MicroRNA-Mediated Antiproliferative Effects of M1 Macrophage-Derived Extracellular Vesicles on Melanoma Cells

ABSTRACT Introduction Research in tumor treatment has shown promising results using extracellular vesicles (EVs) derived from immune cells. EVs derived from M1 macrophages (proinflammatory), known as M1-EVs, have properties that suppress tumor growth, making them a promising treatment tool for immune susceptible tumors such as melanoma. Here, small unaltered M1-EVs (M1-sEVs) were employed in a 3D mouse melanoma model (melanospheres) to evaluate such activity. Methods Macrophages were polarized and EVs were isolated by ultracentrifugation. The EVs obtained were characterized based on size, with measurements performed by dynamic light scattering and electron microscopy, and the expression profiles of microRNAs were analyzed by microarray and PCR. Melanospheres were used to evaluate the cytotoxicity of M1-sEVs. Pondering a possible future transposition from the animal model to the human, human melanoma cells were transfected with a specific miRNA, and the impact on cell proliferation was evaluated. Results The isolated EVs showed a size distribution between 50–400 nm in diameter, but preeminently in a range of 70–90 nm. M1-sEVs demonstrated a remarkable ability to reduce cell proliferation and viability in the melanospheres, leading to a decrease in their volume. M1-sEVs contained unique miRNAs, including miR-29a-3p, which exhibited significant antitumor activities according to bioinformatics analysis. Validation of the antitumor effects of miR-29a-3p was obtained by a functional evaluation, i.e., by inducing miRNA overexpression in human melanoma cells (SK-MEL-28). Conclusion Although further research would be advisable, the study provides evidence supporting the potential of M1-sEVs and their miRNA load as a possible targeted immune therapy for melanoma. GRAPHICAL ABSTRACT


Introduction
Although malignant melanoma has been reported in only 1-5% of all skin cancers, it is the most aggressive and lethal form of them, accounting for up to 65% of deaths (Domingues et al., 2018).The high heterogeneity of melanoma explains the lack of effective therapies, which has led to treatment resistance.Promising therapeutic strategies have been developed recently, but the 10-year survival rate for malignant melanoma is still less than 10% (Marzagalli et al., 2019).It is known that melanoma is susceptible to immune modulation, and the tumor microenvironment may explain the susceptibility of melanoma cells to immune system activation (Leonardi et al., 2020).Therefore, it becomes essential to understand antitumor immune responses and apply immunotherapy as a primary investigative approach to new forms of melanoma treatment (Curti et al., 2021).
Macrophages are one of the most widely studied cell types in immunotherapy.These immune system cells occur in diverse tissues and play essential roles as immune effectors and regulators.External stimulus can induce macrophages to a coordinated gene expression change, altering the functional capacity of the cell to a pro or anti-inflammatory profile (Hao et al., 2012;Mosser & Edwards, 2008).In vitro, these phenotypes have been broadly characterized as either classically activated (M1) to the proinflammatory form or alternatively activated (M2) to the anti-inflammatory state based on surface receptors, gene signatures, and secretion of inflammatory mediators (Gordon & Martinez, 2010;Mantovani et al., 2007).Thus, macrophages may repolarize in response to changes in the local microenvironment to adapt to outside stimuli, demonstrating high plasticity (Hao et al., 2012).In the tumor microenvironment, M1 macrophages play an essential role in identifying and damaging tumor cells, and their presence usually indicates a good prognosis (Weagel et al., 2015).Tumor immunotherapy with macrophages aims to polarize macrophages toward a proinflammatory response (M1), stimulating macrophages and other immune cells to destroy the tumor.Many cytokines and bacterial compounds can achieve this effect in vitro, although side effects are usually too severe when replicated in vivo.For this reason, it would be beneficial to identify substances from M1 with minimal or easily manageable side effects (Xia et al., 2020).
M1 macrophages may communicate with the melanoma microenvironment via extracellular vesicles (EVs).EVs are defined as particles secreted for cells constituted by a lipid bilayer.Their content is mainly composed of nucleic acids, proteins and lipids (Colombo et al., 2014;Jurj et al., 2020).Firstly found in 1946(Chargaff Erwin, 1946) and described as EVs in 1971 (Aaronson et al., 1971), these vesicles are now being studied in terms of biogenesis, content, diagnostic/prognostic and therapy for a range of diseases, including cancer, through the ability to cross membranes and delivering effector molecules as a way of intercellular communication (Y.Q. Chen et al., 2022;Cocks et al., 2021;Dar et al., 2021;Signore et al., 2021;Svedman et al., 2018).Around their content, microRNAs (miRNAs/ miRs) are highlighted and studied to execute functions as biomarkers and gene therapy in cancer (Ingenito et al., 2019;O'Brien et al., 2020;Théry et al., 2018).
This study investigated the antitumor effects of small extracellular vesicles (sEVs) derived from M1 macrophages on a three-dimensional (3D) mouse melanoma microenvironment model.It provides new insights into the potential of M1 macrophage-derived sEVs and their miRNA cargo as a novel targeted therapy for melanoma.
Although further extensive studies are needed to confirm our initial findings, to the best of our knowledge, this is the first study to investigate the antitumor effects of M1-sEVs on a 3D melanoma microenvironment model, contributing to the development of more effective melanoma treatments.

Extracellular vesicles (EVs) isolation and characterization
EVs were isolated by differential centrifugation (Théry et al., 2006).The medium of no/ polarized J774A.1 cells was replaced by a medium containing 2% vesicle-depleted FBS when cells reached 80% confluency.Each 24 h, the cell culture supernatant was harvested for 3 days and centrifuged at 4°C according to the following protocol: 300 × g for 10 min, 2000 × g for 10 min 10,000 × g for 30 min, and 100,000 × g for 70 min (Ultracentrifuge Optima XE-100, Beckman Coulter, Indianapolis, IN, USA).In some experiments, the EVs pellet was resuspended in phosphate-buffered saline (PBS), and the supernatant resultant to the last ultracentrifugation was used as an EVs-free control.The protein concentration was determined by Lowry's method (Lowry et al., 1951).

Transmission electron microscopy (TEM)
Isolated EVs were fixed in 2% paraformaldehyde.Fixed suspensions containing EVs were adsorbed on 200 mesh Formvar/carbon-coated copper grids for 20 min.Samples were washed in water, fixed with 1% glutaraldehyde for 5 min, and negatively stained with 5% uranyl acetate for 5 min.EVs were examined on a JEM-1011 transmission electron microscope (JEOL, Tokyo, Japan).

Dynamic light scattering (DLS)
EVs size was determined in PBS using a Zetasizer Nano ZS instrument (Malvern Instruments, Malvern, UK)

RNA isolation
Total RNA was isolated from EVs and the host cells using the mirVana™ PARIS TM Kit or Trizol Reagent (Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer's instructions.Then, an aliquot of the final RNA eluate was used to measure the RNA concentration and integrity of each sample with Agilent RNA 6000 Pico kit, using the 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), or NanoVue Plus (Biochrom, Holliston, MA, USA), according to the manufacturer's protocol.The results were processed by Agilent 2100 Expert software (Agilent Technologies).

miRNA microarray
Mouse miRNA microarray slides (release 17), which detect 1136 distinct miRNAs (Agilent Technologies) were used according to the manufacturer's instructions to identify miRNAs differentially expressed among M0, M1 and M2 macrophagederived EVs and their host cells.Total RNA was fluorescence-labeled with Cyanine 3 and hybridized onto the arrays for 20 h at 55°C.Slides were scanned on the SureScan Microarray Scanner, and images were processed using Feature Extraction software.Intensity values were evaluated using GeneSpring software (all tools from Agilent Technologies).

Quantitative PCR (qPCR)
Total RNA was reversed-transcribed using miRNA 1st-Strand cDNA Synthesis Kit (Agilent Technologies), according to the manufacturer's protocol.The qPCR assay was performed using miRNA qPCR Master Mix (Agilent Technologies), according to the manufacturer's protocol.All reactions were run on a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA), and Ct values were normalized to the expression of endogenous miR-106b-5p, identified by NormFinder software (Aarhus University Hospital, Denmark), or U6 (Sigma-Aldrich) for mimic expression.The relative quantification method (2−ΔΔCt) quantified relative miRNA and mimic expressions.Primer sequences are listed in Supplementary Table S1.

Bioinformatics
miRNAs were validated by analyzing predicted targets and pathways by bioinformatics.miRDB (Y.Chen & Wang, 2020) and TargetScan (7.2) (Agarwal et al., 2015) databases were used to analyze predicted targets and MirTarBase (7.0) (Huang et al., 2020) databases for the validated target.LinkedOmics (Vasaikar et al., 2018) database was used for data correlation.Gene Ontology resource (Mi et al., 2019) was consulted to identify biological process enrichment.Only human miRNAs and genes were considered in this analysis.

Viability assay
A suspension of increasing concentrations of M0, M1 and M2 macrophage-derived EVs (5-100 µg/mL) in PBS was incubated with the 3D mouse melanoma microenvironment model (melanospheres), as previously developed by our group, with some modifications (Saleh et al., 2021).Briefly, murine cells lines including melanoma cells (B16F10), M2 macrophages (J774A.1),and fibroblast (NIH/3T3) (2:1:1 respective proportion, 300 total cells/ well) were mixed and cultivated on agarose gel (1.5%, 40 µl/well) in a 96-well plate to grow the melanospheres.Melanosphere formation was achieved in 4 days; then, EVs were incubated with the 3D model for 72 h.Melanospheres viability was measured by the acid phosphatase assay (Friedrich et al., 2007;Yang et al., 1996).Data was normalized by Eq. ( 1): where AT is the absorbance of cells incubated with EVs; AC is the absorbance of cells with vehicle/control (PBS); AB is the absorbance of the blank.

Melanospheres size
After 72 h of incubation with the EVs, melanospheres were analyzed using a bright-field Nikon Eclipse TS100 microscope (Melville, NY, USA), and the diameter was calculated from images using ImageJ software (NIH, Bethesda, MD, USA).The mean melanospheres volume was estimated by Eq. ( 2): where R is the mean of vertical + horizontal radii.

Melanospheres proliferation
Melanospheres cell proliferation was evaluated with 5-ethynyl-2′-deoxyuridine (EdU), a nucleoside analog of thymidine, by using the Click-it EdU kit (Thermo Fisher Scientific).After 72 h of incubation with the EVs, melanospheres were trypsinized and incubated for 2 h with EdU, followed by fixation and permeabilization.Detection was made by flow cytometry (BD FACS CANTO II, BD Bioscience, 10.000 events, 633-635 nm Alexa Fluor), and data were determined using Flowing 2.5 software (Turku Bioscience, Turku, Finland).
After 24 h of transfection, cell proliferation was analyzed using the Click-it EdU kit, and flow cytometry analysis was also carried out to explore the cell cycle and death.For cell cycle analysis, cells were trypsinized, fixed with 70% ethanol, incubated in an RNAse solution, and treated with propidium iodide (Thermo Fisher Scientific).Cell death was evaluated by double labeling with Annexin V-FITC (BD Biosciences) and propidium iodide.Cell proliferation, cycle and death were analyzed using Flowing 2.5 software.

Data analysis
Assays were performed in duplicate or triplicate on alternate days.Results were analyzed using GraphPad Prism software version 6 (GraphPad Software, San Diego, CA, USA) and expressed as mean � standard deviation (SD).p ≤ .05 and fold change (FC) ≥ 2 were considered statistically significant.Specific statistical tests were performed as described in figure legends.

Extracellular vesicles (EVs) were found majority as small EVs (sEvs)
In this work, to compare, EVs were isolated from M0, M1 and M2 macrophages.Macrophage polarization was confirmed by detecting IL-6 and IL-10 in macrophage supernatant (Figure 1a).IL-6 content was higher in M1 macrophages than in M0 and M2, as expected and, by contrast, IL-10 content was higher in M2 macrophages, confirming both M1 and M2 polarization.After polarization, EVs were isolated from macrophage supernatants by ultracentrifugation.On microscopic examination, EVs exhibited spherical morphology without any apparent vesicular fusion or aggregation (Figure 1b), EVs measured 50-400 nm in diameter (less than 5% >200 nm), showing peaks in a range between 70 and 90 nm (Figure 1c), characteristic of small EVs (sEVs) when analyzed by DLS (Lyu et al., 2021;Szatanek et al., 2017).RNA analysis revealed a predominance of small RNAs (25-200 nt) and a small proportion, less than 20% of RNAs greater than 500 nt in length, especially in M0 and M2 macrophage-derived EVs (Figure 1d).

M1-sEVs damage the 3D mouse melanoma microenvironment model (melanospheres)
The functional screening was performed to investigate the effects of sEVs on melanospheres and estimate the best EVs concentration (protein basis) for functional cell assays without causing excessive cell death.After 72 h of incubation, M1-sEVs reduced cell viability of melanospheres by 15% to 35% with 5 µg/mL and 25 µg/mL, respectively.Cell viability was maintained at approximately 65% up to 100 µg/mL and no cell viability alteration was found with progressive concentration amounts of M0 and M2 sEVs (Figure 2a).Given these results, subsequent experiments aimed to evaluate further antiproliferative effects using a sEVs concentration of 10 µg/mL of protein content, due to a reduction of 20% in cell viability in this condition of protein concentration.
The melanospheres' size was measured microscopically to monitor growth in response to incubation with sEVs.We noticed a decrease in melanospheres volume by almost 3-fold in samples incubated with M1-sEVs compared with the untreated group (Figure 2b).To complement functional analysis, we assessed cell proliferation, which reduced by about 50% (Figure 2c) in melanospheres incubated with 10 µg/mL M1-sEVs.No significant changes were found with M0 and M2 sEVs (Supplementary Figures S1a  and S1b).

M1-sEVs reveal some differently expressed miRNAs
Upon confirming the antiproliferative function of M1-sEVs, we investigated miRNA content to assess possible differences between the effects of M0, M1 and M2 macrophagederived sEVs on cell proliferation.A total of 436 miRNAs were detected by microarray, including sEVs and their host cells (data not shown).Only 17 miRNAs were found exclusively in sEVs and demonstrated differentially expressed among M0, M1 and M2 (Figure 3a).
Although M1 and M0 presented a very similar expression profile, four miRNAs (miR-20a-5p, miR-22-3p, miR-24-3p, and miR-29a-3p) exhibited upregulation in M1 compared to M0 and M2 sEVs (Supplementary Table S2), and were assessed by qPCR to validate microarray data.These miRNAs were selected according to the following criteria: (i) miRNAs showing the highest FC values and (ii) possessing a homologous human sequence.The results showed that all four miRNAs were highly expressed in M1 (FC > 2) compared with the expression in M0 and M2 macrophage-derived sEVs, corroborating microarray results (Figure 3b).

Prediction of targets and pathways highlighted antitumoral miR-29a-3p found in M1-sEVs
Bioinformatic tools were used to predict target genes and pathways associated with miRNAs of M1-sEVs that showed significant antitumor effects against melanospheres.Of the four  selected M1-sEVs miRNAs, miR-29a-3p demonstrated more associations with antitumor effects than the others.A total of 86 common genes were observed in databases for miR-29a-3p.Of the genes related to antiproliferative effects, 11 were associated with negative regulation of cell proliferation (such as PTEN and CDK6), 15 were associated with cell cycle regulation (such as CCNT2, CCNA2, and CCND2), and 12 were associated with apoptosis (such as FOXO3 and PIK3R1).It was also observed possible regulatory effects of miR-29a-3p on tumor microenvironment genes, 4 of them associated with negative regulation of the response to cytokine stimulus, 8 related to the extracellular matrix organization, and 5 associated with regulation of cell-matrix adhesion (Figure 4).miR-20a-5p was associated with some genes related to negative regulation of cell proliferation (Supplementary Figure S2a) and miR-22-3p with positive regulation of apoptosis (Supplementary Figure S2b).We have identified a total of 38 common genes among the databases, associated with miR-24-3p.However, no statistically significant results were generated when subjected to Gene Ontology evaluation (Supplementary Figure S2c).miR-29a-3p was most associated with antiproliferative effects, and, for this reason, it was chosen to be assessed in functional assays with human cells.All data lists can be found in the Supplementary Data I (A, B, C, D).

miR-29a-3p derived from M1-sEVs exhibited antiproliferative effects in human melanoma cells
The miR-29a-3p function was assessed by inducing overexpression of miR-29a-3p mimic via transfection into SK-MEL-28 cells.This procedure increased endogenous miRNA expression up to 200-fold compared with basal (control) and negative miRNA mimic expression levels (Supplementary Figure S3).
Cell functions related to miR-29a-3p predicted pathways, such as cell cycle, cell death and cell proliferation, were analyzed."We observed a propensity toward an increase in G0/ G1 cell population (p = .059)associated with miR-29a-3p overexpression compared with the negative control (Figure 5a).Additionally, we noted an increase in apoptosis/necrosis, indicated by a higher total number of dead cells with miR-29a-3p mimic; however, no statistically significant differences in apoptosis or necrosis were observed (Figure 5b)." Cell proliferation was affected by transfection.miR-29a-3p overexpression reduced the percentage of proliferating cells compared with the negative control (Figure 5c).
In resume, our results showed that M1-sEVs reduced melanospheres volume, cell viability, and cell proliferation.In silico analysis of the M1-sEVs content revealed miR-29a-3p as a key antiproliferative agent.To further validate this finding, miR-29a-3p was transfected into a human melanoma cell line, demonstrating its antitumor effects.

Discussion
This study demonstrated that small extracellular vesicles derived from M1 macrophages (M1-sEVs) have antiproliferative effects against melanoma, evidencing that the mir-29a-3p, found in this type of vesicle, is partly responsible for the effect.M1 macrophages generally produce proinflammatory responses to protect the body from injuries.This behavior can be used as a therapeutic advantage (Cendrowicz et al., 2021;Hao et al., 2012).In addition to releasing cytokines and chemokines for cell communication, M1 macrophages interact with other cells via EVs, transmitting proinflammatory signals that generate immune responses (Ismail et al., 2013;McDonald et al., 2014;Y. Wang et al., 2020).
In this work, to explore the potential of M1-EVs, we polarized mouse macrophages (cell line J774A.1)and polarization was validated by a well-established method (Mantovani et al., 2002;Orecchioni et al., 2019;Weagel et al., 2015).Following polarization, the macrophage supernatant was collected to isolate and characterize EVs.Both isolation and characterization assays were performed in previous studies by our group (Prigol et al., 2021;Rode et al., 2021), followed by well-known characterization studies (Lötvall et al., 2014;Théry et al., 2018).According to them, some criteria for distinguishing vesicle size include morphology, size >200 nm, and small RNA as a principal constituent.Although we have not exhausted the EV characterization, our results taken together showed that most EVs isolated from macrophages met some ISEV criteria and were therefore classified as sEVs.
Through intracellular communication, EVs derived from immune cells can deliver their content to nearby or faraway targets, an ability that has stimulated research on these EVs as therapeutic agents for tumor in the form of vaccines (Besse et al., 2016;P. Wang et al., 2019).One of them uses dendritic cell-derived exosomes (Dex) and has been applied in a clinical trial (Besse et al., 2016).This new tumor immunotherapy strategy appears to be better than the use of immune cells since EVs are more stable, less antigenic and therefore more resistant to tumor immunosuppression (Santos & Almeida, 2021).Furthermore, EVs derived from immune cells use the original cell's biological activity to eliminate tumor cells (Haney et al., 2020).Despite these benefits, there are few studies on unmodified M1-EVs investigation (Cheng et al., 2017;Choo et al., 2018;P. Wang et al., 2019).Therefore, we decided to investigate their potential antitumor properties in a melanoma microenvironment and associate the effects with their miRNA expression profile.
To assess the antitumor effects of M1-sEVs, we used a 3D mouse melanoma microenvironment model previously developed by our group (Saleh et al., 2021).The 3D model uses cell co-culture of melanoma, M2 macrophages and fibroblasts, being more representative of the tumor microenvironment (Pinto et al., 2020).Given the model's-controlled size, the surrounding collagen-based extracellular matrix (ECM), and the presence of proliferative cells in the peripheral region with a necrotic core resembles a solid tumor and can be used to assess the effectiveness of new therapies, resembling the in vivo environment (Charoen et al., 2014).
The functional screening revealed an indirect relationship between melanospheres cell viability and the amount of sEVs.Surprisingly, cell viability remained constant under high sEVs concentrations.A lack of change in tumor cell viability after incubation with M1derived vesicles was also reported earlier (Choo et al., 2018).These effects seem to be specific to M1-EVs since it was found that EVs derived from natural killer and dendritic cells were highly cytotoxic to tumor cells (Bu et al., 2011;Laura et al., 2020).One possibility for this unaltered profile observed in M1-sEVs-treated melanospheres could be related to macrophage non-expression of Fas ligand (FasL) (G.Wang et al., 2019).EVs released by natural killer cells express the FasL on the membrane and exert potent cytotoxicity against Fas + tumor cells (Lugini et al., 2012;Zhu et al., 2017).However, we did note an expressive reduction in melanospheres volume 3 days after incubation with M1-sEVs, apparently resulting from suppression of cell growth, as observed by a decrease in cell proliferation.In a previous study conducted by G. Wang et al. (2019), a reduction in breast cancer tumor volume (more than 50%) was observed in vivo 27 days after treatment with M1-EVs.Such decrease was even more significant (on the nineteenth day) when M1-EVs treatment was associated with paclitaxel, an important agent in breast cancer treatment (P.Wang et al., 2019), demonstrating that M1-EVs may also potentiate the effects of current therapies.
Among the small RNA in the EVs content with potential antitumor effects, miRNAs are the most interesting for research since their known regulatory functions in gene expression (Zhang et al., 2015).As previously shown, EVs derived from polarized macrophages have similar miRNA contents to the cell of origin, allowing distinction between M0, M1 and M2 vesicles (Garzetti et al., 2014).These differences can be produced intracellularly by proteins and surface markers (Liu & Su, 2019) or extracellularly by environmental stimuli (Lindenbergh et al., 2019).Choo et al. (2018) confirmed the repolarization of tumorassociated macrophages to M1 macrophages after treatment with M1 nanovesicles.According to the authors, the miRNA profile of M1-EVs can repolarize macrophages, inducing antitumor immune responses (Choo et al., 2018).In the present study, miRNA profile analysis revealed the upregulation of four miRNAs in M1-sEVs.They were selected for validation and bioinformatics analysis to further understand their antiproliferative effects and potential targets.
We crossed our validated miRNAs in four predicted, validated and correlated target analysis databases and analyzed the identified targets and their functions using a biological process database.In our study, we have identified critical oncogenes pertinent to melanoma, such as CDK6 and VEGFA, which are primarily subject to regulation by the 3' untranslated region (3'UTR) (Geng et al., 2021;Zhao et al., 2015), for this reason only miRNAs with a 3'UTR binding were included.Of the four selected M1-sEVs miRNAs, miR-29a-3p demonstrated more associations with antitumor effects than the others.Pathways and genes related to the regulation of cell proliferation, cell cycle, and apoptosis were identified.We also found associations with tumor microenvironment targets regulating ECM adhesion and response to cytokine stimulus.These observations are aligned with the wellestablished role of the miR-29 family as regulators of gene expression primarily associated with the inhibition of tumorigenesis and tumor progression (Kwon et al., 2019).This assertion is reinforced by previous research findings documented by other investigators.As shown by Zhang and collaborators (Zhang et al., 2021), engineered EVs from human mesenchymal stem cells overexpressing miR-29a-3p decrease migration and vasculogenic mimicry formation in the glioma reinforcing the potential antitumor effect of this miRNA carried by EVs.Moreover, miR-29a has exhibited suppressive effects on tumor cells in various cancer types, as evidenced by its impact in laryngeal carcinoma (targeting PROM1) (Su et al., 2017), gastric cancer (targeting the CDK family and p42.3) (Cui et al., 2011;Zhao et al., 2015), hepatocellular carcinoma (targeting IGF1R) (X.Wang et al., 2017), colorectal carcinoma (targeting RPS15A) (Zheng et al., 2019), and breast cancer (targeting the B-Myb gene) (Wu et al., 2013).
The antiproliferative function of miR-29a-3p was confirmed by transfecting the miRNA into human melanoma cells (SK-MEL-28).This approach serves to authenticate the results obtained within the 3D mouse melanoma microenvironment, thereby enhancing the overall scientific substantiation of miR-29's role in governing melanoma cell proliferation.Broseghini et al. (2021) observed downregulation of miR-29a-3p in melanoma samples of 20 patients compared with benign nevus samples (Broseghini et al., 2021).Apparently, miR-29a-3p is commonly low expressed in melanoma and is related to the progression of the disease.Inhibiting cell proliferation in melanoma seems to be correlated with miR-29 family overexpression.Using melanoma cells, Xiong et al. (2018) demonstrated that miR-29a suppresses growth, migration, and invasion by negatively regulating Bmi1 (Xiong et al., 2018).Likewise, cell proliferation was decreased when miR-29a targeted MAFG and MYBL2 (Vera et al., 2021).As well, for the first time, it was confirmed the relationship between the interferon-γ-released by macrophages and overexpression of the miR-29 family, resulting in CDK6 downregulation and G1 arrest (Schmitt et al., 2012), showing the relation of macrophages, miR-29 expression and melanoma cells.Even though no significant differences in cell cycle were observed in the current study, probably due to the short assay time (24 h), it was possible to identify a propensity toward G0/G1 arrest.Nevertheless, as assessed using the EdU kit, the decrease in cell proliferation is strong evidence, given that this type of assay directly measures cell activity during active DNA synthesis (Salic & Mitchison, 2008).

Conclusion
Extracellular vesicles and their miRNAs content might be an innovative and valuable tool in clinical oncology.We identified differences between the miRNA expression profiles of sEVs derived from M0, M1 and M2 macrophages.M1-derived macrophage sEVs were found to exert antitumor effects, reducing cell proliferation in a 3D mouse melanoma microenvironment model.We also found an indirect relationship between miR-29a-3p, a miRNA highly expressed into the M1-sEVs, upregulation and the reduced expression of predicted targets related to tumor cell proliferation, which was validated by the observation of a decrease in cell proliferation following transfection of the miR-29a-3p mimic into human melanoma cells.These findings bring new possibilities for immunological/targeted treatment of diseases such as melanoma, which are susceptible to the immune system.
Furthermore, the present work provides a library of miRNAs, targets and pathways that can be exploited for therapeutic purposes.Moreover, in this work we analyzed only the miRNAs from the unmodified M1-EVs.While we focused exclusively on the miRNA content of unmodified M1-EVs, further studies may elucidate the role of other vesicle components in mediating antitumor effects.

Figure 1 .
Figure 1.Polarization of macrophages and characterization of EVs isolated from cell supernatants.(a) the concentration of cytokines (interleukin-6 and −10) released in macrophage supernatant after polarization.(b) microphotographs illustrating EVs morphology (scale bar = 100 nm) and (c) histogram of EVs size obtained by DLS.(d) Electropherograms and respective virtual gels show the concentration and quality of total RNA extracted from EVs.Data were generated with the agilent 2100 bioanalyzer system using the agilent RNA 6000 pico kit.M0, macrophage without the polarizing agent.M1, macrophage polarized with lipopolysaccharide (LPS).M2, macrophage polarized with dexamethasone (DEX).IL, interleukin.n = 3. Mean ± standard deviation.Unpaired t-test.*p < .05;***p < .001.

Figure 4 .
Figure 4. Venn diagram of predicted and validated targets correlated with LinkedOmics data of hsa-miR-29a-3p including the pathways and genes involved in antitumoral processes.hsa, human.miR, microRNA.