Exosomal MiR‐500a‐3p promotes cisplatin resistance and stemness via negatively regulating FBXW7 in gastric cancer

Abstract Chemoresistance has been a major challenge in advanced gastric cancer (GC) therapy. Exosomal transfer of oncogenic miRNAs implicates important effects in mediating recipient cell chemoresistance by transmitting active molecules. In this study, we found that microRNA‐500a‐3p was highly expressed in cisplatin (DDP) resistant GC cells (MGC803/DDP and MKN45/DDP) and their secreted exosomes than that in the corresponding parental cells. MGC803/DDP‐derived exosomes enhance DDP resistance and stemness properties of MGC803 recipient cells via exosomal delivery of miR‐500a‐3p in vitro and in vivo through targeting FBXW7. However, reintroduction of FBXW7 in MGC803 cells reverses miR‐500a‐3p‐mediated DDP resistance as well as stemness properties. Furthermore, elevated miR‐500a‐3p in the plasma exosomes of GC patients is correlated with DDP resistance and thereby results in poor progression‐free prognosis. Our finding highlights the potential of exosomal miR‐500a‐3p as an potential modality for the prediction and treatment of GC with chemoresistance.


| INTRODUC TI ON
Gastric cancer (GC) is a serious global public health problem that rank the sixth most common malignancy and the third leading cause of cancer-related deaths over the world. 1 Because of a majority of GC patients diagnosed in advanced or metastatic stages, 2 chemotherapy has been the pivotal therapeutic strategy to improve prognosis. 3 In this regard, cisplatin (DDP) is one of the most important and basic drugs in chemotherapy regimen against advanced GC. 4 However, chemoresistance-whether intrinsic or acquired-remains an inevitable obstacle in most GC patients and represents the most important cause of recurrence and mortality in GC. 5 Exosomes have been identified as an important group of 30-100 nm sized extracellular vesicles with lipid membranes and cupshaped constructio. 6 When endosomal multivesicular bodies merge into cell membrane, exosomes containing biomolecules can be discharged into the extracellular surroundings. [7][8][9] Later, exosomes that uptake by surrounding or distant recipient cells would carry out various biological activities such as immunomodulation, 10 angiogenesis, 11 autophagy, 12 stem cell differentiation 13 and intercellular communication. 14 While in cancer research, a plethora of recent evidence shows exosomes participate in tumour microenvironment remodelling, development, metastasis and chemoresistance through initiating or suppressing various signalling pathways in the recipient cells. 15 RNA cargo that protected by exosomes from digestion has garnered much attention from researchers, especially microRNAs (miRNAs). MiRNAs are a class of 18-22 nucleotides small single-stranded non-coding RNA molecules that promote mRNA cleavage and subsequent degradation by binding to the complementary 3′ untranslated region (UTR) of the mRNA and thereby regulate protein regulation. 16 Emerging evidence demonstrated that exosomal miRNAs from cancer cells played important roles in mediating tumour growth, metastasis and chemoresistance. [17][18][19][20] However, the mechanisms of exosomeal miRNAs in DDP resistant GC are still waiting for exposure.
In this study, the effects of exosomal miRNAs on DDP resistance in GC cells were investigated. Furthermore, we identify exosomal miR-500a-3p promote DDP resistance and CSCs properties in GC cells by down-regulating FBXW7. Clinically, miR-500a-3p expression correlated positively with DDP resistance as well as recurrence and might be a potential therapeutic predictor of DDP-based chemotherapy in GC patients.

| Cell culture
GC cell lines MGC803 and MKN45 were obtained from the ATCC.

| Exosome isolation and purification
An ExoQuick precipitation kit (System Biosciences, LLC, Palo Alto, CA, USA) was used to extract and purify exosomes in accordance with the manufacturer's instruction. Briefly, the culture medium or plasma was harvested and centrifuged at 3000 g for 15 minutes.
Then, the obtained supernatant was mixed with ExoQuick precipitation solution and incubated at 4°C for 30 minutes, centrifuged at 1500 g for 25 minutes. After removing the supernatant, the exosome pellets were centrifuged for another 10 minutes at 1500 g to discard the extra liquid. Finally, the exosomes were conserved in PBS.

| Characterization of exosomes
The morphology of exosome was observed by transmission electron microscopy. Briefly, exosomes were fixed by 1% glutaraldehyde and incubated at 4°C. Next, 10 μL of the medium was placed onto formvar/carboncoated copper grids, followed by dyeing with 3% aqueous phosphotungstic aid for 35 seconds. Subsequently, exosomes were observed with a transmission electron microscopy (Tecnai 12; Philips, Amsterdam, Netherlands). Size distribution of exosomes was analysed by NanoSight LM10 system which was equipped with a fast video capture and particle-tracking software (NanoSight, Amesbury, UK). Western blot analysis was performed to detect exosome markers CD63 and CD81.

| Exosomes and miR-500a-3p internalization assays
Exosomes were labelled with PKH-67 green fluorescent Cell Linker Kit (Sigma-Aldrich, USA) according to the manufacturer's protocol. The labelled exosomes were co-cultured with MGC803 cells for 30 hours at 37°C. For the transfer of exosomal miR-500a-3p, PKH-67 labelled miR-500a-3p was transfected to MGC803 cells by liposome 2000 (Invitrogen). The PKH-67-miR-301a-expressing MGC803 cells were grown on the 0.4 mm pore size transwell (Thermo Fisher Scientific), and then co-cultured with MGC803 cells that had been grown on the cover slips in the bottom well of the transwell for 30 hours. The uptake of labelled exosomes or miR-500a-3p by the recipient MGC803 cells was observed using a Nikon Eclipse fluorescence microscope (Nikon, Tokyo, Japan).

| Immunofluorescence assay
Transfected or exosomes-treated GC cells were fixed in 4% paraformaldehyde for 10 minutes, blocked with PBS buffer containing 5% bovine serum albumin. Then, those cells incubated with antibodies at 4°C overnight, followed by incubation with fluorescein isothiocyanate (FITC)-conjugated secondary antibody and the nuclear counterstain diaminophenylindole (DAPI).
After rinsing, the cells were analysed using immunofluorescence microscopy.

| Western blot assay
Proteins were extracted with a lysis buffer and then quantified by a bicinchoninic acid protein assay. Equivalent amounts of cell lysates were separated using SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Roche Applied Sciences, Indianapolis, IA, USA). Membranes were immunoblotted overnight at 4°C with corresponding antibodies (Table S1). The bands were visualized using Pierce ECL Western Blotting Substrate (Thermo Fisher Scientific).
Image density of the immunoblotting was determined by Gel densitometry (Bio-Rad).

| RNA extraction and real-time qRT-PCR
Total RNA for cultured cells and exosomes were extracted with using Trizol Reagent (Takara Bio, Inc., Shiga, Japan). The mRNA expressions were detected by the PrimeScript RT Reagent Kit and SYBR Premix Ex Taq (Takara Bio, Inc.). GAPDH was used as control.
All the primers designed for qPCR were listed in Table S1. All-in-One microRNA qRT-PCR Detection Kits (GeneCopoeia, Inc., Rockville, MD, USA) were used to detect miRNA expression and U6 used as a control. Every experiment was repeated three times according to the manufacturer's protocol. Final data were analysed with the 2 −ΔΔC t method.
Each assay was repeated in three independent experiments.

FBXW7) and transfected to GC cells via Lipofectamine 2000
(Invitrogen; Thermo Fisher Scientific, Inc.) as the delivery agent, according to the manufacturer's protocol.

| Abdominal tumorigenicity assay in vivo
All animal experiments were conducted in accordance with the prin-

| Statistics
All in vitro experiments were repeated at least in triplicate. The data were represented as either a scatter plots or bar graphs with means ± standard error deviation of the mean (SEM). The statistical analysis was performed using SPSS software (version 13.0, New York, NY, USA). Statistical significance between two groups was determined using a two-tailed Student's t test. To compare multiple groups, one-way analysis of variance (ANOVA) followed by a Bonferroni-Dunn test was performed. The GC patients were divided into high expression group and low expression group according to the median ofmiR-500a-3p expression and Kaplan-Meier survival analysis was implemented to compare GC patient progression-free survival by logrank test. The receiver operating characteristic (ROC) curve was applied to determine the area under the curve (AUC) values for exosomal miR-500a-3p in plasma by the GraphPad Prism software (GraphPad Software, Inc.). P < 0.05 was considered statistically significant.

| DDP resistant GC cells exhibited higher tumorigenesis and CSCs properties
To explore the underlying molecular mechanism of GC DDP resist-

| MGC803/DDP-derived exosomes conferred DDP resistance and promote CSCs properties in recipient MGC803 cells
Recent studies indicated that exosomes derived from cancer cells were implicated in chemotherapy resistance. 19,20 We speculated Besides, the sphere formation capability ( Figure 2J) and CSCs properties ( Figure 2K) of MGC803 cells increased when co-culturing with exosomes from MGC803/DDP rather than MGC803. These results indicated that the exosomes isolated MGC803/DDP contributed to disseminate DDP resistance and promote CSCs properties in recipient MGC803 cells.

| MGC803/DDP-derived exosomes enhance DDP resistance of MGC803 recipient cells via exosomal miR-500a-3p delivery in vitro and in vivo
To identify DDP resistance associated miRNAs in GC cells, we compared the miRNAs expression profile between MGC803/ DDP and MGC803 cells by a miRNA microarray. The Figure S1 showed that a total of top 10 up-regulated and 10 down-regulated miRNAs between MGC803/DDP and MGC803 cells. Of these miRNAs, miR-500a-3p exhibited the most considerable degree of up-regulation. Then, we validated the expression of miR-500a-3p in MCG803, MGC803/DDP and their secreted exosomes.
To visualize miR-500a-3p transfer, MGC803 and MGC803/DDP cells transiently transfected with PHK67-tagged miR-500a-3p were co-cultured with MGC803 cells for 30 hours in a transwell system, as depicted in Figure 3H. As a result, the green fluorescently labelled miR-500a-3p was observed in the lower chamber cells through confocal microscopy ( Figure 3H), further suggesting that miR-500a-3p could be transferred by exosomes. In abdominal tumorigenesis model, MGC803/DDP exosomes promoted tumour growth and dissemination under DDP therapy but down-regulating miR-500a-3p in MGC803/DDP exosomes could abolished its tumour promoting effect ( Figure 3I-K). These findings revealed that functional exosomal miR-500a-3p from DDP resistant GC cells could be transferred to recipient ones, which subsequently became resistant to DDP in vivo and in vitro.

| FBXW7 reversed the DDP resistance of exosomal miR-500a-3p by inhibiting CSCs properties
To study the mechanisms of FBXW7 in abrogating the DDP resistance induced by miR-500a-3p, we investigated the CSCs properties in GC. In sphere formation assay, MGC803/DDP exosomes induced more number and size of sphere formation were abrogated by FBXW7 overexpression (Figure 5A,B). Additionally, up-regulation of cell stemness markers CD133, CD44 and SOX2 by MGC803/DDP exosomes could be inhibited by reintroduction of FBXW7 ( Figure 5C-F).
Parallel results were also observed in MKN45 cells ( Figure S2). These above data demonstrated that exosomal miR-500a-3p/ FBXW7 axis enhances DDP resistance in GC cells by CSCs properties activation.

| Plasma exosomal miR-500a-3p is related to DDP resistance in III stage GC patients
Clinically, we investigated the miR-500a-3p level in plasma exosomes and GC tissues of III stage GC patients who would receive DDPbased chemotherapy. As presented in Figure 6A and Figure S3A, the miR-500a-3p levels were significantly higher in both plasma exosomes and GC tissues from DDP resistant patients than in those from DDP sensitive patients, while the FBXW7 expression was lower in DDP resistant patients compared with sensitive patients.
These results further supported that FBXW7 was the downstream target of miR-500a-3p. Moreover, Kaplan-Meier analysis revealed that high expression of exosomal miR-500a-3p levels in III stage GC patient plasma was negatively correlated with prognosis ( Figure 6B). F I G U R E 4 MGC803/DDP-derived exosomal miR-500a-3p confers DDP resistance in recipient MGC803 cells via inhibiting FBXW7. A, Sequences of miR-500a-3p and the potential miR-500a-3p-binding sites at the 3′UTR of FBXW7, including nucleotides mutated in FBXW7-3′-UTR. Seed sequences are marked. Also shown effects of Blank, mimic or anti-miR-500a-3p and corresponding control on the luciferase activity of FBXW7 3′UTR-Wild-Type and FBXW7 3′UTR-Mutant by dual-luciferase reported assay in 293T cells. B-D, Expression of FBXW7 in MGC803 or FBXW7 overexpressed MGC803 cells transfected with mimic-control or mimic-miR-500a-3p by real-time qRT-PCR (B) and Western blot analysis (D). Expression of FBXW7 in MGC803 or FBXW7 overexpressed MGC803 cells co-cultured with CM without exosomes or MGC803/DDP exosomes by real-time qRT-PCR (C) and Western blot analysis (D). E, Proliferation assay analysis and IC 50 value in MGC803 or FBXW7 overexpressed MGC803 cells transfected with mimic-control or mimic-miR-500a-3p. F, Proliferation assay analysis and IC 50 value in MGC803 or FBXW7 overexpressed MGC803 cells co-cultured with CM without exosomes or MGC803/DDP exosomes. G, Representative bioluminescent images and microscopy observations showed effects of CM without exosomes, MGC803/DDP exosomes on abdominal tumorigenesis by MGC803 or FBXW7 overexpressed MGC803 cells abdominal tumorigenesis under DDP treatment. H, J, Comparison of bioluminescent signals and abdominal metastatic nodules after indicated treatment, five nude mice in each group. *P < 0.05 and **P < 0.01 Importantly, receiver operating characteristic (ROC) curve analysis demonstrated that the ability to discriminate between the resistant and sensitive group with the plasma exosomal miR-500a-3p level was acceptably accurate (AUC = 0.843, Figure 6C). Above all, the plasma exosomal miR-500a-3p might be applied as the non-invasive biomarker for DDP resistance in GC.

| D ISCUSS I ON
In spite of DDP-based chemotherapy is still effective in a large number of malignancies, the emerge of DDP resistance is still an unavoidable difficulty for cancer patients, 21,22 especially in GC. [23][24][25] Clinically, the overall 5-year survival rate for GC patients who received DDP-based chemotherapy after surgery remains dismal, while for late-stage cases, DDP has shown little benefits because of dissatisfactory treatment efficiency, resulting in tumour progression and reduced prognosis. 5 Therefore, figuring out the molecular mechanisms underlying DDP resistance may be of great assistance for improving GC patient outcome. 4,26,27 In current study, the effects and mechanism of exosomal miR-500a-3p in DDP resistance were explored in GC cell. Our data suggested that miR-500a-3p abundance was elevated in DDP resistant GC cells and their secreted exosomes.
Moreover, we found that exosomal miR-500a-3p could contribute to DDP resistance in recipient GC cells by down-regulating FBXW7 expression via enhancing stemness cells properties.
There have been several reports showing that chemotherapy is capable to stimulate cancer cells to release more exosomes. 28 could release more exosomes than their parental ones. The higher exosomes release induced by chemotherapy is probably because of the cellular stress and damage resulted from chemotherapy. This process is resembling to how cells release damage-associated molecular patterns (DAMPs). 34,35 Recently, miRNAs have been reported to be encapsulated in tumour-derived exosomes to avoid degradation and subsequently those exosomal miRNAs would transfer to recipient cells to regulate genes expression, including angiogenesis, invasion and F I G U R E 5 FBXW7 reverse the DDP resistance of exosomal miR-500a-3p by inhibiting CSCs properties in MGC803 cells. A, B, Effect of CM without exosomes, MGC803/DDP exosomes on sphere formation in MGC803 or FBXW7 overexpressed MGC803 cells. C, Relative mRNA expression of stemness markers CD133, CD44 and SOX2 in MGC803 or FBXW7 overexpressed MGC803 cells co-cultured with CM without exosomes or MGC803/DDP exosomes. D-F, Expression level of stemness markers in indicated GC cells by confocal microscopy (D) and Western blot (E, F). *P < 0.05, **P < 0.01 and ***P < 0.001 metastasis. 36,37 While for chemoresistance, those cells may release exosomal miRNAs into the extracellular environment and induce drug resistance to surrounding cells. 38,39 Exosomal miR-196a derived from cancer-associated fibroblasts result in head and neck cancer resistance to DDP. 38 Exosomal miR-126a have been reported to be involved in the doxorubicin resistance of lung cancer. 19 Our results proved that exosomes from DDP resistant GC cells enhance recipient cells resistance to DDP by miR-500a-3p/FBXW7 pathway in vitro and in vivo.
MiR-500a-3p has been reported to be involved in the chemoresistance, invasion and migration via GSK-3β and LY6K in different types of cancers. [40][41][42] In this study, we found that miR-500a-3p was elevated in exosomes from DDP resistant GC cells and clinical up-regulation of miR-500a-3p in exosomes from III stage GC patients' plasma correlated with DDP-based chemoresistance and GC progression, which might be used as a non-invasive predictor of chemotherapy in GC Patients. Furthermore, FBXW7 was identified as the target of miR-500a-3p in GC. FBXW7 (F-box with 7 tandem WD40) is one of the crucial components of ubiquitin ligase that aids in the degradation of many oncoproteins via the ubiquitin-proteasome system. FBXW7 is regarded as a potent tumour suppressor in different human cancers, as most of its target substrates can function as potential growth promoters. 43 For instance, FBXW7 inactivation sensitized cancer cells to radiation or etoposide by stabilizing p53 to induce cell-cycle arrest and apoptosis. 44 While in GC, low expression of FBXW7 was observed in primary GC and contributed to the poor survival and minimal response to adjuvant therapy. 45 Down-regulation of FBXW7 by miR-223 in GC cells promote proliferation, invasion and chemoresistance to trastuzumab in vitro. 46,47 Here, we found that overexpression of FBXW7 suppressed exosomal miR-500a-3p induced CSCs properties and thus reversed exosome mediated DDP resistance in GC.
In conclusion, we provide evidence that DDP resistance GG cells can secret miR-500a-3p enriched exosomes to promote stemness and DDP resistance by targeting FBXW7 in GC cells ( Figure 6D).
Moreover, exosomal miR-500a-3p is up-regulated in the plasma of F I G U R E 6 Plasma exosomal miR-500a-3p is related to DDP resistance in III stage GC patients. A, Plasma exosomal miR-500a-3p level was detected in III stage GC patients responding or not responding to DDP treatment by real-time qRT-PCR. B, Kaplan-Meier analysis of 3 y progression-free survival rate in III stage GC patients from high (n = 30) and low (n = 25) miR-500a-3p groups, according to the median exosomal miR-500a-3p level in pre-gastrectomy plasma. C, ROC curve analysis of plasma exosomal miR-500a-3p expression for discriminating the DDP resistant group (n = 30) from the sensitive group (n = 25). AUC, area under the curve. D, Schematic diagram of the potential roles of exosomal miR-500a-3p in GC DDP resistance. Briefly, the DDP-resistant GC cell-secreted exosomes containing miR-500a-3p can be taken up by surrounding DDP sensitive GC cells and subsequently down-regulate FBXW7 to enhance DDP resistance as well as CSCs properties of the recipient GC cells. ***P < 0.001 GC patients with DDP resistance, which thereby results in poor progression-free prognosis. We assume that inhibiting exosomal miR-500a-3p could be used as a potential modality for the prediction and treatment of GC with chemoresistance.

CO N FLI C T O F I NTE R E S T
The authors declare they have no competing interests.

E TH I C A L A PPROVA L
The study was approved by the medical ethics committee of XuZhou Central Hospital.

CO N S E NT FO R PU B LI C ATI O N
We have received consent from individual patients who have participated in this study. The consent forms will be provided upon request.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support findings of the study are available from the corresponding author upon request.