miR-329-containing exosomes derived from breast tumor cells suppress VEGF and KDM1A expression in endothelial cells

1department of Cellular and Molecular Biology, Faculty of Biological Sciences, Islamic azad University-tehran North Branch, tehran, Iran; e-mail: dr.nargesmaleki@yahoo.com; 2Gynecology and reproductive biology department, kowsar poly-clinic, tehran, Iran; 3Cancer research Center, Shahid Beheshti University of Medical Sciences, tehran, Iran; 4appletree Medical group, 275 dundad W (Grange), toronto, Ontario, Canada; e-mail: ghazalehparnian1@gmail.com


the exosomal transfer of mirNas from tumor cells is considered to modulate VeGF expression and angiogenesis in endothelial cells. the aim of our investigation was to focus exclusively on the ability of specific exosomal miR329 to regulate angiogenesis within breast tumor. All experiments were done on MCF-7 and HUVeC cell lines. exosomes were derived from MCF-7 cells both untreated and treated with tamoxifen
that is an effecrive suppressor of hormone receptor-positive breast cancer. The level of miR32 and its targeted genes VEGF and lysine (K)-specific demethylase 1A (KDM1A) expression was estimated with q-RT-PCR. The PKH26 red fluorescent labeling kit was used to label the isolated exosomes and monitor their uptake. It was shown that the relative amount of mir-329 in exosomes was twice as large as in breast cancer cells. Fluorescence microscopy imaging presented that exosomes from MCF-7 cells were able to penetrate into endothelial cells and concentrate in the cytoplasm. It was observed that exosomes derived from untreated breast cancer cells induced kdM1a and VeGF gene expressions whereas exosomes from tamoxifen-treated cancer cells induced time-dependent decrease of kdM1a and VeGF expression in endothelial cells. It is assumed that the transfer of mir-329 containing exosomes from tamoxifen treated breast cancer cells to the endothelial cells could repress angiogenic molecular signaling pathway and be used as a supplementary strategy in breast cancer treatment. k e y w o r d s: breast cancer, angiogenesis, exosome, mir-329, tamoxifen, VeGF, kdM1a. B reast cancer (BC) has the highest mortality prevalence in females [1]. BC tumor is surrounded by a micro-environment of various stromal cell types, which have a role in tumor progression [2], the onset of which is via angiogenesis; the new blood vessel formation process facilitating metastasizing [3].
One of the mechanisms tamoxifen takes advantage of in suppressing breast tumors is its antiangiogenic feature. ERRα-dependent induction of VEGF [4] in the tumor microenvironment by stimulating endothelial cell proliferation and angiogenesis would lead finally to tumor progression, metastasis, and drug resistance [5][6][7]. Tumor neovascularization is originated and developed by the key stimulating factors of endothelial cells, VEGF and its receptors, VEGFR-1 and VEGFR-2 [8]. Although the presence of estrogen receptors on the endothelial cells can assume as the antiangiogenic function of tamoxifen, it may directly inhibit the proliferation of VEGFdependent angiogenesis surrounding the tumor environment [9].
The micro-environment and tumorigenesis of breast cancer tissue is regulated through microRNAs and diffusion of exosomes containing microRNAs [10]. Some microRNAs found in exosomes are thought to be specifically involved in breast tumor progression; for instance, TRPC5, miR-1246, miR-21, miR-34a, Let-7 family, miR-494, miR-210, miR-200 family, miR-221, miR-222, miR-23b, miR-125b, miR-105, miR-29, to name but a few. Since exosomes are a rich and protected source of microRNAs, it is presumed that the exosomal transfer of miRNAs from tumor cells to endothelial cells may be a factor in modulating VEGF expression and regulating angiogenesis. This provides the feasibility of using these regulatory non-coding, endogenous, evolutionarily-conserved RNAs as adjuvant therapy with anti-cancer treatments to enhance therapeutic efficacy [11]. Estrogens regulate miRNAs transcription through estrogen receptors α and β in a tissuespecific and cell-dependent manner. It has been shown that tamoxifen alters the pattern of miRNAs by either downregulating or upregulating them [12]. Hence, the question arises whether the antiangiogenic function of tamoxifen was mediated by altered miRNA expression. MiR-329, as a deterrent factor for angiogenesis, is an "anti-angiomir", which has a negative regulatory effect on endothelial cells functions. The direct target gene of miR-329 as named kdM1a has been shown to have a direct role in the neoangiogene sis in tumor growth [13].
Considering the biological functions of exosomes, we initially evaluated whether tumor exosomes carrying microRNAs can per se regulate VEGF-related angiogenesis in vascular endothelial cells, and subsequently, we sought to determine whether tamoxifen could change the breast cancer angio-miR-329 level in breast cancer cell line MCF-7 to represent potential use of tamoxifen for developing new therapeutic strategies in breast cancer patients.

materials and methods
ethics approval and consent to participate. This work has been carried out in accordance with European Convention for the protection of vertebrate animals used for experimental and other scientific purposes (Strasbourg, 1986).
tamoxifen treatment and cell viability assay (Mtt). In order to reach the appropriate dosage of tamoxifen, cell viability was evaluated by the halfmaximal inhibitory concentration (IC 50 ) through applying 3-(4,5dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to breast cells. In brief, almost 10,000 cells/well were seeded into a 96well plate in RPMI-1640 medium and were allowed to attach overnight. Subsequently, cells were treated with different ascending doses of tamoxifen and incubated for 48 hours (37°C, 5% carbon dioxide in the air). The effects of all doses were assessed in triplicate. Cell growth inhibition was determined using the MTT assay. The medium was removed and 20 µl of the MTT (5 mg/ml) (Sigma, St. Louis, MO, USA) was added to each well of the 96-well plate and were incubated for 4 h at 37°C. Composed crystals dissolved by adding 200 µl of DMSO to each well. Finally, the optimal absorbance level was measured by the ELISA-Reader machine at 570 nm.
Exosome isolation and purification. The cells were cultured in T75 flasks with 5×10 4 cells/cm 2 , complemented with 10% exosome-depleted serum (EXO-FBS; System Biosciences) for 48 h. When the cells reached 80-90% confluency, at passage 3, differential centrifugation was performed. In short, around 700 ml of the media was centrifuged at 3000×g for 10 min at 4°C and then at 10,000×g for 30 min at 4°C to eliminate residual cells and cellular debris, respectively. Extracellular vesicles were pelleted by ExoQuick™ Exosome Precipitation Solution kit (System Biosciences) with a ratio of 5 to 1. Exoquick (system biosciences) solution was added to the supernatant with a ratio of 5 to 1. Isolation steps were followed per the manufacturer's instruction. Finally, exosome pellets were suspended in 50 µl PBS and stored at -20°C until use.
dynamic light scattering analysis. The dynamic light scattering (DLS) was utilized for the determination of exosomes' size, for which Zetasizer Nano ZS (Malvern Instruments, Worcestershire, UK) was used based on the given protocol.
Scanning electron microscopy. The morphology and size of the purified exosomes were measured by scanning electron microscopy (Digital FESEM, KYKY-EM3200, China). For this purpose, an aliquot of the above-mentioned exosomes was fixed in 2.5% glutaraldehyde on a microscope slide, washed by PBS, and an ascending amount of ethanol was utilized for critical point dehydration. The slide was then dried on a glass substrate and coated with a thin layer of gold.
exosome labeling assay. The PKH26 red fluorescent labeling kit (Sigma-Aldrich, USA) was utilized to label the isolated exosomes. The labeled exosomes were incubated with endothelial cells for 12 h. An inverted fluorescence microscope (Olympus CKX41) was finally utilized for monitoring the fluorescence uptake.
transfer of mirNa. In order to assess the exosome transfer of miR-329 into endothelial cells, almost 3×10 4 cells/well were co-incubated at different times with either 100µg/ml exosomes and transcription inhibitor α-amanitin (Sigma, 50 µg/ml) or just with α-amanitin. Subsequently, endothelial cells' total RNA was isolated at time 0, 12, and 48 h post-stimulation. The difference in Ct values between α-amanitin stimulated cells in the presence and absence of exosomes at each intervention time was evaluated in order to measure the exosome transfer of miRNA [15].
Statistical analysis. Data were assessed by oneway ANOVA with Graph-Pad Prism software v.5.0. All data are expressed as mean ± SD for three experiments. P-value < 0.05 was considered statistically significant for all statistical assessments.

Results
Initially, breast cancer cells were treated with various ascending dosages of tamoxifen over 48 h to obtain IC 50 . The sensitivity and tolerable dosage of tamoxifen were determined using the MTT assay. Finally, the concentration of 60μm was considered as IC 50 of tamoxifen for subsequent experiments.
Isolation, characterization and cellular uptake of purified exosomes. Exosome isolation was performed to discover the paracrine effects of breast cancer exosomes on endothelial cell angiogenesis. FESEM analysis indicated that the purified exosomes had a spherical shape with an approximate diameter of ~50-150 nm (Fig. 1, a). To examine whether exosomes could be taken up by endothelial cells, PKH26-labeled exosomes were incubated with endothelial cells for 12 h. The results of fluorescence microscopy imaging presents that exosomes can be concentrated in the endothelial cells' cytoplasm ( Fig. 1, B, C). In addition, the particle size distribution of exosomes by dynamic light scattering revealed a peak of almost 90 nm (Fig. 1, d).
Breast cancer cell-derived mir-329 promote in vitro angiogenesis in endothelial cells through modulating kdM1a and VeGF expression. In order to have a clear description of the molecular mechanisms by which tumor-derived exosomes promote endothelial cell angiogenesis, this paper initially hypothesized that miRNAs shuttled by breast cancer cell-derived exosomes might have a role in the pro-angiogenic effects of tumor-derived exosomes on endothelial cells. Therefore, since miR-329 is considered as a critical miRNA involved in cancer pathogenesis, as well as being enriched in breast cancer cell-derived exosomes [17], this miR has been evaluated in both the donor and the recipient cells. As shown in Fig. 2, it is highlighted that miR-329 was detectable in both breast cancer cell-derived exosomes and the endothelial cells. Noticeably, no considerable amount of miR-329 was detected in tamoxifen-treated breast cancer cell-derived exosomes down-regulated the gene expression of angiogenesis. After tamoxifen-treated/untreated exosomes being encountered with breast cancer cell-derived endothelial cells, the expression level of two angiogenic factors, VeGF and kdM1a, were quantified. It was observed that tumor cell-derived exosomes, which are tamoxifen-untreated, induced more kdM1a and VeGF gene expressions in a time-dependent manner. However, endothelial cells confrontation with tamoxifen-treated exosomes had a time-dependent downward kdM1a and VeGF expression. It can be concluded that the high amount of miR-329 in the exosomes from tamoxifen-treated breast cancer cells, could significantly reduce kdM1a and VeGF expression in the endothelial cells after 24 and 48 h. As are shown in Fig. 3 (a  and B), both kdM1a and VeGF genes were downregulated time-dependently.

discussion
Intercellular interaction plays a vital role in microenvironment equilibrium. Numerous studies support the participation of exosomes in this bioactivity, and since they are filled up with various miRs, there is an ongoing notion that miRs partake in tissue hemostasis and reciprocation [17]. Tumor cells are not excluded from this creed. In other words, in a carcinogenic microenvironment, normal unaffected cells are surrounded by tumor-associated fibroblasts, vas-  cular endothelial cells, inflammatory agents, to name but a few, which have proved negative consequences on their normal adjacent cells [18]. Neovascularization is an essential part of tumor development since newly invaded parts are in need of blood for maintenance and metastasis, which is modulated fundamentally by the tumor microenvironment. Consequently, the alteration of normal cells to cancerous cells occurs [19]. Regulation of vascular endothelial cells' roles is performed by the extracellular matrix (ECM). Vascular Endothelial Growth Factor (VEGF) acts as a mitogen for vascular endothelial cells derived from arteries, veins, and lymph vessels, increases angiogenesis and metastasis, increases vascular permeability, leads to the development of extracellular membrane, and therefore, it helps tumor growth [20].

Fig. 2. exosome transfer of mir-329 up-regulates the in vitro angiogenesis in endothelial cells through modulating kdM1a and VeGF expression. the rela tive amount of mir-329 in tumor cells is considered as the base, and then the relative amount of mir-329 in tumor-derived exosomes has been detected as two times more than the tumor cell-line. In contrast, the relative amount of mir-329 in normal non-cancerous breast cell line and non-tumor exosomes were greatly lower than the base line. all results have been produced with RT-qPCR
Lately, the positive impact of tumor-derived exosomes on VEGF expression and consequently vascular endothelial cells' development has been identified [16] and some papers have suggested the modulating role of microRNAs, one of the exosomes' secretions in this pathway, through the discharge of which messages are transferred from tumorigenic cells to their neighbor's cells [21]. KDM1A is involved in carcinogenesis as one of the important epigenetic regulators. Moreover, it is a direct target gene for miR-329. KDM1A is used as a cancer diagnostic biomarker and if it has a positive expression algorithm, the condition would result in a poor prognosis of cancer. There is a positive and direct relationship between the expression levels of KDM1A and VEGF-A in angiogenesis [22]. Among the predicted targets of miR-329, KDM1A and VEGF are the genes that have been found to be down-regulated in human breast cancer post-tamoxifen-treatment while being up-regulated if being left untreated [23]. In this regard, we initially evaluated the impact of exosomal miR-329 on VEGF and KDM1A signaling in endothelial cells and deduced that tamoxifen-treated, transferred breast cancer exosomes significantly decline the tumor proliferation level in angiogenesis via down-regulation of VEGF and KDM1A in endothelial cells in a time-dependent manner (Fig. 3). To the best of our knowledge, our research was the first to demonstrate the exosomal miRNA-329 from human breast cancer cells may suppress in vitro angiogenesis through modulating angiogenic factors and receptors. In this relation, it was seen that the control subgroup upregulated target gene expression, KDM1A and VEGF, in the endothelial microenvironment while in the intervention group the target genes' expression were downregulated (Fig. 3). The results of our study lend support to the notions of other previous researchers' works, which declare that tamoxifen-treated breast cancer cellular exosomes may suppress in vitro angiogenesis through downward adjusting of VeGF gene via an angiogenic molecular signaling pathway in breast cancer [24,25].
Conclusion. In conclusion, we deduced that exo somes derived from breast cancer cells have the capability to up-regulate VEGF and KDM1A expression and signaling in endothelial cells. Novelty of this project was assessing the impacts of breast cancer cell-derived exosomal miR-329 on behavior of endothelial cells, specifically angiogenesis progression. Finally, by further research on the functions of cancer-derived exosomes and broaden the horizons of tumor microenvironment, evaluation of different new drugs with better tolerance and the least resistance could be sought.