Enriched Environment Cues Suggest a New Strategy to Counteract Glioma: Engineered rAAV2-IL-15 Microglia Modulate the Tumor Microenvironment

Several types of cancer grow differently depending on the environmental stimuli they receive. In glioma, exposure to an enriched environment (EE) increases the overall survival rate of tumor-bearing mice, acting on the cells that participate to define the tumor microenvironment. In particular, environmental cues increase the microglial production of interleukin (IL)-15 which promotes a pro-inflammatory (antitumor) phenotype of microglia and the cytotoxic activity of natural killer (NK) cells, counteracting glioma growth, thus representing a virtuous mechanism of interaction between NK cells and microglia. To mimic the effect of EE on glioma, we investigated the potential of creating engineered microglia as the source of IL-15 in glioma. We demonstrated that microglia modified with recombinant adeno-associated virus serotype 2 (rAAV2) carrying IL-15 (rAAV2-IL-15), to force the production of IL-15, are able to increase the NK cells viability in coculture. Furthermore, the intranasal delivery of rAAV2-IL-15 microglia triggered the interplay with NK cells in vivo, enhancing NK cell recruitment and pro-inflammatory microglial phenotype in tumor mass of glioma-bearing mice, and ultimately counteracted tumor growth. This approach has a high potential for clinical translatability, highlighting the therapeutic efficacy of forced IL-15 production in microglia: the delivery of engineered rAAV2-IL-15 microglia to boost the immune response paves the way to design a new perspective therapy for glioma patients.

INTRODUCTION Among brain malignancies, glioma is the most diffuse tumor, and its prognosis is significantly correlated with pathological grades. Malignant glioblastoma (GBM) (classified as >IV grade of glioma by the World Health Organization (WHO)) creates an immunosuppressive microenvironment that supports its own growth, leaving few disease-free months to patients (1). The pattern of cytokines produced by glioma and infiltrating cells dramatically alters the tumor-suppressive functions of innate and adaptive immune cells, hampering their antitumor activity (2). Microglia/macrophages (M/M F ) invade glioma with high efficiency, representing up to 30% of tumor mass (3), but are "tricked" to favor tumor growth: tumor-derived cytokines cooperate to polarize M/M F toward a pro-tumoral phenotype, characterized by the production of anti-inflammatory, proinvasive cytokines promoting tumor proliferation, angiogenesis, and invasion. By contrast, M/M F with a pro-inflammatory phenotype are endowed with antitumor activities (4).
Consistently to a pro-tumoral microenvironment, also natural killer (NK) cells are silenced by glioma. The maturation and survival of NK cells require IL-15, which also stimulates interferon (IFN)-g production (5), and mice lacking IL-15 or one of its signaling components are devoid of NK cells (6). Activated NK cells are able to exert antitumor function by direct cytotoxic activity, producing several cytokines and chemokines (7)(8)(9). Among the cytokines, IFN-g contributes to the antitumor activity in different neoplasms by inhibiting tumor cell proliferation and angiogenesis (10,11). Furthermore, IFN-g counters the acquisition of a pro-tumor phenotype by M/M F and reverses toward a pro-inflammatory phenotype (11,12). Nevertheless, invasion of glioma by NK cells is weak and conflicting results are reported on the impact of NK cells on GBM clinical outcome (13)(14)(15). In addition, the presence of glioma cells attenuates NK cell cytotoxic activity (16).
Over the past few years, many pieces of evidence highlighted the beneficial effects of housing animals in an enriched environment (EE) in physio-pathological conditions, obtained with prolonged voluntary physical exercise, social interactions, and sensory stimulation. EE has been shown to shape neuronal circuits and to reduce the impact of several neurodegenerative diseases and cancers both in animal models and in patients (11,(17)(18)(19)(20). In addition, environmental stimuli also shape immune functions, modulating NK cells (20,21), T helper cells (22), and myeloid cells (23).
We have recently shown that EE exposure reduces the tumor size and proliferation rate of glioma cells, with increased survival in human and murine glioma mouse models (20). Exploring the mechanisms involved in the effects induced by EE, we described an increased production of IL-15 by microglial cells, with effects on NK cell cytotoxicity and IFN-g production. EE exposure also affects the phenotype of microglia, switching them toward an antitumor state (11). Consistently, we demonstrated that local administration of IL-15 in tumor mass directly modulates glioma microenvironment, re-boosting the immune response against tumor cells and activating the interaction between NK cells and microglia (11).
In this study, we explored the potential use of microglia engineered to express IL-15 upon infection with a recombinant adeno-associated virus serotype 2 (rAAV2) carrying IL-15 (rAAV2-IL-15), to simulate the effect of EE on glioma. AAV are considered safe for human gene therapy and have been successfully used to target several cell types within the central (CNS) and peripheral nervous systems (PNS), including neurons, oligodendrocytes, astrocytes, Müller glia, Schwann cells, and also microglia (24). We demonstrated that rAAV2-IL-15 microglia increase the viability of cocultured NK cells, without affecting their activation state in vitro. Furthermore, the intranasal (i.n.) administration of rAAV2-IL-15 microglia to glioma-bearing mice reaches the tumor mass, with effects on the tumor volume, NK cell recruitment, and M/M F phenotype.
Altogether, our data support the key role of microglia-derived IL-15 in regulating the glioma microenvironment and boosting immune cell activation.

Mice and Cell Lines
The experiments described in the present work were approved by the Italian Ministry of Health in accordance with the guidelines on the ethical use of animals from the European Community Council Directive of September 22, 2010 (2010/63/EU). We used C57BL/6 mice from Charles River Laboratories (Calco, Italy). The heterozygous Cx3cr1 +/GFP was from The Jackson Laboratory. We always used 2-month-old male mice.

Primary Microglial Cultures and Microglial Infection
Microglial cultures were obtained from mixed glial cultures derived from the cerebral cortices of postnatal day 0-1 (P0-P1) C57BL/6 mice, as described in [20]. In brief, cortices were chopped and digested in 15 U/ml papain for 20 min at 37°C. Cells (5 × 10 5 cells/cm 2 ) were plated on flasks coated with poly-L-lysine (100 mg/ml) in DMEM supplemented with 10% FBS, 100 U/ml penicillin, and 0.1 mg/ml streptomycin. After 7-9 days, cells were shaken for 2 h at 37°C to detach and collect microglial cells. These procedures gave almost pure microglial cell populations. The remaining attached cells were astrocytes. Microglia (18×10 4 ) were incubated for 7 days in 24 wells with serial dilutions of rAAV2-GFP or rAAV2-IL-15.

Coculture of Primary Microglia-NK Cells
To analyze NK cell degranulation, purified NK cells from the spleen of C57BL/6 mice (using Miltenyi MACS Isolation Kit #130-115-818) were activated O/N in IL-15 (50 ng/ml) and then coincubated with primary culture of microglia infected with rAAV2-GFP/IL-15 for 7 days for 4 h at the ratio 1:1. NK cells were collected from the non-adherent fraction and analyzed by FACS.

Flow Cytometry FACS
NK cells were collected from the microglial coculture and stained for 20 min with fluorochrome-conjugated antibodies. All cells were analyzed by flow cytometry using FACSCanto II (BD Biosciences), and data were elaborated using FlowJo Version 9.3.2 software (TreeStar).

Intracranial Injection of Glioma
Male C57BL/6 mice were anesthetized with chloral hydrate (400 mg/kg, i.p.) and placed in a stereotaxic head frame. Animals were stereotactically injected with 7.5 × 10 4 GL-261: a median incision of~1 cm was made, a burr hole was drilled in the skull, and cells were injected 2 mm lateral (right) and 1 mm anterior to the bregma in the right striatum. Cell suspensions in PBS (4 µl) were injected with a Hamilton syringe at a rate of 1 ml/min at a 3-mm depth. After 17 days, animals were sacrificed for different analyses.

Intranasal Microglial Delivery
Starting at 7 days after glioma injection, mice were randomly grouped for the treatments. Thirty minutes before the treatment, mice were anesthetized using isofluorane and treated with 6 µl of hyaluronidase from bovine testes 10 mg/ml per nostril. Then, mice were anesthetized as described before and rAAV2-GFPand rAAV2-IL-15-infected microglia and saline solution were intranasally administrated at the concentration of 1×10 6 cells/ 24 µl of saline solution. Mice were treated with 24 µl of the three treatments at a rate of 6 µl per nostril every 2 min after 7 and 10 days from gliomal inoculation.

Histopathological Evaluation of Tumor Volume
After 17 days from glioma cell injection, brains were isolated for morphological evaluation of tissues and fixed in 4% buffered formaldehyde. Coronal brain sections (20 µm) were prepared by standard procedures and stained with hematoxylin and eosin. A section every 100 µm was collected, and the tumor area was evaluated using ImageTool 3.00.

Immunostaining
Seventeen days after injection of GL261 cells, mice were sacrificed and the brains fixed in 4% formaldehyde. Subsequently, the brains were incubated in 30% sucrose solution for 48 h and then snap frozen. Cryostat sections (10 µm) were boiled for 10 min in citrate buffer, pH 6.0, at 95°C-100°C, then rinsed in PBS and incubated with 3% goat serum in 0.3% Triton X-100 for 1 h at room temperature and then overnight at 4°C with specific antibodies in PBS containing 1% goat serum and 0.1% Triton X-100. The sections were stained with the following primary Abs: anti-IL-15 (1:100), anti-Iba1 (1:500), anti-NKp46 (1:50), and anti-Arg1 (1:500). After several washes, sections were stained with the fluorophore-conjugated antibody and Hoechst (1:3,000) for nucleus visualization and analyzed by fluorescence microscopy.

Image Acquisition and Data Analysis
Images were digitized using a CoolSNAP camera (Photometrics) coupled to an ECLIPSE Ti-S microscope (Nikon) and processed using MetaMorph 7.6.5.0 image analysis software (Molecular Devices). Brain slices were scanned by consecutive fields of vision (×10 objective lens) to build a single image per section. The percentage of positive cells was measured as the ratio of the area occupied by fluorescent cells versus the total tumor area (by converting pixels to square millimeters). For comparison between different treatments, at least 12 coronal sections per brain around the point of injection were analyzed.

Morphological Analysis of Iba1+ Cells
Brain slices from perfused GL261-beraing mice mice were analyzed by confocal microscopy and skeleton analysis to assess Iba1+ cell morphology. Twenty-millimeter z-stacks were acquired at 0.5-µm intervals using an FV1000 laser scanning microscope (Olympus) at ×60 objective. Maximal intensity projections of each image were generated, binarized, and skeletonized using the Skeletonize 2D/3D plugin in ImageJ, after which the Analyze Skeleton plugin (http://imagej.net/ AnalyzeSkeleton) was applied. The average branch number (process end points per cell) and length per cell were recorded for each image with a voxel size exclusion limit of 150 applied. The number of single-and multiple-junction points was additionally calculated to give an indication of branching complexity. The areas of the soma and scanning domain were measured for each cell. Acquisition files were then processed with ImageJ software for two-dimensional analysis.

Environmental Enrichment
Mice were housed for 5 weeks in EE: 10 mice for cage (36 cm × 54 cm × 19 cm), in the presence of an assortment of objects, including climbing ladders, seesaws, running wheel, balls, plastic and wood objects suspended from the ceiling, paper, cardboard boxes, and nesting material. Toys were changed every 2 days, and the bedding every week. Control SE mice were in pair, in standard cages: 30 cm × 16 cm × 11 cm.

Statistical Analysis
Data are shown as the mean ± SEM. Statistical significance was assessed by Student's t-test or one-way ANOVA for parametrical data, as indicated; Holm-Sidak test was used as a post hoc test; Mann-Whitney rank test and Kruskal-Wallis for nonparametrical data; and Dunn's or Tukey's post hoc tests. For multiple comparisons, multiplicity-adjusted p-values are indicated in the corresponding figures (*p 0.05, **p 0.01). Statistical analyses comprising calculation of degrees of freedom were done using SigmaPlot 11.0 and Prism 7 software.  Figures 1A, B). Co-incubation of rAAV2-IL-15 microglia with NK cells almost doubled the number of NK cells after 72 h (NK cells with rAAV2-GFPmicroglia: 25.7 ± 5 per well; NK cells with rAAV2-IL-15microglia 50.0 ± 7 per well, n = 6 *p < 0.05 one-way ANOVA). This effect was due to the release of IL-15, because it was completely abolished in the presence of an IL-15neutralizing antibody ( Figure 1C). Interestingly, rAAV2-IL-15 microglia did not increase the number of NKG2D+ and CD69+ NK cells in culture ( Figure 1D), indicating a specific effect on proliferation.

Intranasally Delivered Microglia Enter the Brain and Reach Glioma
We then used microglia isolated from CX3CR1 GFP/mice to express IL-15 (rAAV2-IL-15-CX3CR1 GFP/microglia); these cells were then delivered through intranasal administration to gliomabearing mice to locally increase IL-15 expression. First, we followed the route of intranasally delivered CX3CR1 GFP/microglia in the brain of wild-type mice: data shown in Figure 2A report that after 3 days, GFP fluorescence was detectable in different brain structures, with a descending gradient from the point of administration (olfactory bulb). Interestingly, GFP/IBA1 double-positive cells accumulated in the tumor mass of glioma-bearing mice, representing approximately 20% of total M/M F infiltrating the tumor ( Figure 2B).

Intranasally Delivered rAAV2-IL-15 Microglia Reduce Glioma Growth and Modulate the Tumor Microenvironment In Vivo
To investigate the efficacy of IL-15 produced by engineered microglia against glioma, we intranasally delivered these cells every 3 days starting 7 days after GL261 glioma cell transplantation in mice (see scheme in Figure 3A). We reported that mice that received rAAV2-IL-15 microglia had a higher number of IL-15/IBA1 double-positive cells in the tumor area, 17 days after glioma transplantation ( Figure 3B). Furthermore, these mice had reduced glioma volume, similar to that observed in EE-housed mice, and in contrast with mice receiving vehicle or rAAV2-GFP microglia (n = 5-7 mice, veh 4.45 ± 0.72 mm 3 ; rAAV2-GFP 4.52 ± 0.75 mm 3 ; rAAV2-IL-15 1.47 ± 0.34 mm 3 ; EE 0.72 ± 0.37 mm 3 ) ( Figure 3C). We previously demonstrated that EE exposure increases the accumulation of NK cells in the tumor mass with an IL-15-dependent mechanism (20). Here, we demonstrated that the delivery of rAAV2-IL-15 microglia enhances the NK cell accumulation in the tumor area, boosted also with respect to the effects of EE ( Figure 3D).
Since NK cells in glioma modulate the microglial phenotype (11,20), we investigated the M/M F phenotype in mice that  Figure 4A). Furthermore, M/M F cell morphology (stained with the marker Iba1) was analyzed by twophoton microscopy, revealing alterations of cell branching and territory (i.e., mean area covered by single cells). In detail, our data show that in the tumoral region of rAAV2-IL-15 microglion-treated mice, GFP+ cells have an increased number of branches and cover a wider parenchymal region ( Figure 4B).

DISCUSSION
In the last years, one big effort of biomedical research has been to try to identify the mediators of the effects of the correct lifestyle on human health. The possibility to mimic the effects of environmental signals would pave the road for new therapeutic approaches for patients suffering from non-communicable diseases (29,30).
In addition to the known effects of the EE on brain functions and cognition (31,32), recent evidence also demonstrated the possibility to reduce the progression of several types of cancer in murine and humanized mouse models, modulating the immune response and the pattern of cytokines released from the parenchyma and tumor cells (tumor microenvironment) (33)(34)(35). In humans, the notion that environmental stimuli can affect cancer progression has long been suspected (36). Clinical and epidemiological studies have recognized that specific psychosocial factors such as stress, chronic depression, eustress, and social support can affect the development and progression of cancer (37)(38)(39). Unfortunately, to date, despite the relevance of the issue, the translation of the environmental stimuli to pharmacological therapy for cancer patients is far to be reached.
In mouse models of melanoma and colon cancer, it was reported that living in an enriched housing environment reduces tumor growth and increases remission (19). Mechanistically, EE increased the neurothrophic brain-derived neurotrophic factor (BDNF) in the hypothalamus of tumor-bearing mice that, in turn, selectively reduced letin production via sympathoneural b-adrenergic signaling with direct effect on cancer cells (19). In glioma, we have recently demonstrated that 5 weeks of EE exposure reduce the tumor volume, further hampering the glioma insurgence in mouse models (20). One key molecule produced by microglial cells upon EE exposure is IL-15, which modifies the tumor microenvironment boosting the immune system to counteract glioma growth (11). IL-15 is a crucial cytokine for the development, maturation, and activation of NK cells and CD8+ T cells, with no effect on the expansion of the T regulatory cell population involved in suppressing immune responses, and thus has a potential therapeutic use in cancer immunotherapy (20,40,41). It has been shown that IL-15 also enhances the antitumor efficacy of the extracellular vesicles derived from NK cells, showing higher cytolytic activity against GBM (42). However, the therapeutic efficacy of IL-15 in glioma has not been tested until recently, partially due to its short half-life and low biological activity in vivo (43), and the systemic, non-local route of administration (44). Accordingly, a continuous and local action of IL-15 in the tumor microenvironment would be fundamental to counteracting GBM growth and progression.
In the last 10 years, few pharmacological approaches have been tested to take advantage of IL-15 for the treatment of GBM patients: a novel double-controlled oncolytic adenovirus driven by the Ki67 promoter and armed with IL-15 selectively infects and kills GBM cells, while sparing normal cells, and significantly enhances the activation of microglial cells (45). Furthermore, the administration of an IL-15 superagonist complex (ALT-803) leads to long-term survival and antitumor immune response in murine models of GBM (41).
In this paper, we investigated the possible therapeutic efficacy of the adoptive transfer of microglia, modified with the rAAV vector carrying IL-15, to mimic the EE effects on glioma. The possibility to exploit microglial cells intranasally delivered as a source of IL-15 directly in the tumor mass could be an attractive alternative for GBM patient treatment, due to the high potential for clinical translatability, as one can obtain the microglia starting from patients' iPSCs, as recently reported (46). The local delivery of cells as therapeutic carriers against glioma has produced encouraging results but encounters obstacles with regard to the repeatability and invasiveness of administration.
Intranasal delivery of cells could overcome these obstacles as a non-invasive and easily repeatable route of administration (47).
Here, we demonstrated that the infection of microglia with an AAV-serotype 2 carrying IL-15 functionally induces the release of IL-15. AAV are considered optimal for human gene therapy as they are small and non-replicative, can transduce dividing and nondividing cells, are non-pathogenic in humans, and can determine long-lasting changes in gene expression. Importantly, AAV have been used to target several cell types, including microglia (24,48).
In our studies, the rAAV2-IL-15 microglia activated the interplay with NK cells both in vitro and in vivo. In particular, these infected microglia increased the number of NK cells in coculture, without affecting their activation state. This could be due to the dose-dependent effect of IL-15: in the in vitro system, the rAAV2-IL-15 microglia produce a concentration of IL-15 with effects on proliferation and viability, without reaching the higher doses needed to activate the NK cells (49,50). We recently reported that microglial production of IL-15, in glioma bearing in EE-housed mice, recruits IFN-g+ NK cells, with effects on glioma growth (20). Attempting to reproduce the effect of EE on glioma, we showed that intranasally delivered rAAV2-IL-15 microglia reach glioma and increase NK cell accumulation in glioma-bearing mice, highlighting the fundamental role of IL-15 in the tumor core to boost the immune reaction.
Since recruited NK cells could revert the GBM microenvironment instructing microglia toward a pro-inflammatory/antitumoral phenotype (11), we investigated the M/M F state in tumor mass. We showed that rAAV2-IL-15 microglia consistently modulate the M/M F state with the reduction of arginase levels and an increased number of branches and covered the parenchymal region, suggesting the switch in an antitumoral phenotype (11,28). Further, these data suggest that the recruited NK cells in the tumor core are activated, thus explaining the modulation of the M/M F phenotype (11). Altogether, our results demonstrated that i.n. delivery of IL-15-engineered microglia is able to reproduce the effect of EE on NK cell and tumor volume in glioma mouse models. These preclinical data support the use of rAAV2-IL-15-infected microglia as a tool to vehicle and translate the EE signals inside the tumor mass and offer a new perspective to use them as Trojan horses to modify the tumor microenvironment.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

ETHICS STATEMENT
The animal study was reviewed and approved by the Italian Ministry of Health in accordance with the guidelines on the ethical use of animals from the European Community Council Directive of September 22, 2010 (2010/63/EU).

AUTHOR CONTRIBUTIONS
AM: performed most of the experimental work and wrote the paper. GC: contributed to many experimental activities for mouse manipulation. GB: performed FACS analyses. NC: sequenced and provided adeno-associated viral particles and wrote the paper. CP: sequenced and provided adeno-associated viral particles and wrote the paper. AS: supervised the experiments on NK cells. CL: supervised all the experimental work and wrote the manuscript. SG: ideated and supervised all the experimental work and wrote the paper. All authors contributed to the article and approved the submitted version.