Exosome-mediated delivery of miR-9 induces cancer-associated fibroblast-like properties in human breast fibroblasts

Abstract

It is established that the interaction between microenvironment and cancer cells has a critical role in tumor development, given the dependence of neoplastic cells on stromal support. However, how this communication promotes the activation of normal (NFs) into cancer-associated fibroblasts (CAFs) is still not well understood. Most microRNA (miRNA) studies focused on tumor cell, but there is increasing evidence of their involvement in reprogramming NFs into CAFs. Here we show that miR-9, upregulated in various breast cancer cell lines and identified as pro-metastatic miRNA, affects the properties of human breast fibroblasts, enhancing the switch to CAF phenotype, thus contributing to tumor growth. Expressed at higher levels in primary triple-negative breast CAFs versus NFs isolated from patients, miR-9 improves indeed migration and invasion capabilities when transfected in immortalized NFs; viceversa, these properties are strongly impaired in CAFs upon miR-9 inhibition. We also demonstrate that tumor-secreted miR-9 can be transferred via exosomes to recipient NFs and this uptake results in enhanced cell motility. Moreover, we observed that this miRNA is also secreted by fibroblasts and in turn able to alter tumor cell behavior, by modulating its direct target E-cadherin, and NFs themselves. Consistently with the biological effects observed, gene expression profiles of NFs upon transient transfection with miR-9 show the modulation of genes mainly involved in cell motility and extracellular matrix remodeling pathways. Finally, we were able to confirm the capability of NFs transiently transfected with miR-9 to promote in vivo tumor growth. Taken together, these data provide new insights into the role of miR-9 as an important player in the cross-talk between cancer cells and stroma.

Figure 1

MiR-9 expression in primary NF/CAF couples. qRT-PCR analysis performed on CAFs and their counterpart NFs isolated from patients affected with different breast cancer subtypes. Data are presented as the mean±S.D. (*P<0.05)

Figure 2

MiR-9 affects cell motility in NFs and CAFs. (a) Migration assays, by transwell (upper panel) and wound healing (lower panel), of NFs after transient transfection with control or miR-9. (b) Invasion assay of NFs transiently transfected with control or miR-9. (c) CAF migration (left panel) and invasion (right panel) after transient transfection with control or LNA-9. The migrated or invaded cells are shown by histograms. Data are presented as the mean±S.D. of three views (*P<0.05; ***P<0.0005). Scale bars, 100 μm

Figure 3

MiR-9 is delivered to NFs via exosomes and promotes cell motility. (a and b) qRT-PCR analysis to evaluate miR-9 level in MDA-MB-231 transiently transfected with control or miR-9 and in exosomes purified from tumor cell supernatants, respectively. (c) MDA-MB-231-secreted exosomes were fed on NFs for 72 h, then RNA was extracted from the recipient cells and analyzed for miR-9 level by qRT-PCR. The data are shown as normalized relative to miR-21 (exosomes) or RNU44 (MDA-MB-231 and NFs, respectively) (***P<0.0005). (d) Migration by transwell (left panel) and invasion assays (right panel) of recipient NFs after miR-9 internalization. Quantitative analysis of the experiments was shown in the lower histograms. Data are presented as the mean±S.D. of three views (*P<0.05). Scale bars, 100 μm

Figure 4

MiR-9 released by microenvironment to neoplastic cells enhances tumor progression. (a) Migration assay of MDA-MB-231 (left panel) and MDA-MB-468 (right panel) co-cultured with conditioned medium derived from NFs transiently transfected with control or miR-9. Quantitative analysis of the experiments was shown in the histograms. Data are presented as the mean±S.D. of three views (*P<0.05). (b) Western blot analysis of E-cadherin expression in MDA-MB-468 after miR-9 internalization. (c) Migration (left panel) and invasion (right panel) assays of NFs after incubation with conditioned medium from NFs transiently transfected with control or miR-9. The migrated or invaded cells are shown by histograms. Data are presented as the mean±S.D. of three views (*P<0.05; **P<0.005). Scale bars, 100 μm

Figure 5

Differentially expressed genes in NFs overexpressing miR-9. (a) Hierarchical clustering analysis of miR-9 exogenous expressing in NFs. Heatmap: rows correspond to differentially expressed genes and columns to samples. Red represents elevated and green downmodulated expression. (b) Validation through qRT-PCR analysis of the differentially expressed genes related with cell motility and ECM organization. The relative expression levels are shown as fold change of NFs/miR-9 versus NFs/control. (c) Interaction network of the significantly enriched gene ontologies and pathways of the differentially expressed genes in the miR-9 transient transfection model. Green and red edges represent the down- or upmodulated pathways, respectively, according to the expression of the connected genes (blue node). (d) Boxplots showing the expression levels of the three selected genes in two public gene expression data sets of tumor (TS) and normal stroma (NS) from human breast specimens. P-values from two-tailed Student's t-test are reported

Figure 6

NFs overexpressing miR-9 promote in vivo tumor growth. Evaluation of tumor volumes in SCID mice co-injected with MDA-MB-468 cells and NFs transiently transfected with control or miR-9. The control group was injected with MDA-MB-468. Data are presented as the mean±S.D. (n=6; *P<0.05)