Vitamin A-Modified Berberine Liposomes Potentiate Anti-Fibrotic Effects Through Enhanced Hepatic Stellate Cell Uptake

Abstract

Background

Activated hepatic stellate cells (HSCs) are key drivers of liver fibrosis, which progresses irreversibly without effective intervention. The natural alkaloid berberine (BBR) has been reported to possess anti-fibrotic activity. However, the poor solubility and injection-associated systemic toxicity of free BBR confine it to oral administration, resulting in its low bioavailability. Nanocarrier encapsulation addresses these drawbacks, and we therefore aimed to develop an HSC‑preferential delivery system with loaded BBR to improve its anti-fibrosis effects without causing systemic toxicity.

Methods

In this study, vitamin A-modified lipids were synthesized to fabricate HSC‑preferential liposomes to encapsulate BBR and enhance delivery efficiency. This targeted BBR liposomal system was physiochemically characterized, followed by in vitro assessments of cellular uptake and inhibition of HSCs activation. In vivo animal studies were performed to validate the therapeutic efficacy of targeted BBR liposomes against hepatic fibrosis.

Results

VA-modified BBR liposomes had an average size of 165.9 ± 7.1 nm with the desired size distribution. Compared to untargeted BBR liposomes, VA-modified liposomal BBR showed significantly increased HSC uptake, thus indicating enhanced inhibition of HSC activation by dramatically reducing HSC migration and proliferation abilities and decreased collagen1 expression in vitro. Intriguingly, contrary to earlier literature, we detected elevated α‑SMA expression upon BBR treatment despite the suppression of cell migration, proliferation and collagen production. In hepatic fibrosis mouse models, targeted BBR liposomes at 5 mg/kg significantly decreased aspartate aminotransferase and alanine aminotransferase levels in serum, and markedly reduced fibrotic areas and collagen synthesis and deposition, while equal doses of untargeted BBR liposomes showed negligible effects.

Conclusion

Our results demonstrate that VA-functionalized liposomes facilitate the efficient targeted delivery of BBR to activated HSCs in hepatic fibrosis. Such targeted delivery augments the inherent anti-fibrotic potency of free BBR, highlighting the potential of this nanoplatform for hepatic fibrosis treatment.

Introduction

Hepatic fibrosis is characterized by repeated wound healing driven by the persistent action of various damaging factors, primarily manifested as the activation and excessive proliferation of hepatic stellate cells (HSCs) into myofibroblasts, which produce and deposit excessive extracellular matrix (ECM) components, such as collagen.^1,2 Without timely intervention, it can eventually progress to irreversible advanced fibrosis stages, including cirrhosis and liver cancer, which are associated with high morbidity and mortality rates, thus making early intervention of hepatic fibrosis crucial.³ Currently, clinical treatment strategies for hepatic fibrosis are limited and lack specificity. The core therapeutic principles mainly focus on etiological control, such as antiviral therapy for viral hepatitis,⁴ lifestyle intervention (eg, alcohol abstinence) for alcoholic liver disease,⁵ and symptomatic support to alleviate complications. However, no approved drugs can directly inhibit HSC activation, block ECM synthesis, or reverse the established fibrotic process. Moreover, existing etiological treatments are often accompanied by poor compliance, drug resistance, and limited efficacy in advanced stages.^6,7 Thus, the development of safe, effective, and targeted therapeutics has become an urgent demand for ameliorating liver fibrosis.

Many natural extracts, such as curcumin,⁸ silymarin,⁹ and matrine¹⁰ may also serve as sources of therapeutics for anti-fibrotic fibrosis. Berberine (BBR) is an isoquinoline alkaloid extracted from Coptis Chinensis and Phellodendron, and other traditional Chinese herbs. Extensive studies have demonstrated that BBR exhibits multiple potential pharmacological activities, including antibacterial, antiviral, antidiarrheal, anti-inflammatory, and anticancer properties, as well as the ability to exert anti-hepatic fibrosis activity.^3,11–14 Despite its wide therapeutic potential, the clinical application of BBR is severely limited owing to its poor aqueous solubility, chemical stability, and low bioavailability, although berberine hydrochloride improves solubility to some extent. In addition, potential acute cardiac toxicity from free BBR injections also limits the systemic administration of BBR.^15–17 Kheir et al reported that the median lethal doses (LD50) of BBR in mice via oral, intravenous, and intraperitoneal administration are >20.8 g/kg, 9.04, and 57.6 mg/kg, respectively.¹⁵ As a result, poor bioavailability and cardiac toxicity, requiring high administration doses to achieve therapeutic effects, thus largely confining BBR to the treatment of gastrointestinal disorders via oral administration, which is currently the only approved administration route for berberine.

Using nanocarriers such as liposomes to encapsulate these drugs provides a promising strategy to overcome the inherent limitations of free drug administration and significantly optimize the therapeutic effect and safety of drugs. In addition, liposomes, as approved nanocarriers for clinical use for many years, have been widely applied in pharmaceutical preparations owing to their unique structure and functional advantages.^18–20 Liposome encapsulation can reduce the toxicity of BBR by decreasing the concentration of free drug in the systemic circulation.²¹ Meanwhile, conventional liposomes tend to accumulate preferentially in the liver due to recognition by the mononuclear phagocyte system.²² However, at the cellular level, these liver-accumulated conventional liposomes are almost exclusively taken up by Kupffer cells and liver sinusoidal endothelial cells, and are barely available for uptake by hepatic stellate cells.^22–24 Due to the lack of specificity of nanocarrier-encapsulated BBR on HSCs, relatively high doses are still required to maintain efficacy, which may cause unexpected side effects.^3,25 A promising strategy to overcome this limitation is to achieve HSC targeted delivery, such as functionalizing nanocarrier with HSC-specific ligands. Recent studies have reported targeted delivery approaches directed toward receptors and endogenous proteins overexpressed in activated HSCs. For instance, Zhang et al developed a PDGF‑derived pPB peptide that specifically recognizes and binds to PDGFR‑β on activated HSCs to mitigate HSC‑driven fibrotic progression.²⁶ In another study, Zhang et al conjugated Aminoethyl anisamide onto nanoparticles to realize improved HSC‑targeted antifibrotic delivery.²⁷ Shinn and coworkers modified nanoparticles with hyaluronic acid to target CD44 overexpressed on activated HSCs, so as to enhance HSC‑specific cargo delivery.²⁸ Nevertheless, these alternative targeting ligands are only validated in scattered individual reports and suffer from inherent drawbacks such as limited translational maturity, potential immunogenicity for peptide ligands, or questionable cell‑type specificity.^28–31 By comparison, vitamin A (VA)‑functionalized nanocarriers have been extensively established for HSC‑targeted therapy, relying on its well‑characterized physiological HSC‑targeting property, favorable biocompatibility, and substantial pre‑clinical validation.^32–37 HSCs are capable of taking up VA via specific ligand-receptor binding between retinol-binding protein 4 (RBP4) and its receptor, STRA6.^32,36 Notably, STRA6 expression is markedly increased in activated HSCs,^38,39 which endows VA-modified nanocarriers with great potential for targeted therapeutic effects on fibrotic HSCs at the cellular level.

VA-modified liposomes loaded with BBR can penetrate the dense fibrotic microenvironment via receptor-mediated endocytosis, significantly increasing the intracellular drug concentration in target cells while reducing non-specific accumulation in healthy tissues. Thus, VA-modified liposomes were rationally designed in this study to improve the bioavailability of berberine to target cells and reduce its systemic toxicity. BBR was encapsulated in VA-functionalized liposomes for targeted delivery. The resultant VA-BBR liposomes exhibited enhanced cellular uptake by activated HSCs and exerted prominent anti-fibrotic effects in vivo at a relatively low dose without compromising therapeutic efficacy.

Materials and Methods

Materials

Hydrogenated soybean phospholipid (HSPC), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-(amino(polyethylene glycol)-2000 (DSPE-PEG2000), and DSPE-PEG-NH2 were purchased from AVT (Shanghai, China). Vitamin A acid, cholesterol, N-hydroxysuccinimide (NHS), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC·HCl), berberine hydrochloride, and Picro-Sirius Red stain were purchased from Yuanye Biotechnology (Shanghai, China). Recombinant TGFβ protein was purchased from PeproTech. DiI dye and Hoechst 33342 were purchased from Beyotime Biotechnology (Shanghai, China). The quantitative polymerase chain reaction (qPCR) Master Mix Kit was purchased from Vazyme (Nanjing, China). CCK-8 assay kit and rat tail collagen type I were purchased from GLPBIO (Montclair, CA, USA). Hematoxylin-eosin and Picro Sirius Red staining kits were purchased from Beyotime Biotechnology (Shanghai, China). Aspartate aminotransferase (AST), alanine aminotransferase (ALT), and Masson’s trichrome stain kits were purchased from Solarbio (Beijing, China). All antibodies used for Western blot or immunofluorescence assay were purchased from Proteintech (Wuhan, China) or Cell Signaling (Danvers, MA, USA) and summarized in Table 1.

Table 1.

Summary of Antibodies Used for Different Assays

Antibody	Dilution for Application	Brand (Cat. No).
Alpha smooth muscle actin polyclonal antibody.	1:2000 for Western blot assay; 1:200 for immunofluorescence assay;	Proteintech (14395-1-AP)
Collagen type I polyclonal antibody.	1:800 for Western blot assay; 1:200 for immunofluorescence assay;	Proteintech (14695-1-AP)
CoraLite488-conjugated goat anti-rabbit IgG (H+L).	1:400 for immunofluorescence assay	Proteintech (SA00013-2)
CoraLite594-conjugate goat anti-rabbit IgG (H+L).	1:400 for immunofluorescence assay	Proteintech (RGAR004)
GAPDH monoclonal antibody, Clone No. 1H18.	1:10000 for Western blot assay	Proteintech (81640-5-RR)
DyLight 680 Conjugate anti-rabbit IgG (H+L).	1:20000 for Western blot assay	Cell signaling (5366P)

Synthesis of DSPE-VA Lipid

DSPE-VA was synthesized according to previous report.⁴⁰ Briefly, vitamin A acid, EDC·HCl and NHS were dissolved in chloroform at the molar ratio of 1:2:2, for reaction for 24 hours at 40 °C in a dark environment, followed by adding DSPE-PEG-NH2 (DSPE:VA at molar ratio of 1:2) for another 12 hours at 40 °C. The resulting solution was dialyzed in methanol aqueous solution with a volume ratio of 1:1 and pure water for 2 days, respectively, using a dialysis bag with MWCO of 1kDa. Subsequently, the resultant products were lyophilized for 3 days to obtain yellow-colored DSPE-VA lipids. Fourier transform infrared (FTIR) spectroscopy was used for characterization (Agilent Cary 630).

Preparation of Liposomal Formulations and Characterization

VA-liposomes were composed of HSPC, cholesterol, DSPE-PEG, and DSPE-VA at a molar ratio of 60:30:4:6 and were prepared by the thin lipid film hydration method, followed by heated extrusion as described previously.⁴¹ Briefly, 100 µmol of lipids was dissolved in methanol/chloroform (1/9 v/v), followed by evaporation under vacuum and nitrogen flush. The lipid film was hydrated in PBS (blank liposomes) or in berberine hydrochloride solution at a concentration of 3 mg/mL (BBR-loaded liposomes). Small unilamellar liposomes were obtained by extrusion using Nuclepore filters with pore sizes ranging from 400 nm to 200 nm. A PD-10 column was used to separate the unencapsulated BBR. Similarly, non-VA liposomes were prepared with HSPC, cholesterol, and DSPE-PEG at a molar ratio of 66:30:4. Fluorescently labelled liposomes were prepared by adding DiI at a molar ratio of 0.1% during lipid film formation. The BBR encapsulation efficiency (EE) was calculated using the following equation:

where W_load and W_total are the weight of BBR encapsulated in liposomes and the total weight of BBR added, respectively.

Dynamic light scattering (DLS) using a Zetasizer Nano ZS90 instrument (Malvern, UK) was used to measure particle size, polydispersity index (PDI), and zeta potential of the prepared liposomal formulations. Transmission electron microscopy (TEM; JEOL-2100F, Japan) was used to examine the morphology of liposomes stained with 2% uranium acetate.

Cell Culture of Mouse HSCs

C57BL/6 mouse hepatic stellate cells (HSCs) were ordered from the Shanghai Cell Bank (Shanghai, China). Cells were cultured in DMEM/HIGH GLUCOSE medium (HyClone) supplemented with 10% fetal bovine serum (FBS), 100 μg/mL streptomycin, and 100 units/mL penicillin. Cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO₂. All stellate cells used in this study were maintained within 10 passages, as their proliferation tended to arrest beyond this range according to information provided by the supplier.

VA-Liposome Targeting Assay

DiI-labeled liposomes were used to assess the targeting ability of HSCs. For non-activated cell targeting, HSCs were seeded at a density of 7.5×10⁴ cells per well into a 6-well plate. After cell adhesion, the cells were starved for 12 h in 5% FBS medium, followed by the addition of VA-liposomes or non-VA liposomes at a concentration of 6 nM (calculated based on DiI concentration) for 2 h. After cell adhesion, HSCs were starved for 12 h in 5% FBS medium, followed by TGFβ 10 ng/mL activation for 24 h. DiI-liposomes were then added and co-incubated for 2 h. Subsequently, these HSCs were fixed with 4% paraformaldehyde and Hoechst staining before fluorescence inverted microscope imaging (Nikon Ti2-U, Japan).

VA-Modified Liposomal BBR Release and Stability Evaluation

After concentration, the BBR content of the VA-modified liposomes was determined using UV spectroscopy. An appropriate amount of the liposome sample (0.6mg BBR) was loaded into 1 kDa MWCO dialysis bag, which was placed in 100 mL PBS and incubated at 37 °C with constant stirring. Samples were withdrawn at scheduled intervals for 5 days, followed by the measurement of BBR in the release solution by using ultraviolet spectrophotometry. Briefly, a series of BBR standard substances with accurately weighed masses were dissolved in PBS. The absorbance of standard solutions was determined at a wavelength of 345 nm to establish the calibration curve with the coefficient of determination at 1,⁴² for released BBR quantification.

For the leakage test of VA-modified BBR liposomes, the concentrated liposome suspension was aliquoted into portions of 500 μL each. An equal volume of PBS (500 μL) was added to each aliquot followed by thorough mixing. Then the Aliquots were stored at 4 °C. Samples were collected every 5 days during the experimental period. After gentle mixing, each sample was subjected to centrifugation at 22,000 rpm for 2 hours at 4°C. The supernatant was then analyzed to quantify the BBR leakage using the same BBR release quantification method described above.

Cellular Uptake Assay with BBR Extraction

HSCs were seeded in 6-well plates at a density of 2×10⁵ cells per well. After attachment, VA-modified liposomal BBR and unmodified liposomal BBR were separately added at a BBR equivalent dose of 800 μg per well and incubated for 1 h. Subsequently, cells were rinsed with PBS and digested with trypsin. The cell suspensions were collected in centrifuge tubes and washed again with PBS before discarding the supernatant. 75% ethanol (600 μL) was added to each tube to extract the BBR. Samples were incubated at 40 °C for 15 min with shaking and ultrasonication, then kept at room temperature for 4 h. After centrifugation at 5000 rpm for 10 min, the supernatant of the cell sample was transferred to a new tube. Quantitative analysis of BBR in the supernatant was performed as described below by using ultraviolet spectrophotometry. Briefly, a series of BBR standard substances with accurately weighed masses were dissolved in 75% ethanol solution. The absorbance of standard solutions was determined at a wavelength of 345 nm to establish the calibration curve, whose coefficient of determination reached 0.9995.⁴² The intracellular concentrations of BBR in all tested samples were subsequently calculated based on this calibration curve.

Cell Migration Assays of VA-Modified BBR Liposomes on HSCs

Different cell migration assays were used to evaluate the migration ability of the BBR liposomes on activated HSCs. The first cell scratch assay was performed by seeding HSCs into 12-well plates at a density of 5×10⁴ cells/well. After the cells were fully grown, linear scratching was induced using a tip and washed 1 time to remove the detached cells. Blank liposomes (VA-modified liposome without BBR encapsulation) or VA-modified BBR liposomes at a concentration of 40 µg/mL of BBR were added for 1 h co-incubation, followed by co-culture with TGFβ at a concentration of 10 ng/mL for 48 h. Scratches of these HSCs were imaged at 0 h, 24 h, and 48 h using fluorescence microscopy.

A transwell assay was used to further evaluate cellular migration capacity. HSCs (3 × 10⁴) were seeded into the upper chamber of Transwell inserts (8.0 μm pore size). After cell attachment, the upper chamber was replaced with serum-free medium and the lower chamber was filled with complete medium containing 10% FBS to form a chemotactic gradient. The cells in the upper chamber were treated with blank liposomes or VA-modified BBR liposomes for 1 h, followed by incubation with TGFβ for another 24 h. Subsequently, non-migrated cells on the inner surface of the upper membrane were gently wiped off with a cotton swab. The inserts were fixed with 4% paraformaldehyde and stained using a transwell staining kit. The cells that migrated to the outer membrane were counted under a microscope.

Cell Proliferation Assays of VA-Modified BBR Liposomes on HSCs

HSCs were seeded at a density of 1.5×10⁴ cells/well in a 24-well plate. After cell attachment, the cells were starved in a medium containing 5% FBS for 12 h, followed by incubation with blank liposomes or VA-modified BBR liposomes (40 µg/mL) for 1 h. These cells were then activated with TGFβ for 48 h. The CCK-8 kit was then used to measure the cell quantity from different groups at a wavelength of 450 nm.

A cell growth cycle assay was performed to further evaluate the cell proliferation ability. HSCs were seeded into 6-well plates at a density of 1×10⁵ cells/well and cultured to allow full attachment. After starvation in medium containing 5% FBS for 12 h, the cells were treated with blank liposomes and VA-modified BBR liposomes (40 µg/mL) for 1 h. HSCs were further activated with TGFβ for 24 h. Subsequently, the cells were digested with trypsin, collected, and rinsed twice with PBS. Cell pellets were fixed overnight at 4 °C. After removal of the fixative, cells were resuspended and stained with PI/RNase working solution (Cell Cycle Detection Kit, Beyotime) at 37 °C for 40 min in the dark. Cell cycle profiles (G0/G1, S, and G2/M phases) were analyzed using flow cytometry, with 30,000 events acquired for each sample. The proportion of cells in each phase was quantified using the ModFit LT 5.0.

Quantitative Polymerase Chain Reaction (qPCR) and Western Blot Analyses of VA-Modified Liposomal BBR on HSCs

qPCR analysis was performed to evaluate the therapeutic effects of liposomal BBR on activated HSCs. HSCs were seeded at a density of 9×10⁴ cells/well in 6-well plates. After cell attachment overnight, 5% FBS medium was used for starvation for 12 h, followed by treatment of HSCs with blank liposomes, non-targeted BBR liposomes, or targeted BBR liposomes (40 µg/mL BBR) for 1 h. Except for the control well, TGFβ (10 ng/mL) was then used in all wells for activation for 24 h. Cells were then collected, and RNA was extracted from different wells, followed by quantification and reverse transcription with HiScript II Q RT SuperMix (Vazyme). Subsequently, qPCR was performed using ChamQ SYBR qPCR Master Mix (Vazyme) in a real-time PCR detection system. The relative mRNA expression levels of target genes (including α‑SMA and Col1α1, primer sequences were summarized in Supplementary Table 1) were normalized to those of GAPDH and calculated.

Western blot analysis was performed in parallel. HSCs were seeded in 6-well plates for overnight and starved for 12 h with 5% FBS medium, followed by treatment with blank liposomes or BBR liposomes (40 µg/mL BBR) for 1h. TGFβ (10 ng/mL) was added in all wells excluding the control group for 48 h activation. Cells were then collected, and total protein was isolated. Subsequently, protein samples were separated by SDS-PAGE and subjected to quantitative Western blot detection.

Cell Contraction Assays of BBR on HSCs

Collagen gel contraction assay was carried out to assess the contractile capability of BBR‑treated HSCs. Briefly, HSCs were seeded at a density of 7 × 10⁵ cells per well in a 24‑well plate embedded within collagen gel. After gel solidification, the gel was carefully detached from the well wall. Subsequently, cells were pre‑incubated with or without BBR (40 µg/mL) for 1 h, and then stimulated with or without TGF β (10 ng/mL) for 48 h. Finally, the contraction area of each gel disc was quantified for contractile‑capacity evaluation.

Animals and Ethics Statement

C57BL/6J (male, 10 weeks old) mice were purchased from GemPharmatech, LLC. (China). All animal experiments were approved by the Animal Care and Use Committee of Xi’an Jiaotong-Liverpool University (Suzhou, China) (approval no.: ER-SRR-11000152820241217191941). All procedures were performed in strict accordance with the guidelines of the American Veterinary Medical Association (AVMA) for the euthanasia of animals and the Guide for the Care and Use of Laboratory Animals to minimize animal suffering throughout the study. The sample size for animal study was determined based on our experimental experience and previous published hepatic fibrosis studies,^39,43 balancing adequate statistical power and the 3R principles to minimize animal consumption.

In vivo Efficacy Study in CCl4-Induced Mouse Hepatic Fibrosis Models

Mouse models of liver fibrosis were established by CCl₄ intraperitoneal injection (i.p). (corn oil and CCl₄ mixed at a ratio of 9:1 by volume) at a dosage of 5 mL/kg body weight twice weekly for consecutive 8 weeks. CCl₄ treated mice were then randomly divided into three groups with five mice per group, which received tail vein administration of VA-modified BBR liposomes, non-VA modified BBR liposomes, or blank liposomes. All liposomal formulations were administered at a BBR-equivalent dose of 5 mg/kg twice weekly for 5 weeks; these three groups continued receiving one weekly i.p. CCl₄ injection to maintain liver injury throughout the therapeutic period. Another five mice receiving an equal volume of corn oil i.p. were defined as the normal control group. The body weight of each animal was recorded for each administration. After 24 h of the final treatment, all animals were sacrificed for sample collection. Animals were anesthetized with intraperitoneal injection of sodium pentobarbital at dosage of 150 mg/kg (Sigma, USA) before surgery and tissue harvesting. Peripheral blood was collected immediately after anesthesia, for the detection the serum ALT and AST levels using commercial assay kits. Liver and other tissues were isolated and fixed in 10% formaldehyde for histological staining.

Histological Staining and Immunofluorescence Analysis of VA-Modified BBR Liposomes

Harvested liver tissues were fixed, embedded in paraffin, and serially sectioned at 7 µm. Tissue slices were subjected to hematoxylin-eosin (HE), Sirius Red, and Masson’s trichrome staining, and immunofluorescence staining was used to assess the in vivo therapeutic efficacy of targeted BBR liposomes. HE staining was conducted to observe hepatic histopathological injury, while Sirius Red, Masson’s trichrome, and immunofluorescence staining were used to quantify collagen deposition and characterize the activation of hepatic stellate cells in vivo. All staining was performed according to standard published protocols.

Statistical Analysis

All data were obtained from at least three independent experiments and are presented as mean ± SD. Normality of data distribution was first assessed for each experimental group. Data passing the normality assumption were analyzed by Student’s t-test or one-way ANOVA with Tukey’s test, where applicable. All differences among and between groups were considered statistically significant at p < 0.05, (*), p < 0.01(**), p < 0.001(***), and p < 0.0001(****); values without significant differences are marked as ns.

Results

Preparation and Characterization of VA-Modified Liposomes

The synthesized DSPE-VA lipids were characterized by Fourier transform infrared spectroscopy (Figure 1A). As evidenced by the presence of characteristic absorption bands from both DSPE-PEG-NH2 and VA acid in the final product within the measured range of 1000 to 4000 cm⁻¹. In the free VA acid, the absorption band at 1678 cm⁻¹ was assigned to the stretching vibration of -COOH groups, whereas the band at 1590 cm⁻¹ corresponded to the C=C stretching vibration of the VA acid. In the spectrum of conjugate product, the characteristic band of -COOH in VA acid at 1678 cm⁻¹ disappeared, indicating the consumption of VA-COOH due to amidation. A newly appeared absorption band was observed at 1650 cm⁻¹, demonstrating the stretching vibration of C=O in the amide. Meanwhile, the C=C absorption originating from VA acid at ~1590 cm⁻¹ is still observable in the product. Besides, the absorption bands at 1720 cm⁻¹ and 1090 cm⁻¹ observed in the conjugate product were assigned to the glyceride C=O and P-O stretching vibration of phosphate groups in the DSPE phospholipid, respectively, which were consistent with the characteristic absorption peaks of free phospholipid. These together indicate the successful conjugation of DSPE-VA products for further experiments.

Figure 1.

Characterization of VA-modified liposomes. (A) FTIR images of the synthesized DSPE-PEG-VA conjugated lipid compared with VA acid and DSPE. (B) Liposome size and PDI of empty liposome with VA modification (n = 4). (C) TEM image of VA-modified liposomes, scale bar = 200 nm. (D) VA-liposome size and PDI stability over time (n = 3). Data represented as mean ± SD.

After formulation optimization, VA-modified liposomes were successfully prepared by incorporating DSPE-VA and other lipid components in a proper ratio, exhibiting a uniform particle size of 171.1 ± 4.9 nm with an average zeta potential of 1.8 ± 0.7 mV (Figure 1B and Supplementary Table 2). In addition, TEM imaging verified the morphology and size of VA liposomes (Figure 1C). An excess amount of VA-lipids might disrupt the integrity of the liposomal membrane, leading to the production of lipid aggregates. The long-term particle size stability of VA-modified liposomes was assessed over 90 days of storage at 4 °C. No aggregation was observed, with negligible variations in particle diameter and a favorable size distribution, suggesting acceptable stability for subsequent studies (Figure 1D).

In vitro Cellular Uptake of VA-Modified Liposomes by HSCs

Subsequently, we performed an in vitro cellular targeting assay using the DiI-labeled liposomes. Compared to unmodified liposomes, VA-modified liposomes achieved remarkably higher cellular uptake in quiescent HSCs, with a mean cellular fluorescence intensity of approximately 1200 (Figure 2A and B). Notably, TGFβ-activated HSCs took up even more VA-modified liposomes, with an average fluorescence intensity of approximate 1400 per cell. This phenomenon is likely due to the increased expression of RBP receptors in activated HSCs.³⁹ By contrast, unmodified liposomes showed negligible internalization, even in activated HSCs (Figure 2A and C). Collectively, these results demonstrate that VA-modified liposomes can be specifically recognized and internalized by HSCs regardless of their activation status.

Figure 2.

Cell targeting evaluation of liposome modified with VA. (A) DiI-labeled targeted or untargeted liposomes uptake by HSCs during quiescent or activated states. Scale bar = 50 µm. (B and C) Quantification of DiI-labeled targeted and untargeted liposome uptake by quiescent HSCs and activated HSCs (n = 5). Data represented as mean ± SD, ****p < 0.0001.

Preparation and Characterization of BBR Loaded VA-Modified Liposomes

Given the validated HSC-targeting capability of VA modification, berberine was encapsulated into liposomes via passive loading. No obvious difference in encapsulation efficiency was observed between VA-modified liposomes (11.8 ± 1.4%) and non-targeted liposomes (10.9 ± 2.9%, Supplementary Table 1). VA-modified BBR-loaded liposomes exhibited a uniform particle size of 165.9 ± 7.1 nm, with a PDI below 0.1 (Figure 3A and Supplementary Table 2). TEM revealed no obvious morphological alterations after drug loading (Figure 3B). The in vitro BBR release of VA-modified liposomes was investigated in PBS at 37 °C. Approximately 30% of berberine was released within the first 12 h, followed by a slower‑release phase. The cumulative BBR release reached around 40% at 72 h, and the total cumulative release was approximately 50% by 120 h (Figure 3C). Moreover, the liposomes maintained long-term stability at 4 °C with negligible BBR leakage over 45 days of storage (Figure 3D).

Figure 3.

Characterization of VA-modified liposomes with BBR encapsulated. (A) Size and size distribution of VA-modified BBR liposomes, n= 4. (B) TEM image of VA-modified BBR liposomes, scale bar = 100 nm. (C) BBR release profile from VA-modified liposomes in PBS at 37 °C, n = 3. (D) BBR leakage from VA-modified liposomes at 4 °C for long term storage, n = 3. (E and F) HSCs treated with VA-modified BBR liposomes or non-targeted BBR liposomes showing BBR uptake difference in cell pellets, n = 3. Data represented as mean ± SD, **p < 0.01.

To further verify HSC targeting ability after BBR loading, VA-modified and non-modified BBR liposomes were incubated with HSCs for 1 h, followed by drug extraction and quantification. The VA-modified liposomes exhibited significantly higher cellular uptake in HSCs, which was approximately 3.5-fold greater than that of their non-VA counterparts (Figure 3E and F). Subsequently, VA-modified liposomes were used for all further investigations.

In vitro Evaluation of VA-Modified Liposomal BBR Therapeutic Effect in Activated HSCs

Next, we evaluated the inhibitory effect of BBR-loaded liposomes on fibrotic HSCs. The qPCR results showed a significant decrease in collagen expression after VA-modified BBR liposome treatment compared to non-VA BBR liposomes or TGFβ alone (Figure 4A). Interestingly, α-SMA expression was positively correlated with liposomal BBR treatment, especially in the VA-modified liposomes, showing a significant increase in expression. Similarly, treatment with non-VA BBR liposomes induced a slight increase in α-SMA expression (Figure 4B). Western blot assay was performed to illustrate the expression at protein level. After liposomal BBR treatment, a significant decrease in collagen expression was detected in activated HSCs. However, consistent with the qPCR findings, no decrease in α-SMA was observed at the protein level (Figure 4C). These contradictory findings raise doubts regarding the anti-fibrotic effect of BBR on activated stellate cells.

Figure 4.

In vitro therapeutic evaluation of VA-modified BBR liposomes on activated HSCs. (A and B) Gene expression level of Col1α1 and α-SMA in HSCs treated differently, n = 6. (C) Protein expression level of Col1α1 and α-SMA in HSCs treated differently, n = 4. (D and E) Cell scratch assay to show migration changes in activated HSCs treated with VA-modified BBR liposomes or blank VA-liposome, n = 3. (F and G) Transwell assay to further indicate activated HSCs migratory capacity decreased after VA-liposomal BBR treatment, n = 4. (H and I) Reduced cellular growth rate on activated HSCs after treatment with VA-liposomal BBR, n = 3. (J) Cell cycle arrest at the S and G₂ phases after treated with VA-modified BBR liposomes, n = 3. Data represented as mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Note: All Ctrl and TGFβ groups received identical blank VA-liposomes treatment; all liposomal BBR groups received identical TGFβ treatment as TGFβ group.

Thus, we further investigated the effect of liposomal BBR on the migration of activated HSCs, as fibrotic cells are thought to have a quick migration capacity. The wound healing assay showed markedly slower wound closure in the VA-modified BBR liposome-treated group than in the untreated group, with differences emerging at 24 h and becoming statistically significant at 48 h (Figure 4D and E). In addition, the transwell assay demonstrated that VA-modified BBR liposome treatment obviously decreased the number of migrated cells compared with the untreated group, showing an approximately 2-fold difference (Figure 4F and G).

Furthermore, the proliferation rate of activated HSCs is an indicator of fibrosis level. VA-modified liposomal BBR demonstrated a significantly slower growth on activated HSCs without cell death or detachment after 48 h post-treatment compared with untreated HSCs (Figure 4H and I). Additionally, cell cycle distribution analysis also revealed significant accumulation of cells at S and G2/M phases after VA-modified liposomal BBR treatment, which resulted in failure of mitosis and inhibition of growth (Figure 4J and Supplementary Figure 1).

In vivo Therapeutic Evaluation of VA-Modified Liposomal BBR

Animal studies were performed to evaluate the in vivo efficacy. The treatment schedule is shown in Figure 5A. After 8 weeks of CCl₄ administration, one mouse was randomly selected and euthanized to verify the success of hepatic fibrosis modeling before BBR treatment was started. Serum AST and ALT levels were measured to evaluate liver function after treatment with liposomal BBR or PBS. Compared with control groups, CCl₄ induced dramatically increased AST and ALT levels in serum. A significant decrease in ALT was observed in mice treated with VA-modified liposomal BBR (p < 0.05) compared to that in untreated mice or those treated with non-VA-modified liposomes (Figure 5B). No significant difference was observed between mice treated with non-VA liposomal BBR and PBS, indicating a lower therapeutic effect of untargeted BBR liposomes. A similar trend was observed for the AST levels. Mice treated with VA-modified liposomal BBR showed significantly decreased AST activity compared with untreated mice (p < 0.05). In addition, non-VA-modified BBR liposomes were less effective than VA-modified liposomes (p = 0.0509), showing no significant effect on fibrotic mice (Figure 5C).

Figure 5.

In vivo therapeutic evaluation of VA-modified BBR liposomes on fibrotic mouse models. (A) In vivo administration protocol for animal study. (B and C) Targeted BBR therapy shows higher reduction of ALT and AST level in serum compared to untargeted or untreated groups. (D) HE, Sirius red and Masson staining comparison of mouse liver sections, green arrows indicate hydropic degeneration. Black arrows indicate inflammatory cell infiltration. (E and F) Quantitative analysis of fibrotic areas in liver sections from mice treated differently from Sirius red and Masson staining, respectively. Data represented as mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n = 5, scale bar = 200 µm.

Histopathological examination of mouse livers was performed. HE staining showed more severe hepatic steatosis and hydropic degeneration in untreated mice, whereas these lesions were rarely observed in the liposomal BBR-treated groups. Inflammatory cell infiltration was observed in all groups, except in the healthy control group (Figure 5D). Histochemical examination by Sirius red staining demonstrated a significant reduction in collagen expression when VA-modified BBR liposomes were used compared to untargeted liposomal BBR. Similar results were confirmed by Masson staining, which showed a significantly reduced fibrotic area after targeted liposomal BBR therapy (p<0.0001) (Figure 5D–F).

To further verify whether the administration of VA-modified liposomal BBR to the liver exerted further inhibitory effects on HSC activation of hepatic stellate cells, we performed immunofluorescence staining to examine the levels of collagen I and α-SMA in the liver. As shown in Figure 6A, CCl4 induced mice model showed a fibrotic liver with markedly increased expression of collagen and α-SMA. However, a significant decrease in the collagen area was observed in the liver after treatment with VA-modified liposomal BBR (p<0.01). Likewise, VA-liposomal BBR exhibited significantly stronger inhibition of collagen than non-VA liposomal BBR, although the untargeted group also showed a slight collagen inhibition in comparison with the untreated group (Figure 6A and B). Interestingly, α-SMA staining did not demonstrate a significant decrease between untreated mice and liposomal BBR-treated mice, regardless of VA-modification. Although the difference was not statistically significant, the VA-modified BBR liposome group showed a slight reduction in α-SMA levels compared with the CCl4 and unmodified groups (Figure 6C and D).

Figure 6.

Immunofluorescent staining of mouse liver sections. (A and B) VA-modified BBR liposomes treatment significantly inhibited Col1 expression compared to untargeted BBR treatment. (C and D) α-SMA expression is less significant but slightly reduced in VA-liposomal BBR treatment group compared to other groups. Data represented as mean ± SD, *p < 0.05, **p < 0.01, n = 5, scale bar = 100 µm.

Previous studies have suggested the potential toxicity of BBR for intravenous injections. In this study, we encapsulated BBR in liposomes at a relatively low dose for in vivo applications. HE staining of the heart, spleen, lung, and kidney did not show any damage when liposomal BBR was applied, regardless of the targeted delivery application (Supplementary Figure 2A). In addition, mouse body weight monitoring throughout the experimental period showed no significant changes, indicating that intravenous administration of BBR liposomes exerted no obvious toxic and side effects (Supplementary Figure 2B).

Discussion

Hepatic stellate cell activation is the core pathological driver of liver fibrosis, which subsequently undergoes vigorous proliferation and excessive collagen secretion. This leads to excessive cross-linking and deposition of the extracellular matrix, ultimately causing liver tissue sclerosis and progressive damage to liver function. Although most drugs and nanoparticles tend to accumulate in the liver, they are predominantly captured by Kupffer cells, sinusoidal endothelial cells, or hepatocytes, rather than HSCs. Owing to the low population and unique anatomical location of HSCs in liver tissue, conventional drug carriers fail to deliver enough drugs to HSCs, thereby limiting the overall therapeutic outcomes. Therefore, hepatic stellate cell-specific drug delivery is required to achieve targeted anti-fibrotic therapy, especially when using potentially toxic and less stable therapeutics. Herein, we successfully conjugated VA to lipids and constructed a VA-modified liposomal delivery system that took advantage of receptor-mediated recognition of the VA-RBP complex by receptors on activated HSCs. By encapsulating berberine into these VA-liposomes, we achieved enhanced targeted delivery and improved anti-fibrotic effects. Although the VA‑modified liposomes showed BBR‑loading content of only ~10%, which can be largely attributed to the intrinsic limitations of the passive loading strategy that generally yields relatively low encapsulation performance,^44,45 the liposomal formulation exhibited favorable stability with minimal drug leakage. The in vitro cumulative release of berberine slowly reached approximately 50% within 5 days, which is not unexpected. Complete drug release is commonly difficult to achieve for conventional liposomes in protein-free buffer without exogenous triggers. Comparable observations have been reported in our earlier work and other published reports, which is explained by phospholipid‑lipoprotein interactions in serum that promote sustained drug release.^46–48 Such stability also enabled VA-modified BBR liposomes to remain relatively intact in vivo for efficient targeted uptake.

Our results demonstrated that VA-modified BBR liposomes significantly increased BBR internalization by activated HSCs compared with non-targeted liposomes, resulting in improved therapeutic efficacy against liver fibrosis. It should be noted that this study did not directly perform side‑by‑side cellular uptake comparisons with Kupffer cells and hepatocytes to rigorously verify the cell‑selective profile of VA‑modified liposomes. Nevertheless, numerous reports have adopted nanocarriers with various VA decoration for HSC‑targeted delivery in liver disease research.^32–37 In particular, Sato et al developed VA‑adsorbed nanocarriers for nucleic‑acid delivery against liver fibrosis.³² Their results demonstrated prominent uptake of VA‑modified nanoparticles by activated HSCs, whereas negligible internalization was observed in Kupffer cells and hepatocytes.³² This published evidence also supports the notion that VA-modification enables nanocarriers to achieve targeted uptake by HSCs. Even so, the overall therapeutic effect may also originate from combined pharmacological actions of BBR liposomes taken up by multiple hepatic cell populations. Employing different hepatic cells to quantitatively characterize the cell‑type selectivity of our VA-modified liposomes is needed for the further study.

Given the potential toxicity risks of parenteral BBR compared with oral administration, therapeutic efficacy at reduced injection dosage is highly desirable.¹⁵ Variable dosages have been reported for BBR-based nanoparticles against liver disorders. Zha et al developed nanomicelles with BBR encapsulated to treat liver fibrosis, which showed promising effects at an intravenous BBR dosage of 20 mg/kg in fibrotic mice.³ In another study, Suman et al constructed a ROS-responsive nanocarrier with BBR loaded. They observed alleviation of liver injury in mice following intraperitoneal injection of BBR nanoparticles at 6mg/kg.⁴⁹ In comparison, VA-modified liposomal BBR yielded obvious antifibrotic effect at a relatively low injection dosage of 5 mg/kg, while non‑targeted liposomal BBR exhibited substantially weaker antifibrotic activity at the same dose. These findings suggest that VA-modified delivery facilitated effective BBR accumulation within lesion-associated cells and thus enhance inhibition of fibrotic progression.

Notably, our study revealed an unexpected increase in α-SMA expression in mouse HSCs after liposomal BBR treatment at both the gene and protein levels. This finding appears contradictory to most previous studies that reported BBR-mediated downregulation of this activation marker.^3,50–54 Additionally, we observed a similar effect using free BBR to treat these activated HSCs, which also showed increased α-SMA expression with an increase in BBR concentration at both the gene and protein levels (Supplementary Figure 3A and Supplementary Figure 3B). It should be noted that most previous studies adopted permanently immortalized rat or human HSC cell lines, whereas the HSCs used in our study had a finite proliferation lifespan, which may underlie the differing α-SMA expression profiles after BBR treatment in vitro.^3,51,54,55 It generally acknowledged that elevated α-SMA expression indicates fibrogenic potential and activated HSCs with enhanced migration, proliferation, and increased contractile stress, which aggravates fibrotic progression with increased of stiffness of liver tissues.^56,57 However, a substantially weakened migration capability was observed in HSCs treated with VA-modified BBR liposomes by the wound healing assay, which was confirmed by the Transwell study, indicating a reduction in HSC activation. In addition to inhibiting migration, targeted BBR liposomes slowed HSC growth. Specifically, they induced cell cycle arrest at the S and G2 phases without causing death in activated HSCs, demonstrating the effective suppression of HSCs proliferation. Besides, we conducted a contraction assay using free BBR‑treated HSCs to assess whether elevated α‑SMA expression would translate into enhanced cellular contractility via the assembly of α‑SMA assembles into stress fiber. Intriguingly, although BBR induced α‑SMA up‑regulation, HSC contractile capacity was reduced irrespective of activation status (Supplementary Figure 3C). These findings indicate that the elevated α‑SMA does not exert functional effects. It is presumably not assembled into cellular stress fibers due to disrupting of the cellular stress fibers by BBR,⁵⁸ and consequently fails to drive fibrotic progression.⁵⁹

Furthermore, VA-modified BBR liposomes significantly decreased collagen production, a key indicator of therapeutic efficacy, in both in vitro and in vivo studies, indicating inhibition of fibrotic progression despite α-SMA upregulation in HSCs. This cell phenotype and functional dissociation suggests that α-SMA expression alone may not invariably predict fibrosis progression. The slight reduction in α-SMA expression detected by immunofluorescence in the targeted treatment group may be attributed to the indirect effects of BBR, the released BBR exerted anti-inflammatory and immunomodulatory effects on immune cells, in turn modulating α-SMA expression in HSCs.^60,61 Taken together, VA-modified liposomes enhanced delivery efficiency and BBR uptake in activated HSCs, thus achieving improved anti-hepatic fibrosis effects by inhibiting HSC profibrotic behaviors, without causing systemic side effects associated with conventional BBR injections.

This study has several limitations. First, this study used in vitro HSC models and a short-term in vivo fibrosis model, which may not fully illustrate the complexity of human liver fibrosis. Second, we cannot fully exclude partial uptake of the VA-modified liposomes in this study by other liver resident cells, which produce multiple effects from different hepatic cell populations. This needs more detailed assessment in the future work. Third, the long-term safety and efficacy of VA-modified BBR liposomes requires further evaluation in animal studies. Finally, one of the most intriguing observations of the present study is that BBR treatment up‑regulates α‑SMA expression in activated HSCs, which appears to conflict with most previous reports. The exact mechanism underlying this contradictory phenomenon remains to be further elucidated. It is still unclear whether this represents an isolated case under our experimental conditions or arises from other unrecognized factors. Further investigations are required to dissect the deeper mechanisms governing BBR‑mediated antifibrotic activity.

Conclusion

In summary, this study successfully constructed VA-modified liposomal carriers to improve the delivery efficiency of berberine in HSCs for hepatic fibrosis treatment. Functionally, the optimized nanosystem exhibited prominent inhibitory effects on HSC migration, proliferation, and cell cycle arrest. Moreover, it effectively suppressed collagen synthesis both in vitro and in vivo, achieving enhanced anti-fibrotic progression and reducing the systemic toxicity of free BBR. Collectively, these findings demonstrate that VA-functionalized liposomal BBR could be an effective targeted therapeutic strategy for liver fibrosis with favorable in vivo tolerability, providing a promising nanoplatform for anti-fibrotic drug delivery. Future work will focus on elucidating the detailed mechanisms and evaluation of the long-term efficacy of this nanoplatform is essential to support further translational potential.

Funding Statement

This work was financially supported by the Youth Program of the National Natural Science Foundation of China (No.32201256), SIP High-Quality Innovation Platform for Chronic Diseases (YZCXPT2022203), and XJTLU Research Development Fund (RDF-23-02-043).

Data Sharing Statement

All data needed to support the conclusions are presented in the paper and/or Supplementary Materials.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

Prof. Dr. Lei Fu reports Support for the manuscript from SIP High-Quality Innovation Platform for Chronic Diseases, during the conduct of the study.

Prof. Dr. Tao Lu reports Support for the manuscript from National Natural Science Foundation of China; SIP High-Quality Innovation Platform for Chronic Diseases; XJTLU Research Development Fund, during the conduct of the study.

The authors declare that they have no competing financial interests or other potential conflicts of interest related to this study.