All experimental procedures were approved by the Institutional Animal Care and Use Committee at the University of Illinois Urbana–Champaign, under protocol #19068 and were performed in accordance with guidelines of the National Institute of Health. Male and female C57BL/6J (Jackson Laboratories No. 000664) mice aged 8-12 weeks old were used for all experiments. Breeders were group-housed in solid-bottom caging with standard bedding (Teklad 1/8" corncob) under temperature-controlled conditions (23 ± 1°C) with a 12-hour reversed light/dark cycles (10am-10pm). Rodent diet (Teklad No. 8640) and water were available ad libitum.

Oral treatment of rodents with GW2580 has been well characterized (40). Specifically, in vitro treatment with GW2580 at 5µM inhibited CSF1 mediated cell proliferation by nearly 100%, and an oral dose of 80mg/kg increased circulating plasma GW2580 concentrations to 5.6µM. Oral treatment was subsequently shown to inhibit disease progression in rodent models of multiple sclerosis [40mg/kg/d (42)], amyotrophic lateral sclerosis [75mg/kg/d; (45)] Alzheimer’s disease [75mg/kg/d; (44)] and to attenuate depression-like behavior in MRL/lpr mice [100mg/kg/d; Chalmers et al. (47)]. As such, in the current study, mice were orally gavaged with either 80 mg/kg of GW2580 (LC Labs, Cat No. G-5903) diluted in 200µl of 0.1% Tween 80, 0.5% hydroxymethyl propyl-cellulose or vehicle alone (44) using 18G needles (Instech, Cat No. FTP-18-30-50) for a total of eight days (40). Animal weight and food disappearance were record daily. Burrowing activity, an indicator of sickness behavior, was measured every 48h (48). On day 8 post treatment, mice were euthanized by CO₂ asphyxiation, blood was collected via cardiac puncture and the mice were perfused with 20-30 mL of sterile phosphate buffered saline (PBS; pH=7.4).

Following euthanasia blood samples were collected by cardiac puncture with EDTA-coated syringes. Whole blood (100µl) was added to ammonium chloride potassium (500µl) solution for 10 min. at room temperature in order to lyse the red blood cells. The cells were subsequently washed then suspended in ice-cold flow cytometry staining buffer (sterile PBS containing 2% FBS) and counted using an automated cell counter (Nexcelom Bioscience). Next, 1x10⁶ cells were labeled on ice with fluorophore conjugated antibodies to CD4 (Pacific Blue; clone RM4-5; BioLegend, Cat. No. 116008), CD8a (APC-Cy7; Clone 53-6.7; BioLegend, Cat. No. 100714), CD11b (APC; Clone M1/70; BioLegend, Cat. No. 101212), B220 (PE-Cy7; Clone RA3-682; BioLegend, Cat. No. 103224), Ly6G (FITC; Clone 1A8; BioLegend, Cat. No. 127608), Ly6C (PE; Clone HK1.4; BioLegend Cat. No. 128044), and FCblock (CD16/32) for 20 min. Cell were washed, suspended in flow buffer and data were acquired on a LSRII Flow cytometer (BD). Gates were determined using unstained and single-stained samples obtained from the same tissue of origin. Results were analyzed using FlowJo version 10.6.2 flow cytometry software (BD).

Adult brains were collected from mice, minced with a sterile razor blade and an enzymatic digestion for 45 min. in 5ml of StemPro Accutase (Gibco, Cat No. A1110501) at 37°C. Then the brain homogenate was filtered through a 70µm filter using a 3ml syringe plunger, then pelleted by centrifuging at 400xg for 5 min. Red cells were removed with a red blood cell lysis buffer (Biolegend, Cat No. 420302) followed by another centrifugation at 400xg for 5 min. Next, cells were suspended in 5ml of PBS solution containing 35% Percoll (VWR, Cat. No. 899428-524), underlay with 3ml of PBS containing 70% Percoll, then centrifuged for 20 min. at 2000xg with no brake. The cells at the 35/70% Percoll interface were collected and in some cases these cells were washed with magnetic activated cell sorting (MACS) buffer (PBS with 3% FBS and 10mM EDTA) to enrich for CD11b positive cells following the manufacturer’s instructions (ThermoFisher Scientific, Cat No. 8802-6860-74). Flow cytometry was also used to confirm the purity of microglia subjected to MACS cell enrichment. Here, microglia from non-enriched and enriched populations were stained with antibodies for CD45 (APC; clone 30-F11) (eBioscience, Cat No. 17-0451-82), CD11b (FITC; clone M1/70) (eBioscience, Cat No. 11-0112-82) and viability dye (eFlour 780) (eBioscience, Cat No. 65-0865-14) as markers.

Brains were collected and placed in fixation buffer (PBS containing 4% paraformaldehyde, pH 7.4) at 4°C for 24h. The tissue was then cryoprotected by replacing the fixation buffer with PBS containing 30% sucrose (w/v) until the tissue sank. The brain tissue was frozen in optimal cutting temperature compound (O.C.T.) with dry ice, then stored at -80°C. Next, 15µm coronal sections were collected using a cryostat (Leica CM3050 S). Tissue sections were then blocked and permeabilized with PBS containing 5% goat serum and 0.3% Triton X-100, then incubated with an anti-Iba-1 specific antibody derived from rabbit (diluted 1/200; Rabbit; Wako Inc. Richmont, VA, Cat No. 019-19741). After washing with PBS three times for 5 min. the tissues were incubated with an Alexa Fluor 488 conjugated goat anti-rabbit antibody (Invitrogen, Cat No. A11034) diluted in 1/1000 in blocking buffer for 1h at room temperature. After washing, the tissue was coverslipped. Sections were imaged and morphological changes were determined by Imaris surface analysis. Specifically, for the Imaris three-dimensional (3D) modeling and analysis sets of 50 serial images for Iba-1⁺ cells in the cortex at 500 nm steps in the Z direction, 1 µm per pixel in the X, Y-plane, and 4.92 µsec pixel dwell time were acquired with the Zeiss LSM 710 Confocal Microscope and 10x objective with 2.0 zoom, resulting in whole datasets of 425 µm x 425 µm in the X, Y-plane and 25 µm in the Z direction. The images were rendered into 3D with Imaris software (version 9.3.1; Bitplane, Oxford Instruments) and analyzed with the software’s automated Filament Tracer module. Seed point and starting point thresholds were manually adjusted to fit each data set and local contrast threshold was set to three for all data sets. The number of Iba-1⁺ cells from the medial prefrontal cortex, cortex and hippocampus were counted using ImageJ (NIH).

Brains were immediately extracted following perfusion with cold PBS and kept on ice in Hanks Balanced Salt Solution (HBSS) buffer (Fisher, Cat No. SH3058801) until further processing. They were minced using as sterile razor blade and enzymatically dissociated for 45 min. in 5ml of StemPro Accutase (Gibco, Cat No. A1110501) at 37°C. The brain homogenate was filtered through a 70µm filter using a 3ml syringe plunger, then pelleted by centrifuging at 400xg for 5 min. Red cells were lysed by osmotic shock (red blood cell lysis buffer; Biolegend, Cat No. 420302) then the cells were washed in sterile PBS by centrifugation at 400xg for 5 min. Cells were suspended in 5ml of PBS solution containing 35% Percoll (VWR, Cat. No. 899428-524), underlaid with 3ml of PBS containing 70% Percoll, then centrifuged for 20 min. at 2000xg with no brake. The cells at the 35/70% Percoll interface were collected and washed with magnetic activated cell sorting (MACS) buffer (PBS with 3% FBS and 10mM EDTA). Microglia were enriched by positive selection using antibodies directed towards CD11b, as previously described. RNA was isolated using the GeneJET™ RNA Purification kit (ThermoFisher Scientific, Cat No. 0732) by following the manufacturer’s instructions. SMART-Seq v4 PLUS Kit (Takara, Cat No. 634888) was used to generate high-quality cDNA libraries from ultra-low amounts of total RNA (10pg–10 ng). Subsequently high-quality Illumina sequencing-ready libraries were prepared and sequencing was performed using with Illumina Hi-seqlane. The fastq read files were generated and demultiplexed with the bcl2fastq v2.20 Conversion Software (Illumina, San Diego, CA, USA). The quality of the resulting fastq files was evaluated with the FastQC software, which generates reports with the quality scores, base composition, k-mer, GC and N contents, sequence duplication levels and overrepresented sequences. On average 47.9 million 150nt paired-end reads per sample were obtained, with a minimum FastQC score of 34. Alignments and counts were performed on the Carl R. Woese Institute for Genomic Biology Biocluster of the University of Illinois High-Performance Biological Computing. Pair-end reads were first filtered using Trimmomatic 0.33 (49) with a minimum quality score of 28 (i.e., base call accuracy of 99.84%) leading and trailing with a minimum length of 30 bp long and subsequently checked using FastQC 0.11.6 (Babraham Institute, Cambridge, UK). No reads were filtered as all had scores greater than 28. Reads were then mapped to the Mus musculus reference genome (GRCm38 e!Ensembl, downloaded in September 2019) using default settings of STAR 2.6.0 (50). Uniquely aligned reads were quantified using feature Counts (51) in the Subread package (v1.5.2) based on the Refseq gene annotation.

Further data analysis was conducted using R. 3.5.1 (R Core Team, 2018). Reads uniquely assigned to a gene were used for subsequent analysis. Genes were filtered if 3 samples did not have > 1 count per million mapped reads. A TMM (trimmed mean of M-values) normalization was applied to all samples using edgeR (52). After data were log2-transformed, edgeR was used to conduct differential expression analyses. The applied statistical model included treatment (GW2580 vs Control) as fixed effect.

The genome index was prepared using STAR and the Mus musculus DNA FASTA file from e!Ensembl. The reads were aligned to the reference genome to both reads. Reads were counted using a FeatureCount in STAR. Statistical analysis of differentially expressed genes (DEG) was performed using R studio. Gene ontology was determined by DAVID Functional AnnotationBioinformatics Microarray Analysis (v6.8). Data is publically available through Gene Expression Omnibus (GEO) from NCBI at (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE185564) and can be accessed using the accession no. GSE185564

Glial cultures were prepared from the brains of 4-10 C57BL/6 neonatal mice per experiment using the differential attachment method as described previously (3, 53). In brief, P1–2 mouse pups were decapitated with scissors, brains dissected and meninges removed under a dissection microscope (Leica). Brain tissue was disassociated in Accutase at 37°C for 45 min., passed through a 70μm filter, washed with Dulbecco’s Modified Eagle Medium (DMEM), seeded onto poly-d-lysine-coated T75-flasks, and grown to confluence (7–10 days) in a humidified incubator at 37°C and 5% CO₂. Microglia were isolated by shaking the T-flasks at 37°C for 1h at 177rpm in an orbital shaker (ThermoScientific Max Q 4000).

Primary microglia were seeded at a density of 1x10⁵ cells per well in a 96-well plate in DMEM supplemented with 10% FBS, 100U pen/strep and glutamax. To determine the effect of CSF1 and GW2580 on cell proliferation, microglia were treated with recombinant mouse CSF1 (0-30ng/ml; R&D Systems; Cat No. 416-ML-10) for 48h with increasing concentrations of GW2580 (0-5μM). After treatment, cells were fixed with 4% paraformaldehyde for 20 min. at room temperature, washed with PBS then blocked and permeabilized in PBS containing 0.3% Triton-X 100 (PBST) and 5% goat serum for 1h at room temperature. Next, the cells were incubated with a rabbit anti-Ki-67 antibody (Dilution 1:200; Cell Signaling; Cat No. 9129S) overnight at 4°C. After washing with PBS, cells were incubated with Alexa Fluor 488 conjugated goat anti-rabbit IgG diluted 1/1000 in PBST for 1h. (ThermoScientific). After washing with PBS, proliferation was determined by counting the number of microglia per well, as well as the number of Ki-67⁺ cells per well using ImageJ (NIH).

To validate the effect of GW2580 on the expression of select genes involved in the regulation of ROS, qRT-PCR was used. Here, primary microglia were stimulated with 5µM of GW2580 or media alone for 72h in a humidified incubator at 37°C and 5% CO₂. RNA was isolated using TRIzol protocol (Ambion by Life Technologies, Cat No., 15596018) and further purified using the GeneJET RNA Purification Kit (Thermo Scientific, Cat No., K0731). cDNA was created using the reverse transcription system (Promega Corporation, Cat No., A3500), and amplified using Sybr Green qPCR master mix (Applied Biosystems, Cat No., A25742), and primers specific for Ndufs8 (Fwd-CGCAGCACTTCAAGATGTATCG; Rev-TCTTGGCTCAGCCTCAATGG), Prdx2 (Fwd-CAACCACCGCCAGAATTGC; Rev-AACATTGTGGATGGCTTGGC), Prdx5 (Fwd-AGGGTACCCAACCCTGTTCT; Rev-GCACAGTAGTACACAGCCGA), Gpx4 (Fwd-CGCCAAAGTCCTAGGAAACG; Rev-TATCGGGCATGCAGATCGAC), and Actb (Fwd-GATTACTGCTCTGGCTCCTAG; Rev-GACTCATCGTACTCCTGCTTG) in a 7300 Real-Time PCR System (Applied Biosystems, ABI) at 95°C for 10min, followed by 40 cycles of 95°C for 15secs and 60°C for 1min. Expression was normalized to Actb expression and control samples. Fold change was calculated using the formula 2^-ΔΔCt.

Primary microglia (1x10⁶ per condition) were stimulated with 100ng/ml of LPS for 0, 6 or 24h in a 1.5ml tube in complete RPMI media at 37°C and 5% CO₂. Cells stimulated with media containing 75µM hydrogen peroxide were used as a positive control. After the stimulation, light-protected cells were stained with 5µM of CellROX Green Reagent (ThermoFisher Scientific, Cat No. C10444) for 30 min. at 37°C and 5% CO_2. The cells were washed twice with 1ml of cold flow buffer (PBS, 2% FBS) then suspended in 100µl of flow buffer. The cells were stained with viability dye (APC Alexa Flour 780, Invitrogen, Cat No. 50-112-9035) for 20 min on ice, washed, then suspended in 500µl of flow buffer and analyzed using flow cytometer. Gates were determined using a complete unstained sample.

To assess the effect of GW2580 on cell viability, microglia were incubated with or without GW2580 at a concentration of 5µM for 45 min. prior to stimulation with vehicle (medium) or lipopolysaccharide from Escherichia coli O111:B4 (100ng/ml) for 48h. TNF production was measured by ELISA (ThermoFisher Scientific, Cat No. 88-7324-88) following the manufacturer’s instructions. For some experiments, microglia were pre-treated with the ROS inhibitor YCG063 (10µM, Sigma Aldrich, Cat No. 557354). Specifically, cells were pre-treated with either YCG063 or YCG063 plus GW2580 for 45 min. prior to LPS stimulation for 48h. To determine if GW2580 treatment of microglia increased sensitivity to ROS, cells were pretreated with GW2580 (5μM) for 48h then challenged with 0-100µM of hydrogen peroxide (H₂O₂; Fisher Scientific, Cat No. H325-100) for 4h. Cell death was assessed by measuring supernatant lactate dehydrogenase (LDH) levels as indicated by manufactures instructions (Sigma Aldrich, Cat No. 4744926001) as well as by counting the number of cells per field.

Primary microglia (1x10⁵ per condition) were pre-treated with either Nec-1 (30nM) or Nec-1 plus GW2580 for 45 min. prior to LPS (100ng/ml) stimulation for 48h in a 96 well plate. After treatment, cells were fixed with 4% paraformaldehyde for 20 min. at room temperature, washed with PBS then blocked and permeabilized with PBS containing 0.3% Triton-X 100 (PBST) and 5% goat serum for 1h at room temperature. Next, the cells were incubated with a rabbit anti-Iba-1 antibody (diluted 1/200; Rabbit; Wako Inc. Richmond, VA, Cat No. 019-19741) overnight at 4°C. After washing with PBS, cells were incubated with Alexa Fluor 488 conjugated goat anti-rabbit IgG diluted 1/1000 in PBST for 1h. (ThermoScientific). Cell death was assessed by measuring supernatant lactate dehydrogenase (LDH) levels as indicated by manufacturer’s instructions (Sigma Aldrich, Cat No. 4744926001).

Data are expressed as means ± standard error (S.E.). For the RNA-seq experiment differentially expressed genes (DEG) across time points were determined with a combination of fold-change (> 2.0 or < -2.0) and P-value (< 0.05) thresholds to balance for reproducibility, sensitivity, and specificity or results (54, 55). Statistical analysis of all other data was performed using GraphPad Prism (version 8.0 or higher) software. Student’s t‐tests were used to compare differences between two groups. Analysis of variance (ANOVA) followed by Bonferroni’s post hoc test was used to compare differences between more than two groups. Statistical significance was determined by p-value <0.05. Data are expressed as means ± standard error (S.E.).

Treatment with GW2580 neither reduced body weight nor the degree of food disappearance ( Figures 1A, B ). Furthermore, GW2580 did not alter burrowing behavior ( Figure 1C ). No effect of gender was found for weight, food intake or burrowing behavior ( Supplemental Figure 1 ). Because macrophages and monocytes also express Csf1r (56), we questioned whether treatment affected the percentages of circulating immune cells. Flow cytometry analysis performed on blood samples indicated that treatment did not alter percentages of circulating B cells, T cells, monocytes or granulocytes ( Figure 2 ). Together, these data indicate that treatment with GW2580 does not overtly affect behavior or hematopoiesis, nor does it decrease blood leukocyte viability.

We next questioned whether treatment with GW2580 altered microglia morphology, a commonly used proxy for activation. Imaris™ software was used to non-subjectively determine effects of GW2580 on various morphological characteristics of Iba-1 labeled microglia in 20x images including: mean filament area, filament length, filament diameter, filament volume, mean number of branches, mean number of branch points, mean number of segments and mean number of terminal points ( Figure 3A and Supplemental Table 1 ). The cortex was chosen as a representative region of morphological changes after GW2580 treatment (57). We found that GW2580 treatment altered morphological characteristics of brain microglia in the cortex. Specifically, the filament length was increased, but filament diameter, number of branches, number of branch points, the number of segments and the mean number of terminal points were decreased as a result of GW2580 treatment ( Figure 3B ). Interestingly, morphological analysis of Iba-1⁺ microglia in 40x images showed similar results, where the number of branch points were decreased in the treated mice, as well as the number of segments ( Supplemental Figure 2 ). However, as in previous studies (40), GW2580 treatment for 8 days did not change the number of microglia in medial prefrontal cortex, cortex or cerebellum ( Figure 3C ).

To access the in vivo effects of CSF1R inhibition, we performed RNA sequencing on freshly isolated microglia from GW2580 treated mice. Microglia were enriched by MACS using CD11b microbeads. This produced greater than 95% enrichment of cells phenotypically characterized as being CD11b⁺CD45^int by flow cytometry ( Figures 4A, B ). The majority of the contaminating cells (5%) were found to be CD11b⁺CD45^hi monocytes/granulocytes ( Figures 4A, B ). To validate the purity of microglia after RNA-seq was performed, we examined the expression of genes specific for microglia (Tmem119, Hexb, Fcrls, Siglech), astrocytes (Gfap, Aldh1l1, Aqp4), oligodendrocytes (Olig2, Plp1, Sox10), endothelial cells (Pecam1, Cldn5, Tie2) and neurons (Sp9, Reln) ( Supplemental Table 2 ). This analysis revealed that the RNA was enriched for microglia specific genes, confirming that the sequenced cells were microglia ( Figure 4C and Supplemental Table 2 ). In total, we identified 577 differentially expressed genes (DEGs), of which 144 were upregulated and 433 were downregulated. Since GW2580 is a CSF1R antagonist we hypothesized that genes involved in the CSF1/CSF1R signaling pathway may be affected by treatment. Therefore, we examined the expression values of the CSF1R ligands Csf1 and Il34 as well as the expression of Csf1r itself. The RNA sequencing analysis showed that Csf1 was upregulated in the GW2580 treated group. Furthermore, while the mean expression value for Csf1r trended towards being higher in the treated group, the effect did not reach statistical significance (p-value of 0.08). Consistent with the fact that Il34 is predominantly expressed by pericytes and neurons (Tabula Muris Consortium, 2018) but not microglia, we found the expression values for Il34 were low. There was no difference in the level of Il34 expression. Analysis of the gene ontology terms indicated that genes encoding proteins involved in glutathione metabolic process, response to oxidative stress, removal of superoxide radicals and response to estradiol were decreased by treatment ( Figure 4D and Supplemental Table 3 ). Additionally, ontology analysis of the upregulated genes showed that GW2580 affects expression of genes involved in G-protein coupled receptor signaling pathway and sensory perception of smell ( Supplemental Table 4 ).

We sought to determine if GW2580 treatment affected microglia in culture. Unlike microglial cell lines, such as BV-2 cells, primary mouse microglia monocultures do not readily proliferate in culture. However, as reported by others (58, 59) we found that CSF1 treatment for 48h increased cell number as well as the number of Ki-67⁺ cells per well in a dose-dependent fashion without affecting cell viability ( Figures 5A, B ). These data indicate that primary microglia are responsive to CSF1. To establish the dose at which GW2580 affects microglial function, primary cultures were pretreated with increasing concentration of GW2580 then stimulated with either media or CSF1. As before, CSF1 treatment increased the number of Ki-67⁺ cells. However, concurrent treatment with GW2580 ameliorated this effect, without affecting cell viability ( Figures 5C, D ). These findings indicate that cultured primary microglia proliferate in response to CSF1 and confirm that treatment with GW2580 at a concentration of 5μM is sufficient to inhibit the effects of CSF1 signaling.

To confirm the effects of GW2580 on cultured microglia, we examined the expression levels of select genes that comprised the gene ontology term “response to oxidative stress” (Ndufs8, Prdx5), “removal of superoxide radicals” (Prdx2), and “glutathione metabolic process” (Gpx4), identified by our RNA-seq analysis. Treatment of primary microglia cultures with GW2580 suppressed the expression of Ndufs8, Prdx2, Prdx5 and Gpx4 without affecting viability ( Supplemental Figure 3 ).

Stimulation with LPS upregulates ROS production in BV-2 cells (60) and primary microglia (61). In order to validate ROS production by primary microglia after LPS stimulation, we analyzed ROS production by flow cytometry using CellRox. For these experiments treatment with H₂O₂ served as a positive control. As reported by others, we found that both LPS and H₂O₂ treated microglia displayed increased fluorescence after 24h ( Figures 6A, B ). Since our RNA-seq data suggested that GW2580 suppressed genes associated with regulation of ROS, we questioned the effects of treatment on primary cultures stimulated with LPS. For these experiments microglia cultures were stimulated with either media, LPS, GW2580 or LPS and GW2580 together. Treatment with GW2580 or LPS alone had minimal effects on cell viability. In contrast, when microglia were concurrently treated with LPS and GW2580 we observed a marked decrease in cell viability which was validated by increased levels of LDH in the supernatant ( Figures 6C, D ). Notably, this effect was dependent on the concentration of GW2580 added to the culture ( Figure 6D ). As reported for primary peritoneal macrophages treatment did not affect the acute production of TNF from stimulated cultures ( Figure 6E ) (40). Taken together, these data indicate that LPS induces ROS in primary microglia and that CSF1R antagonism decreases viability when microglia are activated by LPS.

Next, we sought to examine a potential relationship between cell death resulting from treatment with LPS and GW2580 and ROS production. To determine if ROS was responsible for microglia death, we replicated the experimental design from previous experiments and pretreated cultures with or without YCG063, a ROS scavenger (62–64). As in previous experiments, microglial cell viability was decreased only when they were cultured in the presence of GW2580 and LPS. This effect was prevented in the presence of the ROS inhibitor YCG063 ( Figures 7A, B ), indicating that ROS was involved in the reduction of cell viability observed when microglia were treated with GW2580 and LPS.

To determine if pretreatment with GW2580 increased the sensitivity of microglia to ROS, cultures were challenged with increasing concentration of H₂O₂ in the presence or absence of drug treatment. Microglia pretreated with GW2580 had increased supernatant levels of LDH when stimulated with 50µM, 75µM and 100µM concentrations of H₂O₂ compared to control cultures ( Figures 7C, D ), indicating that GW2580 increased susceptibility of cells to a direct ROS challenge. Taken together, these data indicate that CSF1 antagonism by GW2580 suppresses the capacity for microglia to respond to oxidative stressors and sensitizes them to ROS when they become reactive.

Treatment mediated dysregulation of ROS may promote cell death by either apoptosis or necroptosis (65–67). However, a time course experiment showed that treatment did not cause appreciable changes in the number of cells positively stained for cleaved caspase-3, indicating apoptosis is not likely to be the mechanism of cell death (data not shown). It was recently shown that Toll-like receptor activation in the context of caspase-8 inhibition caused microglia cell death by necroptosis in a manner contingent upon activation of RIP1/RIP3 complex and ROS production (66). CSF1R ligation by CSF1 also promotes caspase-8 activation (68), which may function to inhibit RIP1/RIP3 mediated necroptosis. To test whether necroptosis was involved in cell death resulting from co-treatment with LPS and GW2580, primary microglia were stimulated as before in the presence or absence of Nec-1, a RIP1 antagonist and potent necroptosis inhibitor ( Figure 8A ). Our data show that treatment with Nec-1 inhibited cell death in LPS and GW2580 stimulated cells ( Figure 8B ), indicating an involvement of necroptosis in treatment mediated cell death.