MicroRNA Cargo of Extracellular Vesicles from Alcohol-exposed Monocytes Signals Naive Monocytes to Differentiate into M2 Macrophages*

Membrane-coated extracellular vesicles (EVs) released by cells can serve as vehicles for delivery of biological materials and signals. Recently, we demonstrated that alcohol-treated hepatocytes cross-talk with immune cells via exosomes containing microRNA (miRNAs). Here, we hypothesized that alcohol-exposed monocytes can communicate with naive monocytes via EVs. We observed increased numbers of EVs, mostly exosomes, secreted by primary human monocytes and THP-1 monocytic cells in the presence of alcohol in a concentration- and time-dependent manner. EVs derived from alcohol-treated monocytes stimulated naive monocytes to polarize into M2 macrophages as indicated by increased surface expression of CD68 (macrophage marker), M2 markers (CD206 (mannose receptor) and CD163 (scavenger receptor)), secretion of IL-10, and TGFβ and increased phagocytic activity. miRNA profiling of the EVs derived from alcohol-treated THP-1 monocytes revealed high expression of the M2-polarizing miRNA, miR-27a. Treatment of naive monocytes with control EVs overexpressing miR-27a reproduced the effect of EVs from alcohol-treated monocytes on naive monocytes and induced M2 polarization, suggesting that the effect of alcohol EVs was mediated by miR-27a. We found that miR-27a modulated the process of phagocytosis by targeting CD206 expression on monocytes. Importantly, analysis of circulating EVs from plasma of alcoholic hepatitis patients revealed increased numbers of EVs that contained high levels of miR-27a as compared with healthy controls. Our results demonstrate the following: first, alcohol increases EV production in monocytes; second, alcohol-exposed monocytes communicate with naive monocytes via EVs; and third, miR-27a cargo in monocyte-derived EVs can program naive monocytes to polarize into M2 macrophages.

cell-derived vesicles that are released by almost every cell type into their microenvironment. EVs have a lipid composition similar to those present on the plasma membranes of the parent cells (1)(2)(3). EVs act as signaling conveyors in cell-to-cell communication and contain bioactive molecules, including proteins, microRNAs (miRNAs), and messenger RNAs (mRNAs). Based on the mode of biogenesis, EVs can be categorized into two main classes as follows: shedding MVs (originating from plasma membrane) and exosomes (derived from multivesicular bodies) (2). Exosomal markers are CD63, CD81, and CD9, whereas microvesicles are annexin V-positive (3). Based on size distribution and CD63 expression, exosomes are defined as 40 -150 nm and MVs as 150 -1000 nm (4,5). Recently, EVs, including exosomes, have emerged as important signaling organelles in mediating local and systemic communication between cells and in the pathogenesis of different diseases (6 -9).
The pathogenesis of alcohol-induced liver disease is complex and is associated with liver abnormalities ranging from steatosis and inflammation to cirrhosis and hepatocellular carcinoma (10 -12). The intercellular signaling during alcoholic liver disease (ALD) is key to its progression and pathogenesis (13,14). Our previous studies explored the role of exosomes in various liver diseases and showed that circulating miRNAs in exosomes may serve as biomarkers to differentiate between different types of hepatocyte injury and inflammation (15). Recent data showed a role for exosomes in communication between hepatocytes and monocytes/macrophages via horizontal transfer of hepatocyte-derived miR-122 to monocytes in the presence of alcohol (16). Innate immune cells, including monocytes and macrophages, play an important role in the pathogenesis of ALD. However, the role of immune cell-derived EVs in intercellular signaling and communication has yet to be explored in alcohol-induced immunoregulation.
Monocytes are the precursors of macrophages. In response to various signals, monocytes differentiate into macrophages with a polarized phenotype in the spectrum of M1 (pro-inflammatory) and M2 (anti-inflammatory) phenotypes (17)(18)(19). Increasing evidence suggests that miRNAs, a class of non-coding RNAs, play an important role in macrophage polarization (7, 20 -22). We and others have shown that miRNAs can regulate cellular differentiation and functions in immune cells (21,23,24). miRNAs not only function within the cells of origin but can also be packaged into EVs and released into the extracellular environment, where they are transferred in an active form to other cells and exert functional effects (16,(25)(26)(27)(28). The possible contribution of circulating miRNAs transported via EVs to signal between monocytes and activate the process of monocyte differentiation and/or polarization has yet to be explored.
In this study, we investigated the effect of alcohol on EV secretion by human monocytes and studied the role of alcoholinduced EVs as an alternative mode of cell-to-cell communication between monocytes. Our data demonstrate that the total number of EVs secreted from alcohol-treated monocytes is significantly increased as compared with untreated monocytes. EVs originating from alcohol-exposed monocytes induced differentiation and polarization of naive monocytes into M2 macrophages. We demonstrated the role for miR-27a in EVs from alcohol-treated monocytes in inducing activation, polarization, and increased phagocytic activity of naive monocytes. These in vitro observations were validated in vivo where we found an increased number of EVs in the circulation of patients with alcoholic hepatitis, which had increased levels of miR-27a in the EV cargo. Our results suggest that in the pathogenesis of alcoholic hepatitis, EVs play an important role in cell-cell communication between monocytes/macrophages.

Experimental Procedures
Human Studies-Confirmed cases of alcoholic hepatitis (n ϭ 8) and healthy individuals (n ϭ 8) were enrolled. The diagnosis of alcoholic hepatitis was performed by expert clinicians based on the patients' medical history, physical examination, and laboratory data and according to the guidelines of the American College of Gastroenterology (29). Healthy controls were defined as being free of any systemic and non-systemic diseases based on patients' history and routine laboratory findings performed by primary care physicians and a history of "only social" alcohol consumption. To avoid selection bias, healthy controls and patients were enrolled consecutively. The study protocol was approved by the Institutional Review Board for the Protection of Human Subjects in Research at the University of Massachusetts Medical School (Worcester, MA), and written informed consents were obtained from all subjects. We obtained plasma samples from controls and patients for the study.
Cell Culture and EV Isolation-Primary human monocytes and THP-1 cells were cultured in RPMI 1640 cell culture media containing antibiotics, nonessential amino acids, and FBS. For EV isolation and quantification experiments, the cells were cultured in the presence of RPMI 1640 medium plus 10% exosome-depleted FBS (Exo-FBS TM ) and 1% penicillin/streptomy-cin. For ethanol treatments, 25, 50, and 100 mM ethanol were added to the cells for various time points (24 and 48 h). At desired time points, culture media were harvested, and EVs were isolated or quantified using a NanoSight NS300 system (NanoSight, Amesbury, UK). For EV isolation from ethanoltreated and normal THP-1 cells, supernatants were centrifuged at 1500 ϫ g for 10 min to remove cells followed by 10,000 ϫ g for 20 min to deplete residual cellular debris, and then the supernatant was filtered through a 0.8-m filter. The filtered supernatant was used to precipitate EVs with ExoQuick-TC TM , according to the manufacturer's guidelines (System Biosciences, Mountain View, CA). After isolation, EVs were resuspended in PBS. The concentration of EVs was determined by Bradford assay (30).
Co-culture Experiments-For co-culture experiments, EVs isolated from THP-1 cells or primary monocytes (ethanoltreated or non-treated) or from plasma of patients with alcoholic hepatitis or from healthy controls were added to normal primary human monocytes in the concentration of 25-50 g/ml (50 l). This concentration was comparable with EVs secreted by primary monocytes in the presence of ethanol. The monocytes were stimulated with EVs (25 l) on day 1 and day 3 and cultured for 5 days.
EV Measurement Using Nanosight System and Nanoparticle Tracking Analysis (NTA)-The concentration and diameter of EVs derived from culture supernatant from primary monocytes, THP-1 cells, and patient plasma samples were identified by a NanoSight equipped with a fast video capture and NTA software. Before using the samples, the instrument was calibrated with 100 nm polystyrene beads (Thermo Scientific, Fremont, CA). The samples were captured for 90 s at room temperature. NTA software was used to measure particle concentration (particles/ml) and size distribution (in nanometers). For each sample, three measurements were taken, and the average value was determined.
Scanning Electron Microscopy (SE)-THP-1 human monocytes were plated into 12-well plates and were left untreated or treated with 50 mM ethanol for 24 h. After 24 h, ethanol-treated THP-1 cells and untreated cells were fixed with 2.5% glutaraldehyde in Sorenson Phosphate buffer (0.1 M, pH 7.4) for 2 h. The cells were then given three 5-min washes with PBS and fixed with 1% osmium tetroxide (OsO 4 ) in 0.1 M PBS for 1 h. Next, the samples were rinsed and put through an alcohol dehydration series. The samples were mounted on aluminum, and the mounted specimens were coated with gold/palladium using a gold sputter coater. Then the samples were analyzed and examined with a MKII FEI Quanta 200 FEG MKII scanning electron microscope (FEI, Netherlands).
Western Blotting-EVs were isolated from THP-1 cells, and Western blot analysis for the exosomal marker CD63 was performed. EVs were resuspended in RIPA buffer, and the proteins were extracted. The proteins were run on a 10% SDS-polyacrylamide gel, transferred to a nitrocellulose membrane, and blocked for 1 h in TBS containing 5% nonfat dry milk and 0.1% Tween 20 at room temperature (TBST). The blot was then incubated overnight with primary CD63 antibody at 4°C on a rocker. The following day, the blot was washed three times with TBST and then incubated with the HRP-conjugated secondary miR-27a Increase in Plasma EVs of AH Patients antibody (goat anti-mouse IgG-HRP antibody, Santa Cruz Biotechnology, Dallas, TX) for 1 h at a dilution of 1:10,000. Clarity TM Western ECL substrate kit (Bio-Rad) was used to develop the blot, and the CD63 protein band was visualized with a Fujifilm LAS-4000 luminescent image analyzer.
Flow Cytometry-Cell surface marker analysis on monocytes/macrophages was performed by flow cytometry as described previously (31). FcR blocking reagent (BD Biosciences) was used to inhibit nonspecific binding of antibodies. The cells were incubated with appropriate antibody or isotype control for 30 min at 4°C. Cells were washed with FACS buffer (PBS ϩ 2% FBS) and acquired on a BD-LSR II (BD Biosciences). Data analysis was performed on FlowJo software (Tree Star, Ashland, OR). For CD14 ϩ monocytes, we gated on the various markers and plotted the percentage of positive cells. The mean fluorescence intensity (MFI) is shown by histogram plots.
Enzyme-linked Immunosorbent Assay (ELISA)-Protein levels of TNF␣, IL-10, IL-12, and TGF␤ were measured in cell-free culture supernatant by ELISA. Whole blood cells isolated from human subjects were incubated overnight at 37°C, and the levels of IL-10 and TGF-␤ secretion were determined by ELISA. Levels of TNF␣, IL-12 (BD Biosciences), IL-10, and TGF␤ (eBioscience, San Diego) were measured based on the manufacturers' recommendations and quantified using an ELISA reader.
Phagocytosis Assay-Mannose receptor-mediated endocytosis was measured as the cellular uptake of FITC-dextran and quantified by flow cytometry (32)(33)(34)(35). For the phagocytosis assay, ϳ2 ϫ 10 5 cells per sample were incubated in medium containing 1 mg/ml FITC-dextran beads for 120 min at 4 or 37°C. After incubation, cells were washed with PBS containing 2% FBS. Phagocytic assays performed at 4°C were used as negative controls (i.e. a reduced phagocytosis condition) and accounted for the particles bound to the cells' extracellular surface. The cells were acquired in the LSR II (BD Biosciences) instrument to analyze the uptake of FITC-dextran beads by the cells. The phagocytosis was calculated by normalizing with the phagocytosis at 4°C. The phagocytosis index was calculated as described previously (36).
Loading miR-27a Mimic into the EVs and Monocyte Treatment-We loaded THP-1-derived EVs with miR-27a mimic and scrambled mimic according to our previously optimized protocol (4). Briefly, EVs were diluted in Gene Pulser electroporation buffer (Bio-Rad) at a ratio of 1:1. miR-27a mimic or scrambled mimic at a final amount of 400 pmol was added to EV samples containing 1 g/l EV protein. For the dose-dependent experiments, 300, 100, and 40 pmol were loaded into the EVs. The mixtures (200 l) were transferred into cold 0.2-cm electroporation cuvettes and electroporated at 150 kV and 100 microfarads with a Gene Pulser II System (Bio-Rad). The EVs were treated with 1 unit of RNase H to eliminate free-floating miR-27a mimic or scrambled mimic outside the EVs and were re-isolated using ExoQuick-TC TM . Based on the loading efficiency that we established earlier (4) and dilution of EVs, a final concentration of 10, 25, and 75 pmol of miR-27a mimic was used in the co-culture experiments with primary human monocytes.

RNA Isolation and miRNA Analysis (Quantitative Real Time
Polymerase Chain Reaction (qPCR))-EVs were lysed in QIAzol lysis reagent (Qiagen), and total RNA was isolated using Directzol TM RNA MiniPrep isolation kit (Zymo Research Corp, Irvine, CA). Optical density (260/280 and 260/230 ratios) was measured to check RNA quality and quantity. 50 l of EV suspension from the supernatant of ethanol-treated cells or normal cells was mixed with 500 l of QIAzol lysis buffer, and the mixture was processed according to the manufacturer's protocol. The extracted RNA was eluted with 25 l of RNase-free water.
For miRNA analysis of exosomes, synthetic Caenorhabditis elegans (cel)-miR-39 was added during the total RNA isolation process and was used to normalize the qPCR data. TaqMan miRNA assays (Applied Biosystems, Foster City, CA) were used for detection of miR-27a, miR-146a, and miR-9 expression according to the manufacturer's protocol, as described previously (4). The miRNA levels were normalized, and the relative expression of the specific miRNAs was presented by 2 Ϫ⌬⌬Ct .
Statistical Analysis-All the data are expressed as mean Ϯ S.E. of the mean (S.E.), which was obtained from three or more independent experiments. For pairwise comparisons, nonparametric Mann-Whitney test or parametric Student's t test was used based on underlying distribution. Comparison between groups was performed with the Kruskal-Wallis nonparametric test or the analysis of variance parametric test. p value of less than 0.05 was considered statistically significant. GraphPad Prism version 6.05 (GraphPad Software Inc., La Jolla, CA) was used for statistical analysis.

Dose-and Time-dependent Induction of EV Secretion in Primary Human Monocytes and THP-1 Monocytic Cells Is Induced
by Alcohol Treatment-Monocytes become activated in the presence of alcohol and are characterized by changes in the expression of cell surface molecules and the secretion of cytokines and chemokines (24,37,38). Recently, we have reported that alcohol-treated hepatocytes secrete increased numbers of exosomes (16). Thus, we hypothesized that ethanol treatment can induce EV secretion by monocytes. To evaluate the EV secretion pattern of monocytes, we treated human blood monocytes with different concentrations of ethanol and measured the EVs using a Nanosight instrument coupled with NTA software. We observed that ethanol treatment (25-100 mM) resulted in a dose-dependent increase in the total number of EVs produced by monocytes (Fig. 1A). We found a significant increase in the number of secreted exosomes with the increasing concentrations of ethanol from 25 to 100 mM (Fig. 1A). These ethanol concentrations are relevant to human alcoholics because the 25 mM in vitro dose mimics binge drinking (ϳ0.1 g/dl) blood alcohol content and 100 mM approximates 0.4 g/dl blood alcohol content seen in chronic alcoholics (39,40). MV numbers did not change between 25 and 50 mM ethanol treatment, but it was significantly higher in the presence of 100 mM ethanol (Fig. 1A). Furthermore, we studied the kinetics of the vesicle secretion from 50 mM ethanol-exposed monocytes and observed significantly increased levels of total EVs and exosomes but not of the MVs at 24 and 48 h (Fig. 1B). These data

miR-27a Increase in Plasma EVs of AH Patients
suggest that there is a dose-and time-dependent increase in the secretion of EVs and exosomes with ethanol treatment in primary human monocytes.
Human Monocyte Cell Line, THP-1, Secretes EVs in the Presence of Ethanol Similar to Primary Monocytes-THP-1 is a widely used monocytic cell line that resembles primary monocytes in its phenotype and function (41,42). Thus, we utilized THP-1 cells to further evaluate the EV secretion pattern of THP-1 cells in the presence of ethanol. We found that there was a dose-and time-dependent increase in the levels of EV secretion by the THP-1 cells treated with different doses of ethanol (25-50 mM) at 24 and 48 h after ethanol treatment (Fig. 1, C and  D). Similar to the findings in the primary monocytes, the numbers of exosomes (ϳ1.96 ϫ 10 10 particles/ml) was a magnitude higher as compared with MV numbers (ϳ4.05 ϫ 10 9 particles/ ml), demonstrating that the majority of EVs from ethanoltreated THP-1 monocytes are exosomes.
Characterization of Monocyte-derived EVs-To visualize the EVs secreted from monocytes, the THP-1 cells were either left untreated or treated with ethanol for 48 h, and images were taken by an electron microscope. THP-1 cells treated with ethanol showed an increased number of shedding EVs (exosomes/ MVs) compared with untreated control monocytes based on the electron micrograph (EM) (Fig. 2, A and B). EVs derived from monocytes expressed the exosomal marker CD63 as shown by the Western blot analysis (Fig. 2C). Using NTA software, analysis of the EVs secreted from monocytes revealed that the average size of the EVs derived from THP-1 cells was 131 nm (Fig. 2D).
EVs Derived from Ethanol-treated THP-1 Cells Stimulate Naive Cells to Differentiate into M2 Macrophages-EVs play a role in cellular communication and cell-to-cell signaling (1,6,43,44). Therefore, we explored the effect of EVs derived from ethanol-exposed monocytes on naive monocytes. Although alcohol increased EV release in both the primary human monocytes and THP-1 monocytic cells, the number of EVs secreted by THP-1 monocytes was 10 times greater as compared with the primary monocytes (Fig. 1). Thus, to investigate the functional effect of EVs, we incubated naive primary human monocytes with ethanol-exposed THP-1-derived EVs and studied the expression of CD14 (a monocyte marker which is highly up-regulated in macrophages) and CD68 (macrophage marker) (27). We found that naive monocytes treated with EVs derived from ethanol-exposed THP-1 cells gained high CD14 expression as compared with control EV-exposed monocytes (Fig.  3A). The expression of the macrophage marker, CD68, was also increased in the CD14 ϩ cells incubated with EVs from ethanoltreated monocytes (Fig. 3B). As shown in Fig. 3, A and B, EVs derived from control THP-1 cells did not change expression levels of CD14 and CD68 on naive monocytes, suggesting control EVs do not activate monocytes. Next we studied the expression of M1 and M2 macrophage markers on EV-treated monocytes. There was a significant increase in the frequency of CD206-and CD163-expressing cells in the presence of ethanol

miR-27a Increase in Plasma EVs of AH Patients
EVs as compared with control EVs (Fig. 3, E and F). In contrast, the frequency of M1 marker-expressing macrophages (CD16 and CD86) did not change (Fig. 3, C and D).
A recent study reported that glioblastoma-derived EV can transform the phenotype of monocytic cells to M2-type macrophages (9). To ascertain that the effects of alcohol-treated THP-1 cell-derived EVs on naive monocytes were not cell linespecific, we isolated EVs from primary monocytes treated or untreated with ethanol and then exposed these EVs to naive monocytes. As shown in Fig. 3G, the ethanol EVs from primary monocytes also promoted CD206 and CD163 expression similar to alcohol-treated THP-1 cell-derived EVs. These results suggest that the EVs from ethanol-treated primary monocytes or THP-1 cells could induce monocyte activation, differentiation, and macrophage polarization, thus demonstrating the role of alcohol in influencing the characteristics of monocyte-derived EVs.
Alcohol EVs Stimulate Naive Monocytes to Secrete Increased Levels of IL-10 and TGF␤-Next, we studied the effect of EVs derived from ethanol-exposed monocytes on the cytokine secretion pattern of naive monocytes. We observed that naive monocytes treated with EVs from ethanol-exposed THP-1 cells had no significant effect on IL-12 secretion but increased TNF␣ secretion, although the TNF␣ levels were in the very low range (1-3 pg/ml) (Fig. 4, A and B). We also evaluated the secretion pattern of other pro-inflammatory cytokines, IL-1␤, IL-6, and MCP-1, and found no significant differences between monocytes treated with alcohol EVs and control EVs (data not shown). In contrast to no change in most of the pro-inflammatory cytokines, the levels of antiinflammatory cytokines, IL-10, and TGF-␤ were significantly increased from monocytes treated with alcohol EVs as compared with control EVs (Fig. 4, C and D). Furthermore, we evaluated the secretion of pro-and anti-inflammatory cytokines by naive monocytes that were incubated with EVs from ethanol-treated or untreated primary monocytes. We observed increased IL-10 and TGF␤ and decreased levels of TNF␣ and IL-12 secretion in alcohol EV-treated cells as compared with control EVs (Fig. 4E). Thus, these results suggested that alcohol EVs modulate naive monocytes to secrete anti-inflammatory cytokines.
One of the characteristics of monocytes and macrophages is their property of phagocytosis (45). Because we found an increased expression of the mannose receptor, CD206, a C-type lectin, we studied the uptake of dextran-FITC beads that are mainly taken up through this receptor into macrophages (34,35). We observed that there was a significant increase in the percentage of phagocytic cells in monocytes after alcohol EV treatment as compared with control EVs (Fig. 4F). The phagocytic index was also increased in the alcohol EV-treated monocytes as compared with control EV-treated cells (Fig. 4G). These results demonstrate that EVs derived from ethanoltreated monocytes signal to naive monocytes and modulate both their cytokine secretion as well as phagocytic capacity.
Analysis of the Cargo of Alcohol EVs Revealed Increased Presence of the M2-polarizing miRNA, miR-27a-EVs have been shown to contain miRNAs as their cargo (43). Because we demonstrated a significant effect of monocyte-derived alcohol EVs on the phenotype and function of naive monocytes, we next studied the profile of miRNA cargo in the EVs derived from ethanol (50 mM)-treated monocytes. Based on their functional

miR-27a Increase in Plasma EVs of AH Patients
JANUARY 1, 2016 • VOLUME 291 • NUMBER 1 role in monocyte activation, we first determined the miRNA expression of miRNAs miR-27a, miR-146a, and miR-9 (20,46,47). We found increased expression of miR-27a and miR-146a but not miR-9 in the THP-1 cells exposed to ethanol (Fig. 5A). Next, we studied the expression of these miRNAs in the EVs derived from THP-1 cells and observed a significant increase in the expression levels of miR-27a in EVs derived from ethanoltreated monocytes as compared with control EVs (Fig. 5B). The levels of miR-146a and miR-9 were not significantly changed in the control EVs versus alcohol EVs (Fig. 5B). These data suggested that miR-27a was selectively sorted into the EVs released by the ethanol-treated monocytes.

miR-27a-loaded EVs Have Functional Effects on Naive
Monocytes Similar to Alcohol EVs-Given the selective increase in the levels of miR-27a in ethanol-induced exosomes, we aimed to investigate the functional role of EV-associated miR-27a in monocyte polarization. To achieve this, control EVs were loaded with miR-27a and incubated with naive monocytes. We were successful in loading EVs from untreated THP-1 monocytes with scrambled and various concentrations of miR-27a mimic as shown by the increased levels of miR-27a (Fig. 5C). Next, we incubated naive monocytes with EVs loaded with either miR-27a mimic or scrambled miRNA mimic. We observed a dose-dependent increase in CD68-positive cells in the presence of EVs loaded with miR-27a as compared with scrambled mimic-loaded EVs (Fig. 5D). There was no change in the levels of CD11c expression on the naive monocytes. M1 markers, CD86 and CD16, were differentially regulated as indicated by a decreased expression of CD86 and increased expression of CD16 in the presence of EVs containing the higher doses of miR-27a mimic (Fig. 5D). The expression of M2 macrophage

miR-27a Increase in Plasma EVs of AH Patients
markers, CD206 and CD163, was induced with the miR-27a mimic EVs as compared with scrambled mimic EVs in a dosedependent manner (Fig. 5D). These results demonstrated that miR-27a is present in the EVs, is functional, and effects naive monocytes in a dose-dependent manner that leads to robust M2 macrophage marker expression.
Next, we studied the effect of miR-27a overexpressing EVs on the cytokine secretion capacity of naive monocytes. We noted that there was an increased level of TNF␣ with increased expression of miR-27a in the EVs (Fig. 6A). However, there was no change in the expression level of other pro-inflammatory cytokines IL-12 (Fig. 6A) and IL-1␤, IL-6, or MCP-1 (data not shown). There was a dose-dependent increase in the expression of anti-inflammatory cytokines IL-10 and TGF-␤ with miR-27a-overexpressed EVs as compared with control EVs (Fig. 6B).
miR-27a Mediates the Process of Phagocytosis by Regulating CD206 Expression-miRNAs have emerged as regulators of phagocytosis in myeloid cells (48,49). Thus, we wanted to eval-uate the role of miR-27a in increased phagocytosis induced by alcohol EVs. The addition of various doses of miR-27a-overexpressing EVs to naive monocytes resulted in a significant increase in the percentage of phagocytosis and phagocytic index compared with scrambled mimic-loaded EVs (Fig. 6, C  and D). These results suggested that miR-27a present in EVs can regulate monocyte phagocytosis. Because we observed a high expression of CD206 in the presence of miR-27a-loaded EVs, we reasoned that the increase in CD206 mediated by miR-27a might be responsible for the increased phagocytosis. We used mannan, a CD206 ligand, and anti-CD206 (␣-CD206), a monoclonal antibody directed against CD206, to block the surface expression of CD206 on the monocytes and examined the effect of miR-27a-loaded EVs on naive monocytes (50). We observed that both mannan and ␣-CD206 treatment led to a significant decrease in the phagocytic activity of monocytes (Fig. 6, E and F). Thus, our results demonstrate the functional role of miR-27a present as a cargo in the EVs to mediate naive monocyte differentiation into M2 macrophages with an increase in IL-10 and TGF-␤ secretion and enhanced phagocytic capacity. These results indicate that the effects of EVs derived from ethanol-treated monocytes on the phenotype and function of naive monocytes is mediated primarily by miR-27a present in these alcohol EVs.

Increased Number of Circulating EVs Is Found in the Plasma of Alcoholic Hepatitis Patients with High Levels of miR-27a-
Finally, to validate the in vivo relevance of our in vitro findings, we studied circulating EVs in the plasma of AH patients. We observed a significant increase in the levels of EVs in plasma of AH patients as compared with healthy controls as shown previously by our group (Fig. 7A) (30). Next, we investigated the level of miR-27a in the EVs from the AH patients and healthy controls. Importantly, we found a significant increase in the levels of miR-27a in the EVs of AH patients (Fig. 7B). We also observed increased IL-10 and TGF␤ levels in the plasma of AH patients compared with healthy controls (Fig. 7, C and D). Thus, these in vivo results from AH patients mirror alcohol-induced changes in vitro.
Circulating EVs from Plasma of Patients with Alcoholic Hepatitis Polarize Naive Monocytes into an M2 Phenotype-To study the effect of circulating EVs from patients on naive monocytes, EVs isolated from plasma of AH patients and healthy donors were co-cultured with healthy naive monocytes. We observed a significant increase in CD206 and CD163 expression as well as a slight increase in CD68 and CD11c expression on the monocytes in the presence of EVs isolated from plasma AH patients compared with controls (Fig. 7E). There was no significant change in the expression of CD86 and CD16 (Fig. 7E). Furthermore, we found a significant increase in IL-10 secretion from the monocytes which were incubated with EVs from AH patients as compared with EVs from healthy controls (Fig. 7F). An increase in the phagocytic index in monocytes incubated with EVs from AH patients compared with controls further indicated the functional role of EVs in vivo (Fig. 7G).

Discussion
Innate immune cells, monocytes, and macrophages are targets of alcohol-induced immunomodulation and organ dam- ) E and F, primary monocytes were pretreated with control, mannan, anti-IgG1, and anti-CD206 and then incubated with EVs loaded with scrambled mimic and miR-27a mimic. Phagocytosis assay was performed as described previously using dextran-FITC beads. The data are representative of three independent experiments, where * indicates p Ͻ 0.05.

miR-27a Increase in Plasma EVs of AH Patients
age. Alcohol exposure regulates the production and function of signaling molecules that coordinate the immune response in ALD and influence the cross-talk between liver parenchymal and immune cells (10,51). Interactions between immune cells can occur by direct cell-cell contact and/or cytokine production. EVs derived from different cell types have also been shown to act as signaling organelles (43,52). Recent studies from our group and other groups have evaluated the biological significance of EVs and their miRNA cargo in intercellular communication and signaling (16,43,44). In this study, we investigated the biological relevance of a novel signaling mechanism between the monocytes in the presence of alcohol, involving the role of EV-associated miR-27a in this process. We show that the numbers of EVs released by monocytes are increased in the presence of alcohol and demonstrate that the EVs generated from alcohol-treated monocytes were effective in modulating the phenotype and function of naive monocytes. This in vitro discovery was further validated by our observation of increased numbers of EVs in the circulation of patients with acute alcoholic hepatitis where macrophage activation is a major pathogenic factor. We found that miR-27a was selectively enriched in EVs of alcohol-exposed monocytes, and this finding correlated with increased levels of miR-27a in circulating EVs of patients with alcoholic hepatitis. Furthermore, our investigations revealed a functional role of miR-27a in the induction of an M2 macrophage phenotype. Exosomes released from alcohol-exposed hepatocytes have been shown to selectively accumulate miR-122 (16). Here, we show that the increased levels of miR-27a in EVs from alcohol-treated monocytes were instrumental in inducing monocyte differentiation and M2 macrophage polarization in naive monocytes. We also found that EVs overexpressing miR-27a displayed high phagocytic activity, demonstrating for the first time the role of miR-27a in the process of phagocytosis. Our study demonstrates that the number of EVs released by human primary monocytes or THP-1 monocytic cells is increased with the incubation of the cells with alcohol. The treatment of naive monocytes with these alcohol EVs as well as from EVs from alcoholic hepatitis patients resulted in an increase in expression of M2-polarized macrophages as revealed by the increased CD206 and CD163 (M2 macrophage marker) expression.
Monocytes and macrophages are important players in the innate immune system during ALD. They are highly plastic, are heterogeneous in their phenotype and function, and are regulated by environmental cues to differentiate into macrophages and polarize. We have recently reported the role of alcohol in modulating monocyte differentiation into an M2-type macrophage phenotype (24). In vivo studies have also shown that patients suffering from alcoholic hepatitis have a presence of both M1 and M2 macrophages (53). Our results demonstrate that the EVs secreted from alcohol-exposed monocytes can signal and influence the monocyte phenotype and function and macrophage polarization even when the monocytes were not directly exposed to alcohol. We also observed increased phagocytic capacity of the M2 macrophages generated in the presence of alcohol EVs. Thus, these results bring forth an interesting phenomenon that suggests that monocyte-derived EVs most likely represent a unique and important mode of signaling during alcoholic liver disease.
The contents of the EVs derived from alcohol-treated monocytes revealed the presence of miRNAs as their cargo. The role of miRNAs has been documented in the process of hematopoiesis, differentiation, survival, and function of immune cells, including monocytes and macrophages (23,54). Transfection of miR-29b, miR-125a-5p, or miR-155 mimics in THP-1 cells altered macrophage polarization to M1 macrophages (23). Experiments with human macrophages have identified unique miRNAs associated with polarized macrophage phenotypes such as miR-155 in M1 macrophages and miR-27a and miR-132 expression in M2 macrophages (20). We have recently reported FIGURE 7. Increased level of miR-27a in alcoholic hepatitis patients is associated with increased IL-10 and TGF-␤ secretion from immune cells. A, number of EVs was measured in the plasma of healthy controls and AH patients using Nanosight. Exosomes were defined as EVs with diameter less than 150 nm, and microvesicles were the EVs with a diameter of more than 150 nm and above. B, level of miRNA-27a was quantified in the plasma-derived EVs isolated from healthy controls and AH patients by TaqMan qPCR assay. C and D, IL-10 and TGF-␤ secretion was determined from plasma samples of healthy controls and AH patients. The data are represented as mean Ϯ S.E. (n ϭ 6 -8 individuals), where * indicates p Ͻ 0.05. E-G, EVs were isolated from the plasma of AH patients and healthy controls. Primary monocytes were co-cultured with the EVs for 5 days. E, cell surface expression of various markers; F, IL-10 secretion; and G, phagocytosis measured for the monocytes. The data are represented as mean Ϯ S.E. (4 -5 individuals).

miR-27a Increase in Plasma EVs of AH Patients
that miR-27a was significantly up-regulated in primary monocytes cultured in the presence of alcohol (24). We also demonstrated that alcohol mediates miR-27a induction, which in turn leads to monocyte differentiation and macrophage polarization. In this study, we show that miR-27a induced by alcohol treatment of monocytes is packaged in EVs and released by these cells, which in turn can affect the naive phenotype and function of monocytes. A previous study had demonstrated that miR-223 transfer via microvesicles to monocytes induced differentiation into macrophages (55). In this study we provide evidence that the EVs from alcohol-treated monocytes have miR-27a-enriched cargo, and the functional effects of the alcohol EVs was demonstrated by increased macrophage markers, M2 markers, IL-10, and TGF␤ secretion and increased phagocytosis by the naive monocytes. Similar effects on naive monocytes were observed when they were incubated with EVs overexpressing miR-27a. Additionally, overexpression of miR-27a in EVs increased CD16 expression and TNF-␣ secretion in naive monocytes, which were different as compared with alcohol EVs, demonstrating the complex composition of the alcohol EVs and that miR-27a may have other regulatory effects on monocytes.
In vivo analysis of EVs from acute alcoholic hepatitis patients revealed a significant increase in the circulating EV population as compared with healthy controls. Results from the analysis of circulating EVs derived from an AH patient's plasma showed high levels of miR-27a expression. Interestingly, the fold increase of miR-27a in the EVs from AH patients was almost 100-fold higher than that from the EVs of in vitro alcoholtreated monocytes, suggesting other potential cellular sources of miR-27a in vivo. Acute alcoholic hepatitis is characterized by overactivation of the pro-inflammatory cascade; thus the significant increase in miR-27a levels in the circulating EVs from AH patients, as compared with that of EVs from alcohol-exposed monocytes, suggests that in vivo miR-27a might play additional roles in alcoholic hepatitis.
The role of miRs in monocyte/macrophage phagocytosis is not well studied. miR-15a/16 has been shown to regulate macrophage phagocytosis after bacterial infection (49). miR-21 has also been shown to play a role in the process of engulfment of apoptotic cells by macrophages (56). In this study, we show that miR-27a overexpression in the EVs leads to increased phagocytic capacity of the M2 macrophages that were differentiated from naive monocytes. We showed that miR-27a mediated phagocytosis by up-regulating CD206, as inhibiting the up-regulation by mannan or anti-CD206 led to decreased phagocytosis. Studies have shown that increased mannose receptor expression, a C-type lectin expression, leads to increased phagocytic capacity of monocyte-derived DCs and macrophages (32,33). Previous studies have shown that ethanol inhibits the phagocytic activity of macrophages (57,58). Thus, the increased generation of M2 macrophages with high phagocytic capacity in response to alcohol EVs may represent the next generation of macrophages that are generated when the alcohol is metabolized. EVs from alcohol-exposed monocytes still remain in the system to signal to naive monocytes, and this might be a survival mechanism for the monocytes/macrophages. Our data suggest that EV cargos from alcohol-treated monocytes can signal and communicate with naive monocytes. miR-27a plays an important role in this process. Because EVs function as an important method of cellular communication during diseases, it will be interesting to further explore the therapeutic potential of the EVs during alcoholic liver disease. Our in vivo data from alcoholic hepatitis patients suggest an important role of miR-27a in the circulating EVs during ALD. The data suggest that miR-27a is not only important for macrophage differentiation and polarization, but it may also have other regulatory roles during ALD, which is beyond the scope of this study. Our study highlights a novel communication via EVs between immune cells in the presence of ethanol. Thus, further studies of the molecular mechanisms underlying cell-cell communication via the EVs will facilitate a better understanding of the effects of alcohol on the immune cells.