Protocol 1: isolation and culturing of primary microglia from neonatal mice

Primary microglia were isolated from postnatal P0–3 C57BL/6 mouse neonates. Neonates were euthanized by decapitation. Brains were then excised under aseptic conditions and dissociation was performed using the Neural Tissue Dissociation kit according manufacturer’s instructions (Miltenyi Biotec, Bergisch Gladbach, Germany). The resulting pellet was gently resuspended in complete DMEM (containing 4.5 g/L glucose, 10% FBS, 1% penicillin–streptomycin) and was filtered through a 70 μm cell strainer (Corning, NY, USA) to eliminate any clumped cells or tissue debris. Based on the intended protocol variation, cell suspensions were treated as follows:

Protocol 1A: Cells were cultured in complete DMEM supplemented with 0.5 ng/ml recombinant mouse GM-CSF (R&D Biosystems, Minneapolis, USA) for 10 days.

Protocol 1B: Cells were cultured in complete DMEM (without GM-CSF) for 10 days.

Protocol 1C: Cells were cultured in complete DMEM supplemented with 0.5 ng/ml recombinant mouse GM-CSF for 21 days.

For each of the above protocols, cell suspensions were derived from 5 mice. Cells were plated onto a 175 cm2 flask, and medium was changed every 3 days. Upon reaching confluence, on day 10, microglia were dislodged and harvested from culture by manual shaking for 20–30 s. The medium containing the detached microglia was collected, and cell count was determined. This was followed by centrifugation at 400g for 10 min to pellet the cells. The microglial pellet was resuspended in serum-free X-VIVO medium (Lonza Biosciences, Walkersville, MD), and plated into 48-well Falcon tissue culture plates (Fisher Scientific, Waltham, Massachusetts, U.S.) at 200,000 cells per well. X-VIVO medium is a serum-free medium, which lacks exogenous growth factors, artificial stimulators of cellular proliferation, or undefined supplements. It has been used for myeloid cell growth as well as microglia cultivation [7, 12, 25]. Cells were allowed to adhere overnight and were ready for experimentation the subsequent day. At least 3 biological replicates for each condition were performed, with each replicate representing microglia isolations derived from distinct litters of mice.

Protocol 2: isolation and culturing of primary microglia from adult mice by adherence properties

C57BL/6 mice, at 6–8 weeks, were administered a lethal dose of Euthasol followed by transcardial perfusion with 0.9% ice cold heparinized saline. Brains were excised under aseptic conditions, and tissue dissociation was performed using the Neural Tissue Dissociation Kit, according to the manufacturer’s instructions. Following dissociation, myelin debris was separated and removed using Debris Removal Solution (Miltenyi Biotec, Bergisch Gladbach, Germany), according to manufacturer’s instructions. The cell pellet was resuspended in pre-warmed complete DMEM and cells from 3 brains were pooled together and plated on a T75 flask. After a 3-h incubation period to allow adherence, the medium was gently replaced to discard non-adherent cells. The medium was supplemented with 100 ng/mL macrophage colony-stimulating factor (M-CSF) and 100 ng/mL GM-CSF on the following day. Medium was changed on day 4 with complete DMEM supplemented with M-CSF and GM-CSF, and on day 7, contaminating cells were collected using 0.05% trypsin- ethylenediaminetetraacetic acid (EDTA) for 10 min and discarded. Contaminating cells adhered less strongly to the flasks than microglia when exposed to 0.05% trypsin, so this step allowed for enhanced purity. Microglia were then obtained by incubating flasks in 0.25% trypsin–EDTA for 30 min. Trypsin was quenched with complete DMEM, and cells were removed from the flasks and centrifuged at 400g for 10 min, followed by resuspension in X-VIVO serum-free medium (Lonza Biosciences, Walkersville, MD). Cells were plated in 48-well dishes (Falcon) at 200,000 cells per well, and they were ready for experimentation the following day.

Protocol 3: acute isolation of microglia from adult mice using anti-CD11b MicroBeads

C57BL/6 mice, at 6–8 weeks, were administered a lethal dose of Euthasol followed by transcardial perfusion with 0.9% ice cold heparinized saline. Brains were excised under aseptic conditions, and tissue dissociation was performed using the Neural Tissue Dissociation Kit, according to the manufacturer’s instructions. Following dissociation, myelin debris was separated and removed using Debris Removal Solution, according to the manufacturer’s instructions. CD11b-positive cells were enriched using anti-CD11b MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany), according to manufacturer’s instructions. The resulting cell:bead suspension was centrifuged at 400g for 10 min and resuspended in X-VIVO medium (Lonza Biosciences, Walkersville, MD), or X-VIVO medium supplemented with 1 ug/mL HMGB1 for immediate stimulation.

Cell culture treatment

HMGB1 was obtained as a generous gift from Kevin Tracey, MD, of the Feinstein Institutes for Medical Research. Cells were treated with HMGB1 for 4 h prior to harvesting for analysis of bulk mRNA sequencing and for 24 h prior to analysis of cytokines in cell culture supernatant by ELISA.

RNA extraction

Total RNA was extracted from microglia using the Qiagen RNeasy RNA extraction kit (Qiagen) according to the manufacturer's instructions. Briefly, cells from each well were homogenized in RLT lysis buffer. The homogenate was then passed through QIAshredder spin columns to remove cellular debris, and RNA was purified using RNeasy spin columns. RNA was eluted in 30 μl RNase-free H20.

ELISA

Cell culture supernatant was collected and centrifuged at 500g for 5 min followed by separation of supernatant. The Duoset TNFα, Duoset IL-1β, and Duoset IFNβ ELISAs (R&D Biosystems, Minneapolis, MN) were performed on supernatant according to the manufacturer's instructions.

Flow cytometry

Cells were washed in FACS buffer (1% BSA, 0.1% sodium azide in PBS), then incubated in FACS buffer containing functional viability dye (65-0866-14, ThermoFisher Scientific) along with anti-CD45 (1:80, BioLegend, clone 30-F11), anti-CD11b (1:200, BD Biosciences, clone M1–70) and anti-transmembrane protein 119 (Tmem119; 1:500, Abcam, clone 106-6) antibodies for 15 min at 4 °C in the dark. After staining, cells were washed with FACS buffer, and flow cytometry was performed using the BD LSRFortessa™ Cell Analyze. Data analysis was conducted using FlowJo.

mRNA sequencing analysis

Gene read counts were obtained using featureCounts v1.5.0 [23], and normalized using the DESeq2 package (1.20.0) [26] with variance-stabilizing transformation (VST). Differential gene expression following HMGB1 treatment for each isolation protocol was assessed, using an adjusted p value < 0.05 and an absolute fold change of 1. Raw gene counts from the NCBI Gene Expression Omnibus GEO accession number GSE79819 [27] were downloaded and combined with our data, and we applied batch correction using combat-seq. For gene set enrichment analysis, a list of genes that were DEGs in both the Protocol 1 vs Protocol 3 comparison and the early microglia vs adult microglia comparison were compiled based on log2fold change > 2 and p value < 0.01 in both comparisons, giving us a list of 399 genes which was input into EnrichR [20]. The top 10 terms from GO Biological Processes 2023 were identified.

Statistical methods

Statistical analyses were conducted using R for bulk mRNA sequencing data and GraphPad Prism for ELISA measurements. For neonate-derived cultures, each experimental group comprised three biological replicates from separate mouse litters. One-tailed t-tests were employed to compare cytokine secretion levels (TNFα, IL1β, IFNβ) between control and HMGB1-treated groups for each isolation method. Results are presented as median ± standard deviation.

Data availability

All bulk sequencing data generated and analyzed during this study are publicly accessible in the Gene Expression Omnibus (GEO) repository. The datasets can be retrieved under the accession number GSE242683.