4.1 Materials, cell lines, and animals
Soybean lecithin, cholesterol, and DSPE-mPEG(2000) were purchased from AVT Pharmaceutical Tech Co., Ltd. (China). (R)-(+)-bornylamine was purchased from Sigma-Aldrich Co. (USA). DSPE-PEG(2000)-NHS was purchased from Shanghai Ponsure Biotech, Inc. (China). Pierce BCA Protein Assay Kit was purchased from Thermo Fisher Scientific (USA). Luteolin (98% purity) was purchased from Chengdu Herbpurify Co., Ltd. (China). RIPA lysis buffer, protease inhibitor cocktail, phosphatase inhibitor cocktail, and phenylmethanesulfonyl fluoride (PMSF) were purchased from Beyotime (China). 4% Paraformaldehyde was purchased from Shanghai Biotend Biotech Co., Ltd. (China). Difco™ Nutrient Broth (Cat. No. 234000) was purchased from BD Bioscience (USA). DiD dye was purchased from Dalian MeiLun Biotechnology Co., Ltd. (China). DAPI was purchased from MedChemExpress (USA). DSS was purchased from Yeasen Biotechnology (China). The fecal occult blood detection kit and the GSH assay kit were purchased from Nanjing Jiancheng (China). The MPO activity assay kit and the SOD activity assay kit were purchased from Beijing Solarbio Science & Technology Co., Ltd (China). The mouse IFN-γ ELISA kit, the TNF-α ELISA kit, and the IL-6 ELISA kit were purchased from Shanghai Jianglai Industrial Limited By Share Ltd. (China). The anti-Epcam mouse antibody was purchased from Hangzhou HuaAn Biotechnology Co., Ltd. (China). The anti-mouse IgG conjugated with AF555 was purchased from Cell Signaling Technology (USA). The anti-ZO-1 rat monoclonal antibody (sc-33725) was purchased from Santa Cruz (USA). The Horseradish peroxidase (HRP)-labeled goat anti-rat IgG (GB23302) was purchased from Wuhan Servicebio Technology Co., Ltd. (China). The anti-occludin rabbit monoclonal antibody (ab216327) was purchased from Abcam (UK). The HRP-labeled goat anti-rabbit IgG (RCA054), the TYR-570Plus red dye, and the TYR-520Plus green dye were purchased from Refinebio (China).
Caco-2 cells (Human epithelial colorectal adenocarcinoma cells) were purchased from Cell Bank of Chinese Academy of Sciences (China). They were cultured with Dulbecco’s modified Eagle’s medium (DMEM; Meilunbio, China) containing 10% (v/v) fetal bovine serum (FBS; ExCell Bio., China) and 1% penicillin/streptomycin (Biosharp, Beijing Lanjieke Technology Co., Ltd., China) at 37°C in 5% CO2.
Female C57BL/6J mice (18 ~ 20 g) were purchased from Shanghai SLAC Laboratory Animal Co. Ltd. (China). All animal work complied with the protocols approved by Institutional Animal Care and Use Committee of Shanghai University of Traditional Chinese Medicine (IACUC, No: PZSHUTCM2303240001). All animal experiments were conducted following the “3R” principle (reduction, replacement, refinement).
4.2 Isolation and characterization of bacterial OMVs
The isolation of bacterial OMVs from S.mal. was conducted using the ultracentrifugation method reported previously with minor modifications52,53. S.mal. was inoculated into Nutrient Broth medium and cultured in a rotary shaker (220 rpm) at 32℃ for 12 h. Then, the bacterial suspension was diluted 1:100 with fresh medium and cultured for another 12 h. Next, the bacterial medium was centrifuged at 4℃, 4000 g for 10 min to remove S.mal., followed by filtration with a 0.22-µm vacuum filter (Beyotime, China). Thereafter, the filtrate was concentrated using an ultrafiltration tube (MWCO 100 kDa, Millipore). The concentrated solution was ultracentrifuged at 20,0000 g, 4℃ for 3 h with SW 70 Ti rotor (Beckman coulter, USA) to obtain OMV pellets, which were suspended finally in PBS and stored at -80℃ for further use.
The size distribution and zeta potential of OMVs were detected by a NICOMPTM 380 ZLS Zeta Potential/Particle Sizer (PSS, USA) using DLS. The morphology of OMVs was characterized by 120 kV TEM (FEI, USA). Briefly, the OMV sample was dropped onto 300-mesh carbon-coated copper grids. Following incubation for 1 min, the sample was removed from the grid, which was then stained with uranyl acetate for another 1 min and dried before TEM analysis. In addition, the OMV protein content was measured by a BCA protein assay kit, following the lysis of OMV samples in RIPA buffer supplemented with protease and phosphatase inhibitors.
4.3 Synthesis and characterization of DSPE-PEG-BO
DSPE-PEG-BO was synthesized through NHS easter reaction by conjugating the amino group of (R)-(+)-bornylamine (an amino derivative of borneol) and DSPE-PEG-NHS easter. Briefly, DSPE-PEG-NHS (100 mg, 1 equiv.) was dissolved in chloroform and stirred at 300 rpm, followed by rapidly adding (R)-(+)-bornylamine (5.6 mg, 1.1 equiv.) and N, N-Diisopropylethylamine (DIPEA). The mixture was stirred under a nitrogen atmosphere for 24 h at room temperature. Then, the product was purified by immersing the mixture in the ether at -20℃ for 3 ~ 4 h and collecting the precipitate by centrifuging at 5000 g, 4℃ for 30 min. The purification procedure was conducted three times, and the final powder product of DSPE-PEG-BO was obtained through further lyophilization. The structure of DSPE-PEG-BO was characterized by the 600 MHz 1H-NMR spectrometry (Bruker, Germany).
4.4 Preparation of luteolin-loaded liposomes
In this study, four kinds of luteolin-loaded liposome formulations were constructed, including liposomes without any modification (Lipo@LU), single borneol-modified liposomes (BO-lipo@LU), single OMV-mimetic hybrid liposomes (OMV-lipo@LU), and borneol-modified OMV-mimetic hybrid liposomes (BO/OMV-lipo@LU).
Lipo@LU were initially fabricated using the thin film hydration method43,54. Firstly, a mixture of soybean lecithin, cholesterol, DSPE-mPEG, and luteolin with different weights was placed in a round-bottom flask and ultrasonically dissolved in the organic solvent. Subsequently, the solvent was evaporated entirely under reduced pressure using a rotary evaporator at 45℃ and 125 rpm for 20 min until a thin-lipid film was formed. Then, PBS was added to hydrate the film for 10 min to obtain multilamellar liposomes, followed by probe-sonicating (Ningbo Scientz Biotechnology Co., Ltd., China) at 25% power (2 s pulse on/4 s pulse off) for 5 min to acquire unilamellar liposomes. Next, the suspension was filtered through a 0.45-µm filter to remove large particles, and Lipo@LU was ultimately obtained by extrusion through 400 and 200 nm polycarbonate membranes 9 times using a mini extruder (Avestin, Canada), respectively. Any unentrapped luteolin was removed by ultrafiltration (MWCO 10 kDa, Millipore). To optimize the formulation of Lipo@LU, the single-factor test was performed to investigate effects of the organic solvent type (a 9:1 chloroform-methanol mixture or ethanol), the ratio of soybean lecithin to cholesterol (3:1, 4:1 or 6:1), hydration temperature (25℃ or 37℃) and luteolin dosage (1.2 mg, 2 mg or 3 mg) on the particle size and encapsulation efficiency of Lipo@LU.
Based on the formulation mentioned above of Lipo@LU, BO/OMV-lipo@LU were prepared by replacement of DSPE-mPEG with DSPE-PEG-BO and membrane fusion with OMVs. Briefly, DSPE-PEG-BO was added to lipid materials to form liposomes with targeting moieties of borneol (BO-lipo@LU). Additionally, OMVs were mixed with hydration suspension containing liposomes and then were extruded to induce membrane fusion and generate OMV-mimetic hybrid liposomes (OMV-lipo@LU). BO/OMV-lipo@LU was synthesized by combining these two processes.
4.5 Characterization of luteolin-loaded liposomes
4.5.1 Particle size, zeta potential, morphology, and storage stability
The size distribution and zeta potential of luteolin-loaded liposomes were detected by a NICOMPTM 380 ZLS Zeta Potential/Particle Sizer. The morphology of these liposomes was observed by 120 kV TEM. To evaluate the storage stability at 4 ℃ within one week, their particle size and PDI were detected and recorded daily.
4.5.2 Encapsulation efficiency and drug loading capability
The EE and DLC of luteolin-loaded liposomes were calculated using Eqs. (1) and (2). The content of luteolin entrapped in the liposomes was determined with high performance liquid chromatography (HPLC, Shimadzu, Japan). The HPLC analysis for determining luteolin was conducted with an Agilent ZORBAX SB-C18 column (250 mm × 4.6 mm), mobile phase of the mixture of acetonitrile and 0.2% phosphoric acid (33:67, v/v), flow rate of 1 mL/min, column temperature of 25℃, injection volume of 20 µL, and ultraviolet (UV) detection wavelength of 350 nm. Prior to HPLC analysis, liposomes were mixed with methanol by vigorous vortexing (1 min) and sonication in a bath sonicator (10 min, Shanghai Titan, China), followed by centrifuging at 10,012 g for 10 min to collect the supernate. Additionally, the total dry weight of liposomes was measured to calculate DLC after liposomes were freeze-dried for 24 h.
$$\:\text{E}\text{E}(\text{%})=\text{M}\text{o}\text{u}\text{n}\text{t}\:\text{o}\text{f}\:\text{e}\text{n}\text{c}\text{a}\text{p}\text{s}\text{u}\text{l}\text{a}\text{t}\text{e}\text{d}\:\text{l}\text{u}\text{t}\text{e}\text{o}\text{l}\text{i}\text{n}/\text{M}\text{o}\text{u}\text{n}\text{t}\:\text{o}\text{f}\:\text{a}\text{d}\text{d}\text{e}\text{d}\:\text{l}\text{u}\text{t}\text{e}\text{o}\text{l}\text{i}\text{n}\:\times\:100\%$$
1
$$\:\text{D}\text{L}\text{C}(\text{%})=\text{M}\text{o}\text{u}\text{n}\text{t}\:\text{o}\text{f}\:\text{e}\text{n}\text{c}\text{a}\text{p}\text{s}\text{u}\text{l}\text{a}\text{t}\text{e}\text{d}\:\text{l}\text{u}\text{t}\text{e}\text{o}\text{l}\text{i}\text{n}/\text{T}\text{o}\text{t}\text{a}\text{l}\:\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}\:\text{o}\text{f}\:\text{l}\text{i}\text{p}\text{o}\text{s}\text{o}\text{m}\text{e}\text{s}\times\:100\%$$
2
4.5.3 SDS-PAGE protein analysis
SDS-PAGE analysis was employed to characterize proteins. The OMVs, BO/OMV-lipo@LU, and Lipo@LU samples were lysed in RIPA buffer supplemented with protease and phosphatase inhibitors, mixed with loading buffer, and then incubated at 95℃ for 10 min. Each sample (10 µg) was loaded in the wells of a Bio-Rad electrophoresis system (Bio-Rad Laboratories, USA). To visualize the protein profile, the gel was stained with Coomassie Brilliant Blue G-250 for 15 min and destained for 1 h, and imaged using a Bio-Rad gel imaging system.
4.5.4 Stability and in vitro drug release in simulated gastrointestinal fluids
Simulated gastrointestinal fluids included SGF (pH 1.2, consisting of HCl and pepsin), SIF (pH 6.8, consisting of KH2PO4, NaOH, and trypsin), and SCF (pH 7.8, comprising KH2PO4 and K2HPO4), which were used to simulate the physiological conditions in the stomach, small intestine and colon regions, respectively. To investigate the stability of luteolin-loaded liposomes in the gastrointestinal fluids, their particle size and PDI were determined after the liposome samples were incubated in SGF, SIF, and SCF, respectively, at 37℃ for 2 h.
The in vitro release properties of luteolin-loaded liposomes were investigated via dialysis. To mimic the environment of liposomes during gastrointestinal passage in vivo, the selection of release medium applied a three-stage method according to previous studies55,56. Briefly, Free-LU, Lipo@LU, BO-lipo@LU, OMV-lipo@LU, and BO/OMV-lipo@LU were put in dialysis bags (MWCO 8 ~ 14 kDa, Shanghai Yuanye Bio-Technology, China), respectively. Then, the dialysis bags were fastened and immersed in the release medium in an order of SGF, SIF, and SCF, which contained 0.5% Tween-80. The experiment was performed by a shaker at 37 ℃ and 150 rpm. The samples of release medium were collected at predetermined time points, and the same volume of fresh medium was immediately replenished. Afterward, the luteolin content in the samples diluted with methanol was determined by HPLC analysis to calculate the cumulative release of luteolin.
4.6 Preparation and characterization of DiD-loaded liposomes
Aiming to investigate the cell internation and in vivo bio-distribution of liposomes, DiD dye was encapsulated into liposomes as a fluorescent marker. The preparation process of DiD-loaded liposomes was consistent with that of luteolin-loaded liposomes using the thin-film hydration method, except that those lipid materials and DiD dye were dissolved in chloroform rather than ethanol. The particle size, zeta potential, DLC, and EE of Lipo@DiD, OMV-lipo@DiD, BO-lipo@DiD, and BO/OMV-lipo@DiD were determined as previously mentioned in 4.5.1 and 4.5.2. The DiD concentration was qualified by determining the fluorescence intensity with a microplate reader (Tecan, Switzerland).
4.7 In vitro epithelial uptake assay
The cellular uptake ability of BO/OMV-lipo@LU was investigated in the Caco-2 cell monolayer model. To observe effects of the OMV proportion on the uptake efficiency of liposomes, three ratios of OMV protein to lipids (1:100, 1:50, and 1:20) were applied to prepare hybrid liposomes, respectively, for subsequent cellular uptake experiments. Caco-2 cells were seeded on 24-well plates at a density of 104 cells/well and incubated with Lipo@DiD and hybrid liposomes with various OMV proportions (including 1:100-OMV-lipo@DiD, 1:50-OMV-lipo@DiD, and 1:20-OMV-lipo@DiD) at 37 ℃ for 1 h. Then, the cells were washed with PBS three times and fixed for 10 min using paraformaldehyde, followed by DAPI (10 µg/mL) staining for 10 min. Cellular uptake imaging was acquired using a fluorescence microscope (Leica, Germany), and the fluorescence intensity was semiquantified using Image J 1.53t (USA).
In addition, to investigate the effects of the DSPE-PEG-BO proportion on uptake efficiency, Caco-2 cells were incubated with liposomes modified with DSPE-PEG-BO of different proportions (of lipids, 3.3%, 6.7%, 10%). Briefly, Caco-2 cells were treated with OMV-lipo@DiD and BO/OMV-lipo@DiD with various DSPE-PEG-BO proportions (including 3.3%-BO/OMV-lipo@DiD, 6.7%-BO/OMV-lipo@DiD, 10%-BO/OMV-lipo@DiD). Then, the cells were washed, fixed, stained with DAPI, and observed by the fluorescence microscope to visualize the epithelial uptake.
4.8 DSS-induced UC model
Female C57BL/6J mice (18 ~ 20 g) were cohoused in cages under specific pathogen-free (SPF) conditions of a 12-h light/dark cycle and acclimatized for a week before random allocation to experimental groups. To establish a DSS-induced UC model, the mice received 3% DSS (w/v) supplemented in drinking water for 7 days23,57.
4.9 In vivo imaging
To investigate the targeting ability of BO/OMV-lipo@LU to the inflamed colon, UC mice were divided into 4 groups and treated orally with Lipo@DiD, BO-lipo@DiD, OMV-lipo@DiD and BO/OMV-lipo@DiD, respectively, at an equivalent DiD dose of 1.25 mg/kg). At the predetermined time points, in vivo fluorescent images were observed with the IVIS (Perkinelmer, USA) after the mice were anesthetized using isoflurane. After 8 h of administration, the mice were euthanized. The main organs (including the heart, liver, spleen, lung, kidney, and colon) were excised, followed by the collection of fluorescence images via the IVIS. The fluorescence intensity analysis was performed by the IVIS and living image software (version 4.4).
4.10 In vivo immunofluorescence imaging
The colon tissue was prepared into frozen sections for immunofluorescence staining to further analyze the colocalization of BO/OMV-lipo@DiD and colon epithelial cells. Specifically, the tissues were fixed in 4% paraformaldehyde overnight and dehydrated in 30% sucrose for 24 h. Then, the tissues were embedded in OCT (optimal cutting temperature), stored at -80 ℃, and cut into slices (slice thickness 10 µm) by a freezing microtome (Leica, Germany). Next, the colon frozen sections were stained with Anti-Epcam mouse antibodies (1:100) and subsequent anti-mouse IgG conjugated with AF555 (1:1000), followed by counter staining with DAPI. The fluorescence images were observed with a laser scanning confocal microscope (SP8, Leica, Germany), and the Pearson correlation coefficient (PCC) was analyzed using Image J.
4.11 Treatment of DSS-induced UC model
For exploring effects of the OMV proportion on the anti-UC efficacy of BO/OMV-lipo@LU, the UC mice model was induced by 3% DSS for 7 days and orally treated with 1:20-BO/OMV-lipo@LU, 1:50-BO/OMV-lipo@LU or 1:100-BO/OMV-lipo@LU. Amongst, the proportion referred to the ratio of OMV proteins to lipids in BO/OMV-lipo@LU. All formulation administration (luteolin dose of 17 mg/kg) was performed on predetermined days complying with the treatment regimen as depicted in Fig. 5a. The healthy mice were provided with normal water as a control.
In the therapeutic experiments, C57BL/6J mice were randomly divided into 7 groups, including i) the healthy control group, ii) the DSS control group, iii) the Free-LU-treated DSS group, iv) the Lipo@LU-treated DSS group, v) the BO-lipo@LU-treated DSS group, vi) the OMV-lipo@LU-treated DSS group, and vii) the BO/OMV-lipo@LU-treated DSS group. According to the treatment regimen in Fig. 6a, all formulations (equivalent luteolin dose of 17 mg/kg) were administrated once daily for 5 days, starting on the third day of DSS treatment. After 7 days of DSS treatment, mice received normal water without DSS for 3 days and were euthanized.
During the experimental periods, the body weight, stool consistency, and fecal blood of mice were monitored daily. The DAI score was calculated based on the summation of body weight loss (0: 0%, 1: 1–5%,2: 6–10%, 3: 11–15%, 4: 15%), stool consistency state (0: hard, 1: soft, 3: diarrhea), and fecal occult blood (0: negative, 1: positive, 3: macroscopic)56. The fecal occult blood detection kit was used to test the degree of occult blood by the benzidine method.
At the end of the experiments, the colon tissue was collected and measured for length. Then, the colon was washed with PBS to remove feces, and a segment from each group was fixed with 4% paraformaldehyde for histological analysis and immunofluorescence staining assays. Additionally, the remaining samples were stored at -80℃ for further use. H&E staining sections of colon tissue were prepared for histological assessment through further hydration and paraffin embedding. The colonic damage severity was scored conforming to a previously described scoring criterion as follows: inflammation cell infiltration (0: few inflammatory cells in the lamina propria, 1: enhanced granulocyte infiltration into the lamina propria, 2: extending into submucosa, 3: extending into muscular and serosal layer), crypts lesion (0: intact crypts, 1: loss of the basal one-third, 2: loss of the basal two-thirds, 3: entire crypt loss, 4: alternation of epithelial surface with erosion, 5: confluent erosion) and ulceration (0: absence of ulceration, 1: 1 ~ 2 foci of ulcerations, 2: 3 ~ 4 foci of ulcerations, 3: confluent ulceration)58.
4.12 In vivo ELISA analysis of inflammatory factors
To determine the concentration of cytokines in serum, mice blood was collected in the procoagulant tubes by eyeball extraction and centrifuged at 1000 g for 20 min to isolate serum. In addition, the cytokine concentration in colon tissues was also detected using the tissue homogenate sample. The colon segments were cryogenically homogenized (60 Hz, 4 min) in PBS containing 1% PMSF by a tissue grinder (Shanghai Jingxin Industrial Development Co., Ltd., China). Subsequently, the homogenized solution was centrifuged at 10,000 g for 10 min to obtain supernatant for further detection. Following the preparation of serum and colon tissue samples, the levels of cytokines, including IL-6, TNF-α, and IFN-γ, were quantified by ELISA assay kits according to the manufacturer's instructions.
4.13 In vivo MPO, GSH, and SOD activity assay
For testing the ability of BO/OMV-lipo@LU to alleviate intestinal oxidative stress, the colon tissue homogenate of mice was prepared for MPO, SOD, and GSH activity analysis. According to the manufacturer’s instructions, the activity of MPO, SOD, and GSH was tested using an MPO activity assay kit, a SOD activity assay kit, and a GSH assay kit, respectively.
4.14 In vivo immunofluorescence imaging of ZO-1 and occludin
The immunofluorescence staining method was employed to detect the expression of tight junction proteins (ZO-1 and occludin) in the colon epithelium. Specifically, the paraffin sections of colon tissue were prepared for multiplex immunofluorescence staining based on the method of tyramide signal amplification (TSA). The sections were dewaxed and placed in EDTA alkaline buffer for antigen repair, followed by 3% H2O2 to block endogenous peroxidase. Then, the tissue sections were blocked using blocking buffer to avoid nonspecific binding and stained with anti-ZO-1 rat monoclonal antibodies (1:50), followed by HRP-labeled goat anti-rat IgG (1:200) and TSA buffer containing the TYR-570Plus red dye. Next, the antibodies were eluted in the citrate antigen repair buffer (pH 6.0). The sections were subsequently reblocked and restained with anti-occludin rabbit monoclonal antibodies (1:200), HRP-labeled goat anti-rabbit IgG, and TSA buffer containing the TYR-520Plus green dye. Nuclei were counter-stained with DAPI. Immunofluorescence images were acquired by confocal laser-scanning microscopy.
4.15 In vivo safety test
Once mice were sacrificed, the main organs (including heart, liver, spleen, lung, and kidney) were separated and weighed to calculate the organ index using the following formula:
$$\:\text{O}\text{r}\text{g}\text{a}\text{n}\:\text{i}\text{n}\text{d}\text{e}\text{x}\:\left(\text{%}\right)=\text{O}\text{r}\text{g}\text{a}\text{n}\:\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}/\text{b}\text{o}\text{d}\text{y}\:\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}\:\text{o}\text{f}\:\text{m}\text{i}\text{c}\text{e}\times\:100\text{\%}$$
3
Then, these organs were fixed in 4% paraformaldehyde and embedded in paraffin. The tissues were sliced, stained with H&E, and observed under a microscope for histological assessments.
4.16 Statistical analysis
All data were expressed as Mean ± SD. Statistical analysis was performed using GraphPad Prism software (Version 9.0). The two-tailed T-test and one-way ANOVA were applied to statistically analyze all data. The data were collected from at least three independent measurements. Statistical significance was defined as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. ns indicated that there is no significant difference.