aHSCs-targeted bimetallic nanozymes and luteolin-loaded liposomes: Synergistic reversal of liver fibrosis via antioxidant, cellular senescence, and cellular apoptotic mechanisms

Abstract

Liver fibrosis is a prevalent pathological process in the development of a range of chronic liver diseases. Hepatic stellate cells (HSCs) are known to be highly activated in the Liver fbrosis environment, and the proliferation of activated HSCs (aHSCs) and the secretion of associated extracellular matrix are crucial in the process of Liver fibrosis, which in turn promotes the development of Liver fibrosis. Concurrently, the inhibition of HSCs activation and the induction of aHSCs senescence/apoptosis have been identified as a therapeutic strategy that exhibits numerous synergistic mechanisms. In addition, the efficacy of clinical treatment is constrained by a number of factors, including the limited selectivity of pharmaceuticals and the inefficiency of drug delivery mechanisms. Consequently, the present study proposes a Luteolin-loaded liposome (LUT@LIP-BSA) and Ce/Mn bimetallic nanozyme (CMB) dual nanodelivery system, which has been modified by BSA to target aHSCs. The results of both in vitro and in vivo experiments demonstrated that the aHSCs-targeted dual delivery system combining LUT@LIP-BSA with CMB exhibited effective targeting delivery capacity to aHSCs. In vitro, LUT@LIP-BSA effectively induced senescence, apoptosis, and counteracted oxidative stress in aHSCs. These effects were confirmed in vivo. Consequently, this combination reduced extracellular matrix production and deposition, thereby inhibiting liver fibrosis. This combinatorial strategy provides a promising foundation for the construction and clinical application of hybrid delivery systems that synergize metal nanoenzymes with natural drugs as an effective targeted therapeutic approach for liver fibrosis.

Graphical abstract

1. Introduction

The liver is composed of parenchymal cells and non-parenchymal cells (NPCs). The parenchymal cells primarily consist of hepatocytes (accounting for 60 %–70 % of liver cells) and cholangiocytes (biliary epithelial cells). NPCs constitute approximately 30 %–40 % of the total liver cells and include HSCs, Kupffer cells (KCs), liver sinusoidal endothelial cells (LSECs), and other immune cells. Collectively, these NPCs play a central role in hepatic physiology. The liver microenvironment is critical for hepatocyte survival and function and is essential for maintaining the normal structure and physiological functions of the liver [1,2]. Liver fibrosis (LF) represents a defining pathological feature in the progression of various chronic liver diseases. This process arises within the context of hepatic inflammation, injury, and reparative responses. In the fibrotic liver, overproduction of reactive oxygen species (ROS) and heightened activation of HSCs drive the excessive deposition of extracellular matrix (ECM) [3,4]. Since this process involves multiple cascading reactions such as oxidative stress, apoptosis, and inflammation, a single therapy is insufficient to effectively break the vicious cycle [5]. In addition, recent studies have found that the cGAS-STING pathway can participate in fibrosis regulation by regulating cellular senescence, but the secretion of SASP factors induced by it may exacerbate the inflammatory response [[6], [7], [8], [9], [10], [11]]. This contradictory mechanism suggests that simply inducing aHSC senescence requires a combination of other strategies to balance the risk of inflammation.

Proteins, as natural biopolymers, offer multiple clinical translational advantages: inherent biocompatibility, tunable biodegradability and low immunogenicity [12]. Notably, a large number of empirical studies have confirmed the significant affinity of albumin for the SPARC receptor, which is highly expressed on the surface of aHSCs and cancer cells [[13], [14], [15], [16]]. Meanwhile, bovine serum albumin (BSA) can undergo controlled degradation through lysosomal pathways (e.g., Cathepsin B-mediated enzymatic cleavage) to enable regulated release of encapsulated drugs [12]. Leveraging these properties, albumin-based carriers enable efficient targeting of aHSCs. Following receptor-mediated endocytosis, they achieve site-specific accumulation in fibrotic lesions, establishing an ideal platform for precision drug delivery [15,17].

Luteolin (LUT, C₁₅H₁₀O₆, 3′,4′,5,7-tetrahydroxyflavone), a naturally occurring flavonoid compound, primarily exists in glycoside forms and is commonly found in various traditional Chinese medicines, vegetables, and fruits [18,19]. Substantial evidence demonstrates that LUT possesses marked hepatoprotective effects. In hepatic disease models, this compound alleviates LF by significantly inhibiting TGF-β1-induced phosphorylation of both AKT and Smad pathways, while concurrently enhancing caspase-3 activity and promoting p53 expression to induce apoptosis of aHSCs [20]. Existing studies confirm that LUT demonstrates potent antioxidant effects; however, its antioxidative mechanism in HSCs remains poorly understood. Notably, current research on the anti-fibrotic effects of LUT is predominantly confined to cellular models, with limited evidence from in vivo studies regarding its therapeutic potential against LF [21]. In addition, the anti-inflammatory effect of LUT has been confirmed [22,23].

Cerium ions (Ce³⁺) are extensively utilized in biomaterial applications due to their superior antioxidant properties, particularly for exhibiting remarkable superoxide dismutase (SOD)-mimetic activity alongside excellent biocompatibility [[24], [25], [26]]. Furthermore, Ce³⁺ may alleviate LF by scavenging reactive oxygen species (ROS), reducing hepatic inflammatory infiltration and suppressing HSCs activation [27]. Recent studies reveal that manganese ions (Mn²⁺) can induce cellular senescence by modulating the Stimulator of Interferon Genes (STING) pathway, while promoting the expression of senescence-associated secretory phenotype (SASP) factors to enhance natural killer (NK) cell-mediated phagocytosis of senescent cells. Clinically, the in vivo distribution of Mn²⁺ can be monitored in real-time via magnetic resonance imaging (MRI), providing robust support for improving therapeutic efficacy and clinical monitoring capabilities [[28], [29], [30], [31]].

In this research, we made use of linking chitosan (CS) and bovine serum albumin (BSA) through amide bonds and wrapping them on the surface of LUT-loaded liposomes by electrostatic adsorption (LUT@LIP-BSA). At the same time, Ce3+ and Mn2+ were encapsulated using BSA to form CMB. In LF, heightened aHSCs highly express SPARC receptors, and BSA in our synthesized dual nanosystems can induce nanoparticles to actively target aHSCs rather than hepatocytes. When treating LF, we use CMB and LUT@LIP-BSA injected 12 h tail vein interval into LF mice, they will enter the Disse space through the endothelial window of the hepatic sinusoids, in which the BSA enables them to precisely target the SPARC receptor on the surface of the aHSCs, and thus endocytosis and uptake by the aHSCs. We hypothesized that the dual delivery system would remodel the hepatic microenvironment by attenuating LF through inhibition of the transforming growth factor-β (TGF-β) signaling pathway and reducing collagen deposition and α-Smooth Muscle Actin (α-SMA) expression. Meanwhile, it exerts antioxidant effects and reduces ROS production and aHSCs activation in aHSCs by regulating Kelch-like ECH-associated protein 1 (Keap1)/nuclear factor erythroid 2-related factor 2 (Nrf2). In addition, it induced cellular senescence in aHSCs by modulating the STING pathway, as well as promoted apoptosis in aHSCs, thus synergistically reversing LF in a multifunctional manner (see Fig. 1).

Fig. 1.

Mechanism of dual nanodelivery system for the treatment of liver fibrosis.

2. Materials and methods

2.1. Materials

LUT, Ce(NO₃)3·6H₂O were acquired from Yi'en Chemical Technology Co., Ltd. in Shanghai, China. Mn(CH₃COO)2·4H₂O, CS were acquired from Macklin Biochemical Co., Ltd in Shanghai, China. BSA were acquired from Saiguo Biotechnology Co., Ltd. in Guangzhou, China. N-Hydroxy succinimide(NHS), 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC), Cholesterol (CHOL), Vitamin E polyethylene glycol succinate (TPGS) and coumarin 6 (C6) were acquired from Aladdin Reagent Co., Ltd., located in Shanghai, China. Egg phosphatidylcholine (EPC) were sourced from AVT(Shanghai) Pharmaceutical Tech Co., Ltd. China.

Antibodies against P16 were acquired from Hua'an Biotechnology Co. in Hangzhou, China. Antibodies against STING, Keap1 were acquired from Abmart Co., Ltd. in Shanghai, China. Antibodies against Nrf2, TGF-β1, α-SMA, IL-6, IL-8, p53, BCL-2 and COL-I were acquired from Affinity Biosciences Inc, China. Antibodies against P21, Cy3-labeled goat anti-rabbit IgG were acquired from Servicebio Technology Co., Ltd. in Wuhan, China.

Assay kits for evaluating the activities of (CREA), Creatine Kinase(CK), alanine aminotransferase (ALT) were procured from the Shenzhen Mindray Bio-Medical Electronics Co., Ltd. Assay kits for evaluating the activities of blood urea nitrogen (BUN)、Hydroxyproline (HYP)、superoxide dismutase (SOD), glutathione (GSH), malondialdehyde (MDA) and aspartate aminotransferase (AST) were procured from the Nanjing Jiancheng Bioengineering Institute in Nanjing, China. Additionally, enzyme-linked immunosorbent assay (ELISA) kits for detecting tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) were obtained from Elabscience Biotechnology Co., Ltd., located in Wuhan, China.

2.2. Cell lines and animals

In this study, the cell lines that were employed were HSC-T6 and AML-12, which were obtained from Procell Life Science & Technology Co., Ltd., which is located in Wuhan, China. The HSC-T6 cells were cultured in DMEM supplemented with 10 % FBS (v/v) and 1 % penicillin/streptomycin (w/v). However, AML-12 cells were grown in a mixture of DMEM/F-12 and FBS (v/v), ITS Liquid Media Supplement (v/v) and 4 ‰ Dexamethasone (w/v). All cell lines were kept in a humidified incubator at 37 °C with 5 % CO₂. The ICR mice used were eight weeks old, weighing between 30 and 35 g, and were sourced from Ensiweier Biotech Ltd. in Chengdu, China. These mice were given unrestricted access to fresh water and food. All protocols involving animal experimentation were approved by the Experimental Animal Ethics Committee at Chengdu University of Traditional Chinese Medicine.

2.3. Preparation and characterisation

2.3.1. Preparation and characterization of LUT@LIP-BSA

LUT@LIP were prepared via the film-dispersion and hydration-sonication method, and then formed into LUT@LIP-BSA by electrostatic adsorption with CS-BSA. Specifically, 1 mL LUT (2 mg/mL), 1 mL EPC (20 mg/mL), 1 mL CHOL (5 mg/mL), and 1 mL TPGS (4 mg/mL) were rotary evaporated under vacuum at 45 °C and 120 rpm for 15 min to remove organic solvents.(Rotary evaporator, Shanghai Xiande Experimental Instrument Co.) Subsequently, an additional volume of chloroform was added, followed by rotary evaporation for 30 min. The resulting thin film was hydrated with 5 mL of preheated PBS for 30 min. The suspension was then probe-sonicated for 10 min (Biosafer 1000, 300 W, 2 s on/3 s off pulse cycle), filter-sterilized through a 0.22 μm microporous, and stored at 4 °C for further use.

CS-BSA conjugate are synthesized by amide reaction. Specifically, a BSA solution (1.17 mg/mL) was prepared in Phosphate buffered saline (PBS, pH 7.4). EDC and NHS (at a 1:1 M ratio) were added to 30 mL of this BSA solution. The mixture was gently stirred under nitrogen atmosphere at 10 °C in the dark for 2 h. Subsequently, The CS was added to the reaction mixture, followed by continuous stirring under identical conditions for 24 h. The resulting solution was dialyzed against PBS for 3 days using a dialysis membrane (MWCO: 10 kDa), with the PBS being replaced daily. Finally, the CS-BSA conjugate was stored at 4 °C for further use. The synthesis was verified by ζ-potential measurement and Fourier Transform infrared spectroscopy(FTIR) spectroscopy.

The solution of CS-BSA conjugate was dropped into LUT@LIP, stirred slowly for 40 min and stored at 4 °C for further use. The preparation of BSA-LIP loaded with C6/DIO/DiR was identical. The particle size, zeta potential and polydispersity index (PDI) of the NPs were measured by dynamic light scattering (DLS, Litesizer 500, Anton Paar, Austria). The morphology of the NPs was obtained by transmission electron microscopy (TEM, Tecnai 12, Philips company, Holland). The concentration of LUT was detected by high performance liquid chromatography(HPLC, 1200 series, Agilent Technologies, USA). The drug loading efficiency (DL%) and drug encapsulation efficiency (EE%) were analyzed according to the following equations.

$$DL(\%)=\frac{\text{weight}\text{of}\text{loaded}\text{drug}}{\text{weight}\text{of}\text{the}\text{whole}\text{LIP}\text{nanoparticel}}/\times 100$$

$$\text{EE}(\%)=\frac{\text{weight}\text{of}\text{loaded}\text{drug}}{\text{Weight}\text{of}\text{feeding}\text{drug}}/\times 100$$

The in vitro release profile of LUT@LIP-BSA was investigated using a dialysis method. The release medium consisted of PBS (pH 7.4) containing 2 % (w/v) Tween 80. Aliquots (1.5 mL) of LUT solution, LUT@LIP dispersion, and LUT@LIP-BSA dispersion – each containing 600 μg of LUT – were separately placed into dialysis bags (MWCO 14 kDa). Each bag was immersed in 30 mL of release medium and incubated in a thermostated shaker at 37 °C and 100 rpm. Samples of the external medium were withdrawn for analysis at predetermined time points: 0.25, 0.5, 1, 2, 4, 8, 12, 24, 36, and 48 h. These experiments were repeated three times in each group. PBS (pH 7.4, containing 2 % Tween 80) was selected as the in vitro release medium.

2.3.2. Preparation and characterization of CMB

CMB was synthesized by one-pot method. A mixture of Ce(NO₃)3·6H₂O (44 mg/mL) and Mn(CH₃COO)2·4H₂O (24.5 mg/mL) was prepared. Subsequently, BSA(25 mg/mL) was added dropwise at 80 °C, and the mixture was stirred for 15 min. Thereafter, KOH (1 M) was added dropwise and stirred for 2 h. The system was then left to cooling, and subsequently subjected to centrifugation at 5000 rpm for 5 min. The initial sample volume was reduced to 30 mL, after which the solution was extracted using Millipore ultrafiltration tubes(100 KDa). The samples were then collected and stored at 4 °C for further use.

The morphology of NPs was obtained by TEM and the synthesis of nanoenzymes was verified by X-ray photoelectron spectroscopy (XPS, Thermo Fisher-K-Alpha, America), X-ray diffraction (XRD, Rigaku-Ultima IV, Japan), microplate reader (uPerMax3100, Shanghai Flash Bio-Tech Co.) and FTIR(Thermo Nicolet IS5). The concentration of cerium and manganese in the nanoenzymes was determined by inductively coupled plasma-mass spectrometry (ICP-MS). Subsequently, the samples were collected and stored at 4 °C for subsequent use.

Furthermore, the DIO/DIR-labeled CMB was prepared by combining DIO/DIR, BSA, Ce3+ and Mn2+ in a PBS solution, followed by stirring for 10 min. In order to evaluate the stability, LUT@LIP-BSA and CMB were subjected to incubation in PBS (pH 7.4) for a duration of two weeks, respectively. Meanwhile, we also determined the stability of LUT@LIP-BSA and CMB in 10 % FBS, respectively.

2.4. DPPH free radical elimination assay

The overall antioxidant activity was evaluated by observing the color change of DPPH (1,1-diphenyl-2-trinitrophenylhydrazine radical). To evaluate the potential of scavenging DPPH radicals, a series of Free LUT, LUT@LIP, LUT@LIP-BSA and LUT@LIP-BSA + CMB solutions with different concentrations (2.5, 5, 10, 20, 50, 100, 200 μM) were prepared. The DPPH (0.1 mM, 2 mL) ethanol solution was mixed with 2 mL of each sample solution at the indicated concentration. The mixture was then left to react for 30 min at room temperature (RT) in the absence of light. Subsequently, the absorbances were measured at a wavelength of 517 nm. The DPPH radical scavenging capacity of the sample solutions was calculated using the following equation:

$$\text{The}\text{DPPH}\text{radical}\text{scavenging}\text{capacity}(\%)=\frac{A0-Asample}{A0}\ast 100\%$$

where A₀ is the absorbance of DPPH at 517 nm in the absence of sample, and A_sample indicates the absorbance of DPPH at 517 nm after sample addition.

2.5. ABTS free radical scavenging capacity assay

In preparation for the experiment, a mixture of 2 mM ABTS and 2.45 mM K₂S₂O₈ was combined with Phosphate Buffered Saline (PBS) (pH 7.4) and left to incubate overnight in a dark environment to yield an ABTS solution. Subsequently, the mixture's absorbances at 734 nm were calibrated to a standard value of 0.750 ± 0.025, with this adjustment being made using Phosphate Buffered Saline (PBS). Subsequently, a range of concentrations of Free LUT, LUT@LIP, LUT@LIP-BSA and LUT@LIP-BSA + CMB (2.5, 5, 10, 20, 50, 100, 200 μM) were introduced into the prepared solutions and incubated at 37 °C for 30 min. At the conclusion of the incubation period, the absorbances of the resulting mixtures were measured at a wavelength of 734 nm. The ABTS clearance was calculated for each concentration in comparison to the blank's own absorption. The calculation formula is given below.

$$\text{ABTS}\text{Free}\text{Radical}\text{Scavenging}\text{Capacity}(\%)=\frac{A0-Asample}{A0}\ast 100\%$$

Where A₀ denotes the absorbance at 734 nm of an ABTS solution in the absence of any sample, and A_sample denotes the absorbance of the same solution at this wavelength after the addition of sample.

2.6. SOD-like activity assay

Evaluation of SOD-like activity of preparations by scavenging Superoxide Radical (·O₂⁻) activity. SOD-like activity was measured using the o-triol autoxidation technique. Free LUT, LUT@LIP, LUT@LIP-BSA and LUT@LIP-BSA + CMB solutions were prepared at varying concentrations (2.5, 5, 10, 20, 50, 100, 200 μM). In order to perform the assessment of SOD-like activity, 1 mL (20 μL) of each concentration was mixed with 4.5 mL (90 μL) of Tris-HCl buffer (pH 8.2) and then incubated for 4 min at RT. Subsequently, 3 mL (60 μL) of 7 mmol of catechol was added to the mixture. Following a period of 5 min, the absorbances of the solutions were measured at a wavelength of 325 nm.

$$·{\mathrm{O}}_{\mathbf{2}}^{-}\text{Free}\text{radical}\text{scavenging}\text{capacity}(\%)=\frac{A0-Asample}{A0}\ast 100\%$$

where A₀ represents the absorbance of the sample solution at 325 nm without the sample, and A_sample represents the absorbance of the solution at the same wavelength when the sample is added.

2.7. CAT-like activity

Evaluation of CAT-like activity of formulations by Hydrogen peroxide (H₂O₂) scavenging activity. In advance of the experiment, the following concentrations of free LUT, LUT@LIP, LUT@LIP-BSA and LUT@LIP-BSA + CMB (2.5, 5, 10, 20, 50, 100, 200 μM) were meticulously prepared. In addition, aqueous FeSO₄ (9 mM), ethanol, salicylic acid (9 mM) and 30 % H₂O (8.8 mM) was added to 0.3 ml of the sample solution. The mixture was shaken well and left to stand for 25 min in a water bath at 37 °C for 30 min. The measurement of the absorbances was conducted at a wavelength of 510 nm.

$${\mathrm{H}}_2{\mathrm{O}}_2\text{scavenging}\text{capacity}(\%)=\frac{A0-Asample}{A0}\ast 100\%$$

where A₀ is the absorbance at 510 nm of a sample-free solution containing an aqueous FeSO₄ solution, an ethanol solution of salicylic acid, and 30 % H₂O₂. In contrast, A_sample is the absorbance of the added sample solution.

2.8. In vitro cellular uptake

The HST-T6 cells were inoculated at a density of 1 × 10⁵ cells/well in 12-well plates (with cell crawlers) and cultured for 12 h. Subsequently, the cells were treated with free C6, C6@LIP, C6@LIP-BSA, C6@LIP-BSA + BSA, DIO@CMB and DIO@CMB + BSA, respectively. The cells were fixed with 4 % paraformaldehyde (PFA) for a period of 15 min, after which they were stained with 4′,6-diamidino-2-phenylindole (DAPI) for a further 5 min. The localization of DAPI and C6/DIO fluorescence was observed with a confocal laser scanning microscope (CLSM, TSC SP8 STED, Leica, Germany).

For flow cytometry analysis, HSC-T6 cells were inoculated in 6-well plates at a density of 2 × 10⁵ cells/well, and other treatments were consistent with the confocal laser scanning microscopy observations described above. Then, cells were collected and centrifuged at 1250 rpm for 5 min. The cells were resuspended in PBS solution and detected by flow cytometry (FACS, CantoII, BD, USA). The experiment was repeated 3 times for each group.

2.9. Cytotoxicity assay

The cytotoxic potential of free LUT, LUT-LIP, LUT@LIP-BSA, CMB and LUT@LIP-BSA + CMB was determined through the implementation of a cck-8 assay. The HSC-T6 cells were cultivated in 96-well plates at a density of 1 × 10⁴ cells/well and subsequently cultured for a period of 12 h. The AML-12 cells were cultivated in 96-well plates at a density of 5 × 10³ cells per well for a period of 12 h. Subsequently, various gradient concentrations of free LUT, LUT-LIP, LUT@LIP-BSA, CMB,or LUT@LIP-BSA + CMB (2.5, 5, 10, 25, 50, 100, 200 μM) were employed in conjunction over a period of 24 h. Subsequently, the drug medium was disposed of, 100 μL of a 10 % cck-8 solution was added to each well, and the mixture was incubated at 37 °C for 2 h. The analysis of the extinction coefficient at a wavelength of 450 nm was conducted using a MicroplateReader. The experiment was replicated thrice for each group.

2.10. Apoptosis assay

The apoptosis of the cells was detected using the Annexin V-FITC/PI staining kit. The HSC-T6 cells were cultivated in 6-well plates (3 × 10⁵ cells/well) for 12 h. The cells were then exposed to TGF-β, free LUT + TGF-β, LUT-LIP + TGF-β, LUT@LIP-BSA + TGF-β, or LUT@LIP-BSA + CMB + TGF-β (all at a concentration of 20 μM, TGF-β 10 ng/mL) for 24 h. Subsequently, the cells were collected and stained using Annexin V-FITC and PI according to the instructions. The presence of apoptotic cells was detected by means of flow cytometry. The experiment was replicated thrice for each group.

To investigate the mechanism of action of LUT@LIP-BSA + CMB in inducing apoptosis in aHSCs, we determined the mRNA expression of P53 and BCL-2 in aHSCs using WB.

2.11. Intracellular ROS scavenging activity

Total intracellular ROS levels were quantified using DCFH-DA as a fluorescent probe. HSC-T6 cells were seeded in 12-well plates containing glass coverslips at a density of 1 × 10⁵ cells/well and incubated overnight. To induce ROS production, cells were co-treated with 2 ng/mL TGF-β and one of the following compounds (all at 20 μM): free LUT, LUT-LIP, LUT@LIP-BSA, or LUT@LIP-BSA + CMB. After 24 h of incubation, cells were loaded with 10 μM DCFH-DA for 30 min at 37 °C. ROS-associated fluorescence was visualized using fluorescence microscopy (Leica DM6B, Germany).

For parallel flow cytometry analysis, HSC-T6 cells were seeded in 6-well plates at 2 × 10⁵ cells/well. Cells received identical treatments as described for the microscopy assay. Following incubation, cells were harvested by centrifugation (1250 rpm, 5 min), washed twice with ice-cold PBS, and resuspended in PBS for immediate analysis on flow cytometer (FACS, Meloldy, BD, USA). Each experimental group was analyzed in triplicate independent replicates.

2.12. Evaluation of in vitro antifibrotic effects

The expression of α-SMA in HSC-T6 cells was assessed by immunofluorescence. HSC-T6 cells were seeded onto glass coverslips placed in 12-well plates at a density of 1.5 × 10⁵ cells per well and incubated overnight. Cells were then stimulated with 2 ng/mL TGF-β for 24 h to induce activation. Concurrently, cells were treated with free LUT, LUT-LIP, LUT@LIP-BSA, or LUT@LIP-BSA + CMB (all at a concentration of 20 μM).

Following treatment, cells were permeabilized with 0.1 % Triton X-100 for 5 min at RT and washed with ice-cold PBS. Non-specific binding sites were blocked by incubation with 1 % BSA for 1 h at RT. Cells were subsequently incubated with the primary antibody overnight at 4 °C. After primary antibody incubation, cells were washed with pre-cold PBS. A fluorescently conjugated secondary antibody (Cy3-labeled goat anti-rabbit IgG) was applied and incubated for 1 h at RT. Unbound secondary antibody was removed by washing with 0.1 % PBS-T. Finally, coverslips were mounted using an antifade mounting medium containing DAPI for nuclear counterstaining. The expression of α-SMA was visualized and imaged using a fluorescence microscope (Leica DM6B, Germany)

2.13. Induction of aHSCs senescence

Immunofluorescence was employed to evaluate the expression of cyclic cGAS-STING and IL-6 in aHSC-T6 cells. Cells were seeded onto glass coverslips in 12-well plates at a density of 1.5 × 10⁵ cells per well and incubated overnight. Cells were then either stimulated or not stimulated with 2 ng/mL Transforming Growth Factor-beta (TGF-β) for 24 h to induce activation. Concurrently, cells were treated with one of the following: blank culture medium, TGF-β alone, TGF-β + Mn²⁺, TGF-β + CMB, or TGF-β + LUT@LIP-BSA + CMB (with LUT at 20 μM and TGF-β at 2 ng/mL).

Subsequent immunofluorescence processing steps were performed identically to those described for the α-SMA evaluation. Fluorescence visualization and imaging were conducted using a fluorescence microscope (Leica DM6B, Germany).。

2.14. Organ distribution and targeting capacity in vivo

Liver fibrotic mice (n = 3 per group) were intravenously administered 200 μL of DIR-labeled formulations: LIP, BSA-LIP, Mn-LIP, or CMB via the tail vein. To evaluate the real-time biodistribution kinetics, in vivo fluorescence imaging was performed at 2 h and 4 h post-injection using a multimodal in vivo imaging system (IVIS Spectrum, PerkinElmer). At the 4-h terminal time point, mice were euthanized and major organs (heart, liver, spleen, lungs, kidneys) were harvested for ex vivo imaging. All fluorescence signals were acquired at excitation/emission wavelengths of 748/800 nm. Quantitative analysis of region-of-interest (ROI) fluorescence intensity in target organs was conducted using LivingImage® software (v4.5, PerkinElmer).

Liver fibrotic mice (n = 3 per group) received intravenous injections of 200 μL of DIO-labeled formulations: LIP, BSA-LIP, Mn-LIP, or CMB. At 4 h post-injection, liver tissues were harvested and fixed in 4 % PFA for 24 h. Fixed tissues were paraffin-embedded, sectioned at 5-μm thickness, and subjected to immunofluorescence analysis. Sections were incubated overnight at 4 °C with primary antibody against α-SMA, followed by incubation with Cy3-conjugated goat anti-rabbit IgG secondary antibody for 1 h at RT. Nuclei were counterstained with DAPI. Fluorescence images were acquired using CLSM, and co-localization analysis of DIO (green) and α-SMA (red) signals was performed using ImageJ (NIH) with Pearson's correlation coefficient calculation.

2.15. In vivo antifibrotic effect

LF was induced in ICR mice (male, 8-week-old, n = 6/group) via intraperitoneal injection of CCl₄/olive oil mixture (1:4 v/v; 2 mL/kg) twice weekly for 7 weeks. Commencing at week 5, therapeutic interventions were administered intravenously (2 times a week, 3 weeks) as follows: saline vehicle control, Model, free LUT, LUT@LIP, LUT@LIP-BSA, CMB, and LUT@LIP-BSA + CMB (2 mg/kg LUT, CMB-40 mg/kg). At week 7, mice were euthanized for serum and organ collection (heart, liver, spleen, lungs, kidneys). Serum samples were subjected to biochemical analysis for ALT, CK, CREA, AST, BUN and TNF-α/IL-6. Liver tissues underwent comprehensive assessment including: (1) histopathological evaluation via H&E, Masson's trichrome, and Sirius red staining; (2) confocal microscopy analysis of α-SMA/P16/P21 co-localization (3) immunofluorescence quantification of COL-I, Nrf2, Keap1, and TGF-β1 expression; and (4) oxidative stress marker detection (MDA, SOD, GSH) using standardized commercial assays.

2.16. Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA). Quantitative data are expressed as mean ± standard deviation (SD). Statistical comparisons between two groups were conducted using a Student's two-sided t-test, while multiple group comparisons were performed using one-way analysis of variance (ANOVA). Statistical significance was defined as ∗p < 0.05, with the following annotation conventions applied:∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 vs model group; #p < 0.05 vs control group, ##p < 0.01 vs control group, ###p < 0.001 vs control group.

3. Results and discussion

3.1. Preparation and characterization of LUT@LIP-BSA

The CS-BSA conjugate was synthesized via an amidation reaction. Analysis by FTIR (Fig. 2F) revealed the disappearance of peak a at 1597.2 cm⁻¹ attributed to compound NH₂-. Concurrently, the amide I band (C=O) of compound BSA shifted from 1652.17 cm⁻¹ to 1656.69 cm⁻¹ with increased intensity, while the amide II band (N-H) shifted from 1534.73 cm⁻¹ to 1542.3 cm⁻¹, indicating an alteration in the chemical environment of the moiety corresponding to band N-H. Furthermore, compared to compound BSA, the CS-BSA conjugate exhibited a decrease in the absorption intensity of both the amide I and amide II bands, confirming the formation of a new amide bond. Supporting evidence was provided by zeta potential measurements, which showed a decrease in the zeta-potential of CS-BSA to +57.81 ± 0.03 mV from the +28.19 ± 0.48 mV value observed for compound CS (Fig. 2G). Collectively, these results demonstrate the successful synthesis of the CS-BSA conjugate.

Fig. 2.

Characterization images of nanopatricles. (A) Schematic of LUT@LIP-BSA. (B) TEM image of LUT@LIP. scale bar: 100 nm. (C) TEM image of LUT@LIP-BSA. Scale bar: 100 nm. (D) Average particle size of LUT@LIP, LUT@LIP-BSA and CMB.(n = 3) (E) Zeta potential of LUT@LIP, LUT@LIP-BSA and CMB. (n = 3) (F) FTIR spectra of CS, BSA, CS-BSA, Blank LIP, LUT@LIP, LUT@LIP-BSA. (4000-500 cm-1) (G) Zeta potentials of CS and CS-BSA. (n = 3) (H) In vitro release rate of Free LUT, LUT@LIP, LUT@LIP-BSA in PBS (pH 7.4) for 48 h. (I) Schematic representation of CMB (Ce/Mn bimetallic nanoenzyme). (J,K) TEM images of CMB. Scale bar: 100 nm, 200 nm. (L) UV–Vis full wavelength scan of CMB. (M) FTIR spectra of BSA and CMB. (N) Full XPS analysi of CMB(inset: selected segment related to valence state of cerium element). The data represent mean ± SD; ∗∗P < 0.01, ∗∗∗P < 0.001.

LUT@LIP-BSA was prepared using the rotary evaporation-film ultrasonication method. For comparative purposes, LUT@LIP (without CS-BSA) was also prepared. Dynamic light scattering (DLS) analysis revealed that upon encapsulation of component LUT@LIP by the CS-BSA conjugate, the average hydrodynamic diameter of LUT@LIP-BSA increased from 145.31 ± 0.97 nm to 165.82 ± 1.07 nm, while its zeta-potential shifted from −29.41 ± 0.31 mV to −31.33 ± 0.26 mV (Fig. 2D and E). Furthermore, High-performance liquid chromatography (HPLC) quantification determined that the EE and DL of LUT in the formulation were 98.48 % ± 1.16 % and 6.36 % ± 0.07 %, respectively.

FTIR analysis confirmed the successful encapsulation of LUT within the LIP(Fig. 2F). The emergence of a characteristic absorption band (C=O) at 1655.51 cm⁻¹ and the redshift of the stretching vibration (O-H) from 3293.8 cm⁻¹ to 3288.41 cm⁻¹ indicated the incorporation of LUT into liposomes and its interaction with phospholipids. Subsequent encapsulation of LUT@LIP within the CS-BSA further induced a significant redshift of the from 3365.77 cm⁻¹ to 3272.79 cm⁻¹. This shift likely resulted from strong hydrogen bonding between the encapsulation layer and phosphate groups on the liposome surface, leading to reduced vibrational frequency. Additionally, the liposomal band at 1260.76 cm⁻¹(P=O) either disappeared or shifted upon encapsulation, demonstrating electrostatic and/or hydrogen-bonding interactions between the phospholipid headgroups and the encapsulation layer that altered vibrational modes.

Morphological characterization was further performed using TEM. Liposomal samples were negatively stained with 5 % phosphotungstic acid prior to imaging. Comparative analysis of TEM micrographs revealed distinct structural differences between formulations LUT@LIP and LUT@LIP-BSA. Specifically, the presence of an additional dense layer at the periphery of formulation LUT@LIP-BSA suggests successful encapsulation of component CS-BSA onto the outer surface of LUT@LIP(Fig. 2B and C). This observable surface modification is consistent with the results detected by FTIR and DLS analysis. Furthermore, stability assessment of LUT@LIP-BSA in PBS revealed superior maintenance of physicochemical properties at 4 °C compared to 25 °C over 2 weeks. Specifically, samples stored at 4 °C exhibited significantly less variation in both average hydrodynamic diameter and encapsulation efficiency than those at 25 °C. Collectively, these data demonstrate favorable short-term stability of formulation LUT@LIP-BSA at 4 °C(Fig. S1–5). Additionally, serum stability experiments demonstrated that LUT@LIP-BSA exhibited increased particle size over time in 10 % FBS, due to serum protein attachment to the nanoparticle surface. Nevertheless, overall stability remained superior (Fig. S6 and 7).

The drug release profiles of formulations Free LUT、LUT@LIP and LUT@LIP-BSA were evaluated in PBS (pH 7.4) to simulate in vivo drug release conditions. The cumulative release rate of drug LUT from formulation Free LUT reached 94 ± 2 % within 4 h, significantly higher than that of LUT@LIP-BSA (about 34.33 ± 0.58 %) and LUT@LIP (about 25 %). This distinct release hierarchy demonstrates that formulations LUT@LIP and LUT@LIP-BSA maintain structural integrity under physiological pH conditions (e.g., blood circulation and extracellular environments), thereby preventing premature drug leakage(Fig. 2H).

3.2. Preparation and characterization of CMB

CMB was synthesized via a one-pot method. To investigate the structure of CMB, TEM was employed. TEM analysis confirmed the successful synthesis of A, revealing well-defined nanoparticles (Fig. 2J and K). DLS demonstrated that the average particle size of the CMB was 54.03 ± 4.79 nm and the zeta potential was −42.53 ± 4.80 mV. The microstructure of CMB was further examined using XRD (Fig. 2D and E). The XRD pattern of CMB exhibited distinct peaks compared to that of BSA, indicating its well-crystallized structure, which is consistent with the TEM observations. Furthermore, characteristic diffraction peaks corresponding to both Mn (e.g., (100), (101), (110), (160), (303)) and Ce (e.g., (002), (111), (112), (220), (311)) were identified in the XRD pattern of CMB, providing additional evidence for the successful synthesis of CMB (Fig. 2O). XPS was employed to investigate the relative proportions of different oxidation states of metals Ce and Mn in CMB. The results revealed the coexistence of Ce(III)/Ce(IV) and Mn(IV)/Mn(II) species in CMB. Further quantitative analysis indicated a notably high proportion of Ce(III) in CMB, reaching as high as 55.9 %(Fig. 2N). Previous studies have demonstrated that a higher Ce(III)/Ce(IV) ratio in cerium oxide nanomaterials correlates positively with enhanced superoxide dismutase (SOD)-mimetic activity, which is beneficial for treating diseases associated with oxidative stress or inflammation [24]. UV–vis absorption spectroscopy showed that CMB exhibits the characteristic absorption peak of BSA at 300 nm, along with broad absorbance across the 250–400 nm range. However, no distinct absorption features attributable to the Mn element were observed. This absence may be attributed to the d⁵ electronic configuration of Mn(II), which represents a relatively stable half-filled d-orbital state and is consequently less prone to electronic transitions in the ultraviolet region (Fig. 2L).

FTIR spectroscopy revealed the presence of characteristic peaks associated with BSA (at 1396、1655 and 3300 cm⁻¹) in CMB, providing further evidence of successful synthesis of CMB (Fig. 2M). Additionally, the stability of CMB in PBS (pH 7.4) was evaluated over 14 days at different temperatures. It was found that both the average particle size and zeta potential of CMB remained more stable at 25 °C. However, significant protein precipitation was visibly observed at 4 °C. This phenomenon is likely due to the aggregation propensity of the protein component of BSA at lower temperatures, leading to the formation of macroscopically visible precipitate (Figs. S8–11). Additionally, In addition, the serum stability results of CMB showed consistency with LUT@LIP-BSA (Fig. S12 and 13).

3.3. In vivo and in vitro targeting effects

aHSCs are widely present in LF. Previous studies have demonstrated high binding affinity of BSA towards receptor SPARC on the surface of aHSCs. Using CLSM, we investigated the cellular uptake kinetics of Free C6、C6@LIP、C6@LIP-BSA and DIO@CMB in aHSCs over a 4-h period. As illustrated in Fig. 3A, aHSCs treated with C6@LIP-BSA exhibited significantly stronger green fluorescence intensity compared to those exposed to C6@LIP and Free C6. This indicates that LIP-mediated delivery of C6 enhances cellular internalization, and subsequent loading of LIP-BSA onto the carrier further potentiates uptake efficiency. Competitive inhibition assays revealed that upon addition of free BSA, the uptake of both C6@LIP-BSA and DIO@CMB by aHSCs was markedly reduced. This verifies the receptor-mediated targeting specificity of BSA-functionalized nanoparticles. These findings were further corroborated by quantitative FCM analysis (Fig. 3B–D).

Fig. 3.

Ex vivo and in vivo distribution and uptake of LUT@LIP-BSA and CMB. (A) Confocal images of uptake of Free C6, C6@LIP, C6@LIP-BSA, C6@LIP-BSA + BSA, DIO@CMB and DIO@CMB + BSA by aHSCs. (B,C) Determination of uptake flow cytometry images of aHSCs on Free C6, C6@LIP, C6@LIP-BSA, C6@LIP-BSA + BSA, DIO@CMB and DIO@CMB + BSA. (n = 3) (D) Quantification of uptake flow cytometry of aHSCs against Free C6, C6@LIP, C6@LIP-BSA, C6@LIP-BSA + BSA, DIO@CMB and DIO@CMB + BSA. (n = 3) (E) In vivo distribution of Free DIR, DIR@LIP, DIR@LIP-BSA and DIR@CMB at 2 h and 4 h in liver fibrosis model mice. (n = 3) (F) Heart, liver, spleen, lung and kidney distribution of Free DIR, DIR@LIP, DIR@LIP-BSA and DIR@CMB at 4 h in liver fibrosis model mice. (G) Immunofluorescence co-localization of α-SMA and DIO by Free DIO, DIO@LIP, DIO@LIP-BSA and DIO@CMB in liver fibrosis livers at 4 h. (n = 3) Scale bar: 20 μm. The data represent mean ± SD; ∗∗P < 0.01, ∗∗∗P < 0.001.

Furthermore, to comprehensively evaluate the in vivo targeting capability of the codelivery system toward aHSCs in LF, we performed real-time small-animal in vivo imaging to monitor the spatiotemporal distribution of nanoparticles in diseased mice at 2 h and 4 h post-injection. As depicted in Fig. 3E, the DIR group exhibited rapid clearance due to insufficient circulatory stability, resulting in substantially diminished fluorescence intensity. In contrast, BSA-functionalized DIR@LIP-BSA demonstrated higher fluorescence intensity in target tissues compared to DIR@LIP and DIR groups, respectively. Similarly, DIR@CMB showed significantly enhanced accumulation versus DIR group. Notably, both DIR@CMB and DIR@LIP-BSA displayed desirable liver accumulation, attributable to target effects mediated by ligand BSA (Fig. 3F).

To visually examine nanoparticle distribution within aHSCs of fibrotic livers, ex vivo immunofluorescence co-localization analysis was performed on DIO-labeled nanoparticles and biomarker α-SMA. Fluorescence channels were assigned as: red for α-SMA (aHSCs marker), green for DIO (nanoparticles), and blue for DAPI (nuclear stain). Compared to free DIO and DIO@LIP groups, DIO@LIP-BSA and DIO@CMB groups demonstrated significantly enhanced green-red co-localization signals, providing further evidence for the targeting efficacy of BSA-modified nanoparticles toward aHSCs (Fig. 3G).

3.4. Inhibition of proliferative activity of HSCs and promotion of apoptosis of aHSCs

To investigate the anti-proliferative effects of nanoparticles on cells HSC-T6 and AML-12 and identify optimal dosing concentrations, cells were treated with varying concentrations of Free LUT、LUT@LIP、LUT@LIP-BSA、CMB and LUT@LIP-BSA + CMB for 24 h, followed by cell viability assessment using CCK-8 assay. As shown in Fig. 4A and B, at equivalent concentrations, Free LUT induced the lowest viability in HSC-T6, followed by LUT@LIP or LUT@LIP-BSA. This is attributed to sustained drug release from liposomal encapsulation reducing acute cytotoxicity. Increasing concentrations enhanced LUT@LIP-BSA 's toxicity over LUT@LIP in HSC-T6 due to its positive charge-facilitated cellular uptake, whereas no such difference occurred in AML-12. CMB exhibited minimal cytotoxicity in both cell types. Consequently, 20 μM LUT@LIP-BSA(≥70 % viability) was selected for combination with graded CMB concentrations. The optimized combination of 200 μM CMB and 20 μM LUT@LIP-BSA maintained ≥80 % cell viability, demonstrating favorable biocompatibility for subsequent studies.

Fig. 4.

Cytotoxicity and apoptosis. (A) Cytotoxicity of HSC cells. (n = 5) (B) Cytotoxicity of AML-12 cells. (n = 5) (C) Flow cytometry determination of TGF-β, free LUT + TGF-β, LUT-LIP + TGF-β, or LUT@LIP-BSA + TGF-β induced apoptosis in aHSCs. (D) Quantitative flow cytometry plot of TGF-β, Free LUT, LUT@LIP or LUT@LIP-BSA induced apoptosis in aHSCs. (n = 3) (E) Flow cytometry of TGF-β, LUT@LIP-BSA or LUT@LIP-BSA + CMB induced apoptosis in aHSCs. (F) Flow cytometry quantification of LUT@LIP-BSA or LUT@LIP-BSA + CMB induced apoptosis in aHSCs. (n = 3) (G) TUNEL fluorescence sections of liver tissue from mice with liver fibrosis in different treatment groups. Scale bar: 50 μm. The data represent mean ± SD; ∗P < 0.05, ∗∗∗P < 0.001, ns indicates P > 0.05.

To investigate the pro-apoptotic effect of LUT@LIP-BSA + CMB on aHSCs, flow cytometry was performed to quantify apoptosis after treatment with Free LUT、LUT@LIP、LUT@LIP-BSA or LUT@LIP-BSA + CMB. LUT@LIP-BSA significantly induced apoptosis (9.14 ± 1.02 % vs control 2.27 ± 0.83 %), surpassing Free LUT and LUT@LIP groups (Fig. 4C and D). Mechanistically, LUT@LIP-BSA's effect may derive from BSA modification-enhanced cellular uptake. Meanwhile, LUT@LIP-BSA + CMB exhibited the most potent pro-apoptotic activity (19.50 ± 1.22 % vs control 6.88 ± 0.39 %), representing about 7 % increase over LUT@LIP-BSA (11.99 ± 0.83 %vs control 6.88 ± 0.39 %) (Fig. 4E and F). It's superior efficacy reveals a novel apoptosis-inducing function of CMB.

Current research indicates that LUT can induce cellular DNA damage and apoptosis through the p53/Bcl-2 signaling pathway [32]. To investigate the mechanism of the dual-delivery system on apoptosis in aHSCs, we examined the expression of apoptosis-related proteins p53 and BCL-2 in aHSCs following drug administration. The results indicated that, compared to the control group, p53 expression increased after treatment, while the expression of the anti-apoptotic protein BCL-2 was reduced (Fig. S19)

Additionly, in vivo validation via TUNEL staining of fibrotic liver sections showed significantly higher apoptotic cell density in the LUT@LIP-BSA + CMB group (Fig. 4G). Collectively, combined administration of LUT@LIP-BSA + CMB selectively induces apoptosis of aHSCs through the p53/BCL-2 pathway, thereby treating liver fibrosis.

3.5. ROS scavenging activity in vitro and in vivo

Excessive accumulation of ROS exacerbates inflammatory responses, thereby stimulating the production of pro-fibrotic mediators. Multiple studies have demonstrated that LUT and Ce³⁺ exhibit potent ROS-scavenging capabilities, effectively suppressing ROS generation and consequently mitigating fibrotic progression [3,33].

To evaluate the antioxidative stress capacity of LUT@LIP-BSA + CMB, CLSM was employed to assess ROS scavenging efficiency in TGF-β-stimulated HSC-T6 cells. Free LUT, LUT@LIP, LUT@LIP-BSA or LUT@LIP-BSA + CMB treatments progressively attenuated intracellular ROS fluorescence intensity, with LUT@LIP-BSA + CMB exhibiting the most potent suppression (Fig. 5A). This effect was further validated by FCM of DCFH-DA fluorescence (Fig. 5B), showing strong correlation with microplate reader measurements (Fig. S14).

Fig. 5.

Antioxidant stress in vivo and in vitro. (A) Fluorescence microscopy images of ROS scavenging effect in aHSCs. Scale bar: 20 μm. (B) Flow cytometry quantification of ROS scavenging effect in aHSCs. (n = 3) (C–E) MDA, GSH and SOD content in liver tissue homogenates. (n = 6) (F) Immunohistochemical paraffin sections of liver tissue for Nrf2 and Keap1. Scale bar: 50 μm. (G,H) Quantification of Nrf2 and Keap1 expression in immunohistochemical paraffin sections of liver tissue. (n = 6) The data represent mean ± SD; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.00, ns indicates P > 0.05. # means that compared to Control.

During hepatic fibrogenesis, excessive ROS accumulation in aHSCs dysregulates the Nrf2 signaling pathway. Under physiological activation, Nrf2 dissociates from its cytosolic inhibitor Keap1 and translocates to the nucleus, inducing transcription of downstream antioxidant genes that suppress HSCs activation and ECM deposition [3,34]. To evaluate systemic antioxidative stress capacity, we measured the levels of SOD, GSH and MDA in ex vivo liver tissues from fibrotic mice. The treatment groups significantly reversed the decreased SOD and GSH levels as well as the increased MDA levels observed in model groups, suggesting potential amelioration of LF (Fig. 5C–E). Furthermore, it has been shown that the antioxidant effects of LUT and Ce are related to the Nrf2 pathway [33,35,36] In order to elucidate the anti-oxidative stress mechanism in vivo, we used immunohistochemistry to detect Nrf2 and Keap1 expression levels in the livers of mice in different treatment groups. The results showed that Nrf2 expression was significantly lower in the model group than in the control group. However, the treatment group effectively reversed the low expression of Nrf2 in mice with LF. It is noteworthy that Keap1 expression levels exhibited a significant negative correlation with Nrf2, further confirming their classical antagonism in oxidative stress regulation (Fig. 5F–H).

Collectively, these studies suggest that LUT@LIP-BSA + CMB may ameliorate LF by suppressing ROS generation in aHSC-T6 cells through downregulation of Keap1 expression and concomitant activation of the Nrf2 signaling pathway, thereby exerting antioxidative stress effects.

We further evaluated the antioxidant capacity of LUT@LIP-BSA + CMB. Using DPPH and ABTS as in vitro surrogates for ROS, LUT@LIP-BSA + CMB demonstrated significantly higher scavenging activities against both radicals compared to Free LUT、LUT@LIP and LUT@LIP-BSA at equivalent concentrations. Specifically, at 200 μM, LUT@LIP-BSA + CMB decomposed approximately 88 % of DPPH radicals and 82 % of ABTS⁺ radicals, confirming its potent antioxidant properties (Figs. S15 and S16).·O₂⁻ and H₂O₂, categorized as ROS, mediate cellular signaling at physiological levels but induce oxidative tissue damage upon exceeding homeostatic thresholds. [27] In addition, LUT@LIP-BSA + CMB exhibited superior H₂O₂ scavenging capacity and ·O2–catabolic activity compared to Free LUT, LUT@LIP or LUT@LIP-BSA (Figs. S17 and S18). These findings suggest that LUT@LIP-BSA + CMB has potent SOD-like and CAT-like activities.

3.6. Induction of cellular senescence in vitro and in vivo

Mn2+ has been demonstrated to modulate the STING signaling pathway, which is mechanistically linked to cellular senescence [30,31]. We therefore investigated the regulatory effects of CMB on senescence-associated pathways in aHSC-T6 using immunofluorescence (Fig. 6A–F). We found that Mn2+ treatment significantly upregulated the expression of STING, IL-6 and IL-8 in aHSC-T6, consistent with our hypothesis that STING activation enhances SASP factor secretion [28,30,37].。In contrast, the Mn2+-induced increase in these expressions was reduced by the administration of CMB, an effect that may result from the slowing down of the cellular uptake rate of Mn2+ by CMB, as well as the Ce3+-mediated anti-inflammatory effect [38,39]. Notably, LUT@LIP-BSA + CMB group showed a paradoxical increase in STING expression and a decrease in IL-6 and IL-8 expression. We hypothesize that LUT may specifically enhance STING transcriptional activity and that it has anti-inflammatory properties.

Fig. 6.

Induction of aHSCs cell senescence ex vivo. (A) Immunofluorescence for STING expression in aHSCs. Scale bar: 20 μm. (B) Immunofluorescence quantification of STING in aHSCs. (n = 3) (C) Immunofluorescence for IL-6 expression in aHSCs. Scale bar: 20 μm. (D) Immunofluorescence quantification of IL-6 in aHSCs. (n = 3) (E) Immunofluorescence for IL-8 expression in aHSCs. Scale bar: 20 μm. (F) Immunofluorescence quantification of IL-8 in aHSCs. (n = 3) (G) Elisa determination of IL-6 in liver tissue. (n = 6) (H,I) Co-localization of P21, P16 and α - SMA detected by immunofluorescence in paraffin sections of liver tissue. Scale bar: 20 μm. The data represent mean ± SD; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ns indicates P > 0.05.

Immunofluorescence analysis of murine liver tissues examined the co-localization of senescence-associated proteins p16, p21 and α-SMA. LUT@LIP-BSA + CMB and CMB groups exhibited significantly enhanced expression of both p16 and p21 versus other groups, whereas no upregulation was observed in other treatments. Concurrently, α-SMA expression demonstrated consistent reduction across all treatment groups (Fig. 6H and I). Besides, serum analysis revealed significant decrease in IL-6 levels though the combination group showed higher IL-6 expression than LUT@LIP-BSA group. We speculate that this may be due to increased secretion of SASP factors induced by Mn2+. As expected, the expression of IL-6 was significantly enhanced in the CMB treatment group compared to the combination therapy group (Fig. 6G).

Synthesizing these findings, we hypothesize that CMB may modulate the STING pathway to upregulate p16 and p21 expression while increasing the secretion of SASP factors (IL-6 and IL-8), thereby inducing senescence in aHSCs. Notably, LUT@LIP-BSA group significantly attenuated the expression of these pro-inflammatory mediators.

3.7. Anti-fibrotic effect in vitro and in vivo

To evaluate the therapeutic efficacy of LUT@LIP-BSA + CMB in vivo, a carbon tetrachloride (CCl₄)-induced LF model was established, followed by treatment with Free LUT, LUT@LIP, LUT@LIP-BSA, CMB or LUT@LIP-BSA + CMB.

Histopathological examination via H&E staining demonstrated that LUT@LIP-BSA + CMB treatment substantially attenuated inflammatory cell infiltration in fibrotic livers, but was lower efficacy than LUT@LIP-BSA group (Fig. 7A). This phenomenon may be caused by the aggregation of immune cells (such as natural killer cells (NK) and macrophages) induced by inflammatory factors secreted by CMB-induced senescent aHSCs. In accordance with the theoretical framework positing the accumulation of collagen fibrils in fibrotic microenvironments, quantitative analysis of Masson's trichrome and Sirius Red staining revealed a diminution in collagen deposition and fibrotic area in mice treated with LUT@LIP-BSA + CMB (Fig. 7A–C). Immunohistochemical analysis further confirmed diminished positive expression of collagen-I, indicating LUT@LIP-BSA + CMB ameliorates LF through suppression of collagen-I biosynthesis pathways (Fig. 7A–D).

Fig. 7.

Effectiveness of ex vivo treatment of liver fibrosis. (A) HE staining, Masson ' s trichrome staining, Sirius Red staining, COL1 immunohistochemical staining, and immunofluorescence staining for TGF-β1 and α-SMA in paraffin sections of liver tissue. Scale bar: HE, Masson's trichrome, Sirius Red 200 μm; COL1,50 μm; TGF-β1 and α-SMA 20 μm. (B–F) Quantification of Masson ' s trichrome staining, Sirius Red staining, COL1 immunohistochemical staining, TGF-β1 and α-SMA immunofluorescence staining in paraffin sections of liver tissue (n = 5) (G) Immunofluorescence staining for α-SMA in aHSCs. (H) Quantification of immunofluorescence staining for α-SMA in aHSC. (n = 3) (I) HYP content in liver tissue homogenates. (n = 6) (J,K) Levels of AST and ALT in liver serum. (n = 6) The data represent mean ± SD; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ns indicates P > 0.05. # means that compared to Control. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Within the fibrotic liver microenvironment, activation of the TGF-β signaling pathway drives HSCs activation with concomitant overexpression of α-SMA, leading to aberrant extracellular matrix deposition [40,41]. Additionally, TGF-β1, a family to which TGF-β belongs, also induces HSCs activation in the setting of LF [42]. Mechanistic investigation revealed that LUT@LIP-BSA + CMB treatment significantly suppressed TGF-β-induced α-SMA upregulation in HSCs (From 28.25 ± 1.5 % to 9.46 ± 1.24 %), outperforming LUT@LIP-BSA group (13.18 ± 0.67 %) in immunofluorescence assays (Fig. 7G and H). Meanwhile, histological analysis of liver sections confirmed that α-SMA expression was also significantly elevated when TGF-β1 was highly expressed in the LF group. Both were significantly reduced after LUT@LIP-BSA + CMB treatment, and the effect was stronger than the other groups (Fig. 7A–E,F).

Serological analysis confirmed significant reductions in hepatic function markers (AST and ALT) and pro-inflammatory cytokines (TNF-α, IL-6), accompanied by decreased HYP level in the LUT@LIP-BSA + CMB group (Fig. 6, Fig. 7-L). Collectively, LUT@LIP-BSA + CMB ameliorates LF via dual blockade of the TGF-β signaling axis, thereby inhibiting hepatic stellate cell activation, downregulating α-SMA expression, and reversing aberrant extracellular matrix remodeling.

3.8. Preliminary biosafety evaluation

The in vivo safety profile of formulation LUT@LIP-BSA + CMB was further evaluated throughout the therapeutic regimen. After n weeks of treatment, histopathological assessment via H&E staining revealed no significant tissue damage in major organs (heart, spleen, lungs, kidneys) across all treatment groups. Serum biochemical analysis demonstrated that renal function markers (CREA,BUN) and myocardial enzymes (CK) in Free LUT、LUT@LIP、LUT@LIP-BSA or LUT@LIP-BSA + CMB groups showed no statistically significant differences versus Control. Although the LUT@LIP-BSA + CMB group exhibited elevated BUN levels compared to LUT@LIP-BSA group (p ≤ 0.05), both remained within normal physiological ranges (Fig. 8B–D). Hemolysis assays indicated that Free LUT、LUT@LIP or CMB formulations at concentrations up to 200 μg/mL induced <5 % hemolysis, whereas the LUT@LIP-BSA reached >7 % hemolysis at 200 μg/mL – attributable to chitosan-induced positive surface charge disrupting erythrocyte membrane integrity (Fig. 8E and F). Critically, repeated administrations caused no adverse effects on body weight trends or vital organ functions, confirming the favorable biosafety of the LUT@LIP-BSA + CMB. Additionally, our in vivo distribution study in hepatic fibrosis mice demonstrates that both nanoparticles predominantly accumulate in the liver, with significant co-localization observed at α-SMA expression sites. This targeted distribution profile suggests enhanced systemic safety (Fig. 3E–G).

Fig. 8.

Preliminary Security Evaluation. (A) HE staining of paraffin sections of heart, spleen, lung and kidney tissues. (B–D) Serum levels of CRAE, BUN and CK in mice. (n = 6) (E) Hemolytic activity of different concentrations of Free LUT, LUT@LIP, LUT@LIP-BSA and CMB on mouse blood samples. (F) Quantification of hemolytic activity. (n = 3) The data represent mean ± SD; ∗P < 0.05, ns indicates P > 0.05.

Despite the promising results of our work, there are still some limitations that need to be addressed. This process of cellular senescence induced by nanoenzymes that may be cleared by NK cells needs to be further explored. In addition, in order to better enable the dual delivery system to be used in clinical applications, whether the metal nano-enzymes have in vivo with potential long-term accumulation toxicity should be emphasized.

4. Conclusion

In this study, we designed a therapeutic strategy of a combined dual nanodelivery system aimed at treating LF by promoting apoptosis/senescence and anti-oxidative stress in aHSCs. To enhance the therapeutic efficacy, we designed a dual nanodelivery system targeting aHSCs. The experimental results showed that bimetallic nanoenzymes and LUT nanoliposomes were enriched into the liver after tail vein injection and were effectively internalized by aHSCs. To our satisfaction, they had the ability to co-regulate the Nrf2/Keap1 pathway to exert an antioxidant effect, while significantly reducing the ROS level in aHSCs, which inhibited HSCs activation and extracellular matrix secretion. We also found through mechanistic studies that CMB increased the expression of p21 and p16 in aHSCs by regulating the STING pathway, which induced cellular senescence and produced SASP factors (IL-6, IL-8). The release of such SASP factors may promote phagocytosis of senescent HSCs by NK cells. Notably, LUT@LIP-BSA consistently attenuated CMB-induced inflammation in both in vitro and in vivo models, aligning with our initial hypothesis. In addition, the dual nanocarrier system exhibited a synergistic pro-apoptotic effect in aHSCs, accompanied by regulation of the p53/bcl-2 pathway. In conclusion, our results suggest that the dual nano-co-delivery system works together in vitro and in vivo to alleviate LF through anti-oxidative stress, apoptosis/senescence promotion, and anti-inflammation, and that this multifunctional synergistic therapeutic strategy is promising for research.

CRediT authorship contribution statement

Zihao Sun: Writing – review & editing, Writing – original draft. Chuipeng Liang: Visualization, Validation, Supervision. Yuxin Zhao: Visualization, Validation, Supervision. Jijiao Wu: Software, Resources, Project administration, Methodology. Lin Wen: Software, Resources, Project administration, Methodology. Xiaolian Liu: Software, Resources, Project administration, Methodology. Mingyi Shi: Funding acquisition, Formal analysis, Data curation, Conceptualization. Xiaofang Li: Investigation, Funding acquisition, Formal analysis, Data curation.

Notes

The authors declare no competing financial interest.

Declaration of competing interest

The article is not a declared conflict of interest.

Appendix B. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.