Interactions between macrophages and breast cancer cells result in more invasive cancer cells, metastasis and poor patient prognosis. Preventing malignant breast epithelium-macrophage communication may inhibit the metastatic cascade during cancer progression [2, 9], and thereby, provide important treatment targets for breast cancer therapy. Consequently, the mechanisms underlying breast cancer cell-macrophage interactions require further investigation. Most current studies focus on the cytokines, molecules or enzymes that are secreted by macrophages; however, greater attention is needed on miRNAs due to their regulatory effects on cancer cell progression .
Previously, miRNA profiling has been limited to tissues or cells. Evidence for circulating miRNAs is increasing, and recent studies have detected miRNAs in the peripheral blood of patients . These peripheral blood-isolated miRNAs could potentially be used as biomarkers for certain diseases [33, 34]. In these studies, microvesicles/exosomes were found to serve as mediators of miRNA circulation. Additional studies have suggested that exosomes can be secreted by many cells, including T cells , B cells , mast cells , dendritic cells , cancer cells  and macrophages . Using ultracentrifugation, we isolated exosomes that were released from macrophages. We demonstrated that these macrophage-derived exosomes have classic structures consistent with previous reports . Exosomes have been shown to contain protein, RNA and DNA, and much work has focused on understanding the mechanisms of exosome-mediated cell-cell interactions . Some of these interactions involve communication between different cell types and convey regulatory effectors [41, 42]; whereas, others occur among cells of the same type . Our knowledge regarding communication between macrophages and tumor cells, however, is limited to physical contact and cytokine or chemokine secretion . More recently, human macrophages were found to release exosomes containing migration-promoting enzymes .
Therefore, exosomes may mediate the interactions between macrophages and target cells by shuttling functional miRNAs between them. Within the breast cancer microenvironment, it is possible for macrophages to communicate with cancer cells by transferring miRNAs via exosomes. For this study, we utilized a transwell system to mimic the tumor microenvironment; this system allows macrophages to communicate with breast cancer cells without physical contact. We detected a macrophage-dominant miRNA, miR-223, that showed an increase in breast cancer cells following co-culture with macrophages. To verify miRNA transport between macrophages and breast cancer cells, we transfected Cy3-labeled miR-223 into macrophages and tracked its movement within the co-culture system. Once fluorescently-labeled miRNAs were secreted from the macrophages and captured by breast cancer cells, the fluorescent Cy3 signal could be detected in the breast cancer cells. As demonstrated by confocal microscopy and flow cytometric analyses, Cy3-positive breast cancer cells were observed, which indicates miRNA transfer. Additionally, functional miRNA delivery was confirmed using lin-4, a miRNA that is not expressed in mammalian cells.
Next, we incubated breast cancer cells with exosomes treated with RNase and triton X-100 to degrade the RNA. The depletion of RNAs in exosomes significantly decreased the invasive potential of exosomes derived from IL-4-activated macrophages. In order to determine the specific function of miR-223, we transfected miR-223-ASO into SKBR3 cells and then incubated these cells with exosomes. The invasion-promoting activity of exosomes derived from IL-4-activated macrophages on miR-223-ASO-treated SKBR3 cells was dramatically reduced. Moreover, when miR-223 expression in IL-4-activated macrophages was reduced, the invasiveness of co-cultivated breast cancer cells also decreased. Thus, it is likely that the miR-223 delivered from macrophages, and not the endogenous miR-223, promotes breast cancer cell invasion.
MDMs (monocyte-derived macrophages), prior to activation by IL-4, were not polarized and were relatively quiescent. The invasion potential of breast cancer cells co-cultured with either unactivated macrophages or exosomes released from unactivated macrophages was not different from that observed for the untreated cells. When incubated with IL-4, macrophages became activated and polarized. IL-4, the signature cytokine of a TH-2 type immune response, is a major activator of the TAM phenotype in the tumor microenvironment, and the effects of IL-4 collectively prime TAM with the ability to promote tumor growth, invasion and metastasis . When activated by IL-4, macrophages change their cellular morphology, gene expression profile and secretory pattern . TAMs selectively secrete regulators or effectors that promote tumor proliferation, invasion and metastasis . miR-223 is a miRNA associated with tumor progress. Therefore, when activated by IL-4, macrophages would selectively secrete more invasion-potentiating cytokines, enzymes and miRNAs, including miR-223, that are able to regulate the characteristics of target cells. Our data demonstrate that there are differences between exosomes from unactivated and from IL-4-activated macrophages in terms of size, miR-223 content, and invasion-potential.
There is a profound impact of macrophages on breast cancer cells relative to biological function. Therefore, the identification and verification of these cellular interactions may provide additional strategies to negate the tumor promoting functions of TAMs by targeting those regulators and/or effectors which could have significant potential in anticancer therapies. In our studies, IL-4-activated macrophages efficiently secrete and deliver miR-223 to breast cancer cells, and miR-223 is responsible for macrophage-promoting breast cancer cell invasion. Clinically, based on this data, we could identify macrophage infiltration using CD68 staining, and then separate tumor cells for analysis of miR-223 expression. If greater levels of macrophage infiltration were observed in combination with up-regulation of miR-223 expression in tumor cells, which would indicate a more aggressive breast cancer phenotype according to our data, then the patient may have a poor prognosis and alternative treatments might be necessary. Inhibition of miR-223 in tumor cells and/or TAMs might be helpful. Furthermore, quantification and/or detection of miR-223 in the serum may indicate tumor aggressiveness. Therefore, it may be of great significance to designate miRNAs as diagnostic or prognostic markers for breast cancer. This will require extensive additional work to detect miR-223 in patient serum and calculate the relationship of miR-223 with prognosis, survival rate, etc.
Additionally, we explored the targets of miR-223, and a possible mechanism through which miR-223 may promote breast cancer cell invasion. Our data do not clearly demonstrate the invasion pathway utilized; however, our observations indicate that the involvement of the miR-223/Mef2c/β-catenin pathway is plausible. Additionally, a miRNA has many targets and may function through multiple different pathways. Currently, we are trying to find additional targets and verify their effects using gain- and loss-of-function assays.
The results of this study provide the first evidence suggesting that macrophages can transfer miRNA via exosomes to breast cancer cells. We found that the vesicular miRNA is responsible for macrophage-promoting breast cancer cell invasion, and we have provided a rationale for therapeutically targeting miR-223 in M2 macrophages or exosomal miR-223 from M2 macrophages.