Background
MicroRNAs (miRNAs) are small non-coding RNAs that function as guide molecules in RNA silencing [
1]. miRNAs are also enriched in the central nervous system (CNS) and play crucial roles in the development and plasticity of the brain [
2]. Certain miRNAs regulate differentiation, activation, and polarization of microglia [
3]. miR-124, one of the most abundant miRNAs in neurons, is expressed in microglia and promotes microglial quiescence by inhibiting the C/EBP-α-PU.1 pathway [
4]. miR-155 is elevated in M1-polarized microglia and regulates their pro-inflammatory responses [
5]. Recently, miR-9 was reported to promote microglial activation by targeting MCPIP1 [
6].
Microglia arise from primitive yolk sac macrophages and develop independently of the hematopoietic system [
7]. Furthermore, a recent study demonstrated that other tissue-resident macrophages also originate from yolk sac progenitors [
8]. This evidence suggests that the developmental fate of tissue macrophages is influenced by the environment in which cells are placed. In addition to colony-stimulating factors including interleukin-34, which dictate the developmental fate of microglia [
9], miRNAs enriched in the CNS may also play a role in the acquisition of a distinct microglial phenotype.
A murine bone marrow chimera model suggests that hematopoietic cells have the potential to develop into microglia [
10]. We developed an in vitro model in which lineage-negative bone marrow cells co-cultured with astrocytes differentiated into microglia-like cells [
11,
12]. In this model, bone marrow-derived cells showed two morphological forms, namely, small round (SR) cells having relatively small, round-shape soma or large flat (LF) cells having polymorphic soma and pseudopodia. Only SR cells expressed triggering receptor expressing on myeloid cells-2 (TREM-2) that is predominantly expressed on microglia [
13]. This indicated that SR cells had phenotypical similarity to microglia and thus we defined these cells as microglia-like cells [
12].
In the present study, we sought to identify miRNAs that affect the phenotype of microglia using an in vitro co-culture model and immortalized microglial cell line. We show that miR-101a modulates microglial morphology and inflammation.
Methods
Mice
Six-week-old female C57BL/6J (B6) mice were purchased from the Japan CLEA Laboratory Animal Corporation (Tokyo, Japan). Beta-actin promoter-driven enhanced green fluorescence protein (GFP) transgenic mice on a B6 background were kindly provided by Dr. Masaru Okabe (Osaka University, Japan).
Isolation of murine LN− cells
To isolate lineage negative (LN−) cells, bone marrow cells were collected from B6 mice or GFP mice by flushing the femora and tibiae of the hind limbs with phosphate-buffered saline. Erythrocytes were lysed using ammonium chloride-potassium buffer. LN− cells were negatively selected using antibodies (Abs) against lineage-specific markers (CD3, CD4, CD5, CD8α, CD11b/MAC-1α, B220, Gr-1, and TER-119; R&D Systems, Minneapolis, MN, USA) and immunomagnetic beads (Dynal, Oslo, Norway).
Isolation of microglia from the CNS
To isolate microglia from the CNS of adult mice, we dissociated the brain tissues of 6-week-old mice using Neural Tissue Dissociation Kits (P) (Miltenyi Biotec, Bergisch-Gladbach, Germany). Mononuclear cells were obtained through a density gradient from the interface between 27 and 72% Percoll (GE Healthcare, Waukesha, WI, USA) layers. Microglia were isolated from the single-cell suspension by MACS Technology using CD11b (Microglia) Microbeads (Miltenyi Biotec). The purity of microglia was >98%.
Isolation of peritoneal macrophages and bone marrow-derived macrophages
To isolate peritoneal macrophages, peritoneal lavage fluid was collected from B6 mice by washing the peritoneum with 5 mL Hank’s balanced salt solution (HBSS; Gibco, Carlsbad, CA, USA). Cells were cultured overnight and adherent cells were harvested as macrophages. To obtain bone marrow-derived macrophages, bone marrow cells were isolated from the femur and the tibia of B6 mice and cultured for a week in the presence of recombinant murine M-CSF (20 ng/mL, BioLegend, San Diego, CA).
Induction of microglia-like cells
Primary mixed glial cell cultures were prepared from the brains of postnatal 3–5-day-old (P3–P5) B6 mice, as previously described [
12]. Astrocytes were prepared after removal of CD11b
+ cells using Dynabeads (Dynal) conjugated with anti-CD11b Ab. The purity of astrocytes was >96% as determined by anti-GFAP and anti-CD11b immunofluorescence. Astrocytes were then seeded into culture flasks, 96-well plates, or Lab-Tek II 8-well chamber slides (Thermo Fisher Scientific, Waltham, MA, USA) at a density of 7.5 × 10
6 cells per flask, 2.5 × 10
4 cells per 96-well, or 1.0 × 10
5 cells per chamber slide well and cultured in the medium described above for 5 days to form a confluent monolayer. LN
− cells were seeded on astrocytes at a density of 1.5 × 10
6 cells per flask or 0.5 × 10
4 cells per 96-well and chamber slide well and cultured for 7 days.
Culture of MG6 cells
A murine microglia cell line of MG6 cells was kindly provided by Dr. Hiroshi Kitani (National Institute of Agrobiological Sciences, Japan) [
14]. We cultured MG6 cells in Dulbecco’s modified Eagle’s medium, supplemented with 10% fetal calf serum (FCS), 1% glucose, 1% l-glutamine, 1% penicillin/streptomycin, 10 μg/ml insulin, and 0.1 mM 2-mercaptoethanol at 37 °C in a humidified atmosphere of 5% CO
2 and 95% air.
Screening using microRNA inhibitor library
We screened miRNAs that modify microglial differentiation and activation using the miRCURY LNATM microRNA inhibitor library (Exiqon, Vedbaek, Denmark). A total of 739 miRNA inhibitors were screened. miRNA inhibitors were suspended in Opti-MEM and HiperFect transfection reagent (Qiagen, Hilden, Germany) and were incubated for 15 min at room temperature. Then, miRNA transfectants were added to LN− cell-astrocyte co-culture seeded on 96-well plates at a final concentration of 40 nM on days 0 and 3. On day 7, wells were observed microscopically by two authors (RS and HS) in a blinded manner and morphological findings were independently assessed in a semi-quantitative manner. Wells were scored as hits when both investigators judged that the number of SR cells changed significantly.
miRNA mimic and inhibitor treatment
For immunohistochemistry, LN− cell-astrocyte co-culture seeded on 8-well chamber slides were transfected with miRNA inhibitors and mimics using the same protocol as described above. For ELISA, MG6 cells were seeded at a density of 2 × 104 cells per 96-well. The next day, miRNA inhibitors and mimics were transfected at a final concentration of 40 nM with Lipofectamine RNAiMAX (Thermo Fisher Scientific).
Immunohistochemistry
LN− cells that were isolated from GFP mice and cultured on astrocytes were fixed in 4% paraformaldehyde. After treatment with protein block (Dako, Denmark), samples were immunostained with: (1) biotin-conjugated anti-CD11b Ab (BioLegend, San Diego, CA, USA) followed by DyLight 649-Streptavidin (Jackson ImmunoResearch, West Grove, PA, USA), (2) anti-TREM2 Ab (R&D systems) followed by rhodamine-anti-sheep IgG (Jackson ImmunoResearch), (3) anti-CX3CR1 (R&D systems) followed by Cy5-anti-goat IgG (Jackson ImmunoResearch), or (4) anti-Iba1 (Wako Pure Chemical, Osaka, Japan) followed by rhodamine-anti-rabbit IgG Ab (Jackson ImmunoResearch). Images were acquired using an FV1000-D microscope (Olympus, Tokyo, Japan). To quantify the number of cells, ten fields under a ×20 objective were randomly selected, and GFP-positive or immunostained cells were counted using ImageJ software (National Institutes of Health, Bethesda, MD, USA). SR cells and LF cells were defined as cells with a circularity of >0.8 and ≤0.8, respectively.
Enzyme-linked immunosorbent assay (ELISA)
After stimulation with LPS, we collected the cell supernatants and analyzed them using the BD OptEIATM Set (BD Biosciences, Franklin Lakes, NJ, USA) for mouse IL-10, TNFα, and IL-6 and the Ready-SET-Go! kit (eBioscience, San Diego, CA, USA) for mouse IL-1β.
Real-time PCR analysis
We isolated mRNA from MG6 and brain cells using the RNeasy Mini Kit (Qiagen), and we isolated miRNA using the miRNeasy Mini Kit (Qiagen). We performed reverse transcription for mRNA using PrimeScript RT Master Mix (Perfect Real Time) (Takara Bio, Otsu, Japan) and real-time PCR with SYBR Premix Ex TaqTM II (Tli RNaseH Plus) (Takara Bio). For miRNA expression analysis of miR-101 and snoRNA202, we used TaqMan MicroRNA assays (Applied Biosystems, San Francisco, CA, USA).
MTT assay
Viability of MG6 cells was assessed using the CellTiter 96® Aqueous One Solution Cell Proliferation Assay kit (Promega, Madison, WI, USA). MG6 cells (2.5 × 104 cells per well) were seeded on 96-well plate and cultured for 24 h. CellTiter 96® Aqueous One Solution Reagent (20 μL per well) were added to each well and plate was incubated at 37 °C for 1 h. The quantity of formazan product that is proportional to the number of living cells in culture was measured by absorbance at 490 nm.
Statistical analysis
Results represent at least three independent experiments. Data are presented as mean ± S.D. Statistical significance was determined by one-factor ANOVA, Kruskal-Wallis test, Student’s t test, and Mann-Whitney test.
Discussion
Comparative microarray analyses of microglia have identified a unique transcriptome that is distinct from other macrophage cell types. This includes many neuron-associated transcripts such as synaptic molecules [
15]. Microglial progenitors arise from the yolk sac and migrate to the fetal brain, where they develop under the influence of the CNS environment [
7]. It is thus logical to speculate that some factor that is enriched in the brain facilitate microglia to acquire a distinct phenotype as a CNS-biased cell. miRNAs are a candidate factor for determining the specific phenotype of microglia.
In the present study, we identified miR-101a as a modifier of microglial morphology and function using a miRNA inhibitor library. It has been reported that primary macrophage is difficult to transfect siRNA or plasmid DNA due to reduced cell vitality and severely altered cell behavior [
16]. This also hampers the study on the effect of miRNAs on microglia. Due to this limitation, we screened the effect of miRNAs using LN
− cell-astrocyte co-culture. As a result, we identified miR-101a as a modifier of the shape of LN
− cell-derived cells.
In these experiments, transfection of control mimic or inhibitor significantly changed the character of cells. In general, transfection of control mimic or inhibitor tended to increase the number of LN
− cell-derived cells. Although miR-101a inhibitor decreased the number of total, SR, TREM-2
+ and CX3CR1
+ cells, the differences were not significant when compared to untreated control (Figs.
2b, c and
3e, f). Nevertheless, miR-101a mimic undoubtedly increased the number of these cells and increased cytokine production form MG6 cells (Figs.
2b, c,
3e, f and
5a–c) compared to both medium and control mimic. Although the mechanism of this phenomenon is unknown, we speculated that the transfection of miRNA had non-specific immune activation of these cells. Various RNAi and miRNA reagents, which differ in length and structure, have been reported to cause non-sequence-specific immune responses [
17]. The activation of the cellular sensors of foreign RNA or DNA may lead to the induction of type I interferon and cytokine release. This effect depends mostly on the reagent length, structure, chemical modification, and concentration rather than on its specific sequence [
18]. In addition, it may also depend on the method of RNA delivery into the cells. However, control miRNA inhibitor that can cause such response would not be a good candidate for control when studying immune response of the cells. Thus, control miRNA and their delivery method should be carefully chosen to minimize non-specific effects. The RNAimmuno database (
http://rnaimmuno.ibch.poznan.pl) provides information regarding the non-specific effects generated by RNA interference triggers and miRNA regulators [
17].
miR-101a is downregulated in various types of cancer and functions as a tumor suppressor by repressing the polycomb group protein EZH2 [
19]. miR-101a also inhibits tumor growth by regulating the cyclooxygenase-2 pathway [
20]. miR-101a is enriched in the brain and regulates amyloid precursor protein expression in rat hippocampal neurons cultured in vitro [
21]. However, the role of miR-101a in microglia was previously unknown.
We demonstrated that miR-101a affects microglia in two different ways. First, miR-101a promotes the development of uncommitted hematopoietic progenitors into myeloid cells that resemble microglia. Our previous study indicated that hematopoietic cells had the potential to develop into microglia [
11]. Our model, in which bone marrow cells are co-cultured with astrocytes before differentiating into microglia-like cells, enables us to track the process by which non-microglial cells acquire phenotypes that are characteristic of microglia as well as identify factors that are essential for this process. In this model, miR-101a facilitates induction of TREM-2-positive microglia-like cells from bone marrow lineage-negative cells. The expression of TREM-2 in microglia is significantly higher than in other monocyte-macrophage lineage cells [
13], indicating that miR-101a favored the commitment of LN
− cells to microglia. Nevertheless, it is unlikely that miR-101a plays a crucial role in the development of microglia in the fetal brain because this miRNA is not enriched in the fetal brain. Another possibility is that miR-101a may function to maintain the cellular character of microglia in the adult brain.
Second, miR-101a regulates proinflammatory cytokine expression in microglia. miR-101a increased the production of IL-6 and TNFα from microglia. In contrast, miR-101a decreased the production of IL-1β from LN
− cell-astrocyte co-culture. These results suggest that miR-101a regulates microglial inflammation through several diverse pathways. miRNAs exert their effects by targeting certain downstream molecules. We predicted several pathways enriched in the target genes of miR-101a, and they included MAPK and TGF-beta signaling pathways. With regard to MAPK signaling pathway, miR-101a is reported to enhance the transcription of inflammatory mediators by suppressing MKP-1 activity in RAW264.7 macrophages [
22]. In the present study, miR-101 suppressed the expression of MKP-1 in LPS-treated MG6 cells. It is also reported that miR-101 mimics increased the production of TNFα [
23]. These results suggest that miR-101 augment the production of both IL-6 and TNFα from microglia by inhibiting MKP-1.
Acknowledgements
Not applicable.