Exosomal miRNA Cargo as Mediator of Immune Escape Mechanisms in Neuroblastoma

Abstract

Both natural killer (NK) cells and exosomes released from these cells induce tumor cell cytotoxicity by way of the cell killing proteins perforin and granzyme. TGFβ1 protein in the tumor microenvironment generates an immune escape mechanism rendering NK cells inactive. The tumor-suppressive miR-186 that is downregulated in neuroblastoma and in TGFβ-treated NK cells represses oncogenic proteins in neuroblastoma (MYCN and AURKA) and components of the TGFβ pathway. Restoration of miR-186 levels in neuroblastoma through NK cell–derived exosomes or by nanoparticle delivery reduces tumor burden, promotes survival, and restores the cell-killing abilities of NK cells, demonstrating the therapeutic potential of tumor-suppressive miRNAs in neuroblastoma.

Neuroblastoma is the most common cancer in infants and the most prevalent extracranial childhood cancer. Although neuroblastoma is a relatively rare disease, with approximately 800 new cases per year in the USA, the 5-year survival rate of high-risk neuroblastoma is only 40% to 50%. Although the etiology of neuroblastoma is not completely understood, a number of genetic alterations are present in both familial and sporadic neuroblastoma. Among the genes most frequently altered in neuroblastoma is MYCN, which is amplified in about 25% of all neuroblastoma cases and whose expression positively correlates with advanced disease stage (1). High levels of MYCN protein in neuroblastoma patients correlate with metastasis and other hall-marks of poor prognosis (2). MYCN works in concert with other proteins including the mitotic kinase aurora kinase A (AUKRA). The AUKRA protein binds to MYCN and enhances its activity by preventing protein destabilization (3). Because MYCN protein is believed to be undruggable, novel approaches to therapeutically target MYCN and its affiliated proteins are warranted.

The tumor microenvironment is a milieu that surrounds tumor cells and consists of cellular and noncellular components such as fibroblasts, extracellular matrix, blood vessels, immune cells, signaling pathways, inflammatory cells, and lymphocytes. This dynamic setting contributes to tumorigenesis through complex interactions of these components. For example, CD163⁺ tumor–associated macrophages are abundant in the neuroblastoma microenvironment and produce immunosuppressive cytokines that activate signaling pathways such as TGFβ that promote tumorigenesis and metastasis (4). Activated natural killer (NK) cells in the tumor microenvironment may limit the development of cancer when their cytotoxic activity is maintained by the presence of a supportive profile of cytokines. In high-risk neuroblastoma, however, a type of immune escape mechanism exists whereby high levels of TGFβ1 produce chemokine receptor modulation and compromises the cell-killing activity of NK cells (5).

Exosomes are the nanosized vesicles that are shed from virtually all cells and are present in nearly all biofluids. Exosomes contain a variety of RNA and protein cargo and this cargo may be transferred from one cell to another (i.e., between cells of the immune system and cancer cells) via exosomes. Exosomes released from NK cells possess cell-killing activity toward cancer cells through the cytotoxicity factors perforin 1 and granzymes A and B (6, 7). Although the release of these cell-killing proteins from the exosomes is one source of their cell-killing activity, additional mechanisms to explain the cytotoxicity of NK exosomes is possible.

In a previous issue of Cancer Research, Neviani and colleagues report that noncoding RNA, specifically miR-186, present in NK cell exosomes inhibits tumor growth and regulates TGFβ-dependent immune escape mechanisms in neuroblastoma (8). The relationship between miR-186 and the cytotoxicity of NK cell–derived exosomes was demonstrated in a series of elegant experiments. Exosomes were collected from NK cells that were isolated from the peripheral blood of healthy donors. The cytotoxicity of exosomes isolated from inactivated NK cells was compared with those isolated from NK cells activated by IL15. As anticipated, the exosomes derived from activated NK cells exerted cytotoxicity toward MYCN-amplified neuroblastoma cells that was equivalent to the NK cells. However, when the cell-killing ability of the TGFβ inactivated the NK cells and their exosomes were compared with those exosomes derived from IL15-activated NK cells, the TGFβ-treated NK cells possessed less cytotoxicity, whereas the activity of their exosomes remained intact. Of note, the exosomes from the inactivated NK cells expressed lower levels of classical killer proteins, suggesting that factors other than these proteins contributed to their cell-killing activity.

miR-186 was previously reported to possess tumor suppressive properties in breast cancer (9) and other solid and hematologic malignancies. The expression of tumor-suppressive miRNAs is typically reduced in the tumor compared with normal tissue. miR-186 is downregulated in MYCN-amplified neuroblastomas (10). Neviani and colleagues reported that miR-186 was downregulated in patients with high-risk compared with low-risk neuroblastoma and those with low expression of miR-186 showed significantly shorter event-free survival and overall survival probability. Interestingly, they showed that miR-186 targets not only MYCN and AUKRA but also TGFBR1 and TGFBR2. miR-186 levels were reduced in TGFβ-treated NK cells but were increased in NK cells activated with IL15. NK cell lines transfected with miR-186 mimic had reduced cell proliferation, migration, and invasion compared with control oligonucleotide.

To evaluate miR-186 as a potential therapeutic in neuroblastoma, Neviani and colleagues formulated anionic lipopolyplex nanoparticles that contained the antibody to CD56 on their surface to enhance targeting to NK cells. The CD56 nanoparticles loaded with miR-186 oligonucleotide negated the TGFβ inactivation of the NK cells, suggesting that miR-186 could reactivate NK cells in the patient’s tumor microenvironment that are otherwise idle by TGFβ signaling. Moreover, when miR-186 expression was inhibited in the NK cell exosomes by a miR-186 antisense oligonucleotide, their cytotoxic activity was significantly reduced. In vivo activity was established using an orthotopic mouse model of the neuroblastoma and GD2-positive and MYCN-overexpressing cell line, CHLA-136, implanted into the kidney. Formulation of a nanoparticle containing the anti-GD2 antibody increased intratumoral miR-186 expression. Furthermore, miR-186–loaded GD2 nanoparticles, but not several control nanoparticles, reduced tumor burden and promoted survival.

The findings of Neviani and colleagues are highly significant and suggest that reduced expression of miR-186 in the tumor microenvironment of patients with neuroblastoma not only contributes to the oncogenic activity of MYCN, but also to the immune escape mechanism that the TGFβ pathway has on NK cells. This activity is a direct result of miR-186 repressing MYCN and AURKA proteins and attenuating the NK cell inactivating abilities of TGFβ pathway. Although this study paves the way for additional preclinical exploration of miRNA and exosome therapy for neuroblastoma, it should be noted that the in vivo experiments were conducted using immunocompromised mice. Future studies using immunocompetent mice should be performed to further support the role that miR-186 and NK-derived exosomes has on reducing tumorigenesis of neuroblastoma.

Experiments to assess miR-186 replacement therapy were conducted using nanoparticle delivery of miR-186 oligonucleotide in vitro and in vivo. Although both approaches were effective, it is possible that the in vivo activity could be enhanced even further by using exosomes derived from activated NK cells. In addition, proteins and RNA cargo other than miR-186 could contribute to the exosome’s ability to modulate the immune escape mechanism of NK cells. For example, NK-cell exosomes were previously reported to activate caspase-3, −7, and −9 in the target cells (6). To achieve the potential of activated NK-cell exosomes as a therapy in the clinic, the means to reproducibly manufacture therapeutic exosomes must be achieved. The Fabbri lab and collaborators have made progress in this area by developing the means to upscale and purify the exosomes from activated NK cells (6). NK cell–derived exosomes along with their miR-186 cargo were effective in a preclinical mouse model of neuroblastoma. Thus, ex vivo–derived NK exosomes may be combined along with NK cell–based immunotherapy as a potential therapeutic option for high-risk neuroblastoma.