Materials

NAR (purity: 99%) was purchased from Feiyu Biological Co., Ltd (Jiangsu, China). PEG-PCL (PEG, polyethyleneglycol, MW 2000 Da; PCL, Polycaprolactone, MW 6000 Da), CPP-PEG-PCL (CPP, Cell-penetrating peptides RRRRRRRRRRRR), GA-PEG-PCL (GA, galactose, CH2OH(CHOH)4CHO), Cyanine-5 (Cy5) dye were provided by Ruixi biological Co., Ltd. (Shanxi, China). Phosphate-buffered saline (PBS, pH 7.35), fetal bovine serum (FBS), Dulbecco’s modified eagle medium (DMEM), Roswell Park Memorial Institute 1640 medium (RPMI 1640), trypsin-ethylenediaminetetraacetic acid (Trypsin–EDTA) and penicillin/streptomycin (PS) were bought from Gibco (Carlsbad, CA, USA). Estradiol (E2), 1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI), Methyl-β-cyclodextrin (MβCD), chlorpromazine (CPZ), hypertonic sucrose, and 5-(N-ethyl-N-isopropyl)-amiloride (EIPA) were purchased from Sigma-Aldrich Corp. (St. Louis, MO, USA), and estradiol-3-O-sulfate (E2-3-O-sulfate) was ordered from Meilun Bio-Technology Co., Ltd (Dalian, China). Tamoxifen was ordered from Macklin Co., Ltd (Shanghai, China). 4′,6-Diamidino-2-phenylindole (DAPI), 4% paraformaldehyde (PFA), and Cell counting kit-8 (CCK8) were purchased from Beyotime (Shanghai, China). Lysosome (Lyso)-Tracker™ Deep Red, endoplasmic reticulum (ER)-Tracker™ Red, and mitochondria (Mito)-Tracker™ Red were purchased from Invitrogen (Carlsbad, CA, USA). Methanol and acetonitrile were derived from Merck (Darmstadt, Germany). The deionized water (18 MX) was from a MilliQ water purification system (Millipore, Billerica, MA). HLB solid phase extraction column was sourced from Waters (Milford, MA, USA).

Preparation of NC, NG, NCG, and Cy5-NCG nanoparticles

To prepare the NCG nanoparticles, we first dissolved 18 mg of PEG-PCL, 1 mg of GA-PEG-PCL, and 1 mg of CPP-PEG-PCL in 1 mL of acetone containing 15 mg of NAR to generate the organic phase. Next, 10 mg of PVP K29/32 was dissolved in 50 mL of dd H2O to prepare the aqueous phase, which was then stirred using a magnetic stirrer at 1000 rpm. Subsequently, 0.2 mL of the organic phase was injected into the aqueous phase while maintaining the stirring. The resulting mixture was further stirred for 2–3 min and then subjected to centrifugation to eliminate any unformed precipitate. The resulting nanoparticles were washed three times with deionized water, followed by centrifugation after each wash to remove residual impurities. The preparation procedures for the NC, NG and Cy5-NCG nanoparticles were analogous to that for NCG nanoparticles, except for the use of different PEG-PCL derivatives in the organic phase. For the NC nanoparticles, we dissolved 19 mg of PEG-PCL and 1 mg of CPP-PEG-PCL in 1 mL of acetone containing 15 mg of NAR. For the NG nanoparticles, we dissolved 19 mg of PEG-PCL and 1 mg of GA-PEG-PCL in 1 mL of acetone containing 15 mg of NAR. In case of Cy5-NCG nanoparticles, we dissolved 18 mg of PEG-PCL, 1 mg of GA-PEG-PCL, 1 mg of CPP-PEG-PCL, and 5 mg of Cy5 in 1 mL of acetone to create the organic phase. All other conditions were held constant.

Nanoparticle characterizations

The morphology and size of NCG were evaluated using a transmission electron microscope (TEM, H-7650, Hitachi, Tokyo, Japan). High-resolution TEM images were captured using a FEI F200S microscope (Waltham, USA) at 200 kV. The zeta potential and particle size distribution of NCG nanoparticles were measured by dynamic light scattering (DLS, Nano-ZS Malvern Instruments, Worcestershire, UK) in dd-H2O. The absorption spectra of NCG were obtained using a UV–vis–NIR spectrophotometer (TU-1810, PERSEE, Beijing, China). Additionally, a Fourier transform infrared (FTIR) spectrophotometer (Nicolet iS50, Thermo Fisher Scientific Ltd.) was utilized to record the chemical functions fabricated, which absorbed in the spectral range of 2400 to 4000 cm−1. The synthesized NCG was characterized using proton nuclear magnetic resonance (1H-NMR) spectroscopy on a Bruker Avance NEO 600 instrument. The physical state of NCG was characterized using differential scanning calorimetry (DSC) analysis on a TA DSC2500 instrument. Each sample (5 mg of free NAR and NCG, respectively) was sealed in a standard aluminum pan. The sample was purged with dry nitrogen gas at a flow rate of 20 mL/min during the DSC analysis. The heating rate was set at 10 °C/min, and the heat flow was recorded within a temperature range of 30 to 400 °C. To assess the stability of NC, NG, and NCG nanoparticles under varying pH conditions, the nanoparticles were exposed to simulated gastric juice with a pH of 1.5 for 2 h, simulated intestinal fluid with a pH of 6.8 for 2 h, and pH 7.35 phosphate-buffered saline (PBS) for 30 days to approximate blood pH. Sedimentation at the bottom of the bottle was observed and documented using a camera, while the particle size of NC, NG, and NCG was measured simultaneously using dynamic light scattering.

The gastric juice was prepared by diluting concentrated hydrochloric acid with water to a concentration of 1 mol/L and adjusting the pH to 1.5. Then, 1 g of pepsin was added to 100 mL of the solution. The intestinal juice was prepared by dissolving 6.8 g of KH2PO4 in 500 mL of water and adjusting the pH to 6.8 with 0.4% (w/w) NaOH. Then, 1 g of pancreatin was added to 100 mL of the solution.

To evaluate the release kinetics of NAR from NCG, we employed a dialysis-based in vitro method. NCG was first enclosed within a preconditioned dialysis bag (Mw = 8–14 kDa) and immersed in a beaker containing simulated gastric acid for 2 h, followed by simulated intestinal fluid for another 2 h. The bag was then transferred to a beaker filled with 40 mL of pH 7.4 phosphate-buffered saline (PBS) containing 0.5% (w/v) Tween 80, and maintained for 44 h on a rotary shaker at 37 °C and 70 rpm. At predetermined intervals, aliquots of the release medium were withdrawn and replaced with an equal volume of fresh medium to ensure sink conditions. The dissolved samples were then collected in 2 mL beakers for further analysis. The concentration of NAR was determined by measuring the absorbance at 260 nm and correlating it with an established UV–Vis standard curve.

UV–Vis–NIR spectroscopy was used to determine the drug loading (DL) and encapsulation efficiency (EE) of naringenin (NAR) (λ max = 360 nm). The DL and EE values were then calculated using the following equations: DL (%) = Wa/Wb × 100; EE (%) = Wa/(Wa + Wc) × 100. Here, Wa, Wb, and Wc correspond to the mass of encapsulated NAR, the total weight of NCG, and the mass of free NAR, respectively.

Cell culture

WRL-68, MCF-7, Caco-2, HUVEC, E0771, and HBL-100 were obtained from Nanjing Keygen Biotech. WRL-68, MCF-7, Caco-2, and HUVEC cells were cultured in DMEM cell culture medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin, and E0771 and HBL-100 cells were cultured in the same formulation of RPMI-1640 cell culture media. All cells were maintained at 37 °C in a humidified atmosphere containing 5% CO2 with regular sub-culturing.

In vitro cytotoxicity analysis

WRL-68, MCF-7, E0771, Caco-2, HUVEC, and HBL-100 cells were seeded into 96-well plates at a density of approximately 5 × 103 cells per well. NAR, NC, NG, and NCG (at equal concentrations of NAR) were added to the wells, and a control group was also included. After drug addition, the plates were placed in a CO2 incubator and incubated at 37 °C with 5% CO2 for 24 h. CCK-8 reagent was diluted according to the manufacturer’s instructions (Beyotime Biotechnology, C0038), and the old medium was aspirated from the wells and replaced with fresh medium containing the diluted CCK-8 reagent. After a further 2 h incubation at 37 °C, the absorbance was measured at a wavelength of 450 nm using a microplate reader (Biotek, EON).

Cellular uptake detection

WRL-68, MCF-7, E0771, Caco-2, HUVEC, and HBL-100 cells were seeded separately into 12-well plates containing complete culture medium at a density of approximately 1 × 105 cells per well. Subsequently, the plates were incubated for the specified nanoparticles at 37 °C and 5% CO2 in a CO2 incubator. Upon completion of the incubation period, cells were rinsed thrice with PBS, fixed with 4% paraformaldehyde, and subjected to DAPI staining for visualization of the cellular nuclei. The acquired cellular images will undergo further analysis employing a confocal laser scanning microscope (Zeiss LMS-780, Thornwood, New York). For the evaluation of liver targeting efficacy, an excess of free GA was introduced half an hour prior to the experiment, thereby inducing ASGPR saturation.

Intestinal barrier permeability assay

To evaluate the ability of NCG to cross the intestinal barrier, we established an in vitro intestinal barrier model based on previous literature [60]. The model was constructed as follows: Caco-2 cells were seeded onto the PET membrane of the Transwell upper chamber at a density of 1 × 105 cells/cm2. When the transepithelial electrical resistance (TEER) reached over 150 Ω/cm2, WRL-68 cells were seeded in the lower chamber at 1 × 105 cells/cm2 density and incubated at 37 °C in a 5% CO2 atmosphere for 24 h. Subsequently, NAR and NCG drugs (with an equivalent concentration of 100 nmol/L for NAR) were added to the upper chamber, and the Transwell was returned to the incubator for 10, 30, 60, and 120 min. At specified time points, the Caco-2 cells in the upper chamber and the WRL-68 cells in the lower chamber were fixed and stained with DAPI for nuclear visualization. The distribution of NAR and NCG on both sides of the PET membrane was observed using the Z-axis scanning function of a confocal laser scanning microscope (Zeiss LMS-780, Thornwood, New York), enabling further analysis of drug penetration within the intestinal barrier model.

FRET assay

Förster Resonance Energy Transfer (FRET) assay was applied to evaluate the integrity of NCG in Caco-2 and WRL-68 cells. We encapsulated C6 and DiI fluorophores within NCG and exposed them to Caco-2 and WRL-68 cells, respectively. The cells were incubated at 37 °C in a 5% CO2 atmosphere for 1 h and 2 h. Following incubation, the cells were washed thrice with cold PBS, fixed with 4% PFA, stained with DAPI, and analyzed using a confocal laser scanning microscope for evaluation.

Cellular endocytosis and distribution assay

Caco-2 and WRL-68 cells were cultured on cover slips in a 12-well plate for 24 h. In order to investigate the endocytic pathways of NCG, cells were treated with endocytosis-mediated clathrin-dependent inhibitor CPZ, caveolin-mediated endocytosis inhibitor MβCD, and macropinocytosis inhibitor EIPA for a duration of 30 min. Subsequently, NCG was introduced into DMEM culture medium containing 10% FBS and co-incubated with cells for 30 min. To explore the subcellular distribution of NCG in Caco-2 and WRL-68 cells, following the 30 min co-incubation of NCG with cells, mitochondria tracker, lysosome tracker, and endoplasmic reticulum tracker were individually introduced and co-incubated with cells according to the manufacturer’s specifications. Following incubation, cells were washed with ice-cold PBS and fixed with 4% paraformaldehyde. Cell nuclei were stained with DAPI, and laser confocal microscopy was utilized to assess the uptake of NCG in various treatment groups, as well as its co-localization with distinct cellular sub-organelles.

Luciferase reporter assays in vitro

To investigate the effects of different drugs on the EST expression, the luciferase reporter gene assay was conducted. The EST promoter fragment was amplified and subsequently cloned into the pGL3 basic vector (LvPG04, GeneCopoeia, China). WRL-68 cells were transfected with the EST luciferase reporter plasmid (5 µg/well) in 6-well plates. Following 48 h post-transfection, the cells were seeded in a 96-well plate, and the drugs NAR, NC, NG, and NCG were administered at respective NAR equivalent concentrations (0, 1, 5, 10, 50, 100, 200 nmol/L). After a 24 h drug treatment, the supernatant was collected, and luciferase activity was assessed. Promoter activity was determined using the Secrete-Pair Dual Luminescence Detection Kit (LF001, GeneCopoeia, China), following the manufacturer's instructions. Results were normalized to the activity of renilla luciferase. All transfection experiments were performed in triplicate and independently repeated three times.

Zebrafish experiments

In order to observe the distribution of different NAR formulations in zebrafish, we exposed zebrafish larvae (7 days post-fertilization; dpf) to NAR or NCG at concentrations of 50, 100, or 200 nmol/L. Following exposure durations of 10, 30, or 60 min, the zebrafish were anesthetized and subsequently collected for microscopic imaging utilizing a Leica DMi8 microscope (Leica Microsystems, Wetzlar, Germany). To assess the inhibitory potential of NCG on breast cancer cell proliferation, a zebrafish breast cancer xenotransplantation model was developed. E0771 cells were specifically labeled with 5 μmol/L DiI and microinjected into zebrafish embryos at 48 h post-fertilization. The zebrafish harboring E0771 cells were subsequently distributed into 96-well plates (10 fish per well) and cultured in a medium supplemented with 40 μmol/L estradiol, excluding the control group. Interventions with NAR, NC, NG, and NCG were then administered at a 50 nmol/L concentration for each compound. Following a 48 h incubation period, the growth and metastasis of E0771 cells were observed using an inverted fluorescence microscope (Nikon Eclipse C1, Tokyo, Japan).

Mice experiments

C57BL/6 female mice (6–8 weeks old) were procured from the Guangdong Provincial Experimental Animal Center (Guangzhou, China) and maintained under a 12-h light–dark cycle, a consistent room temperature of 20–22 °C, and a relative humidity of 30–70% (SCXK-2020-100). The C57BL/6 female mice were allowed to acclimate to these conditions for a minimum of 7 days before experimentation commenced. Following the stress treatment period, an in situ breast cancer model was established by injecting 1 × 106 ER-positive E0771 breast cancer cells suspended in a 100 μL PBS-Matrigel (1:1) mixture into the fourth mammary fat pad of female C57BL/6 mice. The mice were then randomly assigned to eight groups (n = 6 per group), comprising a control group, Saline group, NAR L (10 mg/kg), NAR H (20 mg/kg), TAM, NG (10 mg/kg), NC (10 mg/kg), and NCG (10 mg/kg) groups. The corresponding treatments were administered orally everyday. Tumor dimensions were measured using a vernier caliper every 3 days, and tumor volume was estimated employing the formula (length × width2/2). After a 14-day treatment period, all mice were sacrificed, and tumor volume and weight were assessed. All animal studies and experimental protocols were approved by the Animal Ethics Committee of the Guangdong Provincial Hospital of Chinese Medicine (2021074).

In vivo NIR imaging

To assess the intestinal barrier permeability and liver-targeting properties of NCG, we prepared nanocarriers and labeled them with Cy5. Mice were orally administered with Cy5, Cy5-NC, Cy5-NG, and Cy5-NCG (Cy5 = 1 mg/kg) and subsequently subjected to whole-body imaging using an in vivo imaging system at various time points (n = 3). The distribution of the fluorescence signal was observed under excitation at 650 nm and emission at 700 nm wavelength. Additionally, a subset of mice was sacrificed to examine the fluorescence signal in major organs, including the stomach, duodenum, jejunum, ileum, liver, heart, spleen, lung, kidney, and tumor. The fluorescence signal was then quantified using the IndiGo software.

In vivo pharmacokinetics analysis

To perform pharmacokinetic studies using Sprague–Dawley (SD) rats, the animals were randomly assigned to four groups including NAR, NC, NG, and NCG. Each rat received the corresponding treatment orally at an equivalent dose of 6.9 mg/kg NAR. Blood samples (300 μL) were collected from the orbital socket of the rats (n = 6/group) at 0.083, 0.5, 1, 2, 4, 6, 12, 24, and 48 h post-administration. The NAR levels in these samples were quantified using liquid chromatography-tandem mass spectrometry (LC–MS/MS). Furthermore, the Drug and Statistics (DAS) 2.0 platform was employed to determine pharmacokinetic parameters, including elimination half-life (T1/2), peak brain/blood quotients of concentration and time (QCT) levels (Cmax), time to reach peak level (Tmax), and area under the curve.

Western blot

Fresh frozen liver samples were sectioned into small pieces and homogenized in a standard lysis buffer containing 1 mmol/L protease inhibitor. The homogenized suspension was subsequently centrifuged to isolate the supernatant, which was preserved at − 80 °C for future use. The protein concentration in the supernatant was ascertained using the bicinchoninic acid (BCA) protein assay and adjusted accordingly. For protein electrophoresis, the adjusted supernatant was combined with 1×  SDS-PAGE buffer and subjected to electrophoretic separation. The resolved proteins were then transferred to a polyvinylidene fluoride (PVDF) membrane, which was blocked with a blocking solution (5% non-fat milk) to prevent nonspecific binding. The membrane was incubated with primary antibodies EST (ProteinTech, 12522-1-AP), overnight at 4 °C, followed by washing and incubation with the corresponding secondary antibodies. Protein bands were visualized using the ChemiDoc XRS+ Imaging System (Bio-Rad) and quantified employing ImageJ or other image processing software.

Immunofluorescence assay

Immunofluorescence was employed to examine the expression of the EST in WRL-68 cells. WRL-68 cells were seeded on a 12-well plate slide. Following drug treatment, the cells were washed thrice with PBS for 5 min each. The cells were then fixed with 4% paraformaldehyde and blocked with 10% bovine serum. To examine the expression of the EST in the liver, frozen liver sections (4 µm thick) were prepared and submerged in ultrapure water for 2 min. The slides were subsequently fixed in cold acetone at 20 °C for 10 min and then incubated with a blocking buffer for 30 min. After washing the slides three times with PBS at room temperature for 5 min each, they were incubated with primary antibody against EST (ProteinTech, 12522-1-AP) overnight at 4 °C. The slides were rewashed three times with PBS for 5 min each, followed by incubation with Alexa Fluor 488 or Alexa Fluor 555-conjugated anti-mouse IgG at a 1:200 dilution. After stained with 4′,6-diamidino-2-phenylindole (DAPI, Sigma) for nuclear visualization, the slides were imaged using a confocal laser scanning microscope (Zeiss LMS-780, Thornwood, New York).

Detection of E2 and E2-3-O-sulfate level

To investigate the effects of NAR and NCG on estrogen metabolism, WRL-68 cells were seeded in 12-well plates and treated with NAR and NCG for 24 h, respectively. Subsequently, the cells were incubated with Hanks’ Balanced Salt Solution containing 40 μmol/L E2 at 37 °C for 4 h. The extracellular medium was collected and analyzed using UPLC-QTOF/MS. As for blood serum, tumors, and liver samples, tissues were homogenized in 600 μL of ice-cold extraction solvent (90% methanol/water) and incubated on ice for 30 min. After centrifugation at 12,000×g for 15 min at 4 °C, the supernatant was gathered, transferred to a new Eppendorf tube, and dried in a vacuum concentrator. Subsequently, HLB solid-phase extraction (Waters, Milford, MA) was employed to desalt the blood serum. Equal aliquots of 100 μL serum were loaded onto the solid-phase extraction cartridge and eluted with 1 mL of acetonitrile. The acetonitrile eluate was vacuum-dried and reconstituted with 80 μL of acetonitrile/water (1:1). An ACQUITY UPLC-QTOF/MS System (Waters, Milford, MA) equipped with an ACQUITY BEH C18 Column (2.1 × 50 mm, 2.6 µm; Waters, Milford, MA) was employed for the analysis. The mobile phases consisted of 0.1% formic acid (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B). The flow rate was set at 0.25 mL/min. The gradient elution program followed a sequence of 25% B from 0 to 1 min, an increase from 25 to 85% B between 1 and 2.8 min, and a decrease from 85 to 25% B between 2.8 and 4 min.

In vivo biocompatibility assays

Upon completion of the treatment period, blood samples were collected from each group of mice for the analysis of blood cells, biochemical markers, and coagulation factors. The heart, liver, spleen, lung, kidney, and uterus were fixed in 4% paraformaldehyde (PFA) and sectioned into 5 µm-thick slices. For pathological examination, tissue sections were stained with hematoxylin and eosin (H&E) and assessed for potential organ damage. The endometrial thickness was measured to evaluate the impact of estrogen level fluctuations on the uterus.

Micro computed tomography

Specimens were scanned using Bruker Micro-CT Skyscan 1276 system (Kontich, Belgium). Scan settings are as follows: voxel size 6.534165 μm, medium resolution, 70 kV, 200 μA, 0.25 mm Al filter and integration time 350 ms. Density measurements were calibrated to the manufacturer’s calcium hydroxyapatite (CaHA) phantom. Analysis was performed using the manufacturer’s evaluation software. Reconstruction was accomplished by NRecon (version 1.7.4.2). 3D images were obtained from contoured 2D images by methods based on distance transformation of the grayscale original images (CTvox; version 3.3.0). 3D and 2D analysis were performed using software CT Analyser (version 1.20.3.0).

Statistical analysis

Data were presented as mean ± standard deviation (SD). Student’s t-test analysis or one-way ANOVA was applied. Post hoc testing of dose-dependent data was conducted using Dunnett’s test, while Bonferroni's post hoc test was used for other data. Repeated measures analysis of variance was used for repeated measures data. Nonparametric tests were utilized for data that were not normally distributed. A p-value of < 0.05 was considered statistically significant. Statistical analysis was performed using SPSS software (IBM SPSS Statistics, version 29.0).