Helicobacter pylori‐induced exosomal MET educates tumour‐associated macrophages to promote gastric cancer progression

Abstract Helicobacter pylori (H. pylori) infection triggers chronic inflammation that has been associated with gastric cancer (GC). Exosomes are small extracellular vesicles that have become the key mediators of intercellular communication. In this study, we investigated exosome‐mediated communication between H. pylori‐infected GC cells and macrophages, focusing on the transfer of activated mesenchymal‐epithelial transition factor (MET). We observed a significant decrease in MET protein expression in GC cells after infection with H. pylori, whereas MET mRNA levels remained unchanged. Intriguingly, MET expression, specifically the phosphorylated active form, was increased in exosomes released from H. pylori‐infected GC cells. Confocal microscopy and Western blotting analyses showed that these exosomes containing MET were delivered to and internalized by macrophages. Indeed, in human GC tissues positive for H. pylori, we also observed that activated MET was highly expressed in tumour‐infiltrating macrophages. After internalization, exosomal MET then appeared to educate the macrophages towards a pro‐tumorigenesis phenotype. This included exosomal MET‐mediated stimulation of proinflammatory cytokine secretion IL‐1β, which subsequently promoted tumour growth and progression in vitro and in vivo. Taken together, these data were the first to demonstrate H. pylori infection‐induced upregulation of activated MET in exosomes and the pro‐tumorigenic effect on tumour‐associated macrophages.


| INTRODUCTION
Gastric cancer (GC) is the fourth most common malignancy and the third leading cause of cancer mortality worldwide. 1 Helicobacter pylori has been considered a major risk factor for the development of GC 2 and affects almost 80% of GC patients. 3 Infection with H. pylori induces an inflammatory response and aberrant activation of immune cells, which contributed to GC pathogenesis. 4 The cell-surface receptor tyrosine kinase mesenchymal-epithelial transition factor (MET) plays a critical role in tumour development, invasion and angiogenesis in solid tumour malignancies. 5 MET is activated via phosphorylation at Tyr1234/1235, 6 and this activated form functions as a key scaffolding protein in multiple intracellular signalling pathways. 7 Notably, the H. pylori effector protein CagA intracellularly targets the MET receptor, resulting in robust MET phosphorylation, activation of downstream cellular processes and a forceful motogenic response (cell scattering). 8 Intriguingly, MET can be packed in exosomes secreted from melanoma cells and subsequently educated bone marrow progenitor cells towards a prometastatic phenotype. 9 Exosomes are 40-150 nm bilayer membrane vesicles that have recently been recognized as important mediators of intercellular communication, as they contain a wide range of functional lipids, proteins, RNA and DNA that can be transferred to a recipient cell via fusion of the exosome with the target cell membrane. 10,11 It has been shown that EGFR in the exosomes secreted by GC cells regulates liver microenvironment and facilitates the metastasis of GC cells to liver. 12 These tumour-derived exosomes (TEXs) are also known to be involved in the recruitment of neutrophils and the activation of macrophages, 13 and influence the antitumour activity of immune cells through transferring suppressive or activating molecular signals. 14 In this study, we investigated the exosome-mediated communication between H. pylori -infected GC cells and macrophage, focusing on the transfer of activated MET. Furthermore, the downstream effects of this communication on proinflammatory factors IL-1β were also evaluated. To our knowledge, this is the first time the mechanism by which H. pylori infection reshapes the immune microenvironment and contributes to the progression of GC has been evaluated.

| Cell culture and co-culture with H. pylori
The gastric cancer cell lines AGS, MGC-803 and SGC-7901 were purchased from the cell bank of the Chinese Academy of Sciences

| Peripheral blood mononuclear cells (PBMCs) isolation and macrophages differentiation
Peripheral blood mononuclear cells were isolated from buffy coats by Ficoll-Hypaque density-gradient centrifugation. The cells were gently incubated in red blood cell lysis buffer (Sigma-Aldrich) for 2 minutes and washed with PBS (pH 7.4). Subsequently, PBMCs suspended in serum-free RPMI 1640 medium (Gibco) supplemented with 1% penicillin/streptomycin were seeded in a six-well plate for 1 hour in a humidified incubator containing 5% CO 2 at 37°C to allow monocyte adhesion. Nonadherent cells were removed and the adherent monocytes were further incubated in RPMI 1640 medium supplemented with 10% (v/v) heat-inactivated human serum and 1% penicillin/streptomycin for 7 days and media replacement every 3 days to obtain matured macrophages. 17

| Exosome isolation and labelling
Exosomes were isolated from the cell culture media with Total Exosome Isolation Reagent according to the manufacturer's instructions (Thermo Scientific, #4478359). The concentration of exosomal proteins was quantified with a BCA Protein Assay Kit (Thermo Scientific).
The purified exosomes were then labelled with the green fluorescent linker PKH67 (Sigma) according to the manufacturer's guidelines.

| EdU staining
DNA synthesis was analysed using a Cell-Light EdU Apollo488 In Vitro Imaging Kit (RiboBio Co., Ltd, Guangzhou, China) per the manufacturer's instructions.

| Colony formation assay
A total of 200 cells were seeded in 6-well plates and treated with macrophage supernatant as described previously. 18 The cells were then incubated for approximately 12 days until most of the colonies contained more than 50 cells. The colonies were then fixed with methanol and dyed with Giemsa solution. Clone formation efficiency was calculated as the (number of colonies/number of cells inoculated) × 100%.

| Scratch assay
A scratch assay was performed to assess cell migration in vitro. First, cells cultured in macrophage supernatant were seeded in 6-well plates until a confluent monolayer was formed. Then, upon confluence, cells were scratched with a 10 μL sterile pipette tip. Pictures were then taken of the scratch at different time-points under the microscope. The cell migration rate was calculated as (width at 0 hours-width at different time-points)/width at 0 hours.

| Immunofluorescence
A total of two samples were collected from GC patients with H. pylori infection or patients with H. pylori-associated chronic gastritis at the Second Affiliated Hospital of Nanjing Medical University.
Formalin-fixed and paraffin-embedded samples were cut into 5 μmol/ L sections, which were then processed for immunofluorescence. Primary antibodies used for immunoblotting were polyclonal rabbit anti-human-p-cMet (Cell Signaling, cat#3077) and mouse anti-human CD206 (Thermo Scientific, cat#53-2069-41). Tissues were then stained with anti-rabbit Alexa Fluor ® 568 conjugated secondary antibody (Thermo Scientific, cat#A10042) at room temperature for 2 hours as well as 4′, 6-diamidino-2-phenylindole (DAPI) to stain the nucleus. This study was approved by the Ethics Committee of the Second Affiliated Hospital of Nanjing Medical University.

| RNA isolation and qPCR
Total RNA was isolated from cells or mouse tissues using Trizol reagent (Invitrogen), following the manufacturer's instructions. The

| Cytokine secretion measurement
Secreted human IL-1β and IL-6 in culture supernatants were quantified using ELISA kits (Thermo Scientific) according to the manufacturer's instructions. LY294002 (phosphatidylinositol 3-kinase, PI3K inhibitor) and U0126 (Erk inhibitor) were obtained from Beyotime Biotechnology (Shanghai, China). Experiments were performed in triplicate. Tumour volume was calculated as width × length × (width + length)/ 2. The mice were killed 28 days after injection, and the tumours were removed. All of the animal studies performed were approved by the Nanjing Medical University Ethics Review Board. were then co-cultured with H. pylori for 12 hours. The exosomes were isolated from the supernatant of the shRNA-treated AGS cells to stimulate macrophages for 48 hours. Finally, the supernatant from the stimulated macrophages was used to treat GC cells. We then evaluated the proliferation, migration and invasion of these cells.

| Statistical analysis
All of the results reported here are representative of at least three independent experiments. Statistical evaluations were made using Student's t-tests (two-tailed). The data are presented as the means ± standard deviation (SD). P-values < 0.05 were considered statistically significant.

H. pylori-infected AGS cells
As MET activation has been observed in H. pylori-induced gastric tumorigenesis, 19 we investigated the regulatory role of H. pylori infection on MET expression. Our Western blot analysis showed that H. pylori infection significantly reduced MET protein abundance in a timedependent manner ( Figure 1A), whereas MET mRNA levels were unchanged ( Figure 1B). We also observed a significant increase in Rab27b mRNA in AGS cells co-cultured with H. pylori ( Figure 1C). As Rab27b plays an important role in exosomes biogenesis, 20 the exosomes were then isolated from the conditioned media of AGS cells infected with H. pylori for 24 hours. According to our TEM analysis, the purified exosomes appeared to be rounded particles ranging from 40 to 150 nm in diameter ( Figure 1D). We further confirmed the presence of CD63, CD81 and TSG101, three specific exosome markers, 21 and the absence of tubulin, in these AGS cell-derived exosomes ( Figure 1E).
Notably, H. pylori infection induced a time-dependent increase in MET expression in the exosomes. In contrast, MET protein was barely detectable in exosomes from uninfected AGS cells ( Figure 1F).
Phosphorylation at Tyr1234/1235 in the MET kinase domain is critical for kinase activation. 6 Furthermore, our data showed that active p-MET was also present in the exosomes from H. pylori-infected AGS cells (Figure 1F). These results indicated that activated MET could be incorporated into exosomes during GC upon H. pylori infection.

| Exosomes transferred MET into macrophages
Once secreted, exosomes deliver biological information to neighbouring or distant cells, thus modulating communication between tumour cells and the surrounding microenvironment. 22

| Activated MET is preferentially expressed by tumour-infiltrating macrophages
We further evaluated the activated MET expression in macrophages present in human GC tissue positive for H. pylori. Interestingly, using the macrophage marker CD206, 23 we found that macrophages were predominantly enriched in the GC tumour area. Furthermore, activated MET was highly expressed in these macrophages (Figure 3), supporting our theory that H. pylori infection modulates MET transfer to macrophages.
In addition, immunofluorescence analysis shows that the number of CD206 + macrophages co-expressing p-MET was less in the tissues from the patients with H.pylori-associated chronic gastritis than that in the patients with GC (Figure 3), indicating a potential role of MET in maintaining the tumorigenic function of the infiltrating macrophages.

| Exosomal MET mediated the pro-tumorigenic effects of macrophages
To confirm that MET released from exosomes affects macrophages, we used shRNA to silence MET expression in AGS cells (please see   supernatant from exosome-treated macrophages, we performed qPCR to detect the mRNA levels of TNF-α, VEGF, IL-6 and IL-1β. 24,25 Of these genes, IL-6 and IL-1β were significantly increased in THP-1-derived macrophages treated with MET + exosomes compared to macrophages treated with PBS ( Figure 5B), whereas only the mRNA level of IL-1β was observed dramatically enhanced in PBMCderived macrophages treated by MET + exosomes ( Figure 5D). To evaluate protein secretion into the supernatant, an ELISA was conducted and the results show that IL-1β protein expression was greatly elevated in the supernatant of THP-1-and PBMC-derived macrophages treated with MET + exosomes ( Figure 5E,G). Next, we seek to investigate whether the increase levels of IL-1β from macrophages stimulated by MET + exosomes was dependent on PI3K-Akt or mitogen-activated protein kinase (MAPK) pathway. As shown in Figure 5F,H, single treatment with LY294002 (PI3K-Akt inhibitor) or U0126 (Erk inhibitor) did not cause significant changes on the mRNA and protein levels of IL-1β induced by MET + exosomes. However, the combination of LY294002 and U0126 successfully abolished MET + exosomes-mediated elevation of IL-1β levels from THP-1 and PBMCs-derived macrophages ( Figure 5F,H), which suggest that MET + exosomes stimulated IL-1β secretion from macrophages via the Akt and MAPK pathway. As shown in Figure 6D, the supernatant from the macrophages trea-  Gastric cancer contains abundant tumour-supportive macrophages to promote malignant progression. 28 Macrophages infiltration is correlated with poor prognosis of gastric cancer patients. Accumulating evidence has shown that tumours can interfere with the immune system via secreted exosomes. For example, tumour-derived exosomes were observed to transfer activated EGFR to host macrophages, thereby suppressing innate antiviral immunity. 29 In the present study, we clearly demonstrated that macrophages internalized exosomal MET at a high efficiency. In addition, the expression of MET was higher in tumour-associated macrophages in the H. pyloriinfected human GC tissues. We, therefore, focused our attention on determining the effects of macrophages treated with exosomes containing activated MET on GC malignancy and tumour progression.

| DISCUSSION
Notably, macrophages treated with MET + exosomes markedly promoted GC growth and invasion both in vivo and in vitro. In contrast, macrophages incubated with PBS or MET − exosomes did not significantly alter GC progression. Therefore, our data suggest that exosomal MET may be a potential regulator of the pro-tumorigenic effect of macrophages in GC pathogenesis.
To better understand the mechanisms underlying the observed exosomal MET-mediated effects on macrophage function and, subsequently, GC progression, we focused on changes in inflammatory mediators known to play a role in the tumour microenvironment.
Chronic inflammation is recognized as a tumour hallmark and is implicated in nearly all stages of tumorigenesis. 30 In this study, we found that exosome-treated macrophages preferentially secrete the proinflammatory cytokines IL-1β and activated the Akt and MAPK pathways. IL-1β plays a pivotal role in proliferation, migration and invasion in malignant tumours. 31,32 In fact, IL-1β is suspected to be CHE ET AL. survival. While additional work is required to test this, the present study defines the role of exosomal MET in the activation of macrophages and highlights this mechanism as a potential drug target.