Melanoma originates in the epidermis and becomes metastatic after invasion into the dermis. Prior interactions between melanoma cells and dermis are poorly studied. Here, we show that melanoma cells directly affect the formation of the dermal tumour niche by microRNA trafficking before invasion. Melanocytes, cells of melanoma origin, are specialized in releasing pigment vesicles, termed melanosomes. In melanoma in situ, we found melanosome markers in distal fibroblasts before melanoma invasion. The melanosomes carry microRNAs into primary fibroblasts triggering changes, including increased proliferation, migration and pro-inflammatory gene expression, all known features of cancer-associated fibroblasts (CAFs). Specifically, melanosomal microRNA-211 directly targets IGF2R and leads to MAPK signalling activation, which reciprocally encourages melanoma growth. Melanosome release inhibitor prevented CAF formation. Since the first interaction of melanoma cells with blood vessels occurs in the dermis, our data suggest an opportunity to block melanoma invasion by preventing the formation of the dermal tumour niche.

Malignant melanoma is the deadliest of skin cancers. Melanoma frequently metastasizes to the brain, resulting in dismal survival. Nevertheless, mechanisms that govern early metastatic growth and the interactions of disseminated metastatic cells with the brain microenvironment are largely unknown. To study the hallmarks of brain metastatic niche formation, we established a transplantable model of spontaneous melanoma brain metastasis in immunocompetent mice and developed molecular tools for quantitative detection of brain micrometastases. Here we demonstrate that micrometastases are associated with instigation of astrogliosis, neuroinflammation, and hyperpermeability of the blood-brain barrier. Furthermore, we show a functional role for astrocytes in facilitating initial growth of melanoma cells. Our findings suggest that astrogliosis, physiologically instigated as a brain tissue damage response, is hijacked by tumor cells to support metastatic growth. Studying spontaneous melanoma brain metastasis in a clinically relevant setting is the key to developing therapeutic approaches that may prevent brain metastatic relapse.

Breast tumors are characterized by an extensive desmoplastic stroma, abundantly populated by fibroblasts. Cancer-associated fibroblasts (CAF) support tumorigenesis by stimulating angiogenesis, cancer cell proliferation, and invasion. CAF also orchestrate tumor-promoting inflammation in multiple tumor types, including breast cancer. However, the mechanisms through which normal tissue fibroblasts are reprogrammed to tumor-promoting CAFs are mainly obscure. Here, we show that mammary fibroblasts can be educated by breast cancer cells to become activated to a proinflammatory state that supports malignant progression. Proteomic analysis of breast cancer cell-secreted factors identified the secreted proinflammatory mediator osteopontin, which has been implicated in inflammation, tumor progression, and metastasis. Osteopontin was highly secreted by mouse and human breast cancer cells, and tumor cell-secreted osteopontin activated a CAF phenotypes in normal mammary fibroblasts in vitro and in vivo. Osteopontin was sufficient to induce fibroblast reprogramming and neutralizing antibodies against osteopontinblocked fibroblast activation induced by tumor cells. The ability of secreted osteopontin to activate mammary fibroblasts relied upon its known receptors CD44 and a V b 3 integrin. Strikingly, osteopontin silencing in tumor cells in vivo attenuated stromal activation and inhibited tumor growth. Our findings establish a critical functional role for paracrine signaling by tumor-derived osteopontin in reprograming normal fibroblasts into tumor-promoting CAFs. Cancer Res; 75(6); 963-73. Ó2015 AACR.

Melanoma is the leading cause of skin cancer mortality. The major cause of melanoma mortality is metastasis to distant organs, frequently to the brain. The microenvironment plays a critical role in tumourigenesis and metastasis. In order to treat or prevent metastasis, the interactions of disseminated tumour cells with the microenvironment at the metastatic organ have to be elucidated. However, the role of brain stromal cells in facilitating metastatic growth is poorly understood. Astrocytes are glial cells that function in repair and scarring of the brain following injury, in part via mediating neuroinflammation, but the role of astrocytes in melanoma brain metastasis is largely unresolved. Here we show that astrocytes can be reprogrammed by human brain-metastasizing melanoma cells to express pro-inflammatory factors, including the cytokine IL-23, which was highly expressed by metastases-associated astrocytes in vivo. Moreover, we show that the interactions between astrocytes and melanoma cells are reciprocal: paracrine signalling from astrocytes up-regulates the secretion of the matrix metalloproteinase MMP2 and enhances the invasiveness of brain-metastasizing melanoma cells. IL-23 was sufficient to increase melanoma cell invasion, and neutralizing antibodies to IL-23 could block this enhanced migration, implying a functional role for astrocyte-derived IL-23 in facilitating the progression of melanoma brain metastasis. Knocking down the expression of MMP2 in melanoma cells resulted in inhibition of IL-23-induced invasiveness. Thus, our study demonstrates that bidirectional signalling between melanoma cells and astrocytes results in the formation of a pro-inflammatory milieu in the brain, and in functional enhancement of the metastatic potential of disseminated melanoma cells.

The mononuclear phagocytic system is categorized in three major groups: monocyte-derived cells (MCs), dendritic cells and resident macrophages. During breast cancer progression the colony stimulating factor 1 (CSF-1) can reprogram MCs into tumor-promoting macrophages in the primary tumor. However, the effect of CSF-1 during colonization of the brain parenchyma is largely unknown.Thus, we analyzed the outcome of anti-CSF-1 treatment on the resident macrophage population of the brain, the microglia, in comparison to MCs, alone and in different in vitro co-culture models. Our results underline the addiction of MCs to CSF-1 while surprisingly, microglia were not affected. Furthermore, in contrast to the brain, the bone marrow did not express the alternative ligand, IL-34. Yet treatment with IL-34 and co-culture with carcinoma cells partially rescued the anti-CSF-1 effects on MCs. Further, MC-induced invasion was significantly reduced by anti-CSF-1 treatment while microglia-induced invasion was reduced to a lower extend. Moreover, analysis of lung and breast cancer brain metastasis revealed significant differences of CSF-1 and CSF-1R expression.Taken together, our findings demonstrate not only differences of anti-CSF-1 treatment on MCs and microglia but also in the CSF-1 receptor and ligand expression in brain and bone marrow as well as in brain metastasis.