ORIGINAL RESEARCH published: 20 May 2021 doi: 10.3389/fnmol.2021.582497 Microglia Stimulation by Protein Extract of Injured Rat Spinal Cord. A Novel In vitro Model for Studying Activated Microglia Joaquim Hernández, Isaac Francos-Quijorna, Elena Redondo-Castro, Rubén López-Vales and Xavier Navarro* Group of Neuroplasticity and Regeneration, Department of Cell Biology, Physiology and Immunology, Institute of Neurosciences, Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Universitat Autònoma de Barcelona, Red de Terapia Celular (TerCel), Bellaterra, Spain Research on microglia has established the differentiation between the so-called M1 and M2 phenotypes. However, new frameworks have been proposed attempting to discern between meaningful microglia profiles. We have set up an in vitro microglial activation Edited by: model by adding an injured spinal cord (SCI) lysate to microglial cultures, obtained Amanda Sierra, from postnatal rats, in order to mimic the environment of the spinal cord after injury. Achucarro Basque Center for Neuroscience, Spain We found that under the presence of the SCI lysate microglial cells changed their phenotype, developing less ramified but longer processes, and proliferated. The SCI Reviewed by: Daniel Garcia-Ovejero, lysate also led to upregulation of pro-inflammatory cytokines, such as IL-1β, IL-6, and National Paraplegic Hospital, Spain TNF-α, downregulation of the anti-inflammatory cytokines IL-10 and IL-4, and a biphasic Bente Finsen, University of Southern Denmark, profile of iNOS. In addition, a latex beads phagocytosis assay revealed the SCI lysate Denmark stimulated the phagocytic capacity of microglia. Flow cytometry analysis indicated that *Correspondence: microglial cells showed a pro-inflammatory profile in the presence of SCI lysate. Finally, Xavier Navarro characterization of the microglial activation in the spinal cord on day 7 after contusion xavier.navarro@uab.cat injury, we showed that these cells have a pro-inflammatory phenotype. Overall, these Specialty section: results indicate that the use of SCI lysates could be a useful tool to skew microglia This article was submitted to Molecular Signalling and Pathways, towards a closer phenotype to that observed after the spinal cord contusion injury than a section of the journal the use of LPS or IFNγ. Frontiers in Molecular Neuroscience Keywords: microglia, spinal cord injury, microglia culture, lysate, phagocytosis, mRNA expression, rat Received: 12 July 2020 Accepted: 14 April 2021 Published: 20 May 2021 INTRODUCTION Citation: Hernández J, Francos-Quijorna I, Spinal cord injury (SCI) leads to partial or complete loss of motor, sensory and autonomic functions Redondo-Castro E, López-Vales R below the injury level, due to damage to the local circuitry of the spinal cord and interruption and Navarro X (2021) Microglia of ascending and descending neural pathways (Ahuja et al., 2017). Traumatic SCI causes direct Stimulation by Protein Extract of tissue damage that initiates a cascade of secondary events resulting in expanded tissue loss. There is Injured Rat Spinal Cord. A Novel In vitro Model for Studying a coordinated change in gene and protein expression profile associated with physiopathological Activated Microglia. events, including hemorrhage, excitotoxicity, oxidative stress, neuronal activity imbalance, and Front. Mol. Neurosci. 14:582497. inflammation (Tator and Fehlings, 1991; Popovich, 2014; Siddiqui et al., 2015; Silver et al., 2015; doi: 10.3389/fnmol.2021.582497 Greenhalgh et al., 2020). Frontiers in Molecular Neuroscience | www.frontiersin.org 1 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia Microglia are one of the most important cells that participate the ethical committee of the Universitat Autònoma de Barcelona in the inflammatory response that occurs after SCI. Although in accordance with the European Directive 86/609/EEC. Under they are often considered to be macrophages of the CNS, anesthesia with ketamine (90 mg/kg; ImalgeneR 1000; Boehringer recent studies documented microglia as a unique cell population, Ingelheim, Germany) and xylazine (10 mg/kg; RompunR ; with distinct lineage and molecular signature than macrophages Bayer, Germany) and aseptic conditions, a longitudinal dorsal (Salter and Beggs, 2014). Microglia play many physiological incision was made to expose the T6-T10 spinous processes. A functions in the CNS. For instance, they track tissue environment laminectomy of T8-T9 vertebra was made and a cord contusion from insults or pathogens (Nimmerjahn, 2005), maintain tissue of 100 Kdynes was induced using an Infinite Horizon Impactor homeostasis and play a key role in the development of CNS device (Precision System and Instrumentation, Fairfax Station, (Prinz et al., 2019), control neuronal activity, synaptic maturation VA, USA). The wound was sutured by planes and the animals and plasticity, drive programmed cell death in CNS development, allowed to recover in a warm environment. An intraperitoneal undergo central players in the innate immune response following (i.p.) bolus of saline solution (B.Braun Vetcare, UK) was injury (David and Kroner, 2011; Salter and Beggs, 2014). administered immediately after surgery. To prevent infection, Microglia rapidly change their phenotype in response to amoxicillin (500 mg/L; Normon, Spain) was given in the drinking any disturbance of tissue homeostasis towards a commonly water for 1 week. Postoperative analgesia was provided with referred to as activated phenotype on the basis of changes in buprenorphine (0.05 mg/kg; B.Braun Vetcare, UK) for 48 h. morphology or expression of cell surface antigens (Ransohoff and Bladders were expressed twice a day until reflex voiding was Perry, 2009; David and Kroner, 2011; Franco and Fernández- re-established. Suárez, 2015; Gensel and Zhang, 2015; Prinz et al., 2019). Extracts were obtained from the intact spinal cord or from The specific activity of microglia after injury includes a wide the injured spinal cord harvested at 7 days after the contusion. range of functions, performed through specific roles of polarized Fresh tissue was removed from the lesion site (around 1 cm), microglia at different stages of the injury (Kigerl et al., 2009). and an ultrasonic probe applied to disaggregate it in HBSS It is usually considered as the concept of two different forms buffer (Gibco, USA). The extract was spun down (10,000 g) of microglia/macrophage polarization induced by either Th1 and filtered. Quantification of the protein content from the (IFNγ, LPS) or Th2 (IL-4 and IL-13) mediators. IFNγ/LPS- supernatant was made by the BCA method (Thermo Fisher polarized microglia/macrophages are called M1 or ‘‘classically’’ Scientific, Germany), and a range of 15–20 µg protein/µl was activated microglia and macrophages and are characterized obtained by colorimetry (Bio-Tek Instruments Inc., Germany). by expressing high levels of iNOS, TNF-α, or IL-1β, and Extracts were water-soluble and were solubilized in HBSS buffer. the surface markers CD16/32. On the other hand, IL-4 or IL-13 promotes M2 or ‘‘alternatively activated’’ microglia and Microglial Cultures macrophage. M2 microglia/macrophages are characterized by Glial cell cultures were prepared from 1-day-old Sprague–Dawley expressing Arg1, CD206 and anti-inflammatory cytokines, such rats, as previously described (Redondo-Castro et al., 2013). as IL-10 or TGF-β (Gordon and Taylor, 2005; David and Briefly, following the removal of meninges, brain tissue was Kroner, 2011; Francos-Quijorna et al., 2016). The current view minced and incubated for 10 min at 37◦ C in Ca2+ -free Krebs- of M1/M2 polarization is a simplified model that only represents Ringer buffer containing 0.0025% trypsin (Sigma–Aldrich, two extremes of activation states (Ransohoff, 2016). However, St. Louis, MO, USA). Enzyme dissociated cells were mechanically after CNS injury, microglia are not stimulated by a single factor, triturated through a glass pipette and filtered through a 40- such as IFNγ or IL-4, but by multiple mediators that lead µm nylon mesh (Corning, USA) in the presence of 0.52 mg/ml microglia to adopt different activation states and functions to soybean trypsin inhibitor (Gibco, USA) and 170 IU/ml DNAse those observed in cell culture conditions. For this reason, there (Roche, Switzerland). After centrifugation (500 g), the cells is a need to develop novel in vitro approaches that could drive were stained with Trypan Blue exclusion dye (Sigma–Aldrich, microglia to a closer activation state to that observed in vivo after St. Louis, MO, USA), counted in a Neubauer chamber, a CNS challenge. For this purpose, here we studied the effects and then resuspended in 90% Dulbecco’s modified Eagle of a spinal cord injury lysate on microglia activation as a more medium (DMEM; Gibco, USA), 10% foetal bovine serum (FBS; physiopathological approach than classical addition of IFNγ/LPS Sigma–Aldrich, St. Louis, MO, USA), 20 U/ml penicillin and or IL-4. By studying microglia polarization in vitro under the 20 mg/ml streptomycin (Sigma–Aldrich, St. Louis, MO, USA) at effect of an SCI lysate, we aim to gain knowledge about microglia 3 × 105 cells/ml. phenotype changes and to have a screening assay for comparison Cells were plated and incubated at 37◦ C in a humidified with the in vivo microglia reactivity after SCI. atmosphere of 5% CO2 and 95% air (HeracellTM 150i CO2 incubator; Thermo Fisher Scientific); medium was MATERIALS AND METHODS replaced every 3 days. After 14–15 days in vitro, the culture became confluent and microglial cells were obtained by Spinal Cord Injury shaking the flasks for 2 h at 200 rpm. Floating cells were Adult female Sprague–Dawley rats (9 weeks old; 250–300 g) were spun down to obtain the dislodged cells and cultured at used. The animals were housed with free access to food and 150,000 cells/ml. This enriched fraction (around 95–97%) water at a room temperature (RT) of 22 ± 2◦ C under a 12:12 of microglia was seeded on poly-D-lysine-treated well or light-dark cycle. The experimental procedures were approved by glass cover-slips in 24-well plates (PDL from Sigma–Aldrich, Frontiers in Molecular Neuroscience | www.frontiersin.org 2 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia St. Louis, MO, USA), and maintained in DMEM medium condition, from three biological replicates, were used to perform supplemented with 10% FBS, 20 U/ml penicillin and 20 µg/ml these analyses (ranging between 600 and 800 total cells per streptomycin. After 3 days, microglial cells were ready to replicate). be treated. Agents used for microglia activation were lipopolysaccharide Proliferation Assay at 10 ng/ml (LPS; Escherichia coli 0127:B Sigma–Aldrich, At 24 h upon microglia stimulation, cells were fixed with 4% St. Louis, MO, USA), IL-4 at 10 ng/ml (Affymetrix eBioscience, paraformaldehyde (PFA) for 15 min and immunohistochemistry USA), IFNγ at 10 ng/ml (Affymetrix eBioscience, USA) or an against Iba1 (1:500; Wako, Japan) and Ki67 (1:200; Abcam, extract lysate obtained from the intact and 7 days injured spinal Cambridge, MA, USA) was performed as indicated above. cords (SC lysate and SCI lysate). Different concentrations of the Several images were randomly taken and total cells and labeled lysate were tested depending on the experiment (10, 25, 50 or Ki67 cells were counted. Microphotographs were analyzed using 100 µg/ml). ImageJ software (NIH, USA); Ki67 was represented as percentage vs. total cells. Three biological replicates per condition were used Immunocytochemistry for these analyses. After treatments, coverslips were fixed in 4% paraformaldehyde Survival Assay for 20 min at RT and washed in PBS. Blockade of nonspecific Propidium iodide (PI) is a standard red-fluorescent probe binding was performed with 5% foetal calf serum (FCS) in 0.1% enabling distinguishing viable cells from the dead ones. Cells Triton X-100/PBS. Then, coverslips were incubated in the same were trypsinized and resuspended in 100 µl staining buffer solution at 4◦ C overnight with primary antibodies: anti-rabbit containing PI (50 µg/ml; Sigma–Aldrich, St. Louis, MO, USA), iba1 (1:500; Wako, Japan) or anti-goat iba1 (1:200; Abcam, and incubated for 15 min at RT in the dark (Calderwood and Cambridge, MA, USA), used to detect macrophage/microglial Prince, 2018). Once setting conditions were defined, samples cells; anti-mouse GFAP (1:200; Sigma–Aldrich, St. Louis, MO, were analyzed by flow cytometry (BD Bioscience, USA). There USA), used to detect astrocytes; anti-rabbit Tubulin β-3 (1:200; 10,000 events from each sample were analyzed. Dead cells were Biolegend, USA), used to detect neurons; anti-mouse O4 defined as PI+ cells. Percentages of PI+ cells were calculated from (1:200; Sigma–Aldrich, St. Louis, MO, USA), used to detect three biological replicates. oligodendrocytes; or anti-rabbit Ki67 (1:200; Abcam, Cambridge, MA, USA), used to detect proliferation. After several washes Metabolic Activity in 0.1% Triton X-100/PBS, samples were incubated for 2 h MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium at RT with donkey anti-rabbit Alexa Fluor 488-conjugated bromide (Sigma–Aldrich, St. Louis, MO, USA), was used to antibody (1:200; Invitrogen, USA), donkey anti-mouse Alexa evaluate mitochondrial activity as an indirect measure of cell Fluor 488-conjugated antibody (1:200; Invitrogen, USA), donkey metabolic activity, as previously described (Redondo-Castro anti-goat Alexa 488-conjugated antibody (1:200; Invitrogen, et al., 2013; Alé et al., 2015). Microglial cells were incubated USA), donkey anti-rabbit Cy3-conjugated antibody (1:200; with LPS, extract lysate or extract lysate with ibuprofen Jackson IR, USA) or donkey anti-mouse Cy3-conjugated (Sigma–Aldrich, St. Louis, MO, USA). After 24 h incubation, antibody (1:200; Jackson IR, USA), in 0.1% Triton X-100/PBS 0.15 mg/ml of MTT was added and the cells were incubated for with 1.5% FCS at RT. In the last wash, cells were incubated with 3 h at 37◦ C. The formazan crystals were dissolved in 200 µl of DAPI (5 µg/ml; Sigma–Aldrich, St. Louis, MO, USA) to perform dimethyl sulfoxide (DMSO; Sigma–Aldrich, St. Louis, MO, USA) nuclear staining. and 150 µl were passed to 96-well plates. The optimal density was determined with a microculture plate reader (Bio-Tek Morphological Analysis of Microglia Instruments Inc., Germany) at 570 and 620 nm (to counteract Microglia morphology was quantified in fluorescent images the noise of the plastic) and analyzed with KCjuniorTM software from Iba1-immunostained cultures by using ImageJ software (Bio-Tek Instruments Inc., Germany). Three biological replicates (NIH, USA) and skeleton analysis plugin [Analyze Skeleton were done. (2D/3D) from http://imageJ.net/AnalyzeSkeleton]. Analysis was performed according to a previously described procedure RNA Isolation, Reverse Transcription, and (Young and Morrison, 2018; Blasco et al., 2020). For this purpose, Real-Time PCR digital photomicrographs were transformed to 8-bit grayscale Cells were collected in RTL buffer reagent (Qiagen, Germany) and then binarised to obtain a black and white image by and RNA extracted using the RNeasy Mini kit (Qiagen, means of a formerly established threshold. Every image was Germany), according to the protocol of the manufacturer. RNA manually edited to obtain an image with a continuous set of was treated with DNase I (Qiagen, Germany) to eliminate pixels and gaps between processes belonging to neighboring cells. genomic DNA contamination. One microgram of RNA obtained The image was then saved, and the plugin Skeleton (2D/3D) from microglia was primed with random hexamers (Promega, was run. This step results in a tagged skeleton image from USA) and reverse transcribed using the Omniscript RT kit which the number of endpoints and branch length can be (Qiagen, Germany). RNase inhibitor (Roche, Switzerland) was summarized from the resulting output files. Endpoints and added (1 U/µl final concentration) to avoid RNA degradation. process length data are then used to estimate the range of Primer sequences for real-time PCR are specified in Table 1. microglia extensions in the photomicrograph. Four images per Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used Frontiers in Molecular Neuroscience | www.frontiersin.org 3 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia TABLE 1 | Primer sequences for real-time PCR analysis. Gene (Rattus norgevicus) Accession number Primer sequence 50 -30 Size Product size (bp) GAPDH XM_573304.3 Forward: AATTCAACGGCACAGTCAAGGC 22 116 Reverse: TACTCAGCACCGGCCTCACC 20 Interleukin-1β NM_031512.2 Forward: TCCCAAACAATACCCAAAGAAG 22 162 Reverse: CCGACCATTGCTGTTTCC 18 Interleukin-6 NM_012589.1 Forward: ATCTGCCCTTCAGGAACAGCTATG 24 110 Reverse: ACTTGTGAAGTAGGGAAGGCAGTG 24 Interleukin-10 X60675.1 Forward: TCCTTTCACTTGCCCTCATC 20 159 Reverse: CGAGACTGGAAGTGGTTGC 19 TGF-β1 NM_021578.2 Forward: AGGACCTGGGTTGGAAGTG 19 127 Reverse: GTGTTGGTTGTAGAGGGCAAG 21 TNF-α NM_012675.3 Forward: GCGTGTTCATCCGTTCTC 18 190 Reverse: CAGCGTCTCGTGTGTTTC 18 Arginase 1 BC091158.1 Forward: TGGAACGAAACGGGAAGG 18 118 Reverse: CTGGTTCTGTTCGGTTTGC 19 iNOS NM_012611.3 Forward: GAGTGAGGAGCAGGTTGAG 19 158 Reverse: TGCTGTAACTCTTCTGGGTG 20 IL-4 NM_201270.1 Forward: AACAAGGAACACCACGGAG 19 200 Reverse: TTCAGTGTTGTGAGCGTGG 19 as a housekeeping gene. Gene-specific mRNA analysis was Bovine Serum (FBS). Primary antibody labeling was performed performed by SYBR-green real-time PCR using the MyiQ5 real- for 1 h at 4◦ C. Cells were labeled with mouse anti-rat time PCR detection system (Bio-Rad Laboratories, Spain). conjugated antibody CD45-FITC (1:250; BD Bioscience, USA). Previously, we had determined the optimal concentration of For intracellular analysis of iNOS and ArgI, cells were fixed, the cDNA to be used as a template for each gene analysis to washed, permeabilized and incubated with rabbit antibodies obtain reliable CT (threshold cycle) values for quantification against iNOS (1:250; Abcam, Cambridge, MA, USA) and goat (between 800 ng and 1 µg/µl). The thermal cycling conditions antibodies against ArgI (1:200 Santa Cruz, CA, USA). After comprised 3 min polymerase activation at 95◦ C, 40 cycles of 30 min of incubation, cells were washed and stained with 10 s at 95◦ C for denaturation and 30 s at 60◦ C for annealing PE (1:500; Affymetrix eBioscience, USA) or Alexa647 (1:500; and extension (62◦ C in the case of IL-6), followed by a Abcam, Cambridge, MA, USA) conjugated donkey secondary DNA melting curve for determination of amplicon specificity. antibodies against goat or rabbit respectively for 30 min. Finally, CT values were obtained and analyzed with BioRad Software. the samples were washed and fixed in 1% PFA. Fold change in gene expression was estimated using the CT Immune cells from the injured spinal cord were analyzed comparative method (2−∆∆CT ; Livak and Schmittgen, 2001) by flow cytometry. Rats were terminally anaesthetized with an normalizing to GADPH CT values and relative to control overdose of sodium pentobarbital and intracardially perfused samples. Three samples were used per condition from three to with PBS. The spinal cord thoracic segment was harvested, five biological replicates. mechanically triturated and then passed through a cell strainer of 70 µm (BD Bioscience, USA), with DMEM containing Phagocytosis Assay 10% of FBS. The cell suspension was centrifuged at 300 g One day after the treatments, cells were incubated with latex for 10 min at 4◦ C. Samples were split into several tubes microbeads (1 µm, 0.0025%, Sigma–Aldrich, St. Louis, MO, and immunostained. Primary antibody labeling was performed USA) for 30 or 120 min at 37◦ C. After this, the medium was for 1 h at 4◦ C, using DMEM+10% FBS as a buffer. Cells removed, and the cells rinsed twice with glucose-phosphate- were labeled with the following mouse anti-rat conjugated buffered saline (PBS) to wash out the free beads. Then cells were antibodies: CD45-PE-Cy7 (1:250; BD Bioscience, USA) and fixed with 4% paraformaldehyde (PFA) for 15 min. After rinsing, CD11b-FITC (1:250; BD Bioscience, USA). For intracellular labeling against Iba1 (1:500; Wako, Japan) was used. Several staining, cells were fixed with 4% PFA and permeabilized images were randomly taken and the number of cells with beads with Permeabilization Wash Buffer (Invitrogen, USA), followed inside as well as the number of beads in each cell were counted. A by staining with unconjugated rabbit antibodies against iNOS minimum of 300 cells were measured for each condition, from (1:200; Abcam, Cambridge, MA, USA) and goat antibodies three independent biological replicates, using ImageJ software against Arg1 (1:150; Santa Cruz, CA, USA), and rabbit (NIH, USA); the number of beads per cell and percentage of cells (1:1,000 Wako, USA) or goat (1:500 Abcam, Cambridge, MA, with beads were calculated (based on Silva et al., 2011). USA) antibodies for Iba1 for 45 min. After that, we stained with PE (1:500; Affymetrix eBioscience, San Diego, USA) or Fluorescent Activated Cell Sorting (FACS) Alexa 647 (1:500; Abcam, Cambridge, MA, USA) conjugated Analysis donkey secondary antibodies against goat or rabbit respectively The cultured cell suspension was centrifuged twice at 300 g for 30 min. The following combination of markers was used for 10 min at 4◦ C with DMEM containing 10% Foetal to identify microglia (CD45low, CD11b+), peripheral myeloid Frontiers in Molecular Neuroscience | www.frontiersin.org 4 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 1 | Microglia purity. (A) Representative image of confluent and mixed culture after 14 days in vitro. Immunochemical image outlining labeling of GFAP (red), Iba1 (green), and DAPI (blue). Microglial cells were obtained by shaking of mixed culture. Scale bar: 100 µm. (B) Representative images of enriched microglial culture showing labeling of Iba1 (green), GFAP (green), tubulin β-3 (red), O4 (red), and DAPI (blue). Scale bar: 100 µm. cells (CD45high, CD11b+), as previously described (Francos- results analyzed using FlowJor software version 10.0.7 (BD Quijorna et al., 2016). Peripheral myeloid cells were further gated Bioscience, USA). for Iba-1 to distinguish macrophages (CD45high, CD11b+, Iba- 1+) from neutrophils since Iba-1 is not expressed in granulocytes. Statistical Analysis Cells were analyzed on a fluorescent activated cell sorting Data are shown as the mean ± SEM. Statistical comparisons (FACS) Canto flow cytometer (BD Bioscience, USA) and the between groups were made using two-way ANOVA with Frontiers in Molecular Neuroscience | www.frontiersin.org 5 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia Bonferroni post hoc test in mRNA studies. One-way ANOVA of endpoints per cell (Figure 5A). However, the length of these with Tukey post hoc test was used for the assessment of processes was markedly increased by the lysate (Figure 5B). cell counts, MTT assay, proliferation assay, survival assay, phagocytosis studies and microglial phenotype in vitro. Finally, mRNA Expression in Microglial Cultures t-test analyses were used in microglia morphology analysis Treated With SC and SCI Lysates and phenotype studies in vivo (IBM SPSS Statistics software, In order to study the influence of SC and SCI lysate (50 µg/ml) USA). Differences between groups were considered statistically on the expression of inflammatory mediators in microglial cells, significant at p < 0.05. we analyzed the transcript levels of IL-1β, IL-4, IL-6, IL-10, TNF- α, TGF-β1, and iNOS and Arginase 1 at different time points (0, 2, 4, 8, 12 and 24 h) as shown in Figure 4. RESULTS The transcripts of the pro-inflammatory cytokines IL-1β, Purity of Cultured Cells IL-6, and TNF-α reached peak expression at 2 h upon Mixed cell culture exhibits a network of cells, mainly astrocytes, stimulation and then decreased with time. The addition of with microglia around not firmly attached to the plate at 14 days the non-injured SC lysate produced a significant increase of in vitro (Figure 1A). However, once microglia were purified the same pro-inflammatory cytokines, but to a much lower by shaking the mixed culture, this was highly enriched in extent as compared to SCI lysate (Figure 6). Oppositely, the microglial cells and we only observed occasional negligible cell anti-inflammatory cytokines IL-10 and IL-4 were downregulated contamination, mainly by astrocytes and neurons (Figure 1B). significantly by the addition of SCI lysate, whereas TGF- β1 showed a slight, but significant, upregulation at 2 h, and then a downregulation at 12 h (Figure 6). The reduced expression Effect of SCI Lysate on Microglial of IL-4 and IL-10 in microglial cells was also observed by the Morphology, Metabolic Activity, addition of the SC lysate, however, this was observed at later time Proliferation and Survival points as compared to the SCI lysate. Resting microglial cells in culture displayed a small round Interestingly, inducible nitric oxide synthase (iNOS), a morphology with some short processes (Figures 2A–C). marker of pro-inflammatory microglia, showed a biphasic Activation with IFNγ induced many ramified extensions expression after the addition of the SCI, showing a significant compared with the processes appearing with IL-4 activation increase in mRNA expression at 4 h and a later significant (Figure 2A). The addition of lysates from the injured spinal cord decrease from 8 h to 24 h. This early up-regulation in iNOS produced qualitatively greater cell counts and morphological expression mediated by the SCI lysate was not observed by the changes (Figure 2). addition of the SC lysate. Finally, the transcripts of Arginase 1 in We therefore, assess the effects of the SCI lysates in cell microglia did not change upon SCI or SC lysate stimulation. counts, metabolic activity, and morphological changes. These These data provide clear evidence indicating that the SCI lysate experiments revealed that microglial cell counts increased drives microglia to adopt a pro-inflammatory phenotype. after the stimulation with the SCI lysates in a concentration- dependent manner (Figures 2B–D), reaching a plateau at Microglial Phagocytosis 50µg/ml concentration (Figures 2B,D). The effect of the SCI A latex bead phagocytosis assay was performed in microglial lysate on microglia metabolic activity was examined by MTT. cultures in the presence of SC and SCI lysates. Between 300 and When analyzing MTT activity in microglial we found that 900 cells were analyzed from three independent experiments to the addition of SCI lysate produced a significant increase in avoid variability in the number of engulfed beads. Both lysates MTT activity (Figure 2D). Interestingly, ibuprofen, which is an produced an increase in the number of beads detected inside the anti-inflammatory drug (Redondo-Castro and Navarro, 2014), microglial cells. Representative images are shown in Figure 7A. reduced the effects of the SCI lysates on microglial cell counts Whilst in the control conditions, we observed that microglia and MTT activity (Figure 2D). No changes in cell counts and engulfed approximately two beads per cell, either at 30 or MTT activity were observed in microglia upon LPS stimulation 120 min after adding beads in the culture medium, microglia (Figure 2D). phagocytosed an average of four and five beads per cells after the Since microglia counts and MTT activity reached maximum addition of the SC and SCI lysate, respectively (Figure 7B). No levels at 50 µg/ml concentration of the SCI lysate, this significant changes were found in the number of engulfed beads concentration was used for the following studies. per microglia after the addition of SC or SCI lysates. However, the We assessed whether the effect of the SCI lysates on SCI lysate significantly increased the percentage of microglia that microglia counts was due to the ability of the lysate to phagocytosed beads as compared to the SC lysate (Figure 7B). induce cell proliferation (Figures 3A,B). Immunostaining against Ki67 revealed that the SCI lysate triggered a significant Activated Microglial Pattern In vitro proliferation of microglia (Figure 3B). However, the lysate did We analyzed the effects of microglial activation by IFNγ, not lead to microglial cell death (Figures 4A–C). IL-4 or SCI lysate in vitro using flow cytometry. IFNγ We then assessed the morphological changes induced by the was used as a positive control for the pro-inflammatory SCI lysate in microglia. These analyses revealed that microglia phenotype, while IL-4 was used as positive control for the became less ramified by the SCI lysate as revealed by the number anti-inflammatory profile. At 24 h after the treatment, the Frontiers in Molecular Neuroscience | www.frontiersin.org 6 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 2 | Effects of lipopolysaccharide (LPS) and spinal cord injury (SCI) lysates in microglial cell count and metabolic activity. (A) Representative images of microglial cells activated by IFN-γ at 10 ng/ml, IL-4 at 10 ng/ml and SCI lysate at 50 µg/ml, 24 h after treatment. Scale bar: 50 µm. (B) Immunochemical images depicting labeling of Iba1 (green) and DAPI (blue) with increasing concentration of SCI lysate in the medium at 10, 25 and 50 µg/ml. Scale bar: 100 µm. (C) Higher magnification images of control microglia, and microglia activated by LPS or SCI lysate (50 µg/ml), labeled with anti-Iba1 (green) and DAPI (blue). Scale bar: 100 µm. (D) Graph showing microglia cell counts and MTT activity at 24 h upon being exposed to LPS (10 ng/ml), different concentrations of SCI lysate (10, 25, 50 and 100 µg/ml), and a combination of SCI lysate (50 µg/ml) with ibuprofen at 0.5 and 1 µM. Values are expressed as mean and SEM (∗ P < 0.05). n = 3 biological replicates per condition. Frontiers in Molecular Neuroscience | www.frontiersin.org 7 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia expression of anti-inflammatory marker (Arg1) was restricted to a very small population of microglia and macrophages (Figure 9). The expression of iNOS was detected in a low proportion of microglial cells in the uninjured and injured spinal cords (∼27%). However, the proportion of iNOS+ macrophages increased significantly at 7 days post-injury as compared to those found in physiological conditions at very low numbers. These results, together with previous reports in the literature, indicate that microglia and macrophage are skewed towards a more pro-inflammatory phenotype. DISCUSSION Spinal cord injury produces an inflammatory response that includes the rapid activation of microglia and their release of pro-inflammatory mediators (Alexander and Popovich, 2009; David et al., 2015). Events that occur after spinal cord injury are characterized by hemorrhage, apoptosis, inflammation, and changes in the blood-spinal cord barrier and the extracellular matrix. All these processes produce a harmful and inhibitory environment that impairs endogenous regeneration and remyelination (Siddiqui et al., 2015). In addition, the balance of inflammatory and intrinsic repair processes influences the resolution of the neuroinflammatory reaction (DiSabato et al., 2016). The immune cell subsets, as major players involved in the immune response, play a key role in the pathological events after SCI (Plemel et al., 2014; David et al., 2015). The microenvironment of the injured spinal cord in the acute phase contains many mediators, such as cytokines, chemokines, reactive oxygen species and secondary messengers (DiSabato FIGURE 3 | Effects of injured spinal cord (SCI) lysates in microglial et al., 2016). The SCI lysate obtained 7 days after lesion contains proliferation. (A) Representatives immunocytochemical images of microglial a profile of cytokines, pro-inflammatory mediators and other cells activated by SC and SCI lysate at 50 µg/ml, 24 h after treatment. factors that come from the extracellular matrix and cellular Images depicting labeling of Iba1 (green), DAPI (blue), and Ki67 (red). Scale bar: 100 µm. (B) Results from counting Ki67 + cells after treatments. Values contents of the lesion site and surrounding. We used this are expressed as mean and SEM (∗ P < 0.05). n = 3 biological replicates per cocktail to study the response of microglia in culture. Microglia condition. responded rapidly to the addition of the SCI lysate, showing marked changes in morphology, activity, and proliferation. Changes in microglia under the presence of SCI lysate in the expression of Arg1 was barely detectable in microglial cells medium were compared with classical administration of LPS, stimulated with IFNγ or the SCI lysate, whereas ∼20% of the or the addition of IL-4 and IFNγ. In summary, SCI lysate cells expressed Arg1 upon IL-4 treatment (Figures 8A–C). In induces changes in cell morphology (less ramified but longer contrast, ∼90% of microglia expressed iNOS upon IFNγ or SCI processes), increases proliferation but does not affect survival. lysate stimulation, whereas this enzyme was observed in ∼75% In a previous study, we studied the phenotype of neural stem of microglia after IL-4 (Figures 8B,C). Therefore these data cells (NSCs) derived from human induced pluripotent stem cells indicate that phenotypically, microglial cells exposed to the SCI (iPSCs) under the presence of an SCI lysate (Lopez-Serrano lysate were more similar to microglia stimulated with IFN-γ, and et al., 2016), showing increased proliferation of iPSC-derived thus, further support the microglia adopt a pro-inflammatory NSCs without changes in their immature stage, due to unknown phenotype after the addition of the SCI lysate. factors present in the SCI lysate. It is well known that ischemia and immune infiltration can lead to oxidative stress and free Microglial Phenotype After Spinal Cord radical production in the lesion area (Donnelly and Popovich, Injury Analyzed by Flow Cytometry 2008). Ionic dysregulation following SCI contributes to cell By using FACS, we assessed the changes in expression damage and loss (Vosler et al., 2009). Furthermore, increasing of iNOS and Arg1 markers in microglia and macrophages glutamate levels, upregulating production of cytokines, matrix 7 days after spinal cord contusion injury in the rat. We metalloproteinases (MMPs) or superoxide dismutase result found that microglia and macrophages expressed mainly ultimately in excitotoxic cell death (Donnelly and Popovich, the pro-inflammatory marker iNOS after SCI, whereas the 2008; Siddiqui et al., 2015; DiSabato et al., 2016). Frontiers in Molecular Neuroscience | www.frontiersin.org 8 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 4 | Effects of SCI lysates in microglial cell death. (A) Representative setting from propidium iodide (PI) in fluorescent activated cell sorting (FACS) density plot and histogram, showing how live and dead microglia were gated. (B) Representative plots from control and microglial cells activated by SC and SCI lysate at 50 µg/ml, 24 h after treatment. (C) Graphs showing the quantification of (B) expressed as PI+ cells (dead cells in each condition). Values are expressed as mean and SEM. n = 3 biological replicates per condition. Frontiers in Molecular Neuroscience | www.frontiersin.org 9 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 5 | Effects of SCI lysates in microglia morphology. (A) Example of original photomicrograph from control and cells treated with SCI lysate (50 µg/ml), and conversion to binary and skeletonized image. Cropped cell corresponds to the box in the skeleton image. Scale bar: 100 µm. Scale bar: 20 µm in cropped images. (B) Summary data of microglia endpoints/cell and process length/cell in control and after SCI lysate treatment. Values are expressed as mean and SEM (∗ P < 0.05). n = 3 biological replicates per condition. The addition of a SCI lysate was a strategy to mimic the pro-inflammatory cytokines was transient and restricted in time, in vivo environment under controlled culture conditions. While due to the SC lysate content, without free radicals and oxidative microglia have been classically classified as either resting or stress compounds. The in vitro phagocytosis assay (Redondo- activated (Streit et al., 1988), recently, some reports have used Castro et al., 2013) revealed in our study that the SCI lysate transcriptome profiles to identify genes selectively expressed induced a phagocytic phenotype in the microglia, indicated by in microglia. This microglial signature revealed a unique the high number of cells with engulfed beads, compared with combination of genes, distinct from other immune cells (Wes control cells or microglia with the SC lysate. et al., 2016). Our results revealed the response of microglial cells We also analyzed the microglial profile in vitro by flow to an SCI lysate in vitro, showing a specific mRNA expression cytometry. iNOS and Arg1 intracellular markers were used pattern, with a rapid and significant increase of IL-1β, IL- to characterize the microglia phenotype after CD11b+ CD45+ 6, and TNF-α expression, downregulation of IL-10 and IL- gating. In the presence of SCI lysate microglial cells showed 4, and no changes in Arginase 1. iNOS showed a biphasic an increased iNOS expression, and a slight decrease in Arg response with an early upregulation and a late downregulation. I expression, results that were similar to those found with This cytokine profile is dynamic in time, between 2 and 24 h, IFN-γ treatment and in agreement with previous studies in indicating a phenotypic adjustment to the environment. The macrophages and microglia (Mantovani et al., 2004; Franco and whole expression profile clearly indicates a proinflammatory Fernández-Suárez, 2015). Thus, the SCI lysate promotes the phenotype, in agreement with other studies (David and Kroner, microglia to adopt a proinflammatory phenotype as similar to 2011). Microglia exposed to a non-injured SC lysate showed also what takes place in vivo after injury or in vitro after IFN-γ a specific mRNA pattern profile. In this case, the increase in exposure. Frontiers in Molecular Neuroscience | www.frontiersin.org 10 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 6 | mRNA expression from microglia cultures. Real-time PCR analyses of different mRNAs expressed in microglia cultures, with the addition of SC lysate and SCI lysate (at 50 µg/ml). Samples were taken in a time-course culture at 0, 2, 4, 8, 12 and 24 h after treatment. Values are expressed as mean and SEM (∗ P < 0.05). n = 3–5 biological replicates per condition. Frontiers in Molecular Neuroscience | www.frontiersin.org 11 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 8 | Effects of SCI lysates in microglia phenotype in vitro. (A,B) Representative FACS plots Arg1 and iNOS levels in cultured microglial cells. (C) Graph showing the quantification of cultured microglial cells expressing ArgI and iNOS after IFN-γ at 10 ng/ml, IL-4 at 10 ng/ml and SCI lysate at FIGURE 7 | Effects of SCI lysates in microglia phagocytosis. (A) 50 µg/ml activation. Values are expressed as mean ± SEM (n = 4 biological Representative images of the phagocytic assay (at 30 and 120 min), showing replicates per condition; ∗ P < 0.05). microglial cells labeled against Iba1, and green-labeled beads inside. Cell images are from control, SC lysate and SCI lysate treatment (at 50 µg/ml). Scale bar: 50 µm. (B) Results from phagocytic assay expressed as the mean and CD45 markers by FACS (Chiu et al., 2013), or Percoll of beads engulfed per cell in each condition, and percentage of cells with ingested beads, at 30 and 120 min after treatments. Values are expressed as gradient plus CD11b staining with magnetic beads (Nikodemova mean and SEM (∗ P < 0.05). n = 3 biological replicates per condition. and Watters, 2012). More recently, a modified method was published using Percoll gradient followed by CD11b staining to obtain an enriched neural cell suspension, in order to assess the To further analyze the proportion of microglial cells ratio of proinflammatory and antiinflammatory cells, a useful expressing each phenotypic marker we used flow cytometry tool to study neuroinflammation in traumatic CNS injury (Bedi analysis. Flow cytometry has been used in different studies to et al., 2013). The cell suspension was identified by flow cytometry quantify or characterize microglia from in vivo samples. In with CD11b+ CD45lo and categorized as proinflammatory or one of the first studies, two cell-isolation methods based on antiinflammatory macrophage/microglia based on CD206 and mechanical dissociation and Percoll gradient separation were Fc γRII/III expression. performed, and thereafter CD11b, CD45, and αβ T-cell receptor Our studies on in vivo samples included CD11b and antibodies were employed for analysis (Campanella et al., 2002). CD45 staining to gate properly the population of microglia An SCI model was used to analyze the resolution of acute (CD45low , CD11b+ ) or peripheral myeloid cells (CD45high , inflammation by Prüss et al (Prüss et al., 2011), defining CD11b+ ), as in previous studies in mice (Francos-Quijorna et al., the existence of a resolution plateau index and characterizing 2016). Macrophages were gated from the peripheral myeloid the non-resolving aspects of inflammation after CNS injury. cells based on the expression of Iba1, since this marker is not Some methods have been published to isolate microglia from expressed in neutrophils. Microglia and macrophages in mice mouse brain in order to assay expression profile, based in express predominantly proinflammatory markers for the first separation of microglia with discontinuous Percoll gradient and 2 weeks after SCI, whereas the expression of M2 is scarce (Kigerl flow cytometry isolation using CD45 (Cardona et al., 2006), a et al., 2009; Francos-Quijorna et al., 2016). Here, we observed that modified protocol using magnetic bead purification and CD11b microglia/macrophages profile on day 7 after contusion injury Frontiers in Molecular Neuroscience | www.frontiersin.org 12 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia FIGURE 9 | Flow cytometer analysis of microglia in vivo. (A) Representative fluorescent activated cell sorting (FACS) density plots of naïve and injured spinal cord showing how microglia and macrophages were gated (>10,000 cells cd45+ cd11b+). (B) FACS histogram plots showing iNOS and Arg1 expression in microglia and macrophages. (C) Graphs showing the quantification of (B) microglia and (C) macrophages expressing Arg1 and iNOS markers 7 days post operation (dpo). Note that the percentage of microglia and macrophages expressing iNOS is markedly higher than those expressing Arg1. Values are expressed as mean ± SEM (n = 3 per group; ∗ P < 0.05). in rats was characterized by an increased iNOS+ expression in Many endogenous factors appear involved in the phenotype macrophages, and minimal expression of Arg1+ macrophages switching of inflammatory cells after CNS injury, including and microglial cells, suggesting that these cells also adopt a cytokines, microRNA, or STAT molecules (Hu et al., 2014; proinflammatory phenotype similar to that observed after spinal Akhmetzyanova et al., 2019). Effective models to approach cord injury in mice. In addition, this proinflammatory phenotype the SCI microenvironment will allow at investigating signaling was also acquired in vitro when microglia were exposed to the pathways, and further identification of regulatory molecules SCI lysate. and microglia profiles that could ultimately accelerate research Frontiers in Molecular Neuroscience | www.frontiersin.org 13 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia towards clinical applicability. Here, we show that SCI lysate could RL-V, and XN wrote the manuscript. All authors contributed to be an effective strategy to mimic the SCI milieu conditions and a the article and approved the submitted version. useful tool for studying microglia activation. FUNDING DATA AVAILABILITY STATEMENT This research was supported by funds from Ministerio de Ciencia The raw data supporting the conclusions of this article will be e Innovación, AEI and ERDF EU funds (PCI2018-093029, made available by the authors, without undue reservation. SAF2016-79774-R) to XN and RL-V. This research also received funds from Red de Terapia Celular (TERCEL, RD16/0011/0035) ETHICS STATEMENT and CIBERNED (CB06/05/1105), both from the Instituto de Salud Carlos III of Spain, co-funded by European Union (ERDF The animal study was reviewed and approved by Ethical funds). committee of the Universitat Autònoma de Barcelona in accordance with the European Directive 86/609/EEC. ACKNOWLEDGMENTS AUTHOR CONTRIBUTIONS We thank Monica Espejo, Jessica Jaramillo, Marta Morell, and JH, RL-V, and XN designed the study. JH, IF-Q, and ER-C Cristina Gutiérrez from INc-UAB culture technological platform performed the research, analyzed or interpreted results. JH, IF-Q, for excellence technical assistance. REFERENCES Donnelly, D. J., and Popovich, P. G. (2008). Inflammation and its role in neuroprotection, axonal regeneration and functional recovery after spinal Ahuja, C. S., Wilson, J. R., Nori, S., Kotter, M. R. N., Druschel, C., Curt, A., cord injury. Exp. Neurol. 209, 378–388. doi: 10.1016/j.expneurol.2007. et al. (2017). Traumatic spinal cord injury. Nat. Rev. Dis. Primers 3:17018. 06.009 doi: 10.1038/nrdp.2017.18 Franco, R., and Fernández-Suárez, D. (2015). Alternatively activated microglia Akhmetzyanova, E., Kletenkov, K., Mukhamedshina, Y., and Rizvanov, A. (2019). and macrophages in the central nervous system. Prog. Neurobiol. 131, 65–86. Different approaches to modulation of microglia phenotypes after spinal cord doi: 10.1016/j.pneurobio.2015.05.003 injury. Front. Syst. Neurosci. 13:37. doi: 10.3389/fnsys.2019.00037 Francos-Quijorna, I., Amo-Aparicio, J., Martinez-Muriana, A., and López- Alé, A., Bruna, J., Herrando, M., Navarro, X., and Udina, E. (2015). Toxic effects Vales, R. (2016). IL-4 drives microglia and macrophages toward a phenotype of bortezomib on primary sensory neurons and schwann cells of adult mice. conducive for tissue repair and functional recovery after spinal cord injury. Glia Neurotox. Res. 27, 430–440. doi: 10.1007/s12640-014-9514-8 64, 2079–2092. doi: 10.1002/glia.23041 Alexander, J. K., and Popovich, P. G. (2009). Neuroinflammation in spinal cord Gensel, J. C., and Zhang, B. (2015). Macrophage activation and its role in repair injury: therapeutic targets for neuroprotection and regeneration. Prog. Brain and pathology after spinal cord injury. Brain Res. 1619, 1–11. doi: 10.1016/j. Res. 175, 125–137. doi: 10.1016/S0079-6123(09)17508-8 brainres.2014.12.045 Bedi, S. S., Smith, P., Hetz, R. A., Xue, H., and Cox, C. S. (2013). Gordon, S., and Taylor, P. R. (2005). Monocyte and macrophage heterogeneity. Immunomagnetic enrichment and flow cytometric characterization of mouse Nat. Rev. Immunol. 5, 953–964. doi: 10.1038/nri1733 microglia. J. Neurosci. Methods 219, 176–182. doi: 10.1016/j.jneumeth.2013. Greenhalgh, A. D., David, S., and Bennett, F. C. (2020). Immune cell regulation 07.017 of glia during CNS injury and disease. Nat. Rev. Neurosci. 21, 139–152. Blasco, A., Gras, S., Mòdol-Caballero, G., Tarabal, O., Casanovas, A., Piedrafita, L., doi: 10.1038/s41583-020-0263-9 et al. (2020). Motoneuron deafferentation and gliosis occur in association Hu, X., Leak, R. K., Shi, Y., Suenaga, J., Gao, Y., Zheng, P., et al. (2014). with neuromuscular regressive changes during ageing in mice. J. Cachexia Microglial and macrophage polarization—new prospects for brain repair. Nat. Sarcopenia Muscle 11, 1628–1660. doi: 10.1002/jcsm.12599 Rev. Neurol. 11, 56–64. doi: 10.1038/nrneurol.2014.207 Calderwood, S. K., and Prince, T. L. (Eds) (2018). Methods and Protocols. New Kigerl, K. A., Gensel, J. C., Ankeny, D. P., Alexander, J. K., Donnelly, D. J., and York, NY: Springer Nature. Popovich, P. G. (2009). Identification of two distinct macrophage subsets with Campanella, M., Sciorati, C., Tarozzo, G., and Beltramo, M. (2002). Flow divergent effects causing either neurotoxicity or regeneration in the injured cytometric analysis of inflammatory cells in ischemic rat brain. Stroke 33, mouse spinal cord. J. Neurosci. 29, 13435–13444. doi: 10.1523/JNEUROSCI. 586–592. doi: 10.1161/hs0202.103399 3257-09.2009 Cardona, A. E., Huang, D., Sasse, M. E., and Ransohoff, R. M. (2006). Isolation Livak, K. J., and Schmittgen, T. D. (2001). Analysis of relative gene expression of murine microglial cells for RNA analysis or flow cytometry. Nat. Protoc. 1, data using real-time quantitative PCR and the 2−∆∆CT method. Methods 25, 1947–1951. doi: 10.1038/nprot.2006.327 402–408. doi: 10.1006/meth.2001.1262 Chiu, I. M., Morimoto, E. T. A., Goodarzi, H., Liao, J. T., O’Keeffe, S., Lopez-Serrano, C., Torres-Espin, A., Hernandez, J., Alvarez-Palomo, A. B., Phatnani, H. P., et al. (2013). A neurodegeneration-specific gene-expression Requena, J., Gasull, X., et al. (2016). Effects of the spinal cord injury signature of acutely isolated microglia from an amyotrophic lateral sclerosis environment on the differentiation capacity of human neural stem cells mouse model. Cell Rep. 4, 385–401. doi: 10.1016/j.celrep.2013.06.018 derived from induced pluripotent stem cells. Cell Transplant. 25, 1833–1852. David, S., Greenhalgh, A. D., and Kroner, A. (2015). Macrophage and microglial doi: 10.3727/096368916X691312 plasticity in the injured spinal cord. Neuroscience 307, 311–318. doi: 10.1016/j. Mantovani, A., Sica, A., Sozzani, S., Allavena, P., Vecchi, A., and Locati, M. neuroscience.2015.08.064 (2004). The chemokine system in diverse forms of macrophage activation David, S., and Kroner, A. (2011). Repertoire of microglial and macrophage and polarization. Trends Immunol. 25, 677–686. doi: 10.1016/j.it.2004. responses after spinal cord injury. Nat. Rev. Neurosci. 12, 388–399. 09.015 doi: 10.1038/nrn3053 Nikodemova, M., and Watters, J. J. (2012). Efficient isolation of live microglia with DiSabato, D. J., Quan, N., and Godbout, J. P. (2016). Neuroinflammation: the devil preserved phenotypes from adult mouse brain. J. Neuroinflammation 9:635. is in the details. J. Neurochem. 139, 136–153. doi: 10.1111/jnc.13607 doi: 10.1186/1742-2094-9-147 Frontiers in Molecular Neuroscience | www.frontiersin.org 14 May 2021 | Volume 14 | Article 582497 Hernández et al. Model for Studying Activated Microglia Nimmerjahn, A. (2005). Resting microglial cells are highly dynamic surveillants bilirubin. J. Neurochem. 117, 412–424. doi: 10.1111/j.1471-4159.2011. of brain parenchyma in vivo. Science 308, 1314–1318. doi: 10.1126/science. 07200.x 1110647 Silver, J., Schwab, M. E., and Popovich, P. G. (2015). Central nervous system Plemel, J. R., Wee Yong, V., and Stirling, D. P. (2014). Immune modulatory regenerative failure: role of oligodendrocytes, astrocytes and microglia. therapies for spinal cord injury—past, present and future. Exp. Neurol. 258, Cold Spring Harb. Perspect. Biol. 7:a020602. doi: 10.1101/cshperspect. 91–104. doi: 10.1016/j.expneurol.2014.01.025 a020602 Popovich, P. G. (2014). Neuroimmunology of traumatic spinal cord injury: a brief Streit, W. J., Graeber, M. B., and Kreutzberg, G. W. (1988). Functional history and overview. Exp. Neurol. 258, 1–4. doi: 10.1016/j.expneurol.2014. plasticity of microglia: a review. Glia 1, 301–307. doi: 10.1002/glia.440 05.001 010502 Prinz, M., Jung, S., and Priller, J. (2019). Microglia biology: one century of evolving Tator, C. H., and Fehlings, M. G. (1991). Review of the secondary injury theory of concepts. Cell 179, 292–311. doi: 10.1016/j.cell.2019.08.053 acute spinal cord trauma with emphasis on vascular mechanisms. J. Neurosurg. Prüss, H., Kopp, M. A., Brommer, B., Gatzemeier, N., Laginha, I., Dirnagl, U., et al. 75, 15–26. doi: 10.3171/jns.1991.75.1.0015 (2011). Non-resolving aspects of acute inflammation after spinal cord injury Vosler, P. S., Sun, D., Wang, S., Gao, Y., Kintner, D. B., Signore, A. P., (SCI): indices and resolution plateau. Brain Pathol. 21, 652–660. doi: 10.1111/j. et al. (2009). Calcium dysregulation induces apoptosis-inducing 1750-3639.2011.00488.x factor release: cross-talk between PARP-1- and calpain-signaling Ransohoff, R. M. (2016). A polarizing question: do M1 and M2 microglia exist. pathways. Exp. Neurol. 218, 213–220. doi: 10.1016/j.expneurol.2009. Nat. Neurosci. 19, 987–991. doi: 10.1038/nn.4338 04.032 Ransohoff, R. M., and Perry, V. H. (2009). Microglial physiology: unique stimuli, Wes, P. D., Holtman, I. R., Boddeke, E. W. G. M., Möller, T., and specialized responses. Annu. Rev. Immunol. 27, 119–145. doi: 10.1146/annurev. Eggen, B. J. L. (2016). Next generation transcriptomics and genomics elucidate immunol.021908.132528 biological complexity of microglia in health and disease. Glia 64, 197–213. Redondo-Castro, E., Hernández, J., Mahy, N., and Navarro, X. (2013). Phagocytic doi: 10.1002/glia.22866 microglial phenotype induced by glibenclamide improves functional recovery Young, K., and Morrison, H. (2018). Quantifying microglia morphology from but worsens hyperalgesia after spinal cord injury in adult rats. Eur. J. Neurosci. photomicrographs of immunohistochemistry prepared tissue using ImageJ. 38, 3786–3798. doi: 10.1111/ejn.12382 J. Vis. Exp. 136, 1–9. doi: 10.3791/57648 Redondo-Castro, E., and Navarro, X. (2014). Chronic ibuprofen administration reduces neuropathic pain but does not exert neuroprotection after spinal cord Conflict of Interest: The authors declare that the research was conducted in the injury in adult rats. Exp. Neurol. 252, 95–103. doi: 10.1016/j.expneurol.2013. absence of any commercial or financial relationships that could be construed as a 11.008 potential conflict of interest. Salter, M. W., and Beggs, S. (2014). Sublime microglia: expanding roles for the guardians of the CNS. Cell 158, 15–24. doi: 10.1016/j.cell.2014.06.008 Copyright © 2021 Hernández, Francos-Quijorna, Redondo-Castro, López-Vales and Siddiqui, A. M., Khazaei, M., and Fehlings, M. G. (2015). Translating Navarro. This is an open-access article distributed under the terms of the Creative mechanisms of neuroprotection, regeneration and repair to treatment of Commons Attribution License (CC BY). The use, distribution or reproduction in spinal cord injury. Pro. Brain Res. 218, 15–54. doi: 10.1016/bs.pbr.2014. other forums is permitted, provided the original author(s) and the copyright owner(s) 12.007 are credited and that the original publication in this journal is cited, in accordance Silva, S. L., Osório, C., Vaz, A. R., Barateiro, A., Falcão, A. S., Silva, R. F. M., et al. with accepted academic practice. No use, distribution or reproduction is permitted (2011). Dynamics of neuron-glia interplay upon exposure to unconjugated which does not comply with these terms. Frontiers in Molecular Neuroscience | www.frontiersin.org 15 May 2021 | Volume 14 | Article 582497