Bone mesenchymal stem cells derived extracellular vesicles promote TRAIL-related apoptosis of hepatocellular carcinoma cells via the delivery of microRNA-20a-3p

Abstract

OBJECTIVE:

Bone mesenchymal stem cells (BMSCs) have been widely researched in cancer treatment, including hepatocellular carcinoma (HCC). This study intended to discuss the mechanism of miR-20a-3p in BMSCs-extracellular vesicles (EVs) in HCC apoptosis.

METHODS:

BMSCs were isolated and identified. EVs derived from BMSCs were extracted and identified. After overexpressing or inhibiting miR-20a-3p expression in BMSCs, EVs were extracted and acted on HCC cells and transplanted tumors. HCC cell apoptosis in the treatment of BMSCs-conditioned medium, BMSCs-EVs and/or miR-20a-3p mimic/inhibitor was evaluated, with the detection of levels of TRAIL and TRAIL-related proteins. A functional rescue experiment about c-FLIP was carried out in HCC cells. The target binding relationship between miR-20a-3p and c-FLIP was detected. The subcutaneous tumorigenesis model of mice was established and injected with BMSCs-EVs to estimate the effect of BMSCs-EVs-miR-20a-3p on HCC growth.

RESULTS:

EVs isolated from BMSCs conditioned medium promoted the apoptosis of HCC cells. After BMSCs-EVs treatment, TRAIL levels, downstream proteins and miR-20a-3p were increased significantly, but the expression of c-FLIP was decreased. miR-20a-3p could target c-FLIP. BMSCs-EVs inhibited the growth of HCC cells, decreased c-FLIP expression, increased TRAIL levels, and promote the of HCC cell apoptosis. BMSCs-EVs with overexpressing miR-20a-3p further enhanced the apoptotic effect of HCC cells in vitro and in vivo.

CONCLUSION:

BMSCs-EVs-carried miR-20a-3p targets c-FLIP and increases TRAIL levels in HCC cells, thus promoting TRAIL-related apoptosis.

Keywords

Hepatocellular carcinoma extracellular vesicles bone mesenchymal stem cells miR-20a-39 c-FLIP TRAIL

1. Introduction

Hepatocellular carcinoma (HCC) ranks the sixth most diagnosed primary cancer in the world and the third leading cause for cancer related mortality [1]. It is estimated that the global incidence and related deaths of HCC are rising over the last decades partially due to a growing number of obesity [2, 3]. The major risk factors for HCC are chronic infections with hepatitis B virus and hepatitis C virus, together with alcohol consumption, metabolic syndromes, obesity, diabetes and frequent exposure to aflatoxin [4]. The surgical intervention (excision or transplantation), radio-frequency catheter ablation and percutaneous ethanol injection are potentially effective for about 30 $\sim$ 40% HCC patients at early stages, while unfortunately 25 $\sim$ 70% HCC patients diagnosed at advanced stages have minimal clinical benefits [5]. Despite advances have been made in the molecular mechanisms of HCC, diagnostic and prognostic markers and effective therapeutic targets for HCC still remain unknown and overall survival and prognosis are far from satisfaction [6]. Therefore, confirming the biomarkers may facilitate the development of new therapeutic targets and improve prognosis.

Mesenchymal stem cells (MSCs) are the main components of tumor microenvironment, and play an important role in the progress and metastasis of liver cancer [7]. A previous study has suggested that the communication between MSCs and cancer cells may be a potential therapeutic target to inhibit the growth of cancer cells [7]. Bone marrow MSCs (BMSCs) play a regulatory role in a variety of diseases and tumors [8]. However, the research on BMSC mechanism in HCC is limited, so we choose to explore the role of BMSCs in HCC. The expression of miR-20a-3p in human umbilical cord MSCs is high [9], while the expression of miR-20a-3p in HCC is low [10]. Therefore, we chose miR-20a-3p to study. MSCs could secrete several kinds of extracellular vesicles (EVs) to maintain tissue homeostasis for optimal tissue function [11]. EVs have profound impacts on normal physiology and pathological processes, and EVs contain a broad range of nucleic acids such as mRNA and small non-coding RNAs [12]. EVs derived from BMSCs have great promise to treat neurological and neurodegenerative conditions due to their anti-inflammatory and neuroprotective properties [13]. In addition, as an important member of tumor necrosis factor (TNF) family, TNF-related apoptosis-inducing ligand (TRAIL) is an encouraging anticancer protein, which can specifically kill diverse cancerous cells without damaging the normal ones [14]. HCC cells exhibit major resistance to TRAIL-induced apoptosis [15]. TRAIL is an antitumor protein and is considered as a potential target in HCC due to its promotion of apoptosis in HCC cells [16]. In the combination with chemotherapeutic drugs, like doxorubicin and 5-fluorouracil, TRAIL serves an effective approach for HCC treatment [17]. From all above, we make a hypothesis that there may be underlying mechanisms of TRAIL, miR-20a-3p and BMSCs-EVs in HCC progression. In this regards, we aimed to explore the therapeutic efficiency of BMSCs-EVs in HCC.

2. Materials and methods

2.1 Ethics statement

This study was performed with the approval of the Clinical Ethical Committee of Chengdu First People’s Hospital. Informed consent was received from all subjects. All procedures were strictly conducted in accordance with the code of ethics. Significant efforts were made to minimize the animals used and their suffering.

2.2 Isolation and identification of BMSCs

As previously described [18], the primary BMSCs were obtained from the bone marrow of healthy donors. Then cells were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with antibiotics (100 U/mL penicillin and 100 U/mL streptomycin) and 10% fetal bovine serum (FBS) (Thermo Fisher Scientific Inc., Waltham, MA, USA) at 37 ${{}^{\circ}}$ C with 5% CO ${}_{2}$ . The growth and morphological changes of BMSCs were observed under an inverted microscope (Olympus Optical Co., Ltd, Tokyo, Japan).

BMSCs of the 3 ${}^{\text{rd}}$ generation were treated with 0.25% trypsin and then counted. About 1 $\times$ 10 ${}^{6}$ cells were re-suspended in 200 $\mu$ L of fluorescence-activated cell sorting (FACS) buffer, and respectively added with 2 $\mu$ L of fluorescent antibody and the homotypic control, and then incubated on the ice for 30 minutes. Next, BMSCs were washed with FACS buffer, fixed in 10% formalin, and the positive rate of antigen was determined on a flow cytometer (MoFloAstrios EQ, Beckman Coulter Inc., CA, USA). The antibodies were CD34 (1: 1000, ab78165), CD44 (1: 1000, ab27285), CD45 (1: 1000, ab10558) and CD105 (1: 1000, ab156756) (all from Abcam Inc., Cambridge, MA, USA).

BMSCs of the third generation were cultured in DMEM with 10% FBS to 80% fused monolayer cells. BMSCs were cultivated with 1 $\mu$ L/L of dexamethasone, 0.5 $\mu$ L/L of isobutylmethylxanthine, 60 $\mu$ L/L of indomethacin and 10 $\mu$ L/L of insulin for obtaining of adipocytes, while osteoblasts were developed after coculture of BMSCs with 0.1 $\mu$ L/L of dexamethasone, 10 mmol/L of $\beta$ -phosphoglycerin and 50 $\mu$ L/L ascorbic acid. Three weeks after the induction, alizarin red staining and oil red O staining (Thermo Fisher) were used to observe the osteogenic and lipogenic differentiation.

2.3 Extraction and identification of BMSCs-EVs

BMSCs were seeded into 6-well plates and were adhered for 24 hours. BMSCs were washed with serum-free $\alpha$ -MEM, and cultured with fresh serum-free $\alpha$ -MEM for 24 hours. The conditioned medium (CM) was harvested after centrifugation at 2000 g.

BMSCs-EVs were isolated from CM by differential ultracentrifugation (according to the size and density of particles, the centrifugation force is increased by stages). Then, BMSCs-EVs were identified through nano particle tracking analysis, transmission electron microscope and western blotting to detected the expression of CD63 (1: 1000, ab59479, Abcam), CD81 (1: 1000, ab79559, Abcam), tumor susceptibility gene 101 (TSG101) (1: 1000, ab30871, Abcam), with CM added with EV-targeting phospholipase inhibitor GW4869 (CM-GW, MedChemExpress, LLC, NJ, USA) as the control [19]. The protein content of EVs was determined and standardized by the bicinchoninic acid (BCA) kit (Thermo Fisher).

miR-20a-3p mimic, miR-negative control (NC) and miR-20a-3p inhibitor synthesized by Shanghai Genechem Co., Ltd. (Thermo Fisher) were transfected into BMSCs according to Lipofectamine ${}^{\text{TM}}$ 2000 (Invitrogen Inc., Carlsbad, CA, USA), and then EVs were extracted and named as EV-mimic, EV-NC (Negtive Control) and EV-In (Inhibition), respectively.

2.4 HCC cell culture and treatment

HCC cell lines Huh7 (from human male liver tumor), HLcz01 (from liver tumor of a human male with chronic hepatitis C), SMMC-7721 (from human male liver tumor) and LH-86 (from human male liver tumor) were all purchased from American type culture collection (ATCC, Manassas, Virginia, USA), and cultured in DMEM with 1% penicillin and streptomycin and 10% FBS at 37 ${{}^{\circ}}$ C with 5% CO ${}_{2}$ . Half of the medium was changed every two days.

HCC cells were assigned into blank group (without any treatment), CM group (cultured with CM), CM-GW group (cultured with CM-GW), BMSCs-EVs group (cultured with 5 $\mu$ g/mL or 50 $\mu$ g/mL EVs), BMSCs-EVs-mimic group (cultured with 50 $\mu$ g/mL BMSCs-EVs-mimic), BMSCs-EVs-NC group (cultured with 50 $\mu$ g/mL BMSCs-EVs-NC), BMSCs-EVs-IN group (cultured with 50 $\mu$ g/mL BMSCs-EVs-IN), BMSCs-EVs $+$ NC group (transfected with pcDNA3.1 empty vectors and cultured with BMSCs-EVs), BMSCs-EVs $+$ c-FLIP group (transfected with overexpression c-FLIP vectors and cultured with BMSCs-EVs). The overexpression c-FLIP vector was constructed by cloning c-FLIP cDNA onto pcDNA3.1 (Invitrogen).

2.5 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay

Cells were seeded into 96-well plates (5 $\times$ 10 ${}^{3}$ cells/well). At different time points (0, 24, 48 hours), cells were incubated in 0.5 mg/mL MTT solution at 37 ${}^{\circ}$ C for 4 hours. After the removal of culture medium, dimethyl sulfoxide was added into each well to dissolve methylene. Absorbance at 570 nm was measured and cell viability was calculated.

2.6 Flow cytometry

Cells were seeded into a 12-well plate (8 $\times$ 10 ${}^{4}$ cells/well) and treated according to the instructions of Annexin V-FITC/PI apoptotic kit (Beyotime Biotechnology Co., Ltd, Shanghai, China), and the apoptotic cells were detected by flow cytometer.

2.7 Western blot analysis

The total proteins were extracted from cell and tissue homogenates and protein concentration was determined based on the instructions of BCA kit (Thermo Fisher). After that, the proteins were added with loading buffer and boiled at 95 ${}^{\circ}$ C for 10 minutes. Extracted proteins (30 $\mu$ g) in each sample were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis. The proteins were then transferred into the polyvinylidene fluoride membrane. The membrane was sealed in 5% bovine serum albumin for 1 hour, and incubated with primary antibodies (all from Abcam Inc.) against Bcl-2 (1: 1000, ab182858), Bax (1: 1000, ab32503), TRAIL (1: 1000, ab9959), DR4 (1: 1000, ab8414), DR5 (1: 1000, ab8416), Caspase-8 (1: 1000, ab25901), Caspase-3 (1: 1000, ab13847), c-FLIP (1: 1000, ab8421) and $\beta$ -actin (1: 1000, ab8227) at 4 ${}^{\circ}$ C overnight. After three rinses by tris-buffered saline tween (TBST), the membrane was cultured with the horseradish peroxidase-labeled secondary antibody (ZSGB-Bio Co., Ltd, Beijing, China) for 1 hour. Following fully TBST wash, membranes were visualized with the enhanced chemiluminescence method and developed by the Bio-Rad Gel Dol EZ imager (Bio-Rad Laboratories, Hercules, California, USA). The target band was analyzed using Image J software (National Institutes of Health, Bethesda, Maryland, USA) for gray value analysis.

2.8 Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA in cells was extracted using one-step method of TRIzol (Invitrogen). Then, the extracted high-quality RNA was verified by ultraviolet analysis and formaldehyde deformation electrophoresis. Fluorescent qPCR was performed on the instructions of RT-qPCR kit (Thermo Fisher). All primers of RT-qPCR (Table 1) were designed and synthesized by Sangon Biotech Co., Ltd. (Shanghai, China), with glyceraldehyde-3-phosphate dehydrogenase or U6 as the internal reference. After the reaction, curves of amplification and dissolution were identified, and the 2 ${}^{-\Delta\Delta\text{Ct}}$ method was employed to calculate the relative expression of genes.

Table 1
Primer sequences of RT-qPCR

Gene	Primer sequence
TRAIL	F: 5’-ATGGCTATGATGGAGGTCCAGG-3’
	R: 5’-ATTTTGCGGCCCAGAGCCTTT-3’
GAPDH	F: 5’-GGGAGCCAAAAGGGTCAT-3’
	R: 5’-GAGTCCTTCCACGATACCAA-3’
miR-20a-3p	F: 5’-ACTGCATTATGAGCACTTAAAG-3’
	R: 5’-CTTTAAGTGCTCATAATGCAGT-3’
U6	F: 5’-CGCTTCGGCAGCACATATAC-3’
	R: 5’-AATATGGAACGCTTCACGA-3’

Note: RT-qPCR, reverse transcription quantitative polymerase chain reaction; TRAIL, tumor necrosis factor-related apoptosis-inducing ligand; miR, microRNA; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; F, forward; R, reverse.

2.9 Dual luciferase reporter gene assay

The c-FLIP fragments containing binding sites with miR-20a-3p were cloned to pmirGLO bisoligosaccharide enzyme vector (Promega Corp., Madison, Wisconsin, USA) to construct pmirGLO-c-FLIP-wild type (WT) plasmids. Identical c-FLIP fragments except the mutant ones of miR-20a-3p binding sites were used to construct pmirGLO-c-FLIP-mutant type (MUT). These two kinds of plasmids were respectively transfected with 293T cells (ATCC) and then with miR-20a-3p mimic or miR-NC. Consequently, luciferase activity was determined via dual luciferase reporter assay system (Promega) after 48 hours, with the ratio of relative activity as luciferase activity of firefly to that of renilla luciferase measured.

2.10 Tumor xenograft in nude mice

LH-86 cells collected by centrifugation were re-suspended. Each nude mouse (Nanjing University Institute of model animal research, Nanjing, China) was subcutaneously injected with 20 $\mu$ L cell suspension, containing about 10 ${}^{6}$ cells. Mice were allocated to control, BMSCs-EVs, BMSCs-EVs-NC, BMSCs-EVs-mimic, or BMSCs-EVs-IN groups ( $n=$ 12 in each), injecting with corresponding cells. After 7 days of injection of LH-86 cells, mouse in each group was injected with 100 $\mu$ L PBS or 100 $\mu$ L corresponding EVs at 50 $\mu$ g/mL every other day. The tumor volume of mice was recorded every 3 days. After 28 days, nude mice were euthanized by intraperitoneal injection of 800 mg/kg pentobarbital [20, 21]. After taking out the tumor and weighing, 6 nude mice in each group were fixed in 4% paraformaldehyde, paraffin embedded and sectioned, and the remaining 6 nude mice were made into homogenate for western blotting.

2.11 TUNEL staining

Sections were treated and stained in TUNEL reaction buffer. Then, each section was stained using 4’,6-diamidino-2-phenylindole, and added with fluorescence quencher. The pictures were captured by a fluorescence microscope (BX51, Olympus) within 24 h. Apoptotic cells were calculated and analyzed using Image J software.

2.12 Statistical analysis

Statistical analysis was conducted by SPSS 21.0 (IBM Corp. Armonk, NY, USA). All the data were in normality distribution checked by the Kolmogorov-Smirnov test. Measurement data were expressed as mean $\pm$ standard deviation. The $t$ test was used for comparisons between two groups. One-way or two-way analysis of variance (ANOVA) was used for comparisons among multiple groups, and Tukey’s multiple comparisons test was used for post-hoc test. The $p$ value was obtained by a two-tailed test and $p<$ 0.05 indicated a statistically significant difference.

Figure 1.

Identification of BMSCs. A, BMSCs isolated from the bone marrow of healthy volunteers, left: BMSCs cultured for 1 day, right: BMSCs cultured for 5–6 days; B, flow cytometry screening of cell surface antigens in the third generation of BMSCs showed that CD29 and CD90 were positive, while CD34 was negative; C, Alizarin red staining showed BMSCs had osteogenic differentiation ability; D, oil red O staining showed BMSCs had lipogenic differentiation ability. Replicates $=$ 3.

3. Results

3.1 Identification of BMSCs

The primary BMSCs were in spindle-shaped adherence. After 5–6 days, BMSCs grew rapidly and reached 80% confluence in a whirlpool (Fig. 1A). Then primary cells were cultured around 14 days, and remained relatively stable after long-term passage to 20 ${}^{\text{th}}$ generations. The results of flow cytometry screening of cell surface antigens in the third generation of BMSCs (Fig. 1B) showed that CD29 and CD90 were positive, while CD34 was negative, indicating no mixed hematopoietic stem cells in BMSCs. The 3 ${}^{\text{rd}}$ generations of BMSCs were induced to differentiate into osteoblasts and chondrocytes, which were identified by alizarin red staining (Fig. 1C) and oil red O staining (Fig. 1D), respectively, showing BMSCs had multi-lineage differentiation potential.

3.2 BMSCs-EVs promote HCC cell apoptosis

To explore the role of BMSCs in the apoptosis of HCC cells, we first applied the CM of BMSCs to Huh7, HLCZ01, SMMC-7721 and LH-86 cells, and through flow cytometry we found the apoptosis rate was significantly increased after culture in CM ( $p<$ 0.05; Fig. 2A). Western blotting showed that after CM treatment, Bcl-2 expression was decreased while Bax was increased significantly ( $p<$ 0.01; Fig. 2B), suggesting that the CM of BMSCs promote HCC cell apoptosis.

Figure 2.

BMSCs-EVs promote HCC cell apoptosis. A/G, flow cytometry showed that the apoptosis rate of HCC cells was increased significantly after BMSCs-CM treatment, while the CM with GW4869 (CM-GW) had no significant effect on promoting apoptosis (A); B, western blot analysis detected the protein levels of Bcl-2 and Bax; C, the morphology of BMSCs-EVs was observed by transmission electron microscope; D, the concentration and size of BMSCs-EVs were analyzed by nanoparticles; E, western blot analysis detected the expression of BMSCs-EVs specific markers, with CM-GW as the control; F, MTT detected the proliferation of LH-86 and Huh7 cells treated with EVs of different concentrations; The apoptosis rate of LH-86 and Huh7 cells treated with high concentration BMSCs-EVs (50 $\mu$ g/mL) was increased significantly (G). Compared with the blank group, ${}^{***}$ , $p<$ 0.001. Data in panels A, B and F were analyzed by two-way ANOVA, and data in panel G were analyzed by one-way ANOVA, followed by Tukey’s multiple comparisons test. Replicates $=$ 3.

Figure 3.

BMSCs-EVs promote HCC cell apoptosis by increasing the sensitivity of HCC cells to TRAIL. A, RT-qPCR detected the mRNA expression of TRAIL in LH-86 and Huh7 cells treated with 50 $\mu$ g/mL BMSCs-EVs; B-C, western blot analysis detected the protein levels of TRAIL and proteins related to TRAIL-induced apoptosis in LH-86 and Huh7 cells treated with 50 $\mu$ g/mL BMSCs-EVs. Compared with the blank group, ${}^{**}$ , $p<$ 0.01, ${}^{***}$ , $p<$ 0.001. Data were analyzed by two-way ANOVA, followed by Tukey’s multiple comparisons test. Replicates $=$ 3.

We speculated that the EVs secreted by BMSCs might play a role in this process, so we added the EV inhibitor GW4869 to BMSCs, and then extracted its CM (CM-GW) to treat HCC cells. It was found that BMSCs-CM lost its effect on HCC cells after inhibiting the formation of EVs (Fig. 2A and B).

To further confirm that BMSCs promote HCC cell apoptosis through the EVs, EVs was isolated from the CM of BMSCs by differential ultracentrifugation. Under the transmission electron microscope, EVs were oval in shape and about 100 nm in size (Fig. 2C). The peak particle size was about 100 nm and the concentration was about 3.5 $\times$ 10 ${}^{6}$ particles/mL (Fig. 2D). Western blot showed that CD63, CD81 and TSG101 were positive, while Calnexin was negative (Fig. 2E). The results showed that the separation of EVs was successful.

Next, the separated BMSCs-EVs were applied to LH-86 and Huh7 cells. MTT results showed no significant difference in the effect of low concentration (5 $\mu$ g/mL) EVs on LH-86 and Huh7 cells (all $p>$ 0.05), but high concentration (50 $\mu$ g/mL) of EVs could significantly inhibit the proliferation of LH-86 and Huh7 cells and promote apoptosis (all $p<$ 0.05; Fig. 2F and G). BMSCs-EVs can promote the apoptosis of HCC cells.

3.3 BMSCs-EVs promote HCC cell apoptosis by increasing the sensitivity of HCC cells to TRAIL

HCC, like other tumors, has inherent resistance to TRAIL induced apoptosis [16]. To explore the relationship between BMSCs-EVs and TRAIL-induced apoptosis, we found TRAIL levels in HCC cells (LH-86 and Huh7) were significantly increased, and the expression of DR4 and DR5, Caspase-8 and Caspase-3 related to TRAIL-induced apoptosis were also upregulated (all $p<$ 0.05). Especially in HCC cells, we found that the expression of c-FLIP (a protein that inhibits TRAIL-induced apoptosis by inhibiting fas-associated death domain (FADD) was significantly reduced ( $p<$ 0.01) (Fig. 3A–C), indicating that BMSCs-EVs promote HCC cell apoptosis by increasing the sensitivity of HCC cells to TRAIL.

3.4 BMSCs-EVs carry miR-20a-3p into HCC cells and miR-20a-3p targets c-FLIP

As we mentioned in the introduction the role of EVs and miR, we speculated that BMSCs-EVs might enhance the sensitivity of TRAIL in HCC cells by transferring miR, and thus promote apoptosis. The EV can play its role in regulating disease by transferring miRs [22, 23]. miR-20a is highly expressed in the BMSC-EVs [9]. Through literature review, we found that miR-20a was downregulated in HCC patients [10], and in online analysis of database (http://www.targetscan.org) [24], miR-20a-3p and c-FLIP 3’UTR had a binding site (Fig. 4A; c-FLIP also known as CFLAR), and c-FLIP participates in TRAIL-induced apoptosis of HCC cells [25]. Then we constructed a dual luciferase report vector to verify the binding relationship between miR-20a-3p and c-FLIP. The results showed that the WT sequence could combine with miR-20a-3p mimic to reduce luciferase activity (Fig. 4B). So we speculated that BMSCs-EVs could inhibit c-FLIP expression and promote apoptosis induced by TRAIL pathway by carrying miR-20a-3p into HCC cells. RT-qPCR found miR-20a-3p expression in BMSCs-EVs was significantly higher than that in CM-GW ( $p<$ 0.05), while in LH-86 and Huh7 cells also increased after BMSCs-EVs ( $p<$ 0.05) (Fig. 4C), indicating that BMSCs-EVs carry miR-20a-3p into HCC cells.

Figure 4.

BMSCs-EVs carry miR-20a-3p into HCC cells and miR-20a-3p targets c-FLIP. A, online database (http://www.targetscan.org) analyzed the target relationship between miR-20a-3p and c-FLIP; B, dual luciferase report gene assay verified the target relationship between miR-20a-3p and c-FLIP; C, RT-qPCR detected miR-20a-3p expression in CM-GW AND BMSCs-EVs (i) and in LH-86 and Huh7 cells before and after treatment with BMSCs-EVs (ii); D, RT-qPCR detected miR-20a-3p expression in EVs derived from BMSCs transfected with miR-20a-3p mimic or inhibitor; E, western blot analysis detected the protein levels of c-FLIP in LH-86 and Huh7 cells treated with EVs. ${}^{**}$ , $p<$ 0.01, ${}^{***}$ , $p<$ 0.001. Data in panel C (i) were analyzed by the t test, in panels B and C (ii) were analyzed by two-way ANOVA, and data in panels D and E were analyzed by one-way ANOVA, followed by Tukey’s multiple comparisons test. Replicates $=$ 3.

Figure 5.

Overexpression of c-FLIP reverses the apoptosis promoting effect of BMSCs-EVs on HCC cells. A, western blot analysis detected the protein levels of c-FLIP after transfection; B, flow cytometry detected HCC cell apoptosis; the concentration of BMSCs-EVs was 50 $\mu$ g/mL. ${}^{**}$ , $p<$ 0.01, ${}^{***}$ , $p<$ 0.001. Data were analyzed by two-way ANOVA, followed by Tukey’s multiple comparisons test. Replicates $=$ 3.

Then we transfected miR-20a-3p mimic/inhibitor into BMSCs and extracted the EVs (BMSCs-EVs-mimic and BMSCs-EVs-inhi). miR-20a-3p expression in BMSCs-EVs-mimic and BMSCs-EVs-inhi increased or decreased significantly ( $p<$ 0.05; Fig. 4D). Western blot analysis displayed that after LH-86 and Huh7 cells were treated with these EVs, compared with HCC cells treated with BMSCs-EVs, c-FLIP expression in HCC cells treated with BMSCs-EVs-mimic was reduced, while in HCC cells treated with BMSCs-EVs-inhi was significantly elevated, indicating that miR-20a-3p had a negative regulatory relationship with c-FLIP ( $p<$ 0.05; Fig. 4E).

3.5 Overexpression of c-FLIP reverses the apoptosis promoting effect of BMSCs-EVs on HCC cells

In the functional rescue experiment, we first constructed a vector expressing c-FLIP and transfected it into LH-86 and Huh7 cells. We found c-FLIP expression was increased significantly, indicating the transfection was successful (Fig. 5A). After BMSCs-EVs were used to treat LH-86 and Huh7 cells overexpressing c-FLIP, the apoptotic rate was significantly lower than those only treated with BMSCs-EVs ( $p<$ 0.05; Fig. 5B), indicating the apoptotic effect of BMSCs-EVs on HCC cells was reversed by overexpression of c-FLIP.

3.6 BMSCs-EVs inhibit the growth of HCC tumors in mice

After 7 days, BMSCs-EVs, BMSCs-EVs-mimic or BMSCs-EVs-inhi were injected into the tail vein every other day. The tumor volume and weight of mouse in each group were compared. Compared with the control, BMSCs-EVs effectively inhibited the growth of HCC tumor (Fig. 6A and B), increased TUNEL-positive rate in tumors (Fig. 6C), reduced c-FLIP expression and increased TRAIL expression (all $p<$ 0.01; Fig. 6D). Compared with BMSCs-EVs treatment, BMSCs-EVs-mimic treatment further inhibited tumor growth and promoted apoptosis and TRAIL expression, inhibited c-FLIP expression; while BMSCs-EVs-inhi treatment showed the opposite trends (Fig. 6A–D).

Figure 6.

BMSCs-EVs inhibit the growth of HCC tumors in mice. A, the growth curves of tumor volume in mice in each group; B, representative images of tumors and histogram of tumor weight; C, TUNEL staining detected the apoptosis; D, western blot analysis detected c-FLIP and TRAIL levels in tumor homogenate. $n=$ 12. ${}^{***}$ , compared with the control group, $p<$ 0.001; ${}^{{\#}{\#}{\#}}$ , compared with the BMSCs-EVs group, $p<$ 0.001. Data in panels A and D were analyzed by two-way ANOVA, and data in panels B and C were analyzed by one-way ANOVA, followed by Tukey’s multiple comparisons test.

4. Discussion

HCC almost ran an acute fulminating course and held an extremely terrible prognosis with low rates of resectability and high rates of recurrence after surgery, especially responding awfully to anticancer agents and radiotherapies [26]. A recent research has indicated that ADSCs conditioned medium could suppress HCC cell growth by repressing cell proliferation and migration, and facilitating cell apoptosis [27]. In this study, we evaluated the underlying mechanism of TRAIL and BMSCs-EVs in HCC apoptosis using both in vivo and in vitro experiments to confirm our hypothesis. Consequently, the observations in this study supported that miR-20a-3p carried by BMSCs-EVs targets c-FLIP and increases TRAIL levels in HCC cells, thus promoting TRAIL-related apoptosis.

Study in animal models has indicated that HCC may originate from genetically mutated BMSCs, BMSCs can modulate their microenvironment via secreted cytokines and BMSCs have been recognized as a putative target of molecular therapies to treat and prevent HCC [28]. The first result of our study was that CM of BMSCs promoted HCC cell apoptosis. Then we added the EV inhibitor GW4869 to BMSCs, and then extracted its CM to treat HCC cells. It was found that BMSCs-CM lost its effect on HCC cell apoptosis after inhibiting the formation of EVs. MTT results further showed 50 $\mu$ g/mL of EVs could significantly inhibit the proliferation of LH-86 and Huh7 cells and promote apoptosis. BMSCs are implicated in cancer initiation and metastasis, sometimes by releasing EVs that mediate cell communication [29]. BMSC-EVs induce apoptosis and cell cycle arrest in HepG2 cells and suppresses tumor progression [30]. In conclusion, BMSCs-EVs can promote the apoptosis of HCC cells.

In addition, TRAIL levels in LH-86 and Huh7 cells were increased, and the expression of DR4 and DR5, Caspase-8 and Caspase-3 related to TRAIL-induced apoptosis were also upregulated after BMSCs-EVs treatment. It is highly noteworthy that mice administered soluble recombinant TRAIL exhibited smaller size of breast adenocarcinoma xenograft tumors [31]. TRAIL alone is able to inhibit HCC cell viability and promote cell apoptosis [32]. Additionally, MSC-delivered TRAIL displays an anti-proliferative effect and a tumor-supportive function, showing greater capability to induce apoptosis in primary tumor cells [33]. TRAIL can trigger apoptosis upon binding to cell death receptors (DR4 and DR5) to activate cytoplasmic death domain and recruit FADD assembled in the death inducing signaling complex, which then interacts with caspase-8 and caspase-10 and activates proapoptotic signaling cascades [16]. The use of TRAIL alone in treating HCC may lead to chemotherapeutic resistance or reduced drug efficacy, because chemotherapeutics or biological agents upregulate DR4 and DR5 and enhance TRAIL-induced apoptosis in HCC, simultaneously upregulating decoy receptors [17]. The mice lack of TRAIL are more likely to develop tumors after administration of carcinogens and show faster development of malignant tumors in the blood system, and TRAIL is further confirmed to inhibit solid tumor growth and xenograft tumors [34], which supported the results in the transplanted tumors in nude mice in our study. HCC cells are commonly TRAIL-resistant, while adjuvant agents that can increase the sensitivity of cancer cells to TRAIL may improve the curative effect of TRAIL-based combined therapy [35].

As a gene delivery system in cancer therapy, EVs have attracted more and more attention. They can transfer nucleic acids, proteins and lipids to receptor cells naturally [22]. miR is a key regulatory molecule of biochemical pathway, and many miRs are abnormally expressed in HCC [36]. In this study we verified that BMSCs-EVs carry miR-20a-3p into HCC cells and miR-20a-3p targets c-FLIP. Patients with decreased miR-20a expression have notably shorter relapse free survival and overall survival [37]. Similarly, a previous study provides a new strategy to improve the chemosensitivity of HCC via adipose tissue derived MSCs-EVs mediated miR-199a-3p [38]. After treatment of BMSCs-EVs-mimic, c-FLIP was reduced, indicating that BMSCs-EVs promote HCC cell apoptosis by increasing the sensitivity of HCC cells to TRAIL. c-FLIP is always expressed in human HCC, and its overexpression suggests a shorter period of recurrence-free survival, while silencing c-FLIP upregulates drug-induced apoptosis, and may be a therapeutic potential for HCC treatment [39]. Interestingly, miR-20a targets CFLAR, facilitates activation of caspase-8 and caspase-3, and enhances sensitivity of HepG2 cells towards TRAIL-induced apoptosis, and subsequently reduced cell proliferation [10]. To sum up, BMSCs-EVs could inhibit c-FLIP expression and promote apoptosis induced by TRAIL pathway by carrying miR-20a-3p into HCC cells.

Together these results offer insights into the mechanism of TRAIL and BMSCs-EVs in HCC development, whereby BMSCs-EVs inhibit c-FLIP expression and promote TRAIL-induced apoptosis by carrying miR-20a-3p into HCC cells. The theoretical basis of anticancer therapy should be carried forward with improved understanding of BMSCs-EVs and their interaction with anticancer genes, which can help to develop personalized strategies for optimizing HCC diagnosis and prognosis. Further study should be conducted to find out the exact application approach for HCC based on results obtained from this study. We selected four kinds of cells to explore the role of BMSCs in the apoptosis of HCC cells. To further study whether BMSCs play a role by secreting EVs and the mechanism of EVs, considering the limitation of experimental funds, we only selected two cells with better therapeutic effect of BMSCs for following experiments according to the results in Fig. 2A.

Footnotes

Conflict of interest

The authors declared that they have no competing interests.

References

Niendorf

Spilseth

Wang

and Taylor

, Contrast enhanced MRI in the diagnosis of HCC, Diagnostics (Basel) 5 (2015), 383–398.

Dutta

and Mahato

R.I.

, Recent advances in hepatocellular carcinoma therapy, Pharmacol Ther 173 (2017), 106–117.

Kim

D.W.

Talati

and Kim

, Hepatocellular carcinoma (HCC): beyond sorafenib-chemotherapy, J Gastrointest Oncol 8 (2017), 256–265.

de Martel

Maucort-Boulch

Plummer

and Franceschi

, World-wide relative contribution of hepatitis B and C viruses in hepatocellular carcinoma, Hepatology 62 (2015), 1190–1200.

Park

J.W.

Chen

Colombo

Roberts

L.R.

Schwartz

Chen

P.J.

Kudo

Johnson

Wagner

Orsini

L.S.

and Sherman

, Global patterns of hepatocellular carcinoma management from diagnosis to death: the BRIDGE study, Liver Int 35 (2015), 2155–2166.

Chen

Zhang

Feng

Wang

and Chen

, Progress and prospects of long noncoding RNAs (lncRNAs) in hepatocellular carcinoma, Cell Physiol Biochem 36 (2015), 423–434.

Liu

Ren

Zhou

Shang

Zhang

Yang

and Shi

, The hypoxia conditioned mesenchymal stem cells promote hepatocellular carcinoma progression through YAP mediated lipogenesis reprogramming, J Exp Clin Cancer Res 38 (2019), 228.

Papait

Stefani

F.R.

Cargnoni

Magatti

Parolini

and Silini

A.R.

, The multifaceted roles of MSCs in the tumor microenvironment: interactions with immune cells and exploitation for therapy, Front Cell Dev Biol 8 (2020), 447.

Zhang

Song

Chen

Feng

Fan

Chen

Livingston

M.J.

and Geng

, MiR-20a-containing exosomes from umbilical cord mesenchymal stem cells alleviates liver ischemia/reperfusion injury, J Cell Physiol 235 (2020), 3698–3710.

10.

Wang

Zhao

Y.R.

Zhang

A.Y.

Wang

Z.Z.

and Zhang

, Targeting of miR-20a against CFLAR to potentiate TRAIL-induced apoptotic sensitivity in HepG2 cells, Eur Rev Med Pharmacol Sci 21 (2017), 2087–2097.

11.

Lai

R.C.

Yeo

R.W.

and Lim

S.K.

, Mesenchymal stem cell exosomes, Semin Cell Dev Biol 40 (2015), 82–88.

12.

D’Souza-Schorey

and Schorey

J.S.

, Regulation and mechanisms of extracellular vesicle biogenesis and secretion, Essays Biochem 62 (2018), 125–133.

13.

Kodali

Castro

O.W.

Kim

D.K.

Thomas

Shuai

Attaluri

Upadhya

Gitai

Madhu

L.N.

Prockop

D.J.

and Shetty

A.K.

, Intranasally administered human MSC-derived extracellular vesicles pervasively incorporate into neurons and microglia in both intact and status epilepticus injured forebrain, Int J Mol Sci 21 (2019).

14.

Ding

Polier

Kohler

Giaisi

Krammer

P.H.

and Li-Weber

, Wogonin and related natural flavones overcome tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) protein resistance of tumors by down-regulation of c-FLIP protein and up-regulation of TRAIL receptor 2 expression, J Biol Chem 287 (2012), 641–649.

15.

Chen

K.F.

Tai

W.T.

Liu

T.H.

Huang

H.P.

Lin

Y.C.

Shiau

C.W.

P.K.

Chen

P.J.

and Cheng

A.L.

, Sorafenib overcomes TRAIL resistance of hepatocellular carcinoma cells through the inhibition of STAT3, Clin Cancer Res 16 (2010), 5189–5199.

16.

Jiang

D.B.

S.Y.

Chen

E.Q.

Tang

and Zhou

T.Y.

, Insight into the role of TRAIL in liver diseases, Biomed Pharmacother 110 (2019), 641–645.

17.

Koehler

B.C.

Urbanik

Vick

Boger

R.J.

Heeger

Galle

P.R.

Schuchmann

and Schulze-Bergkamen

, TRAIL-induced apoptosis of hepatocellular carcinoma cells is augmented by targeted therapies, World J Gastroenterol 15 (2009), 5924–5935.

18.

Yang

Chen

Feng

Liu

Wang

Chen

Xie

and Rong

, Amyloid beta peptide promotes bone formation by regulating Wnt/beta-catenin signaling and the OPG/RANKL/RANK system, FASEB J (2020).

19.

Essandoh

Yang

Wang

Huang

Qin

Hao

Wang

Zingarelli

Peng

and Fan

G.C.

, Blockade of exosome generation with GW4869 dampens the sepsis-induced inflammation and cardiac dysfunction, Biochim Biophys Acta 1852 (2015), 2362–2371.

20.

Kopaladze

R.A.

, Methods for the euthanasia of experimental animals – the ethics, esthetics and personnel safety, Usp Fiziol Nauk 31 (2000), 79–90.

21.

Zatroch

K.K.

Knight

C.G.

Reimer

J.N.

and Pang

D.S.

, Refinement of intraperitoneal injection of sodium pentobarbital for euthanasia in laboratory rats (Rattus norvegicus), BMC Vet Res 13 (2017), 60.

22.

Tran

P.H.L.

Xiang

Tran

T.T.D.

Yin

Zhang

Kong

Chen

Sun

Hou

Zhu

and Duan

, Exosomes and nanoengineering: a match made for precision therapeutics, Adv Mater (2019), e1904040.

23.

Zhang

Wang

Zhao

Han

Zhao

and Wang

, Extracellular vesicle-encapsulated miR-22-3p from bone marrow mesenchymal stem cell promotes osteogenic differentiation via FTO inhibition, Stem Cell Res Ther 11 (2020), 227.

24.

Sorinson

S.N.

Korochkina

O.V.

Malysheva

E.B.

Zhaliauskas

A.B.

and Tsybasova

A.I.

, Clinico-immunological variants of extremely severe course of hepatitis B, Sov Med (1989), 8–10.

25.

Zhang

Huang

Wang

Luo

and Wang

, 2-DG-Regulated RIP and c-FLIP effect on liver cancer cell apoptosis induced by TRAIL, Med Sci Monit 21 (2015), 3442–3448.

26.

Kew

M.C.

, Hepatocellular carcinoma: epidemiology and risk factors, J Hepatocell Carcinoma 1 (2014), 115–125.

27.

Xie

Liao

Lan

Cai

Liu

and Liu

, 3D-cultured adipose tissue-derived stem cells inhibit liver cancer cell migration and invasion through suppressing epithelial-mesenchymal transition, Int J Mol Med 41 (2018), 1385–1396.

28.

Gong

Wang

Zhang

and Wang

, Bone marrow mesenchymal stem cells in hepatocellular carcinoma, Front Biosci (Landmark Ed) 18 (2013), 811–819.

29.

Jiang

Guo

Zhuang

Huang

and Mao

, Human bone marrow mesenchymal stem cells-derived microRNA-205-containing exosomes impede the progression of prostate cancer through suppression of RHPN2, J Exp Clin Cancer Res 38 (2019), 495.

30.

Alzahrani

F.A.

El-Magd

M.A.

Abdelfattah-Hassan

Saleh

A.A.

Saadeldin

I.M.

El-Shetry

E.S.

Badawy

A.A.

and Alkarim

, Potential effect of exosomes derived from cancer stem cells and MSCs on progression of DEN-induced HCC in rats, Stem Cells Int 2018 (2018), 8058979.

31.

Dimberg

L.Y.

Anderson

C.K.

Camidge

Behbakht

Thorburn

and Ford

H.L.

, On the TRAIL to successful cancer therapy? Predicting and counteracting resistance against TRAIL-based therapeutics, Oncogene 32 (2013), 1341–1350.

32.

Zhang

Zhao

Liu

Wang

and Lu

, Smac mimetic SM-164 potentiates APO2L/TRAIL- and doxorubicin-mediated anticancer activity in human hepatocellular carcinoma cells, PLoS One 7 (2012), e51461.

33.

Grisendi

Bussolari

Cafarelli

Petak

Rasini

Veronesi

De Santis

Spano

Tagliazzucchi

Barti-Juhasz

Scarabelli

Bambi

Frassoldati

Rossi

Casali

Morandi

Horwitz

E.M.

Paolucci

Conte

and Dominici

, Adipose-derived mesenchymal stem cells as stable source of tumor necrosis factor-related apoptosis-inducing ligand delivery for cancer therapy, Cancer Res 70 (2010), 3718–3729.

34.

Das

Persaud

Dejoie

Happy

Brannigan

De Jesus

and Sauane

, Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) activates caspases in human prostate cancer cells through sigma 1 receptor, Biochem Biophys Res Commun 470 (2016), 319–323.

35.

Jin

Cai

Wang

Deng

Lei

Xiao

Luo

and Sun

, CASC2/miR-24/miR-221 modulates the TRAIL resistance of hepatocellular carcinoma cell through caspase-8/caspase-3, Cell Death Dis 9 (2018), 318.

36.

L.M.

Liu

Z.X.

and Cheng

Q.Y.

, Exosome plays an important role in the development of hepatocellular carcinoma, Pathol Res Pract 215 (2019), 152468.

37.

Fan

M.Q.

Huang

C.B.

Xiao

Sheng

J.X.

and Zhong

, Decrease expression of microRNA-20a promotes cancer cell proliferation and predicts poor survival of hepatocellular carcinoma, J Exp Clin Cancer Res 32 (2013), 21.

38.

Lou

Chen

Xia

Wang

Zhao

Chen

Zheng

and Liu

, MiR-199a-modified exosomes from adipose tissue-derived mesenchymal stem cells improve hepatocellular carcinoma chemosensitivity through mTOR pathway, J Exp Clin Cancer Res 39 (2020), 4.

39.

Bao

Zhao

Qiao

and Ma

, Expression and biological significance of c-FLIP in human hepatocellular carcinomas, J Exp Clin Cancer Res 28 (2009), 24.

Bone mesenchymal stem cells derived extracellular vesicles promote TRAIL-related apoptosis of hepatocellular carcinoma cells via the delivery of microRNA-20a-3p

Abstract

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Ethics statement

2.2 Isolation and identification of BMSCs

2.3 Extraction and identification of BMSCs-EVs

2.4 HCC cell culture and treatment

2.5 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay

2.6 Flow cytometry

2.7 Western blot analysis

2.8 Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Table 1 Primer sequences of RT-qPCR

2.10 Tumor xenograft in nude mice

2.11 TUNEL staining

2.12 Statistical analysis

3.1 Identification of BMSCs

3.2 BMSCs-EVs promote HCC cell apoptosis

3.4 BMSCs-EVs carry miR-20a-3p into HCC cells and miR-20a-3p targets c-FLIP

3.6 BMSCs-EVs inhibit the growth of HCC tumors in mice

Footnotes

Conflict of interest

References

Table 1
Primer sequences of RT-qPCR