Animals, diets, and postnatal inflammatory challenge: LPD/IL-1β rat model
All experiments were carried out in compliance with Inserm ethical rules, approved by the institutional review board (Ministry of Higher Education and Scientific Research, Directorate-General for Research and Innovation, Paris, France), in accordance with the European Communities Council Directive 2010/63/EU. Briefly, Sprague–Dawley dams (Janvier Labs, Le Genest-Saint-Isle, France) were randomly divided into two groups according to their diet: a 22% (normal) protein diet (control, CTRL) or an isocaloric 9% protein diet (LPD) from the day of conception until delivery. On postnatal day 1 (P1) and P2, pups were injected intraperitoneally (i.p.), twice a day, with PBS or IL-1β (20 μg/kg, Miltenyi Biotec, Bergisch Gladbach, Germany), as previously reported (25).
Cortical sample collection and magnetic sorting of microglial cells
Cortical samples were collected from CTRL and LPD/IL-1β pups at P1, P4, P7, P10, and P20. Primary microglial cells were sorted at P1, P4, P7, P10, and P20 using magnetic antibody-based cell sorting (MACS) and CD11b antibody (Miltenyi Biotec, Bergisch Gladbach, Germany), as previously reported (25). The purity of the P4 and P7 MACS CD11b+ fractions was validated by RT-qPCR performed on the positive and negative CD11b cell fractions for Itgam, Gfap, Neun, and Mbp and the arbitrary units normalized to the respective negative fraction in CTRL: P4 CTRL (mean ± SEM): Itgam 99.78 ± 11.29, Gfap 0.13 ± 0.01, Neun 0.08 ± 0.02, Mbp 0.02 ± 0.003; P4 LPD/IL-1β: Itgam 109.15 ± 5.27, Gfap 0.11 ± 0.1, Neun 0.02 ± 0.002, Mbp 0.01 ± 0.01; P7 CTRL: Itgam 203.80 ± 9.95, Gfap 0.07 ± 0.01, Neun 0.04 ± 0.004, Mbp 0.06 ± 0.02; P7 LPD/IL-1β: Itgam 277.5 8± 23.27, Gfap 0.09 ± 0.01, Neun 0.05 ± 0.01, Mbp 0.02 ± 0.003.
RNA purification, cDNA synthesis, and real-time qPCR
Samples were immediately snap frozen after collection. Total RNA from the cortex was extracted using Nucleazol reagent and the NucleoSpin RNA Set for NucleoZol (Macherey-Nagel, Hœrdt, France). Microglial RNA was isolated using NucleoSpin RNA Plus XS (Macherey-Nagel). RNA quantity and quality were determined using the NanodropTM apparatus (Thermofisher Scientific, Waltham, MA, USA). Reverse transcription was performed using the IscriptTM cDNA synthesis kit (Bio-Rad, Marnes-la-Coquette, France). Primers were designed using Primer3Plus software and sequences are available on request. qRT-PCR was performed in triplicate as reported in (26) using the Ribosomal protein L13 (Rpl13) as the reference gene.
RNA preparation from sorted microglial cells, cDNA synthesis, microarray hybridization, and bioinformatics analysis
Total RNA was extracted from CTRL and LPD/IL-1β P4 animals and cDNA synthesis and microarray hybridization performed as previously described (25). For microarray hybridization, labeled cRNA was generated using classical protocols for Affymetrix array hybridization. Three points per condition were analyzed by RaGene-2_0-st microarray hybridization, for which 36,685 probes are examined and directly analyzed using Gene Set Enrichment Analysis (https://www.gsea-msigdb.org/gsea/index.jsp (27)). Networks analysis of the microarray data was also performed using String (https://string-db.org/). The numerical data from the networks obtained were then exported to Cytoscape (https://cytoscape.org/) to identify putative hub genes, using the Network Analysis option. Microarray results were validated as previously reported (25).
Protein extraction and immunoblotting assay
Total protein was extracted from the cortex and microglial cells sorted from P4 and P7 rats of the CTRL and LPD/IL-1β experimental groups. Cortex samples were homogenized in lysis buffer (0.32 M sucrose, 4 mM HEPES pH 7.4, 1% SDS). Complete, Mini, EDTA-free Protease Inhibitor Cocktail and PhosSTOP (Roche, Meylan, France) were added. RIPA buffer (Sigma-Aldrich, St Quentin Fallavier, France), supplemented with protease and phosphatase inhibitors, was used to extract protein from the microglial cells. Proteins were quantified using the Bradford assay (Sigma-Aldrich) and 30 µg from the cortex and 20 µg from the microglia re-suspended in Laemmli Sample Buffer (Bio-Rad) with 2.5% 2-mercaptoethanol (Sigma-Aldrich). Samples were separated on 4–15% Mini-PROTEAN TGX Precast Protein Gels and the proteins transferred to nitrocellulose membranes using a Trans-Blot Turbo Mini (Bio-Rad). Proteins were stained using Ponceau S solution (Sigma-Aldrich) and digital images acquired. The optical density of each well was measured using NIH ImageJ medical imaging software. Blots were incubated in a blocking solution containing Tris-buffered saline (TBS), 0.1% Tween-20 (Sigma-Aldrich), and 5% non-fat milk (Bio-Rad) for 1 h at room temperature. Blots were then incubated overnight with rabbit anti-mGlu3 receptor (1:600; AGC-012, Alomone Labs, Jerusalem, Israël) or mouse Anti-β-Actin (1:10000; clone AC-74, Sigma-Aldrich) in blocking solution at 4ºC. After incubation with the primary antibody, the blots were incubated with horseradish peroxidase-conjugated goat anti-rabbit (1:2000; 0545, Sigma-Aldrich) or horseradish peroxidase conjugated goat anti-mouse (1:2000; 12-349, Sigma-Aldrich), for 1 h at room temperature. Bands were visualized by enhanced chemiluminescence using Clarity Max ECL (Bio-Rad). Digital images were acquired and the optical density of each band measured as described above. The ratio of the target to Ponceau S or β-Actin was then determined and the values compared for statistical significance.
LY 379268, a highly selective group II mGlu receptor agonist (Tocris Bioscience, Rennes, France), was dissolved in sterile water. LY 2389575 hydrochloride, a selective negative allosteric modulator of mGlu3 receptors (Tocris) and Ro 64-5229, selective, non-competitive mGlu2 receptor antagonist (Tocris) were dissolved in DMSO.
Primary microglial cell culture: real-time qPCR and morphological assay
Microglial cells were isolated from CTRL and LPD/IL-1β animals at P4 and P7 and cultured as previously reported (25). CTRL microglial cells were treated with LY 379268 (0.5, 1, or 5 µM) + Ro 64-5229 (25 µM) to establish the optimal drug concentration (1µM). After 48 h, CTRL and LPD/IL-1β microglia were divided into four groups: (1) DMSO (0.30%), (2) LY 379268 (1 µM) + Ro 64-5229 (25 µM), (3) IL-1β (50 ng/ml) + IFNγ (20 ng/ml) for 1 h, and (4) LY 379268 (1 µM) + Ro 64-5229 (25 µM) for 1 h and then IL-1β + IFNγ for 4 h.
A second series of experiments was performed on P4 and P7 CTRL microglial cells. CTRL microglial cells were treated with LY 2389575 (1, 3, or 5 µM) to establish the optimal drug concentration (5 µM). Forty-eight hours after plating, cells were divided into four groups: (1) DMSO (0.30%), (2) LY 2389575 (5µM), (3) IL-1β + IFNγ for 1 h, and (4) LY 2389575 (5µM) for 1 h and then IL-1β + IFNγ for 4 h.
Total RNA was extracted using NucleoSpin RNA Plus XS. RNA quality and quantity were determined, cDNA synthesized, and real-time qPCR performed as reported above.
Morphological analysis was performed after increasing the drug exposure time to 12 h. After fixation with 4% PFA, cells were stained with goat anti-Iba1 (1:500; ab5076, Abcam, Paris, France) and DAPI (1:10000). Iba1+ cells were analyzed using a fluorescent microscope (Nikon Eclipse Ti-E): (i) area, (ii) perimeter, and (iii) cell circularity (4π×(area/perimeter2)) were determined.
Experiments to knock-down Grm3 expression were performed on P7 CTRL microglia using Viromer Blue Reagent (Lipocalyx, Halle, Germany) according to the manufacturer’s protocol. Transfection efficiency and cell viability after Viromer exposure were evaluated using the siGLO Red Transfection Indicator (25 nM or 50 nM) (Dharmacon, Lafayette, CO, USA). Cells were fixed in 4% PFA, stained with anti-Iba1 antibody, and analyzed by fluorescence microscopy. Cells were treated 48 h after plating with (1) ON-TARGETplus Non-targeting Pool (25 nM or 50 nM) or (2) ON-TARGETplus Rat Grm3 (24416) siRNA - SMARTpool (25 nM or 50 nM) (Dharmacon) for 4 h. The medium was replaced by fresh culture medium and the cells incubated at 37ºC for 48 h before proceeding to RNA and protein extraction.
Statistical analysis of all data was performed using GraphPad PRISM version 8.0 (San Diego, CA, USA). Student’s t test was performed for two-group comparisons. A one- or two-way ANOVA, followed by Newman–Keul's post hoc multiple comparison tests, was performed for comparison of more than two groups. A Pearson correlation test was used to analyze relationship between Grm3 and inflammatory cytokines mRNA expression in siRNA experiment. A multiple t-test was performed to analyzed microarray data. Significance was set at p < 0.05 for all tests. Number of samples, statistics, and p values, were reported in Additional Table 1. Morphological data were analyzed by a person who was blind to experimental conditions. Pups from 2-3 litters were used in each experiment.