Modified liposomes carrying TRAIL target activated liver scar cells and reduce liver fibrosis in lab and animal studies

Referring to the previous study,14 minor modifications in the preparation method were made. The diameter and distribution of pPB‐SSL‐TRAIL particles were determined by dynamic light scattering (Nicom™ 380ZLS, Particle Sizer; Particle Sizing Systems Corp., Port Richey, FL, USA).

Human LX‐2 cells and mouse 3T3 cells were purchased from the Cell Bank Type Culture Collection of Chinese Academy of Sciences (Shanghai, China). Mouse 3T3 cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, New York city, NY, USA), high glucose with 10% fetal bovine serum (FBS; Gibco, New York city, NY, USA) at 37°C in a humidified incubator containing 5% CO₂. Human LX‐2 cells were cultured in DMEM with 2% FBS and 1× Glutamin (Millipore, Bedford, MA, USA). To study the anti‐fibrosis effects of pPB‐SSL‐TRAIL, LX‐2 cells were pre‐incubated with TGF‐β1 (transforming growth factor‐β1, 2 ng/mL) for 48 hours.

Cell suspension (100 μL per well) was inoculated in a 96‐well plate and incubated at 37°C in a humidified incubator containing 5% CO₂. Cell counting kit‐8 (CCK‐8) solution (10 μL) was added to each well of the plate, and then the plate was incubated for 4 hours in the incubator. The absorbance at 450 nm was finally measured using a microplate reader. Cell viablitity (%) = [(OD450 (sample)/OD450 (negative control)) × 100].

Cells were collected and detected by routine DAPI (4′,6‐diamidino‐2‐phenylindole) staining using a commercial DAPI dye (Beyotime, Shanghai, China) and Annexin V Apoptosis Detection Kit APC (eBioscience, San Diego, CA, USA) according to the manufacturers’ instructions and analysed by FACS Calibur. Briefly, 1 × 10⁶ cells were washed and resuspended in 1× binding buffer. Fluorochrome‐conjugated Annexin V was added to the cell suspension and incubated at room temperature for 10 minutes. Cells were washed with 1× binding buffer. Propidium iodide (50 μg/mL) was added before flow cytometry analysis.

In order to find out where the rhTRAIL transfected by pPB‐SSL located in 3T3 cells, the cells were analysed using immunofluorescence (IF) staining after incubation with free TRAIL, SSL‐TRAIL or pPB‐SSL‐TRAIL for 4 hours. The cells were labelled with antibody against hTRAIL (ab9959; Abcam, Cambridge, MA, USA), and FITC conjugated secondary antibody (Beyotime) was used to visualize the antigen antibody complex. The images were taken from a fully motorized inverted fluorescent microscope DMi8 (Leica Microsystems CMS, Wetzlar, Germany).

The liver tissues or cells were suspended and lysed in radio‐immunoprecipitation assay buffer (Beyotime) supplemented with protease inhibitor cocktail (Sigma, St. Louis, MO, USA). Protein extractions were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to a polyvinylidene fluoride membrane (Millipore). The membrane was blocked with 5% (w/v) reagent‐grade non‐fat milk (Cell Signaling Technology, Santa Cruz, CA, USA) and incubated with primary antibody against α‐SMA (ab5831; Abcam), p53 (ab1431; Abcam), caspase 3 (ab13847; Abcam), Bid (ab10640; Abcam), DR4 (ab209412; Abcam), DR5 (ab8416; Abcam), uPA (sc‐59727; Santa Cruz, USA), PAI‐1 (sc‐5297; Santa Cruz, USA), β‐crystallin (ab13496; Abcam), collagen I (ab34710; Abcam), collagen III (ab7778; Abcam) or glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH, ab8245; Abcam) at 4°C overnight followed by secondary antibody incubation. The protein bands were visualized using ClarityTM Western ECL substrate (Bio‐Rad, Hercules, CA, USA). The protein level was quantified using Image J software normalized with GAPDH.

Total RNA was extracted using Trizol reagent (Invitrogen). One microgram of RNA was reversely transcribed into cDNA with M‐MLV reverse transcriptase (Promega, Madison, WI, USA). qPCR was performed with SYBR Premix Ex Taq (Takara, Berkeley, CA, USA) on ABI 7500 fast real‐time PCR system (Applied Biosystems, Foster City, CA, USA). GAPDH mRNA was used as an endogenous control for mRNA. The primers used here were shown in Table. S1

Male C57BL/6 mice were purchased from Shanghai SLAC Laboratory Animal Co. Ltd (SLAC, Shanghai, China), and randomly divided into four groups. Liver fibrosis model was induced as previously reported.14 Mice with liver fibrosis were treated with SSL, TRAIL, SSL‐TRAIL and pPB‐SSL‐TRAIL, respectively, for 3 weeks. Mice were executed after 1 or 3 weeks, and liver tissues were collected for immunohistochemistry (IHC), IF and Western blot (WB) analyses. This study was approved by the Ethics Committee of Shanghai East Hospital, Tongji University.

To explore the biodistribution of rhTRAIL in vivo, SSL‐TRAIL‐DiR (liposomes containing both TRAIL and DiR, namely, 1,1′‐dioctadecyl‐3,3,3′,3′‐tetramethyl indotricarbocyanine iodide, a type of fluorescent dye) and pPB‐SSL‐TRAIL‐DiR (pPB modified liposomes containing both TRAIL and DiR) were adjusted to the same intensity and then injected into mice via tail vein. At various time points (0.5, 1, 2, 4, 8, 12, 16, 24 hours), fluorescent images were collected and fluorescent intensities of the bodies were calculated using an in vivo imaging system (FX Pro; Kodak, New York city, NY, USA).

Liver tissues of the mice from each group were sliced and stained with Sirius red. The detailed method has been published before.15

Liver tissues were fixed within 4% paraformaldehyde, embedded in paraffin and cut into sections. The sections were deparaffinized by dimethylbenzene and rehydrated by graded ethanol. For α‐SMA IHC, the sections were placed into boiled citric acid buffer (0.01 mol/L pH 6.0) for 10‐15 minutes to retrieve antigen, then cooled down at room temperature. The sections were incubated with 3% H₂O₂ for 15 minutes and washed with phosphate buffer saline (PBS) three times, then blocked with normal goat serum (Biofavor, Wuhan, China) for 30 minutes at room temperature. The sections were incubated with primary antibody against α‐SMA (ab5831; Abcam) at 4°C overnight. Washed with PBS and incubated with HRP labelled secondary antibody at room temperature for 1 hour. Washed with poly butylene succinate‐co‐butylene terephthalate and incubated with 3, 3‐diaminobenzidine (Dako REAL EnVision Detection System; Dako, Glostrup, Denmark), then incubated with haematoxylin to stain the nuclear. For the IF, the sections were incubated with DAPI (Beyotime) for nuclear staining. The information of antibodies was listed below: α‐SMA (ab5831; Abcam), 1:100 dilution; TRAIL (ab9959; Abcam), 1:100 dilution; TNFRSF10A (CusaBio, Wuhan, China), 1:100 dilution; FITC‐labelled goat anti rabbit IgG (Biofavor), 1:100 dilution.

For in vitro assays, all data were derived from at least three independent experiments performed in triplicates and presented as the mean ± standard deviation. In vivo, biodistribution data of rhTRAIL were collected from three mice in each group, and the data of Sirius red staining, IHC, IF and WB were obtained from 10 mice in each group. Student's t test and analysis of variance were performed where applicable using with SPSS 18.0 software (SPSS, Chicago, IL, USA) or GraphPad Prism Software (GraphPad Software, Inc., San Diego, CA, USA). For all comparisons, differences were considered significant when P <0.05.

As shown in Figure 1A, the morphology of the two types of nanoparticles was observed by TEM (transmission electron microscopy). Both SSL‐TRAIL and pPB‐SSL‐TRAIL presented well‐defined spherical morphology. The carriers SSL and pPB‐SSL without anything in them were also tested by TEM and were similar with SSL‐TRAIL and pPB‐SSL‐TRAIL.13 The average diameters of SSL, pPB‐SSL, SSL‐TRAIL and pPB‐SSL‐TRAIL were 84.9, 83.85, 127 and 131 nm, and their polydispersity distribution index were 0.050, 0.033, 0.097 and 0.047, respectively (Figure 1B).13

After incubation with TGF‐β1 for 48 hours, the mRNA levels of α‐SMA (ACTA2), TGF‐β1 (TGFB1) and collagen I/III (COL1A2 and COL3A1) in LX‐2 cells markedly increased (P <0.05) (Figure S1), suggesting that LX‐2 was activated and therefore qualified to mimic aHSCs in vitro in the following study, where 3T3 fibroblast cells were also qualified because aHSCs are highly proliferative fibroblast‐like cells.16 Meanwhile, the TRAIL receptors DR4 (TRAILR1) and DR5 (TRAILR2) notably (P <0.05) increased while the decoy receptors DcR1 and DcR2 significantly (P <0.01) decreased at transcriptional level, compared with negative control (NC, treated with PBS) (Figure S1). To some degree, this result indicated that the basis of TRAIL targeting therapy in hepatic fibrosis is solid and feasible.

Compared with the group treated with delivering systems (SSL and pPB‐SSL), all other groups showed remarkably cell viability inhibition at certain concentrations of TRAIL. At 0.063 and 0.125 μg/mL, SSL‐TRAIL and pPB‐SSL‐TRAIL were more powerful than free TRAIL in inhibiting the cell viability of 3T3 and activated LX‐2 cells, and pPB‐SSL‐TRAIL owed the strongest inhibitive effect (Figure 2A). Consistent with the results of CCK‐8 assay, the results of cell apoptosis measured with flow cytometry appeared the same profile in various groups: TRAIL, SSL‐TRAIL and pPB‐SSL‐TRAIL induced 15.61 ± 2.88%, 34.39 ± 4.6% and 56.37 ± 6.38% apoptosis of 3T3 cells, respectively; and induced 61.88 ± 4.92%, 76.61 ± 3.73% and 90.72 ± 4.94% apoptosis of activated LX‐2 cells, respectively (Figure 2B). In both 3T3 and activated LX‐2 cells, pPB‐SSL significantly enhanced the apoptotic‐inductive effect of TRAIL and SSL‐TRAIL (P <0.05). More details are shown in the histograms in Figure 2B. To confirm the effect of pPB‐SSL‐TRAIL on cell apoptosis, DAPI staining was performed. The results showed that the apoptotic cells in pPB‐SSL‐TRAIL treatment group, including those at late or early phase of apoptosis, were obviously more than those in other two groups (Figure 3). In addition, SSL and pPB‐SSL could barely influence the viability and apoptosis, and no significant difference existed between them, therefore only SSL was used as a negative control in the following research.

The key proteins in the TRAIL induced apoptotic signalling pathways, including p53, caspase 3, tBid, DR4, DR5, uPA and PAI‐1 were measured with Western blotting. It was found that the levels of the pro‐apoptotic proteins (p53, caspase 3, tBid, DR4, DR5 and PAI‐1) were significantly promoted and the anti‐apoptotic protein uPA was significantly decreased by TRAIL, SSL‐TRAIL and pPB‐SSL‐TRAIL, in which, pPB‐SSL‐TRAIL had the strongest effect on these proteins (P <0.05, Figure 4A,C).

In order to find out whether TRAIL can suppress fibrosis and whether the pPB‐SSL can effectively enhance the anti‐fibrosis effect of TRAIL in vitro, two biomarkers for activated HSCs (β‐crystallin and α‐SMA)17 and collagen I/III were measured with Western blotting. As a result, we found that TRAIL, SSL‐TRAIL and pPB‐SSL‐TRAIL inhibited the expression of β‐crystallin, α‐SMA and collagen I/III to various extents, in which pPB‐SSL‐TRAIL showed the most powerful anti‐fibrosis effect (P <0.01, compared with TRAIL treatment group). At a certain concentration, TRAIL could not markedly (P >0.05) suppress the expression of the fibrosis‐related genes. But SSL modification significantly (P <0.05) enhanced the inhibitive effect of TRAIL on fibrosis gene expression and compared with SSL‐TRAIL treatment group, the protein expression of most fibrotic genes was even significantly (P <0.05) lower in pPB‐SSL‐TRAIL group due to the targeting effect of pPB (Figure 4B,C).

The hepatic fibrosis mice models were successfully established using the method that had been published,13 which was characterized by strong positive staining by sirius red in two randomly chosen mice.

To study where pPB‐SSL delivers rhTRAIL in the liver fibrosis mice models, in vivo optical imaging was adopted. It was demonstrated that the pPB‐SSL‐TRAIL mainly distributed in liver 24 hours after the tail intravenous injection, but SSL‐TRAIL was mainly observed in cranial cavity (Figure 5A). Then we quantitatively analysed the intensity in liver and found that pPB‐SSL‐TRAIL reached >2000 per gram of liver tissue 8 hours after the injection, which was remarkably higher than SSL‐TRAIL (Figure 5B).

As shown in Figure 6A, Sirius red staining was strongest in SSL treatment group. The expression of the hepatic fibrosis biomarker α‐SMA in liver was studied by IHC and WB. The results showed that its expression was notably reduced by TRAIL, SSL‐TRAIL and pPB‐TRAIL, and compared with one another, pPB‐SSL‐TRAIL caused the lowest level of α‐SMA (P <0.05, Figure 6A,B). The immunohistochemical findings showed that the positive staining distributed mainly in Disse, the abluminal side of the sinusoids between liver sinusoidal endothelium and hepatocytes, in the fibrotic liver treated with SSL, because that is the main place where HSCs usually reside.18 Positively stained cells were also found around the Disse space. In the group treated with pPB‐SSL‐TRAIL, the positive staining could not be that frequently seen in Disse or peri‐Disse spaces, while TRAIL and SSL‐TRAIL treatment group displayed partly weakened staining in Disse and/or peri‐Disse cells (Figure 6A).

To assess the apoptosis inductive effect of different TRAIL preparations in vivo, α‐SMA/TUNEL double IF staining was carried out. The results suggested that pPB‐SSL‐TRAIL induced significantly more apoptotic aHSCs than TRAIL or SSL‐TRAIL did (Figure 6C,D; Figure S2). In consistency with the immunohistochemical findings, the IF signal in pPB‐SSL‐TRAIL group was only detected near hepatic sinusoids, what is more, the IF intensity seemed to be the lowest among all groups.

In vitro, the location of rhTRAIL was detected with IF assay after 3T3 cells were incubated with TRAIL, SSL‐TRAIL or pPB‐SSL‐TRAIL for 4 hours. It could hardly be found in the group treated with free TRAIL, but it was mostly observed localized in cell membrane and cytoplasm in SSL‐TRAIL and pPB‐SSL‐TRAIL groups (Figure 7), where the signal detected in cytoplasm was probably from the endosomes containing TRAIL receptors that were transported from membrane.19, 20 Further, when pPB‐SSL‐TRAIL was adopted in vivo, most of the particles were delivered to mice livers (Figure 5), and specifically, on the membrane of DR4 (TRAILR1)‐positive cells. In the group treated with free TRAIL or SSL‐TRAIL, although some rhTRAIL molecules were found co‐localized with DR4 and the double‐positive cell numbers were significantly greater than that in SSL group, the double‐positive cells were significantly less than that in pPB‐SSL‐TRAIL (Figure 8, Figure S3), due to the absence of long circulation or high potent targeting guidance.