Melanoma cell-secreted exosomal miR-155-5p induce proangiogenic switch of cancer-associated fibroblasts via SOCS1/JAK2/STAT3 signaling pathway

Background Cancer-associated fibroblasts (CAFs) have been widely reported to promote tumor angiogenesis. However, the underlying mechanisms of the proangiogenic switch of CAFs remain poorly understood. This study aims to clarify the mechanisms underlying the proangiogenic switch of CAFs. Methods NIH/3T3 cells were treated with B16 and B16F10-derived exosomes. Then the CAFs markers and proangiogenic factors were detected by RT-PCR and Western blot. CCK-8 assay, transwell migration assay, tube formation assay, and in vivo Matrigel plug assay were conducted to determine the proangiogenic capability of CAFs. Western blot and AG490 were used to investigate the role of Janus kinase 2/signal transducer and activator of transcription 3 (JAK2/STAT3) signaling pathway in the proangiogenic switch of CAFs. Bioinformatics analysis, luciferase reporter assay, microRNA mimic and inhibitor, and xenograft models were used to investigate the role of mmu-miR-155-5p (miR-155) in the proangiogenic switch of CAFs. Results In this study, we show that melanoma cell-secreted exosomes can induce reprogramming of fibroblasts into CAFs and that exosomal miR-155 can trigger the proangiogenic switch of CAFs. Mechanistically exosomal miR-155 can be delivered into fibroblasts and promote the expression of proangiogenic factors, including vascular endothelial growth factor A (VEGFa), fibroblast growth factor 2 (FGF2), and matrix metalloproteinase 9 (MMP9), by directly targeting suppressor of cytokine signaling 1 (SOCS1). Downregulation of SOCS1 activates JAK2/STAT3 signaling pathway and elevates the expression levels of VEGFa, FGF2, and MMP9 in fibroblasts. Treatment with exosomes containing overexpressed miR-155 can promote angiogenesis, and the reduction of miR-155 in melanoma cell-secreted exosomes alleviates angiogenesis in vitro and in vivo. Conclusions These results demonstrate that by promoting the expression of proangiogenic factors in recipient fibroblasts via SOCS1/JAK2/STAT3 signaling pathway, melanoma cell-secreted exosomal miR-155 can induce the proangiogenic switch of CAFs. Although tumor angiogenesis is modulated by various factors, exosomal miR-155 may be a potential target for controlling melanoma angiogenesis and used to set up novel strategies to treat melanoma. Electronic supplementary material The online version of this article (10.1186/s13046-018-0911-3) contains supplementary material, which is available to authorized users.


Background
Melanoma is a highly vascularized tumor. As several anti-angiogenic drugs have been approved to treat malignant tumors, the utility of anti-angiogenic strategies in treating melanoma has been confirmed [1]. However, recent studies and clinical trials have demonstrated the complexity of drug resistance to anti-angiogenic therapies in treatment of melanoma [2], driving the pressing demand for thorough investigation of the underlying mechanisms of melanoma angiogenesis.
Cancer-associated fibroblasts (CAFs), the activated form of tissue-resident fibroblasts, can promote tumor angiogenesis by secreting several proangiogenic cytokines, such as vascular endothelial growth factor A (VEGFa), fibroblast growth factor 2 (FGF2) and proteolytic enzymes, such as matrix metalloproteinases (MMPs) [3,4]. However, the process of how tumor cells reprogram normal fibroblasts to proangiogenic CAFs remains incompletely understood.
Exosomes are small cell-released and lipid-bilayerenclosed vesicles containing various bioactive proteins, mRNAs, and microRNAs (miRNAs). It serves as critical mediators in intercellular communication by transferring functional cargos to recipient cells [5]. Our previous study has shown that melanoma cell-secreted microvesicles can mediate the transformation of normal fibroblasts to CAFs and regulate the expression of vascular cell adhesion molecule-1, resulting in enhanced adhesion of melanoma cells and fibroblasts [6]. Tumorreleased exosomal miRNAs have been shown to play a crucial role in reprogramming the tumor microenvironment [7]. Although various functions of tumor-secreted exosomal miRNAs have been well disclosed, the role of these miRNAs in the proangiogenic switch of CAFs remains poorly understood.
The Janus kinase 2/signal transducer and activator of transcription 3 (JAK2/STAT3) signaling pathway is activated in numerous types of tumors and regulates cell proliferation, angiogenesis, and migration of tumor cells. The activation of JAK2 protein triggers the phosphorylation of STAT3. The phosphorylated STAT3 dimerizes and translocates to the nucleus and then binds to targeted DNA elements and activates specific gene translation [8]. Studies have proved that the JAK2/STAT3 signaling pathway regulates the expression of proangiogenic factors, such as VEGFa and FGF2, and proteolytic enzymes, such as MMP9, and mediates numerous aspects of angiogenesis [9][10][11]. The suppressor of cytokine signaling (SOCS) proteins suppress JAK kinase capability and bind to the receptor to block STAT interaction. In particular, SOCS1 is a potent inhibitor of JAK2/STAT3 signaling cascade. The expression of SOCS1 reduces in various human cancers and is tightly associated with tumor angiogenesis [12,13]. However, whether SOCS1 and JAK2/STAT3 pathway participate in the proangiogenic switch of CAFs and whether tumor-secreted exosomal miRNAs regulate both regulators are unclear.
In this study, we demonstrate that highly metastatic (B16F10) and weakly metastatic (B16) melanoma cell lines release and use exosomes to transfer mmu-miR-155-5p (miR-155) in fibroblasts. These exosomes induce CAF activation and elevate the expressions of proangiogenic factors (VEGFa, FGF2, and MMP9) in CAFs. Exosomal miR-155 directly targets SOCS1 and then activates the JAK2/STAT3 signaling pathway, leading to the proangiogenic switch of CAFs. These results may provide a novel therapeutic target for the anti-angiogenic therapy of melanoma treatment.  [14]. All procedures were in accordance with the Ethics Committee of School and Hospital of Stomatology, Wuhan University. STR was performed routinely on these cell lines to confirm their authenticity and Mycoplasma was routinely tested. NIH/3T3, B16F10, A375, HGF and MS-1 cells were cultivated in high-glucose Dulbecco's Modified Eagle's Medium (DMEM) (Hyclone, UT, USA) containing 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA). B16 cells were cultivated in Roswell Park Memorial Institute (RPMI) 1640 medium (Hyclone, UT, USA) containing 10% FBS (Gibco, Carlsbad, CA, USA). All cell lines were cultivated at 37°C in 5% CO 2 . When NIH/3T3 and A375 cells reached 70-80% confluence, melanoma-derived exosomes were added to the medium at 20 μg/mL.

Isolation and analysis of exosomes
For exosome isolation, A375, B16 and B16F10 cells at 80% confluence were washed thrice with phosphate buffer solution (PBS) and then cultivated with growth medium containing 10% extracellular vesicles (EVs)-depleted FBS (prepared by overnight ultracentrifugation of medium-diluted FBS at 100,000 g at 4°C). After 48 h, the conditioned medium (CM) was collected and pre-cleared by centrifugation at 800 g for 15 min and then at 10,000 g for 30 min. Exosomes were isolated by ultracentrifugation at 110,000 g for 70 min and washed in PBS by using the same ultracentrifugation conditions. Ultracentrifugation experiments were conducted with Beckman Optima L-100XP (Beckman Coulter, USA). Exosomes were observed by transmission electron microscopy HT7700 (HITACHI, Japan). The hydrodynamic diameter of exosomes was measured by using Nano-ZS ZEN 3600 (Malvern Instruments, UK).

Exosome tracing
To monitor the interaction between exosomes and fibroblasts, the exosomes were labeled with PKH26 (Sigma-Aldrich, St. Louis, MO). After incubation with PKH26-labeled exosomes for 4 h, NIH/3T3 and HGF cells were observed by using a confocal microscope (Olympus FV1200, Japan).

RNA extraction, RT-PCR and quantitative real-time PCR (qPCR)
Total RNA was extracted from cells and exosomes by using the TRIzol reagent (Takara, Tokyo, Japan), as instructed by the manufacturer. To analyze the expression levels of protein coding genes, RNA was reversely transcribed into cDNA by using PrimeScript RT Reagent Kit (Takara, Tokyo, Japan), followed by qPCR using SYBR® Premix Ex Taq™ II (Takara, Tokyo, Japan). MiRNA primers were synthesized by Sangon Biotech (Shanghai, China). For the quantification of mature miRNAs and U6 by qPCR, RNA was reverse-transcribed by using miRNA first-strand cDNA synthesis (Sangon Biotech, Shanghai, China), and then quantified by using MicroRNA qPCR Kit (SYBR Green Method) (Sangon Biotech, Shanghai, China). All processes were performed by following the manufacturer's instructions. All RT-PCR tests were performed in triplicate. The comparative cycle threshold (Ct) method was used to quantify miRNA or mRNA levels by using U6 or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as the normalization control. QPCR was conducted on a QuantStudio™ 6 Flex (Life Technologies, USA). The Ct values should not differ by more than 0.5 among the triplicates. The sequences of the primers and the synthesized oligonucleotides used were listed (Additional file 1: Table S1).

Protein extraction and Western blot analysis
Nuclear/cytoplasmic fractionation was separated by using Cell Fractionation Kit (Cell Signaling Technology, USA) according to the manufacturer's instructions. The total protein of cells and exosomes was extracted by using M-PER (Pierce Inc., USA) supplemented with protease and phosphatase inhibitors on ice. The protein concentration of every sample was measured by using BCA Protein Assay Kit (Thermo Fisher Scientific Inc., USA). The mixture of the loading buffer (5×) and protein solutions was heated for 10 min at 95°C. Aliquots of 20 μg of protein were added to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis for 30 min at 60 V and for 1 h at 110 V. Afterward, the proteins were transferred to polyvinylidene difluoride (PVDF) membrane in transfer buffer for 2 h at 200 mA. The membranes were blocked with 5% skim milk in Tris-buffered saline containing 0.05% Tween 20 (TBST) at room temperature for 1 h. When phosphorylated protein was detected, the PVDF membrane was blocked with TBST with 5% BSA. Then, the membranes were incubated with antibodies overnight at 4°C. Subsequently, the bound antibodies were detected by horseradish peroxidase-conjugated, anti-mouse IgG or anti-rabbit IgG (Pierce Chemical, Rockford, IL, USA). Western blot analyses were repeated thrice to confirm the results.

Enzyme-linked immunosorbent assay (ELISA)
The concentrations of VEGFa, FGF2, and MMP9 in the culture medium of NIH/3T3 cells were measured by ELISA kits (4A Biotech Co., Ltd., China) following the manufacturer's instructions and analyzed by comparing the optical densities of the samples with the standard curve of the kits.

Cell proliferation assay
MS-1 cells were seeded in 96-well culture plates and cultivated by the CM of fibroblasts, which were stimulated with or without melanoma-secreted exosomes. After incubation for 24, 48, and 72 h, cell proliferation was determined by CCK-8 (Dojindo, Japan) in accordance with the manufacturer's instructions.

Migration and tube formation assays
For the migration assay, 5 × 10 4 MS-1 cells were seeded in 100 μl of serum-free medium in the upper chambers of 24-well plates with inserts (Corning, USA). Fibroblasts treated with or without melanoma-secreted exosomes were seeded in 600 μl of 10% FBS-DMEM in the lower chambers. After 24 h incubation, the cells in the upper chamber were removed, and cells that traversed to the reverse face were fixed and stained with crystal violet in accordance with the manufacturer's instructions. Six random fields were counted. Migration assays were performed in triplicate. For tube formation assay, the MS-1 cells (1 × 10 4 /well) were seeded in 48-well plates precoated with Matrigel (BD Biosciences, San Jose, CA, USA) and cultivated in the CM of fibroblasts stimulated with or without melanoma-secreted exosomes. After 6 h incubation, the capillary-like structure was observed and photographed under a microscope (BHS-313 Olympus). Quantification and analysis of tube formation results were performed using Image-Pro Plus 6.0 (Media Cybernetics, Inc., USA).

In vivo angiogenesis study
Fibroblasts (5 × 10 6 ) stimulated with B16 and B16F10derived exosomes and the untreated fibroblasts were mixed with Matrigel (200 μL) and subcutaneously injected into C57BL/6 mice (male, 8-week-old) (n = 5). In each mouse, three types of fibroblast-Matrigel mixture were injected into the armpits of the left upper, right upper, and right lower limbs. All mice were raised in sterile laminar flow cabinets under appropriate pathogenfree conditions with a 12 h-12 h light-dark cycle. After 1 week, the Matrigel plugs were harvested and processed for analysis. All mice were raised and manipulated following the protocols approved by the Laboratory Animal Care and Use Committee of Wuhan University (approval numbers: S07917110A).
B16 (4 × 10 6 ) and B16F10 (4 × 10 6 ) cells and different types of exosome-treated NIH/3T3 cells were harvested from culture plates, premixed with Matrigel (200 μL) at the ratio of 1:1, and subcutaneously inoculated into the armpits of the left upper, right upper, and right lower limbs of C57BL/6 mice (male, 8-week-old) ( Fig. 7a and g). After 1 week, xenografts were harvested. The xenografts were weighed and measured by using a caliper, measured in accordance with the formula: volume (cm 3 ) = (width 2 × length)/2, and then embedded in paraffin for immunofluorescence staining.

Immunofluorescence
The Matrigel plugs and xenografts were extracted and then fixed in 4% paraformaldehyde for 24 h, dehydrated using graded ethanol, embedded in paraffin, sectioned serially, and incubated with anti-mouse CD31 antibody (GB11063-3, Servicebio, Wuhan), and then with Cy3-conjugated secondary IgG (Servicebio, Wuhan). Analyses were performed with a fluorescent microscope (Biozero BZ-8000, Keyence, Osaka, Japan). Microvessel density (MVD) was quantified from six random microscopic fields. Any single cell or discrete cluster stained for CD31 was counted as one microvessel as reported previously [15].
After incubation with B16-and B16F10-secreted exosomes for 6 h, NIH/3T3 cells were fixed with 4% paraformaldehyde for 10 min, followed by incubation with 0.2% Triton X-100, and blocked with FBS for 30 min at room temperature. The slides were incubated with anti-P-STAT3 antibody overnight at 4°C, followed by incubation with FITC-conjugated goat anti-rabbit IgG antibody for 45 min at room temperature. Nuclear staining was then incubated with DAPI. The stained cells were examined and photographed using a confocal microscope (Olympus FV1200, Japan).

Luciferase reporter assay
Luciferase plasmids (300 ng) (pGL3) encoding wild-type or mutant 3′ untranslated region (3′UTR) of SOCS1 were co-transfected with miR-155 mimic or miR-NC and anti-miR-155 or anti-NC. Lipo6000™ (Beyotime, China) was used as the transfectant. At 48 h after transfection into 293 T cells, luciferase activity was measured by using a Dual-Luciferase Reporter Assay Kit (Promega, USA) in accordance with the manufacturer's instructions.

Statistical analysis
Results are expressed as means ± standard error of the mean (SEM) from triplicates of independent experiments and analyzed by student's t-tests or one-way ANOVA by using the Statistical Product and Service Solutions (SPSS) 21 software (SPSS Inc., Chicago, USA). Results were considered statistically significant when P < 0.05.

Identification of exosomes secreted by melanoma cells
Isolated from the culture supernatant of B16 and B16F10 cells, the EVs were identified by using transmission electron microscope and dynamic light scattering analysis (Fig. 1a, b). Western blot was performed to detect the exosomal markers Hsp90, Tsg101, and CD63 (Fig. 1c). Bilayer-enclosed morphology, diameters of 50-200 nm, and the existence of exosomal markers verified that the vesicles were exosomes.
To investigate the proangiogenic capability of fibroblasts treated with melanoma cell-secreted exosomes in vivo, NIH/3T3 cells treated with B16-and B16F10-secreted exosomes and the nontreated NIH/3T3 cells were injected subcutaneously into C57BL/6 mice. After 1 week, the Matrigel plugs were harvested and processed for Immunofluorescence staining. Then the microvessel density (MVD) was quantified. In comparison with the control group, the MVDs of B16-and B16F10-secreted exosome-treated groups increased significantly (Fig. 2h, i and j).
Melanoma cell-secreted exosomes suppress the expression of SOCS1 and activate the JAK2/STAT3 signaling pathway, which regulates the proangiogenic switch of CAFs After treatment of exosomes extracted from B16 and B16F10 cells, Western blot analysis revealed the significantly elevated phosphorylation levels of JAK2 and STAT3, and increased nuclear accumulation of P-STAT3 in NIH/3T3 cells (Fig. 3a, b). Besides, confocal microscope verified the enhanced nuclear accumulation of P-STAT3 in NIH/3T3 cells after treatment of B16-and B16F10-secreted exosomes (Additional file 1: Figure S3). To investigate whether the activation of JAK2/STAT3 signaling pathway resulted in upregulation of MMP9, VEGFa, and FGF2, JAK2 inhibitor AG490 was used to concurrently treat NIH/3T3 cells with melanoma cellsecreted exosomes. RT-PCR and Western blot showed that treatment with AG490 alleviated the promoting effect of B16-and B16F10-secreted exosomes on the phosphorylation of JAK2 and STAT3, and expressions of MMP9, VEGFa, and FGF2 (Fig. 3c, d and e). Treatment with AG490 also weakened the promoting effect of the CM of NIH/3T3 cells treated with B16-and B16F10-secreted exosomes on MS-1 cell proliferation (Fig. 3f, g), migration (Fig. 3h, i), and tube formation ( Fig. 3h and j). As an endocellular suppressor of the JAK/STAT signaling pathway, SOCS1 was significantly downregulated in NIH/3T3 cells after treatment of B16-and B16F10-secreted exosomes ( Fig. 3a; Additional file 1: Figure S4).

MiR-155 targets 3′UTR of SOCS1
After detailed sequence analysis, we observed that 3′UTR of SOCS1 exhibited a putative binding site of miR-155 at positions 20-27 (Fig. 5a). Dual-luciferase reporter assay revealed that co-transfection of miR-155 mimic significantly inhibited the activity of firefly luciferase reporter carrying the wild-type 3′UTR of SOCS1 (Fig. 5b), whereas co-transfection of anti-miR-155 significantly increased the activity of firefly luciferase reporter carrying the wild-type 3′UTR of SOCS1 (Fig. 5c). Furthermore, treatment with miR-155 mimic significantly suppressed the expression of SOCS1 in NIH/3T3 cells (Fig. 5d, e).

Discussion
In melanoma, EVs, such as exosomes and microvesicles, have been reported to play a predominant role in promoting pre-metastatic niche formation, proliferation, and metastasis [16]. For example, Dror et al. (2016) reported that melanosomal miR-211, which is secreted by melanoma cells and absorbed by fibroblasts, can activate the mitogen-activated protein kinase (MAPK) signaling pathway and induce fibroblast reprogramming into CAFs [17]. Our previous study showed that melanoma cell-secreted microvesicles can induce the transformation of normal fibroblasts into CAFs [6]. Here, our results indicate that melanoma cell-released exosomes can elevate α-SMA and FAP expression in NIH/ 3T3 cells, indicating that melanoma cell-released exosomes can also trigger normal fibroblast reprogramming into CAFs. Interestingly, exosomes secreted by Fig. 4 Delivery of melanoma cell-secreted exosomal miR-155 into fibroblasts. a MiRNAs that might target SOCS1 were predicted with TargetScan, miRanda, and PicTar. b The predicted candidate miRNAs in B16-and B16F10-secreted exosomes were investigated by qRT-PCR. ** P < 0.01 vs. B16-secreted exosomes. Student's t-tests. Independent experiments performed in triplicate. Values are expressed as means ± SEM. c Comparison of the miR-155 levels in NIH/3T3, B16, and B16F10 cells. d The levels of miR-155 in recipient NIH/3T3 cells after treatment with B16-and B16F10secreted exosomes. * P < 0.05, ** P < 0.01, *** P < 0.001. Independent experiments performed in triplicate. Values are expressed as means ± SEM. One-way ANOVA and student's t-tests. e Exosomes from B16F10 cells that were transfected with 40 nM FITC-tagged miR-155 were labeled with fluorescent dye PKH26 (left panel) and then applied to treat NIH/3T3 cells for 4 h, following by imaging under a fluorescence microscope. As negative controls, NIH/3T3 cells were incubated with non-labeled exosomes (exosomes were treated with PBS instead of PKH26, middle panel) or 40 nM naked FITC-tagged miR-155 (FITC-tagged miR-155 was added directly to the NIH/3T3 cells, right panel) for 4 h. FITC: FITC-tagged miR-155 (green), PKH26: PKH26-labeled exosomes (red), DAPI: cell nuclei (blue). Scale bar, 50 μm. DAPI: 4′6-diamidino-2-phenylindole. Exo: exosomes. FITC: fluorescein isothiocyanate. MiR-155: mmu-miR-155-5p. MiRNAs: microRNAs. SEM: standard error of the mean. SOCS1: suppressor of cytokine signaling 1 human melanoma cell line A375 can also induce HGF transforming into CAFs. These results collectively indicate that exosomes play a crucial role in CAF transformation. Yasushi Kojima et al. (2010) reported that the establishment of the self-sustaining TGF-β and SDF-1 autocrine-signaling loops initiate and maintain the differentiation of fibroblasts into tumor-promoting CAFs [18]. CAFs have been reported to promote tumor angiogenesis. In our study, CAFs exhibited upregulated expression and secretion of the angiogenic factors VEGFa, FGF2, and MMP9. The CM of CAFs promoted EC proliferation, migration, and tube formation. In the Matrigel plug assay, CAF groups exhibited high MVD. These results indicate that melanoma cell-secreted exosomes can induce and enhance the proangiogenic capability of CAFs in the tumor microenvironment. VEGFa and FGF2 are two potent proangiogenic factors. Binding to VEGFa and FGF2 results in the downstream activation of various signaling pathways in ECs; this condition can promote EC proliferation, survival, and migration. In melanoma patients, elevated serum levels of VEGF, FGF2, and other soluble proangiogenic factors have been demonstrated and are closely correlated with poor clinical outcome [19]. Increased VEGF and FGF2 expression and accumulation were identified in the tumor microenvironment. MMP9 can also release VEGF from the extracellular matrix (ECM) [20]. By secreting proangiogenic ECM remodeling enzymes such as MMP9, CAFs reconstruct the ECM to facilitate ECs to cross this structural barrier. Therefore, our results confirm that CAFs play an important role in tumor angiogenesis.
Accumulating evidence have shown the strong proangiogenic effects of JAK/STAT signaling pathway [21][22][23][24][25][26]. JAK2/STAT3 pathway has been reported to regulate several critical proangiogenic factors, such as VEGFa, MMP-2, MMP-9, insulin-like growth factor 1 (IGF-1), and FGF2 [11]. However, whether the JAK2/STAT3 signaling pathway participates in the proangiogenic switch of CAFs in melanoma microenvironment has not been reported. In our study, treatment with melanoma cell-secreted exosomes significantly elevated the phosphorylation levels of JAK2 and STAT3 in CAFs. When the JAK2/STAT3 signaling pathway was blocked by AG490, the expressions of VEGFa, FGF2, and MMP9 reduced significantly. In addition, the proangiogenic effect of the CM of CAFs was also remarkably suppressed by AG490. These results collectively indicate that the JAK2/STAT3 signaling pathway can regulate the expressions of key proangiogenic factors (VEGFa, FGF2, and MMP9) and plays a critical role in the proangiogenic switch of CAFs.
Several studies indicated SOCS1 as a tumor suppressor gene. Aberrant methylation of SOCS1 silences SOCS1 and activates the JAK2/STAT3 pathway in hepatocellular carcinoma cells (HCC), resulting in the development of HCC [27]. A previous report from Fengju Huang et al. (2008) indicated that SOCS1 expression is reduced in brain metastasis of melanoma. The loss of SOCS1 expression leads to the activation of the JAK2/STAT3 signaling pathway and overexpression of MMP-2, FGF2, and VEGF and enhanced invasion and angiogenesis of melanoma cells, consequently promoting brain metastasis [13]. Previous studies have focused on the relationship between SOCS and the JAK/STAT signaling pathway. However, the reason for decreased SOCS1 expression in tumor tissues remains incompletely understood. Here, we observed that melanoma cell-secreted exosomal miR-155 suppressed SOCS1 expression in CAFs. Suppression of SOCS1 in CAFs activated the JAK2/ STAT3 signaling pathway and then promoted the expressions of MMP-9, FGF2, and VEGFa, triggering the proangiogenic switch of CAFs.
MiR-155 have been reported to play an important role in promotion of inflammation [28,29], regulation of adipose tissue function [30], modulation of glucose homeostasis [31]. Exosome-delivered miR-155 can also be taken up by recipient cells and modulate several responses in these recipient cells [32]. Silencing of miR-155 by using antisense oligomers (antimiRs) as anti-cancer drug is an evolving therapeutic strategy [33]. Overexpression of miR-155 is identified in primary melanoma and increases in melanoma exhibiting regional progression [34][35][36]. Consistently, our study showed the elevated miR-155 expression in highly metastatic (B16F10) compared with weakly metastatic (B16) melanoma cell lines. A relatively high levels of miR-155 was detected in B16F10-released exosomes compared with exosomes extracted from B16 cells. Previous studies indicated that overexpression of miR-155 can promote melanoma cell proliferation and invasion [37]. However, the effects of overexpression of miR-155 on melanoma angiogenesis have not been reported. Our study showed that miR-155 can be transferred into CAFs by melanoma cell-secreted exosomes. Elevated levels of miR-155 in B16-secreted exosomes can suppress the expression of SOCS1; upregulate the phosphorylation levels of JAK2 and STAT3; enhance the expressions of MMP-9, FGF2, and VEGFa; and enhance the promotive effect of CAFs on EC proliferation, migration, and tube formation. Reversely, the decreased levels of miR-155 in B16F10-secreted exosomes exhibit upregulated expression of SOCS1; suppress the activation of JAK2/STAT3 signaling pathway; downregulate the expressions of MMP-9, FGF2, and VEGFa; and alleviate the promotive effect of CAFs on EC proliferation, migration, and tube formation. In vivo xenograft models, elevated levels of miR-155 in B16-secreted exosomes significantly increased the MVD of xenografts, but decreased levels of miR-155 in B16F10-secreted exosomes lowered the MVD of xenografts. Our results suggest that exosomal miR-155 plays an important role in the proangiogenic switch of CAFs via SOCS1/JAK2/ STAT3 signaling pathway. However, inhibiting miR-155 in melanoma cell-secreted exosomes cannot reduce the proangiogenic factor expression to the original level, suggesting that additional factors in melanoma cell-secreted exosomes can be involved in the proangiogenic switch of CAFs and require further investigation. Further investigations should determine whether treatment with melanoma cell-secreted exosomes promotes primary miR-155 to be processed via the intermediate precursor miR-155 to the functional mature miR-155 in CAFs. In our study, the xenografts with NIH/3T3 treated with exosomes containing elevated levels of miR-155 exhibited a large tumor volume and were heavy (Fig. 7c, d). Conversely, the xenografts with NIH/ 3T3 treated with exosomes containing lowered levels of miR-155 exhibited small tumor volume and were light (Fig. 7i, j). Therefore, additional studies should also examine whether miR-155 overexpression can enhance the effects of CAFs on melanoma proliferation and metastasis.

Conclusions
In conclusion, our study demonstrates that melanoma cell-secreted exosomal miR-155 can trigger normal (See figure on previous page.) Fig. 7 Exosomal miR-155 regulates angiogenesis in vivo. a, g The schematic drawing show the positions where the cells were injected. b, h The pictures show the isolated tumors. The tumors in the same positions in each group are from the same mouse. c, d, i, and j The tumor size and weight (means ± SEM) of different treatment groups. * P < 0.05, ** P < 0.01. e, f, k, and l Representative fluorescence microscopy images and the quantitative analysis of MVD of the xenografts. * P < 0.05. ** P < 0.01. Student's t-tests. Scale bar, 50 μm. Anti-NC: inhibitor negative control. Anti-miR-155: miR-155 inhibitor. DAPI: 4′6-diamidino-2-phenylindole. Exo: exosomes. MiR-NC: miR-negative control. MVD: microvessel density