent limitations of siRNA- and small-molecule–based therapies in renal fibrosis, while elucidating the pathological role of PGRN in fibrotic progression by enabling targeted co-delivery of siRNA against PGRN and quercetin to fibrotic kidneys, thereby remodeling the MMP9-enriched fibrotic microenvironment and attenuating renal fibrosis.

Methods

TIMP-HDL-Nano@siPGRN&QT synthesis

TIMP-HDL-Nano@siPGRN&QT nanoparticles were prepared via a stepwise self-assembly and surface modification strategy based on reconstituted HDL. Phospholipids and cholesterol were co-dissolved with QT in chloroform/methanol (2:1, v/v), and solvent evaporation was used to form a thin lipid film. The film was hydrated with RNase-free buffer containing siRNA targeting progranulin (siPGRN), followed by sonication and gentle stirring to induce HDL self-assembly, yielding HDL-Nano@siPGRN&QT with QT encapsulated in the lipid core and siPGRN loaded within the HDL corona. For fibrosis-targeted and enzyme-responsive modification, a TIMP-derived MMP9-sensitive peptide was conjugated to the nanoparticle surface using EDC/NHS chemistry and incubated with HDL-Nano@siPGRN&QT under mild conditions to form TIMP-HDL-Nano@siPGRN&QT. Excess reagents were removed by ultrafiltration. The nanoparticles were stored at 4 °C prior to use. Detailed synthesis process was presented in Fig. 1A.

Fig. 1.

Material Synthesis and Modeling Process Flowchart. (A) Schematic illustration of the fabrication process of the multifunctional HDL nanocomplex. (B) Schematic illustration of the in vivo experimental timeline

Dynamic light scattering (DLS)

DLS was used to evaluate the hydrodynamic size distribution and uniformity of HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT. Nanoparticles were dispersed in filtered deionized water prior to measurement, and particle size was determined based on light scattering intensity fluctuations caused by Brownian motion. Measurements were performed using a Zetasizer Nano ZS analyzer (Malvern Instruments, UK) at 25 °C and a scattering angle of 173°. Particle size distributions and polydispersity indices were calculated using the cumulant analysis method, with all measurements conducted in triplicate.

Transmission electron microscopy (TEM)

TEM was used to analyze the morphology and ultrastructural features of the synthesized nanoparticles. HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT samples were diluted with deionized water, deposited onto carbon-coated copper grids, and negatively stained with 2% (w/v) phosphotungstic acid prior to imaging. TEM observation was performed using a transmission electron microscope operated at an accelerating voltage of 80–120 kV.

Zeta potential

Zeta potential measurements were performed to assess the surface charge and colloidal stability of the synthesized nanoparticles and to evaluate the effects of surface modification. HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT were dispersed in filtered deionized water at an appropriate concentration and analyzed using a Zetasizer Nano ZS analyzer (Malvern Instruments, UK) at 25 °C.

X-ray photoelectron spectroscopy (XPS)

XPS was used to analyze the surface elemental composition and chemical states of HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT. Lyophilized samples were analyzed using an X-ray photoelectron spectrometer with a monochromatic Al Kα source under ultra-high vacuum conditions. Survey spectra were collected to identify elemental components (C, O, N, S, and P), and high-resolution spectra of C 1 s, N 1 s, O 1 s, and S 2p were recorded to examine chemical bonding states. Binding energies were calibrated to the C 1 s peak at 284.8 eV. XPS analysis confirmed the presence of siRNA, QT, and TIMP peptide components, verifying successful nanoparticle assembly and surface functionalization.

Fourier transform infrared (FTIR) spectroscopy

FTIR spectroscopy was used to analyze the chemical composition and functional groups of HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT. Lyophilized samples were mixed with KBr and compressed into pellets, and spectra were recorded over 4000–400 cm⁻¹ at a resolution of 4 cm⁻¹. Characteristic absorption bands of HDL components, quercetin, and TIMP peptide were analyzed to confirm successful drug loading and surface modification. FTIR spectra were compared between nanoparticle formulations to verify their structural integrity and multifunctional composition.

Encapsulation efficiency (EE%)

The EE% of siPGRN and QT in HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT was determined to evaluate nanoparticle loading performance. Free siPGRN was separated by centrifugal ultrafiltration and quantified using a fluorescence-based nucleic acid assay, while QT content was measured by UV–vis spectrophotometry after nanoparticle disruption. EE% was calculated as [(total amount − free amount)/total amount] × 100%, with all measurements performed in triplicate and expressed as mean ± standard deviation. The EE% values were as follows: For HDL-Nano@siPGRN&QT: siPGRN EE% = 88.3 ± 3.5%, QT EE% = 82.6 ± 4.1%, For TIMP-HDL-Nano@siPGRN&QT: siPGRN EE% = 87.1 ± 3.9%, QT EE% = 81.2 ± 4.5%, No significant differences were observed between the two formulations (p > 0.05), indicating that TIMP surface modification did not compromise siPGRN or QT encapsulation efficiency. This analysis confirmed that TIMP surface modification did not compromise siPGRN or QT encapsulation efficiency.

Observation of the siPGRN marker in the cytoplasm

Encapsulated siPGRN was labeled with Cy5 fluorescent dye. Cellular F-actin cytoskeleton was stained by Alexa Fluor 488-conjugated Phalloidin (Thermo Fisher Scientific, Cat. No. A12379). HK2 cells were co-cultured with Cy5-tagged nanocarrier-siPGRN complexes, and collected and fixed at 3 h, 6 h, 9 h, 24 h, 36 h and 48 h. Following permeabilization and blocking procedures, cells were incubated with phalloidin staining solution. The dynamic intracellular distribution of labeled siPGRN was observed and captured using fluorescence microscope.

Fluorescence semi-quantification of in vivo organ distribution

Fluorescence semi-quantification corresponding to the organ fluorescence images in Fig. 4F was performed after tail-vein administration of DiR-labeled TIMP-HDL-Nano@siPGRN&QT in the UUO mouse model. Major organs, including heart, liver, spleen, lung, and kidney, were harvested at 0.5 h, 2 h, 8 h, and 12 h, and ex vivo fluorescence images were acquired under identical exposure settings. Region-of-interest fluorescence intensity was quantified using ImageJ and normalized to the same acquisition parameters. Quantitative graphs were generated from three mice per time point and are presented as the mean ± SD. Pearson’s correlation coefficient-based colocalization analysis was performed to quantify the spatial overlap between DiR-labeled nanoparticles and MMP9 and α-SMA immunofluorescence in renal fibrotic regions. Regions of interest were selected from comparable cortical and tubulointerstitial fields, background fluorescence was subtracted, and Pearson’s coefficients were calculated using the Coloc 2 plugin in ImageJ/Fiji. At least three randomly selected fields from each biological replicate were analyzed, and the values were summarized as the mean ± SD, n = 6.

Fig. 4.

TIMP-HDL-Nano@siPGRN&QT Alleviates Renal Fibrosis in PGRN-Deficient UUO Mice. (A) H&E staining showing pathological relief in the TIMP-HDL-Nano@siPGRN&QT group, with partial improvement in PGRN⁻/⁻ Model mice. (B) Masson staining showing reduced collagen deposition in PGRN⁻/⁻ Model mice, further decreased by quercetin, and most markedly reduced by TIMP-HDL-Nano@siPGRN&QT. (C) TUNEL staining showing increased apoptosis in UUO kidneys, partially reduced by quercetin and most strongly suppressed by TIMP-HDL-Nano@siPGRN&QT. (D–F) IHC staining of α-SMA, COL1A1, and MMP9 showing progressive downregulation from UUO to PGRN⁻/⁻ Model, quercetin-treated, and TIMP-HDL-Nano@siPGRN&QT groups. (G, H) Western blot analysis showing α-SMA, MMP9, and TGF-β expression highest in UUO mice and lowest in the TIMP-HDL-Nano@siPGRN&QT group. *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant

HK2 cell in vitro uptake assay

HK2 cells were incubated with the indicated fluorescently labeled nanocarrier formulations, including HDL-Nano@siPGRN&QT, TIMP-Nano@siPGRN&QT, and TIMP-HDL-Nano@siPGRN&QT, to evaluate in vitro cellular uptake and intracellular persistence. Cells were imaged at 24 h, 36 h, 48 h, and 72 h, with longer-term observation at 1 d, 7 d, and 14 d where indicated. Nuclei were counterstained with DAPI and the cytoskeletal outline was visualized using phalloidin staining when applicable. Fluorescence intensity and intracellular distribution were assessed from at least three representative microscopic fields per condition using ImageJ.

Time-course serum and tissue concentration analysis

To characterize the metabolic and clearance-related profile of the nanocarrier after administration, serum, liver, spleen, and kidney samples were collected at four predefined time points: 1 d, 3 d, 7 d, and 14 d after intravenous injection. Serum was obtained after blood lysis, and tissue samples were homogenized in an equal volume of lysis buffer. Fluorescence intensity was measured using the same excitation settings used for the labeled nanocarrier and normalized to serum volume or tissue weight. Time-course concentration curves were generated from three biological replicates per time point and are shown as the mean ± SD.

Liver and kidney function biochemical assays

Liver and kidney function were assessed using serum biochemical assays at four predefined time points: 1 d, 3 d, 7 d, and 14 d during the in vivo metabolic evaluation. Alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), and serum creatinine (Cr) were measured using commercial colorimetric biochemical kits according to the manufacturers’ protocols. Measurements were obtained from six C57 mice per group and are reported as the mean ± SD.

Histopathological analysis of major organs

Major organs, including heart, liver, spleen, lung, and kidney, were collected at three predefined time points: 1 d, 7 d, and 14 d for pathological section analysis. Tissues were fixed in 4% paraformaldehyde, embedded in paraffin, sectioned at 4 microm, and stained with hematoxylin and eosin. Histopathological changes, including inflammatory-cell infiltration, necrosis, hemorrhage, tubular injury, and architectural disruption, were evaluated under light microscopy using comparable magnification and imaging settings.

Animal modeling and grouping

For in vivo experiments, a unilateral ureteral obstruction (UUO)–induced renal fibrosis mouse model was established to evaluate the therapeutic efficacy of TIMP-HDL-Nano@siPGRN&QT (TIMP-HDL-Nano@siPGRN&QT nanoparticles were administered via tail vein injection at 10 mg/kg (0.2 mg per 20 g mouse). The nanoparticles contained 30 μg siPGRN and 60 μg quercetin per milligram of particles, yielding 6 μg (0.46 nmol) siPGRN and 12 μg quercetin per mouse per injection. Mice were treated once every day for 5 days. The cumulative doses were 30 μg (2.3 nmol) for siPGRN, 60 μg for quercetin, and 1.0 mg for nanoparticles per mouse). Male C57BL/6 mice (8–10 weeks old) and age-matched progranulin-deficient (PGRN⁻/⁻) mice were randomly assigned to different experimental groups. To induce renal fibrosis, UUO surgery was performed by ligating and transecting the left ureter under anesthesia, as previously described. Sham-operated mice underwent identical surgical procedures without ureteral ligation. Following surgery, mice were randomly divided into the following groups: (1) Control: sham-operated mice without treatment; (2) PGRN⁻/⁻ WT: PGRN⁻/⁻ mice without UUO; (3) Model: UUO mice without treatment; (4) PGRN⁻/⁻ Model: PGRN⁻/⁻ UUO mice; (5) PGRN⁻/⁻ Model + QT: PGRN⁻/⁻ UUO mice treated with quercetin; and (6) Model + TIMP-HDL-Nano@siPGRN&QT: UUO mice treated with TIMP-HDL-Nano@siPGRN&QT nanoparticles. In selected mechanistic studies, additional groups receiving nanoparticle treatment combined with PPARα neutralization or activation, NF-κB inhibition, or lipopolysaccharide (LPS) stimulation were included according to the experimental design. Nanoparticles or control formulations were administered via tail vein injection at defined time points after UUO surgery. Mice were monitored throughout the experimental period and sacrificed at predetermined endpoints, after which kidney tissues were collected for histological, molecular, and transcriptomic analyses. The experimental timeline was presented in Fig. 1B.

Cell culture and grouping

Human renal proximal tubular epithelial cells (HK-2) and human monocytic THP-1 cells were used for in vitro experiments. HK-2 cells were cultured in DMEM/F-12 supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin, while THP-1 cells were maintained in RPMI-1640 medium containing 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C in a humidified atmosphere with 5% CO₂. Upon reaching 80–90% confluence, cells were passaged and used for subsequent experiments. THP-1 monocytes were differentiated into macrophage-like cells by treatment with phorbol 12-myristate 13-acetate (PMA) prior to experimental use. For in vitro biosafety and uptake assessments, HK-2 cells were divided into the following groups: (1) Control; (2) Nano@siPGRN&QT; (3) HDL-Nano@siPGRN&QT; (4) TIMP-Nano@siPGRN&QT; and (5) TIMP-HDL-Nano@siPGRN&QT. For gene-silencing and mechanistic studies, HK-2 cells were treated with the indicated nanoparticle formulations for defined time points. For macrophage polarization and co-culture experiments, LPS-stimulated HK-2 cells were co-cultured with THP-1–derived macrophages. The experimental groups included: (1) Control; (2) LPS; (3) LPS + TIMP-HDL-Nano@siPGRN&QT; and (4) LPS + TIMP-HDL-Nano@siPGRN&QT combined with PPARα or NF-κB modulation, as specified in the experimental design. This grouping strategy enabled systematic evaluation of nanoparticle-mediated regulation of renal tubular injury, macrophage polarization, and inflammatory signaling in vitro.

CCK-8 assay

Cell viability was assessed using the CCK-8 assay to evaluate the cytocompatibility of the nanoparticles. HK-2 cells were seeded in 96-well plates and allowed to adhere for 24 h, followed by treatment with Control, Nano@siPGRN&QT, HDL-Nano@siPGRN&QT, TIMP-Nano@siPGRN&QT, or TIMP-HDL-Nano@siPGRN&QT for 24, 48, and 72 h. At each time point, 10 μL of CCK-8 reagent was added and incubated for 2 h at 37 °C. Absorbance at 450 nm was measured, and cell viability was expressed as a percentage of the control. All experiments were performed in triplicate.

EdU proliferation assay

Cell proliferation was evaluated using an 5-ethynyl-2′-deoxyuridine (EdU) incorporation assay in accordance with the manufacturer’s instructions. HK-2 cells or THP-1–derived macrophages were seeded onto 24-well plates or glass coverslips and allowed to adhere overnight. Cells were then exposed to the indicated treatments, including nanoparticle formulations, LPS, and/or pathway modulators as required. Subsequently, EdU was added to the culture medium and incubated at 37 °C for 2 h to label newly synthesized DNA. Cells were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X-100, and subjected to fluorescent EdU staining, with nuclei counterstained using DAPI. EdU-positive cells were visualized by fluorescence microscopy, and proliferation was quantified as the percentage of EdU-positive cells relative to total nuclei. All experiments were performed in triplicate.

In vivo fluorescence imaging

In vivo fluorescence imaging was used to evaluate the biodistribution and renal accumulation of the synthesized nanoparticles. HDL-based nanoparticles were labeled with a near-infrared lipophilic dye (DiD or DiR) and intravenously injected into UUO model mice via the tail vein at equivalent doses. Whole-body fluorescence imaging was performed at 15 min, 1 h, 2 h, 8 h, and 24 h post-injection using an in vivo imaging system, with mice anesthetized by isoflurane during imaging. Fluorescence intensity and distribution were recorded to assess nanoparticle dynamics in vivo. For ex vivo analysis, major organs, including the heart, liver, spleen, lung, kidney, and brain, were collected at 30 min, 2 h, 8 h, and 12 h, and imaged under identical conditions. Fluorescence signals were quantified to compare organ-specific distribution among treatment groups.

Immunofluorescence (IF) staining

IF staining was performed to assess protein expression, cellular localization, cytoskeletal organization, macrophage polarization, and nanoparticle distribution in renal tissues and cultured cells. Kidney tissues were collected at designated time points, fixed with 4% paraformaldehyde, embedded in OCT, and sectioned. Sections were permeabilized, blocked with BSA, and incubated overnight at 4 °C with primary antibodies against MMP9, α-SMA, COL1A1, F4/80, CD86, and CD206, followed by fluorophore-conjugated secondary antibodies and DAPI nuclear counterstaining. F-actin was visualized using fluorescent phalloidin, and THP-1–derived macrophages were pre-labeled with carboxyfluorescein succinimidyl ester (CFSE) for uptake analysis. For in vitro studies, treated HK-2 cells or macrophages were stained using the same protocol. Images were captured by fluorescence or confocal microscopy, and fluorescence intensity and co-localization were quantified using ImageJ software.

H&E staining

Hematoxylin and eosin (H&E) staining was performed to evaluate renal histopathological changes following different treatments. Kidney tissues were harvested at designated time points, fixed in 4% paraformaldehyde, dehydrated through a graded ethanol series, embedded in paraffin, and sectioned into 4–5 μm thick slices. Tissue sections were deparaffinized, rehydrated, and sequentially stained with hematoxylin to visualize cell nuclei, followed by eosin to counterstain the cytoplasm and extracellular matrix. After staining, sections were dehydrated, cleared, and mounted for microscopic examination. Histological images were captured using a light microscope under identical imaging conditions.

Masson staining

Masson’s trichrome staining was performed to evaluate collagen deposition and the extent of renal interstitial fibrosis. Kidney tissues were collected at designated time points, fixed in 4% paraformaldehyde, dehydrated through a graded ethanol series, embedded in paraffin, and sectioned into 4–5 μm thick slices. Paraffin sections were deparaffinized and rehydrated, followed by Masson’s trichrome staining according to the manufacturer’s protocol. Briefly, sections were sequentially stained to differentiate cell nuclei, cytoplasm, and collagen fibers, with collagen visualized as blue-stained areas. After staining, sections were dehydrated, cleared, and mounted for microscopic observation. Images were captured using a light microscope under identical settings.

TUNEL staining

Terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) staining was performed to assess apoptotic cell death in renal tissues after different treatments. Kidneys were collected at designated time points, fixed in 4% paraformaldehyde, paraffin-embedded, and sectioned into 4–5 μm slices. Sections were deparaffinized, rehydrated, and stained using a commercial TUNEL apoptosis detection kit following the manufacturer’s protocol. Briefly, sections were treated with proteinase K and incubated with terminal deoxynucleotidyl transferase and labeled dUTP to detect DNA fragmentation. Nuclei were counterstained with DAPI, and TUNEL-positive cells were visualized by fluorescence microscopy. Apoptosis was quantified as the percentage of TUNEL-positive nuclei among total nuclei in randomly selected fields.

Immunohistochemical (IHC) staining

IHC staining was conducted to assess the expression and localization of fibrosis- and inflammation-related proteins in renal tissues. Kidneys were collected at designated time points, fixed in 4% paraformaldehyde, paraffin-embedded, and sectioned into 4–5 μm slices. Sections were deparaffinized, rehydrated, and subjected to heat-induced antigen retrieval in citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked with 3% hydrogen peroxide, followed by blocking with 5% bovine serum albumin. Sections were incubated overnight at 4 °C with primary antibodies against α-SMA, COL1A1, MMP9, and other indicated targets, then treated with HRP-conjugated secondary antibodies. Immunoreactive signals were developed using DAB and counterstained with hematoxylin. Images were acquired under identical conditions, and staining intensity was quantitatively analyzed using ImageJ software.

Gene editing

Gene editing and gene-silencing approaches were employed to investigate the mechanistic roles of PGRN and PPARα in nanoparticle-mediated regulation of renal fibrosis and inflammatory signaling. For in vitro studies, siRNAs targeting PPARα or negative control siRNA (siNC) were synthesized commercially. HK-2 cells were seeded to reach 60–70% confluence and transfected with siRNAs using a liposome-based transfection reagent according to the manufacturer’s protocol. After 6 h of transfection, the medium was replaced with fresh complete medium, and cells were further treated with nanoparticle formulations or other indicated stimuli for subsequent analyses. The efficiency of gene silencing was verified by quantitative real-time PCR and Western blotting prior to downstream experiments. Only siRNA sequences achieving effective knockdown were used for functional studies. For in vivo experiments, PGRN⁻/⁻ mice were used as a genetic loss-of-function model to assess the contribution of PGRN to renal fibrotic progression and therapeutic responses.

Co-Immunoprecipitation (Co-IP)

Co-IP assays were conducted to investigate the interaction between progranulin (PGRN) and peroxisome proliferator–activated receptor α (PPARα). Renal tissues or cultured cells were lysed in ice-cold IP lysis buffer containing protease and phosphatase inhibitors, and lysates were clarified by centrifugation at 12,000 × g for 15 min at 4 °C. Protein concentrations were measured using a BCA assay. Equal amounts of protein were incubated overnight at 4 °C with antibodies against PGRN or PPARα, with IgG serving as a negative control, followed by incubation with protein A/G agarose beads. After extensive washing, immunocomplexes were eluted in SDS loading buffer and analyzed by Western blotting. Co-IP was performed using wild-type and PGRN⁻/⁻ samples to confirm interaction specificity and PGRN dependency.

Transcriptomic sequencing and bioinformatic analysis

Transcriptomic sequencing was performed to elucidate the molecular mechanisms underlying the therapeutic effects of TIMP-HDL-Nano@siPGRN&QT on renal fibrosis. Kidney tissues were harvested from UUO model mice and corresponding treatment groups at defined time points. Total RNA was extracted using a commercial RNA isolation kit according to the manufacturer’s protocol. RNA concentration and purity were assessed using a NanoDrop spectrophotometer, and RNA integrity was evaluated to ensure suitability for sequencing. High-quality RNA samples were used for library construction, followed by high-throughput RNA sequencing on an Illumina platform. Raw sequencing data were processed to remove low-quality reads and adapters, and clean reads were aligned to the reference mouse genome. Gene expression levels were quantified and normalized, and differentially expressed genes (DEGs) between groups were identified based on predefined statistical thresholds. Bioinformatic analyses were conducted to interpret transcriptomic changes. Hierarchical clustering and principal component analysis were used to evaluate global expression patterns. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment and Gene Ontology (GO) analyses were performed to identify significantly enriched biological processes and signaling pathways. These analyses were applied to uncover key pathways associated with inflammation, immune regulation, and fibrosis, thereby providing mechanistic insight into nanoparticle-mediated remodeling of the fibrotic microenvironment.

Flow cytometry

Flow cytometry was used to assess nanoparticle uptake, apoptosis, and macrophage polarization in vitro. For uptake analysis, THP-1–derived macrophages were pre-labeled with CFSE and incubated with PE-labeled nanoparticles for indicated time points, followed by washing and analysis. Apoptosis was evaluated using an Annexin V–FITC/propidium iodide (PI) staining kit to quantify early and late apoptotic cells. For polarization analysis, macrophages were stained with fluorophore-conjugated antibodies against CD86 (M1) and CD206 (M2) after different treatments. Stained cells were analyzed using a flow cytometer, and data were processed with FlowJo software. All flow cytometry experiments were performed in triplicate to ensure reproducibility.

qRT-PCR

qRT-PCR was performed to assess PGRN mRNA expression in renal tissues and cultured cells after different treatments. Total RNA was extracted using a commercial RNA isolation kit, and RNA concentration and purity were measured with a NanoDrop spectrophotometer. Equal amounts of RNA were reverse-transcribed into cDNA using a reverse transcription kit. qRT-PCR was conducted with SYBR Green Master Mix on a real-time PCR system using PGRN-specific primers, with GAPDH as the internal control. All reactions were performed in triplicate, and relative PGRN expression levels were calculated using the 2^{^−ΔΔCt} method and normalized to the control group.

Western blot

Western blotting was performed to analyze the expression of fibrosis-, inflammation-, and signaling-related proteins in renal tissues and cultured cells. Kidney samples or cells were harvested at designated time points and lysed in ice-cold RIPA buffer supplemented with protease and phosphatase inhibitor cocktails. Lysates were centrifuged at 12,000 × g for 15 min at 4 °C to remove debris, and protein concentrations were determined using a BCA protein assay kit. Equal amounts of total protein were separated by SDS–PAGE and transferred onto polyvinylidene difluoride (PVDF) membranes. Membranes were blocked with 5% nonfat milk or bovine serum albumin in Tris-buffered saline containing 0.1% Tween-20 (TBST) and then incubated overnight at 4 °C with primary antibodies against target proteins, including α-SMA, COL1A1, MMP9, TGF-β, PPARα, NF-κB pathway components (P65, p-P65, P50, p-P50), CD86, CD206, iNOS and β-actin as loading controls. After incubation with appropriate horseradish peroxidase–conjugated secondary antibodies, protein bands were visualized using enhanced chemiluminescence reagents. Band intensities were quantified using ImageJ software and normalized to the corresponding internal controls.

GST pull-down and PPARα luciferase reporter assays

GST pull-down analysis was performed to examine the interaction between PGRN and PPARα using purified GST-tagged PGRN and His-tagged PPARα. GST-PGRN or GST control protein was immobilized on glutathione agarose beads and incubated with His-PPARα under native binding conditions. After extensive washing, bound proteins were eluted and analyzed by Western blotting using anti-His and anti-GST antibodies, with input and GST-only lanes included as controls. PPARα transcriptional activity was further evaluated using a PPARα-responsive luciferase reporter assay. Cells were co-transfected with the reporter plasmid and Renilla control plasmid together with PGRN overexpression, siPGRN, or the corresponding control constructs as indicated, and firefly luciferase activity was normalized to Renilla luciferase activity.

Statistical analysis

All experimental data are presented as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). Comparisons between two groups were conducted using unpaired Student’s t-tests, while multiple-group comparisons were analyzed by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test followed by appropriate post hoc tests. For experiments involving two independent variables, two-way ANOVA was applied where applicable. A value of P < 0.05 was considered statistically significant. All experiments were repeated independently at least three times to ensure reproducibility.

Results

Characterization and physicochemical evaluation of HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT

HDL-Nano@siPGRN&QT nanoparticles were successfully fabricated, and their physicochemical features, including those following TIMP modification, were systematically assessed. DLS results showed both HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT possessed a narrow size distribution with diameters around ~ 100 nm (PDI: 0.12 for HDL-Nano@siPGRN&QT and 0.15 for TIMP-HDL-Nano@siPGRN&QT), indicating that TIMP conjugation did not markedly affect particle size (Fig. 2A). Zeta potential measurements revealed a further reduction in surface charge after TIMP modification, suggesting enhanced colloidal stability (Fig. 2B). TEM imaging confirmed spherical HDL morphology with a clear core–shell structure, while TIMP-functionalized nanoparticles exhibited slightly increased aggregation yet maintained overall structural integrity (Fig. 2C). FTIR spectra showed characteristic HDL amide and –OH/–NH peaks, with additional absorption at ~ 1450 cm⁻¹ corresponding to quercetin; TIMP-HDL-Nano@siPGRN&QT further presented a new band near 1680 cm⁻¹ associated with TIMP/MMP9, confirming successful functional incorporation (Fig. 2D). XPS analysis detected C, O, N, S, and P elements consistent with HDL components, siRNA phosphate groups, sulfur-containing residues, and quercetin. High-resolution C 1 s, N 1 s, O 1 s, and S 2p spectra verified the chemical environments expected for nucleic acid and protein conjugation (Fig. 2E–H), and the full-scan spectrum further supported complete nanoparticle assembly (Fig. 2I). Overall, these findings confirm that HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT nanoparticles exhibit stable architecture, successful multifunctional loading, and favorable physicochemical properties suitable for renal fibrosis–targeted therapy.

Fig. 2.

Characterization of HDL-Nano@siPGRN&QT and TIMP-HDL-Nano@siPGRN&QT. (A) DLS analysis showing similar narrow size distributions with average diameters of ~ 100 nm. (B) Zeta potential demonstrating reduced surface charge after TIMP modification. (C) TEM images revealing spherical HDL morphology with intact core–shell structures and slight aggregation after TIMP modification. (D) FTIR spectra confirming characteristic HDL, quercetin, and TIMP/MMP9-related absorption peaks. (E–H) High-resolution XPS spectra (C 1 s, N 1 s, O 1 s, S 2p) indicating successful incorporation of HDL components, siRNA, quercetin, and TIMP/MMP9. (I) Full-scan XPS showing high siRNA encapsulation efficiency maintained after TIMP modification

Safety and targeting evaluation of TIMP-HDL-Nano@siPGRN&QT

To assess the safety and targeting ability of the HDL-based nanoparticles, both in vitro and in vivo studies were performed using five defined treatment groups. In vitro, HK-2 cells exposed to Control, Nano@siPGRN&QT, HDL-Nano@siPGRN&QT, TIMP-Nano@siPGRN&QT, and TIMP-HDL-Nano@siPGRN&QT showed no reductions in viability, as demonstrated by CCK-8 assays (Fig. 3A). qRT-PCR revealed time-dependent decreases in PGRN expression from 3–12 h, with TIMP-HDL-Nano@siPGRN&QT exhibiting the strongest decrease (Fig. 3B). EdU staining confirmed preserved cell proliferation in all groups (Fig. 3C), while phalloidin-based uptake imaging demonstrated efficient internalization of all nanoparticle formulations, with enhanced uptake in HDL- and TIMP-modified groups (Fig. 3D). To evaluate the action pathway of si-PGRN on cells, we adopted Cy5-labeled siPGRN and then encapsulated it in nanomaterials. Time gradient observation results indicated Cy5-labeled siPGRN accumulated along the inner side of cell membrane at 3–6 h, and subsequently entered and spread within cytoplasm from 9 to 24 h. The fluorescent signal intensity gradually decreased starting at 36 h, and exhibited obvious attenuation at 48 h. These observations confirmed that loaded siPGRN can successfully enter cytoplasm to perform gene silencing function (Supplementary Fig. 1).

Fig. 3.

Safety and Targeting Evaluation of TIMP-HDL-Nano@siPGRN&QT. (A) CCK-8 assay showing preserved HK-2 cell viability after nanoparticle treatment. (B) qRT-PCR showing time-dependent PGRN downregulation (0–12 h), with the strongest effect in the TIMP-HDL-Nano@siPGRN&QT group. (C) EdU staining indicating maintained cell proliferation. (D) Phalloidin-based imaging showing efficient nanoparticle uptake, enhanced by HDL and TIMP modification. (E) Whole-body fluorescence imaging showing nanoparticle accumulation from 15 min to 2 h and decreased signals at 8 and 24 h. (F) Ex vivo organ fluorescence (heart, liver, spleen, lung, kidney, brain) showing peak signals at 8 h and highest accumulation in the TIMP-HDL-Nano@siPGRN&QT group. (G-H) IF showing co-localization of DiD-labeled nanoparticles with MMP9 in renal tissue. (I-J) Immunofluorescence was used to detect the co-staining of fibrosis markers α-SMA and DiD-labeled nanoparticles in the renal fibrotic tissues of mice

In vivo, DiR-labeled nanoparticles were injected into UUO mice for biodistribution analysis. Whole-body fluorescence imaging showed increasing accumulation from 15 min to 2 h, followed by gradual decreases at 8 h and 24 h for all formulations (Fig. 3E). Ex vivo organ imaging (heart, liver, spleen, lung, kidney, brain) at 0.5, 2, 8, and 12 h indicated peak fluorescence at 8 h and proportional reductions thereafter, with TIMP-HDL-Nano@siPGRN&QT consistently exhibiting the strongest organ and renal signals (Fig. 3F and Supplementary Fig. 2A). Immunofluorescence staining performed at 2 h and 24 h showed that TIMP-HDL-Nano@siPGRN levels significantly increased at 24 h compared to 2 h, whereas the fibrosis marker MMP9 decreased (Figs. 3G–H and Supplementary Figs. 2B–C). DiD-labeled nanoparticles exhibited co-localization with MMP9 in renal tissue, indicating their effective accumulation in fibrotic regions (Supplementary Fig. 2F).Furthermore, to more clearly demonstrate the accumulation of nanoparticles at fibrotic sites, we performed co-staining of DiD-labeled TIMP-HDL-Nano@siPGRN with the fibrosis marker α-SMA. The results (Figs. 3I–J and Supplementary Figs. 2D–E) demonstrate that the red fluorescence of DiD-labeled nanoparticles highly co-localizes with the green fluorescence in α-SMA-positive expression regions (Supplementary Fig. 2F), directly confirming that these nanoparticles precisely accumulate at areas of renal fibrosis with high α-SMA expression and exhibit excellent tissue-targeted enrichment capability for renal fibrosis.With prolonged intracellular retention time, compared to the 2-h group, the fluorescence expression level of α-SMA in renal tissue was significantly reduced in the 24-h group, while the fluorescence intensity of nanoparticles in the renal region increased. These findings preliminarily demonstrate that this delivery system can effectively release active substances at fibrotic sites and contributes to inhibiting the progression of renal interstitial fibrosis. qRT-PCR of renal tissues from 3–12 h further confirmed progressive PGRN suppression across nanoparticle-treated groups, with TIMP-HDL-Nano@siPGRN&QT producing the most sustained silencing effect (Supplementary Fig. 2C). Collectively, these results demonstrate that TIMP-HDL-Nano@siPGRN&QT nanoparticles possess excellent biocompatibility, efficient cellular uptake, strong renal-targeting capability, and enhanced in vivo gene-silencing performance.

We further investigated the cellular uptake, intracellular retention, in vivo biodistribution and biosafety of the fluorescent nanocarrier. As shown in Supplementary Fig. 3, the fluorescent nanocarrier was internalized into the cytoplasm of HK2 cells in a time-dependent manner. The fluorescence gradually faded from 24 to 72 h. In vivo fluorescence tracking over 14 days showed progressive clearance of the nanocarrier from the blood circulation, transient signal retention was observed in the liver and spleen, while renal fluorescence declined over time, demonstrating no long-term accumulation in major detected organs (Supplementary Fig. 4). Biochemical analysis (1, 3, 7, 14 days post-administration, Table 1) and organ histopathology (1, 7, 14 days post-administration, Supplementary Fig. 5–1 to 5–3) of heart, liver, spleen, lung and kidney confirmed normal serum hepatic and renal function without obvious tissue lesions, inflammatory infiltration or organ damage. Collectively, these results verify the effective metabolic clearance of the nanocarrier and negligible acute-to-subchronic organ toxicity within the 14-day observation period.

Table 1.

Liver and kidney function biochemical analysis

	Days	Nano@ siPGRN&QT	HDL-Nano@ siPGRN&QT	TIMP-Nano@ siPGRN&QT	TIMP-HDL-Nano@ siPGRN&QT
ALP (U/L)
	1	173.03 ± 13.80	167.15 ± 19.59	164.45 ± 13.74	175.14 ± 7.48
	3	158.87 ± 24.96	170.79 ± 28.21	160.92 ± 17.22	183.07 ± 12.16
	7	167.06 ± 13.67	169.45 ± 10.30	174.05 ± 24.16	159.41 ± 10.36
	14	165.44 ± 21.41	164.80 ± 11.45	165.76 ± 13.78	160.03 ± 13.00
ALT (U/L)
	1	38.13 ± 2.87	39.82 ± 5.79	38.13 ± 2.00	37.97 ± 4.07
	3	36.70 ± 2.39	40.63 ± 4.76	37.57 ± 4.59	39.19 ± 3.05
	7	40.22 ± 3.53	37.91 ± 4.96	40.83 ± 2.82	37.25 ± 3.42
	14	35.80 ± 3.01	39.71 ± 4.49	38.27 ± 3.63	37.19 ± 4.35
AST (U/L)
	1	129.47 ± 3.85	138.23 ± 16.69	143.03 ± 16.61	136.62 ± 13.79
	3	136.26 ± 9.26	127.02 ± 12.84	123.05 ± 9.99	136.87 ± 7.57
	7	123.26 ± 14.22	123.46 ± 10.41	129.87 ± 13.48	129.05 ± 16.75
	14	129.15 ± 12.82	129.25 ± 12.73	138.54 ± 9.77	128.18 ± 14.99
BUN (mmol/L)
	1	8.08 ± 0.52	7.81 ± 0.88	7.87 ± 0.98	7.45 ± 0.50
	3	8.02 ± 0.73	7.93 ± 1.15	7.75 ± 0.34	7.65 ± 1.02
	7	7.93 ± 0.81	7.66 ± 0.90	8.37 ± 0.77	8.33 ± 1.17
	14	7.91 ± 0.46	8.18 ± 0.86	7.97 ± 0.53	7.94 ± 0.69
Scr (μmol/L)
	1	49.47 ± 3.36	51.01 ± 2.05	48.59 ± 5.52	52.11 ± 2.53
	3	49.30 ± 3.81	47.30 ± 5.98	46.02 ± 3.45	48.89 ± 5.18
	7	48.16 ± 2.74	49.79 ± 6.22	49.66 ± 5.80	47.12 ± 4.52
	14	50.04 ± 4.13	47.73 ± 3.62	48.39 ± 2.40	48.76 ± 5.21

Table 1. Liver and kidney function biochemical analysis. Serum biochemical analysis of hepatic and renal functional indices at four time points: 1 d, 3 d, 7 d, and 14 d during the in vivo metabolic observation period. ALT and AST were measured to evaluate liver function, whereas BUN and serum creatinine were measured to evaluate renal function. Data are presented as the mean ± SD (n = 6)

TIMP-HDL-Nano@siPGRN&QT alleviates renal fibrosis in PGRN-deficient UUO mice model

To evaluate the therapeutic efficacy of TIMP-HDL-Nano@siPGRN&QT, UUO models were established in WT and PGRN⁻/⁻ mice and divided into six groups: Control, PGRN⁻/⁻ WT, Model, PGRN⁻/⁻ Model, PGRN⁻/⁻ Model + QT, and Model + TIMP-HDL-Nano@siPGRN&QT. H&E staining (Fig. 4A and Supplementary Fig. 6A) showed marked tubular dilation, interstitial expansion, and inflammatory infiltration in the Model group, whereas these injuries were partially attenuated in PGRN⁻/⁻ Model mice, indicating that PGRN deficiency alleviates injury severity. Quercetin further improved tissue morphology, while TIMP-HDL-Nano@siPGRN&QT produced the most substantial restoration. Masson staining (Fig. 4B and Supplementary Fig. 6B) revealed extensive collagen deposition in Model mice, which was reduced by PGRN deficiency, further decreased by quercetin, and most effectively diminished by nanoparticle treatment. TUNEL staining (Fig. 4C and Supplementary Fig. 6C) indicated minimal apoptosis in Control and PGRN⁻/⁻ WT kidneys but extensive apoptosis in UUO mice, with lower apoptotic levels in PGRN⁻/⁻ Model mice. Quercetin reduced apoptosis, and TIMP-HDL-Nano@siPGRN&QT achieved the strongest anti-apoptotic effect. IHC staining (Fig. 4D–F and Supplementary Fig. 6D) showed elevated α-SMA, COL1A1, and MMP9 in UUO kidneys, which were modestly reduced in PGRN⁻/⁻ Model mice and further suppressed by quercetin, with nanoparticle treatment yielding the most significant downregulation. Western blotting (Fig. 4G and 4H) confirmed these trends. α-SMA, MMP9, and TGF-β expression levels were highest in the Model group, lower in PGRN⁻/⁻ Model mice, further decreased with quercetin, and lowest in the TIMP-HDL-Nano@siPGRN&QT group, approaching Control and PGRN⁻/⁻ WT values. Collectively, these results demonstrate that TIMP-HDL-Nano@siPGRN&QT effectively alleviates UUO-induced renal fibrosis by reducing tissue injury, apoptosis, and fibrosis signaling, and that PGRN deficiency itself mitigates the severity of renal damage.

Transcriptomic and mechanistic analyses identify the PGRN–PPARα–NF-κB Axis underlying the anti-fibrotic effects of TIMP-HDL-Nano@siPGRN&QT

To elucidate the molecular mechanisms underlying the anti-fibrotic effects of TIMP-HDL-Nano@siPGRN&QT, transcriptomic sequencing was performed on renal tissues from treatment and UUO model groups. Hierarchical clustering demonstrated clear separation between groups (Fig. 5A), indicating pronounced transcriptional reprogramming. A total of 1,427 differentially expressed genes were identified, including 938 upregulated and 489 downregulated genes in the treatment group (Fig. 5B). KEGG enrichment analysis showed that these genes were primarily involved in inflammation- and immunity-related pathways, such as cytokine–cytokine receptor interaction, chemokine signaling, NF-κB signaling, cell adhesion molecules, phagosome formation, and calcium signaling (Fig. 5C). Bar plot analysis further confirmed enrichment of immune-associated pathways, including complement and coagulation cascades, Th17 cell differentiation, and C-type lectin receptor signaling (Fig. 5D). Notably, metabolic and nuclear receptor–related pathways were also activated, with PPARα emerging as a key regulatory axis.

Fig. 5.

Transcriptomic and mechanistic analyses reveal the PGRN–PPARα–NF-κB axis underlying the anti-fibrotic effects of TIMP-HDL-Nano@siPGRN&QT. (A) Hierarchical clustering heatmap showing distinct transcriptional profiles between UUO model and TIMP-HDL-Nano@siPGRN&QT–treated kidneys. (B) Volcano plot showing differentially expressed genes, including 938 upregulated and 489 downregulated genes. (C) KEGG enrichment analysis. (D) KEGG bar plot. (E, F) Western blot analysis showing reduced PPARα expression in UUO kidneys and restoration following TIMP-HDL-Nano@siPGRN&QT treatment. (G) Co-IP showing a specific interaction between PGRN and PPARα, absent in IgG controls. (H) Co-IP under Control and PGRN⁻/⁻ conditions showing PGRN-dependent interaction between PGRN and PPARα. (I, J) Western blot analysis showing activation of NF-κB signaling (P65, p-P65, P50, and p-P50) in UUO kidneys and suppression following TIMP-HDL-Nano@siPGRN&QT treatment. *P < 0.05; **P < 0.01; ***P < 0.001; ns, no significance

To clarify the molecular mechanisms underlying the therapeutic effects of TIMP-HDL-Nano@siPGRN&QT, PPARα expression and downstream inflammatory signaling were analyzed. Western blotting (Fig. 5E and F) showed that PPARα was markedly downregulated in the UUO Model group but significantly restored by TIMP-HDL-Nano@siPGRN&QT, indicating reactivation of PPARα signaling. Co-IP assays (Fig. 5G) demonstrated a specific interaction between PGRN and PPARα, which was absent in IgG controls. This interaction was further confirmed to be PGRN-dependent, as PPARα was detected in anti-PGRN immunoprecipitates from Control samples but not from PGRN⁻/⁻ samples (Fig. 5H). Consistent with PPARα–inflammation crosstalk, NF-κB signaling was markedly activated in UUO kidneys but significantly suppressed following TIMP-HDL-Nano@siPGRN&QT treatment (Fig. 5I and J). Collectively, these findings indicate that TIMP-HDL-Nano@siPGRN&QT restores PPARα activity, disrupts PGRN-dependent regulation of PPARα, and inhibits NF-κB signaling, thereby remodeling the inflammatory fibrotic microenvironment and attenuating renal fibrosis.

Supplementary Fig. 7 included GST pull-down and PPARα luciferase reporter assays to evaluate both biochemical interaction and functional reporter activity. Luciferase reporter analysis further showed that PGRN expression reduced PPARα-responsive transcriptional activity, whereas PGRN knockdown restored the reporter signal under the corresponding experimental conditions (Supplementary Fig. 7A). GST pull-down analysis showed that His-PPARα was detected in the GST-PGRN pull-down fraction but was absent or markedly weaker in the GST-only control lane, while input lanes confirmed the presence of recombinant proteins before pull-down (Supplementary Fig. 7B). Together, these supplemental data support a PGRN-PPARα association at the biochemical level and indicate that PGRN is linked to reduced PPARα reporter activity in this assay system.

Supplementary Fig. 8 provided additional Western blot controls to separate PGRN-specific knockdown from siRNA control, nanocarrier, and quercetin-related effects. In the siRNA-screening panel, Western blotting confirmed that the tested siPGRN sequences reduced PGRN protein compared with the Control and siNC groups, supporting the suitability of siPGRN-mediated PGRN silencing for the subsequent functional experiments (Supplementary Fig. 8A). In LPS-stimulated macrophages, LPS increased the M1-associated marker CD86 and reduced the M2-associated marker CD206, whereas siPGRN treatment attenuated CD86 and partially restored CD206 relative to the LPS/siNC pattern (Supplementary Fig. 8B). The nanocarrier control panel further showed that TIMP-HDL-Nano@siNC did not substantially reverse the LPS-induced CD86/CD206 pattern, while TIMP-HDL-Nano@siPGRN and TIMP-HDL-Nano@QT each shifted macrophage marker output toward lower CD86 and higher CD206. The combined TIMP-HDL-Nano@siPGRN&QT formulation produced the clearest reduction in CD86 and increase in CD206 among the tested nanocarrier groups, supporting cooperative macrophage-state remodeling by PGRN silencing and quercetin delivery (Supplementary Fig. 8C).

PPARα signaling is required for TIMP-HDL-Nano@siPGRN&QT–induced inhibition of macrophage M1 polarization in vitro

To investigate whether TIMP-HDL-Nano@siPGRN&QT regulates macrophage behavior within the fibrotic microenvironment, nanoparticle uptake, proliferation, apoptosis, and polarization were systematically examined. Flow cytometry using CFSE-labeled macrophages and PE-labeled nanoparticles revealed a time-dependent increase in nanoparticle uptake from 2 to 24 h, which was further confirmed by fluorescence microscopy showing progressive intracellular accumulation (Fig. 6A–D). In an LPS-induced HK-2–macrophage co-culture system, EdU staining demonstrated that LPS markedly suppressed macrophage proliferation, whereas TIMP-HDL-Nano@siPGRN&QT significantly restored proliferative capacity (Fig. 6E and F). Flow cytometric analysis further showed increased macrophage apoptosis following LPS stimulation, which was partially attenuated by nanoparticle treatment (Fig. 6G and H). Macrophage polarization analysis indicated that LPS promoted a pro-inflammatory M1 phenotype, characterized by increased CD86 expression and reduced CD206 expression. In contrast, TIMP-HDL-Nano@siPGRN&QT markedly decreased CD86 levels and partially restored CD206 expression, as confirmed by IF, flow cytometry, and Western blotting (Fig. 6I–P).

Fig. 6.

TIMP-HDL-Nano@siPGRN&QT Regulates Macrophage Behavior and Suppresses M1 Polarization. (A) Flow cytometry showing time-dependent uptake of PE-labeled nanoparticles by CFSE-labeled macrophages from 2 to 24 h. (B–D) Fluorescence microscopy showing progressive intracellular accumulation of nanoparticles in macrophages. (E, F) EdU staining showing macrophage proliferation. (G, H) Flow cytometry evaluating apoptosis. (I–K) IF staining for CD86 and CD206 detecting macrophage polarization. (L–N) Flow cytometry evaluating macrophage polarization. (O, P) Western blot analysis measuring CD86 and CD206 expression. *P < 0.05; **P < 0.01; ***P < 0.001; ns, no significance

To determine whether these effects were mediated by PPARα signaling, LPS-stimulated HK-2 cells were transfected with PPARα siRNA prior to co-culture with THP-1–derived macrophages. Western blotting confirmed efficient PPARα knockdown, with siRNA-1 showing the highest silencing efficiency (Fig. 7A and B). PPARα silencing reversed the inhibitory effects of TIMP-HDL-Nano@siPGRN&QT on fibrosis-related proteins, including COL1A1, α-SMA, MMP9, and TGF-β (Fig. 7C and D), and attenuated its ability to restore macrophage proliferation, reduce apoptosis, and promote M2 polarization (Fig. 7E–N). Collectively, these results demonstrate that TIMP-HDL-Nano@siPGRN&QT is efficiently internalized by macrophages and alleviates inflammatory injury by promoting macrophage survival and M2 polarization through a PPARα-dependent mechanism, thereby contributing to remodeling of the fibrotic immune microenvironment.

Fig. 7.

PPARα silencing abrogates the anti-fibrotic and immunomodulatory effects of TIMP-HDL-Nano@siPGRN&QT. (A, B) Western blot analysis confirming the efficiency of PPARα knockdown. (C, D) Western blot analysis of fibrosis-related proteins (COL1A1, α-SMA, MMP9, and TGF-β) expression. (E, F) EdU staining evaluating macrophage proliferation. (G, H) Flow cytometric analysis assessing apoptosis. (I, J) IF staining for CD86 and CD206. (K, L) Flow cytometry detecting macrophage polarization. (M, N) Western blot analyses measuring CD86 and CD206 expression. *P < 0.05; **P < 0.01; ***P < 0.001; ns, no significance

TIMP-HDL-Nano@siPGRN&QT attenuates renal fibrosis by inhibiting M1 macrophage polarization through PPARα/NF-κB signaling In vivo

To determine whether the anti-fibrotic effects of TIMP-HDL-Nano@siPGRN&QT are mediated by PPARα-dependent macrophage regulation in vivo, UUO mice were treated with nanoparticles combined with PPARα neutralization or activation. Western blotting (Fig. 8A–B) showed that UUO increased M1 markers CD86 and iNOS and reduced the M2 marker CD206, while PPARα blockade further aggravated this pro-inflammatory polarization. Although TIMP-HDL-Nano@siPGRN&QT partially reversed these changes, its regulatory efficacy was markedly attenuated. IF staining (Fig. 8C–F) confirmed increased F4/80⁺CD86⁺ and decreased F4/80⁺CD206⁺ macrophages under PPARα neutralization. Under LPS challenge, renal injury and fibrosis were markedly exacerbated, whereas PPARα activation or TIMP-HDL-Nano@siPGRN&QT treatment significantly alleviated these pathological changes, with the most pronounced improvement observed upon combined treatment (Fig. 8G–H, and Supplementary Fig. 9A-B). Western blotting (Fig. 8I–J) further showed that LPS-induced upregulation of α-SMA, MMP9, and TGF-β was partially suppressed by single treatments and most effectively reduced by combined treatment. Consistently, NF-κB signaling was strongly activated in UUO and LPS-treated kidneys but markedly inhibited by PPARα activation and TIMP-HDL-Nano@siPGRN&QT, with maximal suppression in the combined treatment group (Fig. 8K–L). Collectively, these results demonstrate that TIMP-HDL-Nano@siPGRN&QT alleviates renal fibrosis by suppressing macrophage M1 polarization and inflammatory signaling through a PPARα-dependent mechanism, and that disruption of PPARα signaling substantially compromises its therapeutic efficacy.

Fig. 8.

TIMP-HDL-Nano@siPGRN&QT relieves renal fibrosis by inhibiting M1 polarization through PPARα/NF-κB signaling in vivo. (A, B) Western blot analysis showing increased CD86 and iNOS and decreased CD206 expression in UUO kidneys, further aggravated by PPARα neutralization and partially reversed by TIMP-HDL-Nano@siPGRN&QT. (C–F) IF staining for F4/80/CD86 and F4/80/CD206 showing increased M1 and reduced M2 macrophages under PPARα blockade. (G, H) H&E and Masson staining showing that LPS exacerbates renal injury and fibrosis, which are alleviated by PPARα activation and/or TIMP-HDL-Nano@siPGRN&QT, with the greatest improvement under combined treatment. (I, J) Western blot analysis showing suppression of LPS-induced α-SMA, MMP9, and TGF-β expression by PPARα activation and TIMP-HDL-Nano@siPGRN&QT. (K, L) Western blot analysis showing marked inhibition of NF-κB signaling following PPARα activation and TIMP-HDL-Nano@siPGRN&QT treatment. *P < 0.05; **P < 0.01; ***P < 0.001; ns, no significance

NF-κB signaling acts as a downstream effector of PPARα in nanoparticle-mediated reversal of renal fibrosis

To define the role of NF-κB signaling in nanoparticle-mediated PPARα activation during renal fibrosis reversal, UUO mice were treated with an NF-κB inhibitor in combination with PPARα neutralization and/or TIMP-HDL-Nano@siPGRN&QT. Histological analysis (Fig. 9A and Supplementary Fig. 10A, and 9B and Supplementary Fig. 10B) showed that NF-κB inhibition partially alleviated UUO-induced renal injury and collagen deposition, effects that were largely abolished by PPARα neutralization. TIMP-HDL-Nano@siPGRN&QT partially restored renal architecture and reduced fibrosis under PPARα blockade, although with reduced efficacy. Immunohistochemical (Fig. 9C-E and Supplementary Fig. 10C) and Western blot analyses (Fig. 9F and G) demonstrated that α-SMA, MMP9, and TGF-β expression was elevated in UUO kidneys, reduced by NF-κB inhibition, and restored upon PPARα neutralization, while nanoparticle treatment again suppressed these fibrotic markers. Under LPS-induced inflammatory stress (Fig. 9H and I). NF-κB activation further enhanced fibrotic protein expression, whereas PPARα activation and TIMP-HDL-Nano@siPGRN&QT counteracted this effect, with the strongest inhibition observed under combined treatment. Collectively, these findings indicate that NF-κB functions downstream of PPARα in renal fibrotic progression, and that TIMP-HDL-Nano@siPGRN&QT alleviates renal fibrosis by activating PPARα and suppressing NF-κB–dependent fibrotic signaling, establishing the PPARα–NF-κB axis as a key mechanism underlying nanoparticle-mediated remodeling of the fibrotic microenvironment.

Fig. 9.

NF-κB Signaling Acts Downstream of PPARα in Nanoparticle-Mediated Reversal of Renal Fibrosis. (A, B) H&E and Masson staining showing that NF-κB inhibition partially attenuates UUO-induced renal injury and fibrosis, effects largely abolished by PPARα neutralization, while TIMP-HDL-Nano@siPGRN&QT partially restores renal architecture. (C–E) IHC staining for α-SMA, MMP9, and TGF-β showing reduced fibrotic marker expression after NF-κB inhibition, restoration upon PPARα blockade, and re-suppression by nanoparticle treatment. (F, G) Western blot analysis confirming corresponding changes in fibrotic protein levels. (H, I) Western blot analysis showing that LPS-induced NF-κB activation enhances fibrotic protein expression, whereas PPARα activation and TIMP-HDL-Nano@siPGRN&QT suppress NF-κB–dependent fibrotic signaling, with the strongest inhibition under combined treatment. *P < 0.05; **P < 0.01; ***P < 0.001; ns, no significance

Nanoparticles Promote Anti-Inflammatory Macrophage M2 Polarization by Suppressing the PPARα–NF-κB Signaling Axis In Vitro

To determine whether TIMP-HDL-Nano@siPGRN&QT regulates macrophage anti-inflammatory polarization through the PPARα–NF-κB axis in vitro, LPS-stimulated HK-2–macrophage co-culture systems were subjected to NF-κB and/or PPARα modulation. EdU assays (Fig. 10A and D) showed that LPS and NF-κB activation markedly suppressed macrophage proliferation, whereas TIMP-HDL-Nano@siPGRN&QT partially restored proliferative capacity. Combined activation of PPARα further enhanced this recovery, indicating a cooperative effect between nanoparticle treatment and PPARα signaling. IF analysis (Fig. 10B–C and E–F) revealed that LPS and NF-κB activation promoted M1 polarization, characterized by increased CD86 and reduced CD206 expression. TIMP-HDL-Nano@siPGRN&QT significantly decreased CD86 levels and partially restored CD206 expression, even under NF-κB activation, with more pronounced effects observed upon PPARα activation. Conversely, PPARα neutralization markedly attenuated the nanoparticle-mediated rescue of macrophage proliferation (Fig. 10G and L) and weakened suppression of macrophage-associated fibrotic markers, including MMP9, COL1A1, and α-SMA (Fig. 10H–K and M–N), under NF-κB–activated conditions. Collectively, these findings demonstrate that TIMP-HDL-Nano@siPGRN&QT promotes macrophage anti-inflammatory polarization and suppresses macrophage-driven fibrotic signaling by inhibiting NF-κB activity in a PPARα-dependent manner, highlighting the PPARα–NF-κB axis as a key mechanism underlying nanoparticle-mediated remodeling of the fibrotic microenvironment.

Fig. 10.

TIMP-HDL-Nano@siPGRN&QT Regulates Macrophage Polarization via the PPARα–NF-κB Axis in Vitro. (A, D) EdU assays showing that LPS and NF-κB activation suppress macrophage proliferation, while TIMP-HDL-Nano@siPGRN&QT partially restores proliferation, with further enhancement upon PPARα activation. (B–C, E–F) IF staining showing that LPS and NF-κB activation promote M1 polarization (CD86, CD206), whereas TIMP-HDL-Nano@siPGRN&QT reduces CD86 and partially restores CD206, especially with PPARα activation. (G, L) EdU assays showing that PPARα neutralization attenuates nanoparticle-mediated recovery of macrophage proliferation. (H–K, M–N) IF staining showing that PPARα neutralization weakens the inhibitory effects of TIMP-HDL-Nano@siPGRN&QT on macrophage-associated fibrotic markers (MMP9, COL1A1, α-SMA). *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant

Discussion

This study demonstrates that degradable MMP9-responsive HDL nanoparticles enable targeted co-delivery of siRNA against progranulin and quercetin to fibrotic kidneys, thereby effectively attenuating renal fibrosis. TIMP-HDL-Nano@siPGRN&QT alleviates tubular injury, collagen deposition, and apoptosis by suppressing PGRN-dependent profibrotic signaling and inhibiting key fibrotic markers. Moreover, it remodels the fibrotic microenvironment by restoring PPARα activity, inhibiting NF-κB–mediated inflammation, and promoting macrophage polarization toward an anti-inflammatory M2 phenotype, highlighting the therapeutic potential of this enzyme-responsive nanotherapeutic strategy for renal fibrosis.

Quercetin is a naturally occurring flavonoid abundant in fruits, vegetables, and medicinal plants, widely recognized for its broad pharmacological activities, particularly its anti-inflammatory, antioxidant, and anti-fibrotic effects in conditions involving chronic inflammation and tissue remodeling [20–22]. Accumulating pharmacological studies have demonstrated that quercetin exerts pleiotropic biological functions by modulating multiple signaling pathways, including transforming growth factor-β (TGF-β) [23], nuclear factor-κB (NF-κB) [24], and reactive oxygen species–related pathways [21], which are central to fibrotic progression. In the context of renal disease, quercetin has been shown to alleviate renal inflammation and fibrosis in experimental models by suppressing profibrotic cytokine production, reducing oxidative stress, and inhibiting fibroblast activation [25–27]. Chen et al. [27] have reported that quercetin alleviates lupus nephritis–associated renal injury by suppressing renal fibrosis, inflammation, and tubular cell pyroptosis via inhibition of the IL-33/ST2 signaling pathway in both in vivo and in vitro models. Several studies have reported that quercetin attenuates renal interstitial fibrosis by downregulating TGF-β/Smad signaling and inhibiting NF-κB–mediated inflammatory responses, thereby reducing collagen deposition and tubular injury [25, 28–30]. Wang et al. [25] demonstrated that quercetin attenuates renal fibrosis by suppressing TGF-β1–induced tubular epithelial–mesenchymal transition (EMT) through inhibition of AREG/EGFR signaling in both in vivo and in vitro models. Consistently, Gao et al. [29] reported that quercetin ameliorates folic acid–induced renal fibrosis by inhibiting tubular epithelial ferroptosis via the EGFR/ACSL4 signaling pathway. Moreover, quercetin has been implicated in the regulation of immune cell function, including macrophage activation and polarization, which further contributes to its anti-fibrotic efficacy [31–33]. Abudoureyimu et al. [31] reported that quercetin protects against diabetic nephropathy by inhibiting NLRC5/NLRP3-mediated M1 macrophage polarization, thereby reducing renal inflammation and fibrosis. However, despite its promising therapeutic potential, the clinical translation of quercetin remains limited due to its poor aqueous solubility, rapid systemic clearance, and insufficient renal accumulation in vivo. In this study, we demonstrate that nanoparticle-mediated delivery of quercetin markedly enhances its renal targeting and anti-fibrotic efficacy. Our results show that quercetin, when co-delivered with siRNA against PGRN via MMP9-responsive HDL nanoparticles, effectively suppresses fibrotic signaling, reduces inflammation, and contributes to remodeling of the fibrotic microenvironment. These findings support quercetin as a critical pharmacological component of the nanotherapeutic strategy and highlight the importance of advanced delivery platforms to fully harness its anti-fibrotic potential in renal fibrosis.

The nanotherapeutic system developed in this study is a degradable, MMP9-responsive HDL–based nanoparticle designed for targeted delivery in fibrotic kidney disease. HDL nanoparticles are known for their excellent biocompatibility, intrinsic affinity for renal tissue, and capacity to simultaneously transport nucleic acids and hydrophobic small molecules [17, 34]. He et al. [34] have reported that biomimetic HDL nanoparticles selectively target injured renal tubules via KIM-1–mediated uptake, enabling effective co-delivery of anti-inflammatory and anti-fibrotic agents to remodel fibrotic niches and attenuate renal fibrosis. Recent advances in nanomedicine have highlighted HDL-based carriers as versatile platforms for precision therapy, owing to their biomimetic structure, favorable pharmacokinetics, and receptor-mediated cellular uptake [35, 36]. By incorporating siRNA against progranulin and quercetin into reconstituted HDL nanoparticles, this system enables integrated genetic and pharmacological intervention. Importantly, surface functionalization with a TIMP-derived MMP9-sensitive peptide confers enzyme-responsive degradability, allowing selective activation and cargo release within MMP9-enriched fibrotic microenvironments. Matrix metalloproteinase-9 (MMP-9) is a zinc-dependent endopeptidase that degrades extracellular matrix components and plays a critical role in tissue remodeling, inflammation, and the progression of fibrotic diseases. In the context of renal fibrosis, elevated MMP9 expression is closely associated with extracellular matrix remodeling and inflammatory cell infiltration, making it an ideal endogenous trigger for targeted nanotherapy [19, 37, 38]. Shen et al. [19] have reported that MMP-9 is progressively upregulated during AKI-to-CKD transition and drives renal fibrosis and cellular senescence, whereas its inhibition significantly attenuates fibrotic progression. Previous studies have shown that stimulus-responsive nanoparticles can enhance therapeutic specificity and reduce off-target toxicity [39–41]; however, their application in renal fibrosis remains limited. In this study, we demonstrate that MMP9-responsive HDL nanoparticles exhibit efficient renal accumulation, sustained retention, and superior therapeutic efficacy compared with non-targeted formulations. Our results indicate that this nanoplatform effectively overcomes key barriers in siRNA and small-molecule delivery, including instability, poor tissue specificity, and limited bioavailability. Collectively, these findings highlight the potential of enzyme-responsive HDL-based nanotechnology as a powerful and adaptable strategy for targeted modulation of the fibrotic microenvironment and treatment of renal fibrosis.

PGRN and PPARα are both critical regulators of inflammation, immune homeostasis, and tissue remodeling, and have increasingly attracted attention in the context of chronic inflammatory and fibrotic diseases [42, 43]. PGRN is a multifunctional secreted glycoprotein widely expressed in renal parenchymal and immune cells and is known to modulate inflammatory signaling, macrophage activation, and extracellular matrix dynamics [5, 42, 44]. Adachi et al. [44] have reported that progranulin deficiency alleviates UUO-induced tubulointerstitial injury and renal fibrosis by suppressing inflammatory and profibrotic responses. PPARα, a nuclear receptor involved in lipid metabolism and inflammation resolution, exerts potent anti-inflammatory and anti-fibrotic effects through transcriptional regulation of metabolic and immune-related genes [45–48]. Horinouchi et al. [48] have reported that activation of PPARα by the selective modulator pemafibrate exerts significant renoprotective effects in chronic kidney disease by suppressing renal inflammation and fibrosis and improving renal dysfunction in adenine-induced and UUO mouse models. Recent studies have suggested that dysregulation of PGRN and PPARα signaling contributes to persistent inflammation and fibrosis in multiple organs [42–48]; however, their functional interplay in renal fibrosis has remained largely unexplored. In various disease models, PGRN has been reported to influence macrophage polarization and cytokine production [5], while PPARα activation suppresses NF-κB–dependent inflammatory pathways and promotes tissue protection [47]. This study provides mechanistic evidence that PGRN directly interacts with and negatively regulates PPARα signaling in fibrotic kidneys, representing the first demonstration of a direct PGRN–PPARα binding relationship in renal fibrosis. UUO-induced renal fibrosis is associated with increased PGRN expression, suppressed PPARα activity, sustained NF-κB activation, and pro-inflammatory macrophage polarization. Genetic deletion or nanoparticle-mediated silencing of PGRN restores PPARα signaling, inhibits NF-κB activity, and promotes macrophage polarization toward an anti-inflammatory M2 phenotype. Collectively, this newly identified PGRN–PPARα axis plays a central role in coordinating inflammatory and fibrotic responses and represents a promising therapeutic target for remodeling the fibrotic microenvironment and attenuating kidney injury.

Despite the encouraging results, several limitations should be noted. First, the therapeutic efficacy of TIMP-HDL-Nano@siPGRN&QT was mainly evaluated in the UUO model, which does not fully represent the heterogeneity of human CKD, warranting validation in additional disease models. Second, long-term pharmacokinetics, biodistribution, and potential immunogenicity following repeated nanoparticle administration were not assessed and require further investigation. Third, although a novel PGRN–PPARα interaction was identified, the detailed structural basis and regulatory mechanisms of this interaction under different pathological conditions remain to be elucidated.

Conclusion

In conclusion, this study demonstrates that degradable MMP9-responsive high-density lipoprotein nanoparticles enable efficient and targeted co-delivery of siRNA against progranulin and quercetin to fibrotic kidneys, resulting in marked attenuation of renal fibrosis. This nanotherapeutic strategy achieves enhanced renal accumulation and sustained efficacy by integrating enzyme-responsive targeting with combined genetic and pharmacological intervention. Mechanistically, TIMP-HDL-Nano@siPGRN&QT remodels the fibrotic microenvironment by disrupting the PGRN–PPARα interaction, restoring PPARα signaling, suppressing NF-κB–mediated inflammation, and reprogramming macrophage polarization (Fig. 11). These findings highlight the translational potential of enzyme-responsive HDL-based nanomedicine for precision treatment of renal fibrosis and related fibrotic diseases.

Fig. 11.

Schematic illustration of the therapeutic mechanism underlying the antifibrotic effects of the engineered HDL nanoparticles. The degradable HDL nanoparticles are internalized by macrophages within the kidney, where they release Quercetin and siRNA-PGRN. This co-delivery activates PPARα signaling and inhibits the NF-κB pathway, promoting the repolarization of macrophages from the pro-inflammatory M1 phenotype (characterized by high CD86 and iNOS expression) to the anti-inflammatory M2 phenotype (characterized by upregulated CD206 and IL-4). The shift towards an M2-dominant microenvironment subsequently reduces the secretion of pro-fibrotic factors, including MMP9, α-SMA, and TGF-β, thereby preserving the viability of renal tubular epithelial cells and attenuating renal fibrosis

Author contributions

T.W.C. and H.Y. K. performed validation; T.W.C. drafted the original manuscript; H.Y.K. curated data and prepared visualizations. T.W.C. and H.Y.K. conducted methodology; H.Y.K. reviewed and edited the manuscript and provided supervision. Z.Y. conceptualized the study, reviewed and edited the manuscript, and secured funding. All authors contributed to the final version of the manuscript.

Funding

This study was supported by The National Natural Science Foundation of China (No. 82000629).

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study has been approved by the Animal Ethics Committee of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine (Approval No.: RJ2024033).

Competing interests

The authors declare no competing interests.