Recasens et al. Journal of Neuroinflammation (2021) 18:31 https://doi.org/10.1186/s12974-020-02063-1 RESEARCH Open Access Chronic exposure to IL-6 induces a desensitized phenotype of the microglia Mireia Recasens1, Beatriz Almolda1* , Jeús Pérez-Clausell2, Iain L. Campbell3, Berta González1 and Bernardo Castellano1 Abstract Background: When the homeostasis of the central nervous system (CNS) is altered, microglial cells become activated displaying a wide range of phenotypes that depend on the specific site, the nature of the activator, and particularly the microenvironment generated by the lesion. Cytokines are important signals involved in the modulation of the molecular microenvironment and hence play a pivotal role in orchestrating microglial activation. Among them, interleukin-6 (IL-6) is a pleiotropic cytokine described in a wide range of pathological conditions as a potent inducer and modulator of microglial activation, but with contradictory results regarding its detrimental or beneficial functions. The objective of the present study was to evaluate the effects of chronic IL-6 production on the immune response associated with CNS-axonal anterograde degeneration. Methods: The perforant pathway transection (PPT) paradigm was used in transgenic mice with astrocyte-targeted IL6-production (GFAP-IL6Tg). At 2, 3, 7, 14, and 21 days post-lesion, the hippocampal areas were processed for immunohistochemistry, flow cytometry, and protein microarray. Results: An increase in the microglia/macrophage density was observed in GFAP-IL6Tg animals in non-lesion conditions and at later time-points after PPT, associated with higher microglial proliferation and a major monocyte/ macrophage cell infiltration. Besides, in homeostasis, GFAP-IL6Tg showed an environment usually linked with an innate immune response, with more perivascular CD11b+/CD45high/MHCII+/CD86+ macrophages, higher T cell infiltration, and higher IL-10, IL-13, IL-17, and IL-6 production. After PPT, WT animals show a change in microglia phenotype expressing MHCII and co-stimulatory molecules, whereas transgenic mice lack this shift. This lack of response in the GFAP-IL6Tg was associated with lower axonal sprouting. Conclusions: Chronic exposure to IL-6 induces a desensitized phenotype of the microglia. Keywords: Axonal sprouting, Primed microglia, T cell, Monocyte, MHCII Background in which microglial cell activation takes place [45, 61]. It is well established that perturbation in the central ner- Recent progress in cell-specific transcriptome profiling vous system (CNS) as a consequence of an acute or has been contributed in deciphering the role of microglia chronic injury leads to microglial cell activation. This ac- in both physiological and pathological conditions, reveal- tivation is totally dependent on the type, location, and ing microglia acquire different cell activation signatures duration of damage as well as the specific environment associated with an acute injury, chronic neurodegenera- tive diseases, or aging [48]. Among the different factors * Correspondence: beatriz.almolda@uab.cat controlling microglial cell activation, cytokines and espe- 1 Department of Cell Biology, Physiology and Immunology, Institute of Neuroscience, Universitat Autònoma de Barcelona, 08193, Bellaterra, cially the balance between pro- and anti-inflammatory Barcelona, Spain molecules play an essential role [46]. Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 2 of 22 Among the plethora of cytokines that can act on littermates of both sexes were used. Animals were main- microglia, IL-6 is a molecule that plays an important tained at constant temperature (24 ± 2 °C) and housed function as a regulator of inflammatory and immuno- on a 12-h light/dark cycle with food and water ad libi- logical responses not only in the periphery [54, 55, 89] tum during all the experiment. All experimental animal but also in the CNS. In the healthy CNS, neurons, glial work was conducted in accordance to Spanish regula- cells, and endothelial cells express low levels of IL-6 and tions (Ley 32/2007, Real Decreto 1201/2005, Ley 9/2003, IL-6R [23, 34, 79, 83, 84]. However, an upregulation of and Real Decreto 178/2004) in agreement with European these two molecules is observed after acute insults, such Union directives (86/609/CEE, 91/628/CEE, and 92/65/ as traumatic brain injury and ischemia, as well as in CEE) and was approved by the Ethical Commission of chronic neurodegenerative conditions like Alzheimer’s the Autonomous University of Barcelona. All efforts and Parkinson’s diseases [11, 13, 30, 71, 86]. Both detri- were made to minimize the number of animals used to mental and beneficial functions have been described for produce reliable scientific data, as well as animal IL-6 [19, 77]. On the one hand, IL-6 has been classified suffering. as a pro-inflammatory cytokine with a noxious role, based on the evidence that, together with other cyto- Perforant pathway transection and experimental groups kines like TNF-α and IL-1β, acts as a major stimulator GFAP-IL6Tg (n = 76) and WT mice (n = 62) were of the inflammatory response in the CNS [11, 30, 35], subjected to wire-knife unilateral perforant pathway promoting neuronal death [3, 22]. Nevertheless, other transection (PPT). Briefly, animals were anaesthetized studies indicated that IL-6 might have anti-inflammatory with intraperitoneal injection of a solution of ketamine and neuroprotective effects after different kind of injur- (80 mg/kg) and xylazine (20 mg/kg) at dose of 0.01 mL/ ies including spinal cord injury [20], ischemia [64], and g body weight. Anaesthetized mice were placed in a sciatic nerve transection [51]. Concerning microglia, stereotaxic device (Kopf Instruments®) and a small win- in vitro and in vivo studies showed the capacity of IL-6 dow in the skull was created by drilling in the left side of to induce activation, increasing microglial cell prolifera- the skull (4.6 mm dorsal to Bregma and 2.5 mm lat- tion, as well as the production of microglia-derived pro- erally). A folded wire-knife (McHugh Milleux, m121) inflammatory molecules [15, 17, 21, 57, 91]. Moreover, was inserted at an angle of 15° anterior and 10° lateral. IL-6 also can drive lymphocyte differentiation as demon- The knife was unfolded at 3.6 mm ventrally and the per- strated by some studies where, in the presence of TGF- forant pathway (PP) was transected retracting the knife b, T cell responses are polarized toward a T-regulatory 3.3 mm. Finally, the knife was folded and removed from phenotype, whereas the addition of IL-6 changes the the brain. After surgery, the skin was sutured with 2-0 phenotype of T lymphocytes toward a Th17 pathogenic silk and the wound cleaned with iodine. phenotype [12]. Non-lesioned (NL) and lesioned animals were distrib- The objective of the present study was to investigate uted in different experimental groups for immunohisto- the effects of chronic exposure to IL-6 on the phenotype chemistry (IHC), flow cytometry, and protein analysis, as of microglia cells, linked to anterograde axonal degener- detailed in Table 1. ation. For this purpose, perforant pathway transection were performed in the GFAP-IL6 transgenic (Tg) mouse 5′ Bromodeoxyuridine injections model, in which chronic production of IL-6 was select- In order to determine microglia/macrophage prolifera- ively targeted to astrocytes in the CNS [16, 17]. This tion, the labeling of proliferative cells with 5′ bromo- type of lesion produces a dense anterograde and ter- deoxyuridine (BrdU) was used. BrdU is a synthetic minal axonal degeneration, glial cell activation, and thymidine analog that incorporates into the DNA of div- axonal sprouting within the molecular layer of the fascia iding cells during S-phase and can be transferred to dentata, a very specific area of the hippocampus, making daughter cells upon replication. Lesioned WT (n = 5) it a very useful tool to study modifications in these re- sponses associated to variations in the microenviron- Table 1 Experimental groups of animals ment. Hence, the PPT paradigm has been extensively NL 2dpl 3dpl 7dpl 14dpl 21dpl used as a model to study the mechanisms that initiate the innate immune response and the microglial response WT IHC (n = 45) 6 6 8 10 8 7 without blood-brain barrier disruption [6, 32]. FC (n = 22) 8 7 7 Protein (n = 12) 3 3 3 3 Materials and methods GFAP-IL6Tg IHC (n = 53) 7 7 9 9 10 11 Animals FC (n = 24) 8 8 8 For the present study, a total of 94 GFAP-IL6Tg animals Protein (n = 17) 3 5 4 5 [17] (6–7 months old) and 79 wild-type (WT) Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 3 of 22 and GFAP-IL6Tg animals (n = 6) were intraperitoneally Toluidine blue staining (i.p.) injected with BrdU (50 mg/kg) diluted in TB (0.05 Sections were mounted onto gelatinized slides, air dried Trizma base, pH 7.4) every 24 h from the day of lesion at RT for 1 h, and then were incubated for 1 min in a to 7 days post-lesion (dpl), and subsequently euthanized solution containing 0.1% toluidine blue diluted in Wal- at 7dpl. pole’s buffer (0.05 M, pH 4.5). After washes in distilled water, sections were dehydrated in graded alcohols, N- butyl alcohol, and after xylene treatment, coverslipped Tissue processing for histological analysis with DPX mounting media. Animals were anaesthetized as described above, but at 0.015ml/g body weight concentration, and then per- Single stain immunohistochemistry fused intracardially for 10 min with 4% paraformalde- Free-floating sections were processed for the study of hyde in 0.1 M phosphate buffer (pH 7.4). Brains were microglial morphology, distribution, density, and pheno- immediately removed and post-fixed for 4 h at 4 °C type using antibodies against Iba1, Pu.1, CD45, MHCII, in the same fixative and, after phosphate buffer rinses, CD206 (Table 2). For the analysis of microglia prolifera- cryopreserved in a 30% sucrose solution in 0.1 M tion, sections were stained with anti-phospho-histone phosphate buffer for 48 h at 4 °C and frozen in ice- H3 (pH3) and BrdU, whereas for microglial cell death cold 2-methylbutane solution (320404, Sigma-Aldrich). evaluation, anti-caspase 3 antibody was used. Finally, A series of horizontal parallel sections (30-μm-thick) lymphocyte recruitment was analyzed using antibody were obtained using a Leica CM3050 cryostat and against CD3, a pan marker for T lymphocytes. In all stored free-floating in Olmos anti-freeze solution at cases, sections were washed several times with 0.05 M -20°C until used. Tris-buffered saline (TBS) pH 7.4 and with TBS Table 2 Reagents used for IHC Target antigen Host Dilution Cat Number Manufacturer Primary antibodies Iba1 Rabbit 1:3000 019-19741 Wako Pu.1 Rabbit 1:400 2258S Cell Signaling CD45 Rat 1:1000 MCA1031G AbD Serotec pH3 Rabbit 1:3000 06-570 Millipore BrdU Rat 1:120 Ab6326 Abcam Caspase 3 Rabbit 1:1000 AF835 R&D Systems Iba1 Rabbit 1:1000 GTX100042 Genetex CD3 Hamster 1:500 MCA2690 AbD Serotec MHCII (IA) Rat hybridoma 1:25 TIB-120 ATCC CD206 Rat 1:500 MCA2235GA AbD Serotec CD28 Rabbit 1:50 ab203084 Abcam CTLA-4 Rabbit 1:25 ab237712 Abcam Laminin Rabbit 1:500 AMP420 Biorad TMEM119 Rabbit 1:1000 ab209064 Abcam GFAP Mouse 1:3000 63893 Sigma Secondary antibodies Biotinylated Rabbit 1:500 BA-1000 Vector Laboratories Biotinylated Hamster 1:500 BA-9100 Vector Laboratories Biotinylated Rat 1:500 BA-4001 Vector Laboratories Alexa 555 Rat 1:1000 A11006 Invitrogen Alexa 488 Rabbit 1:500 A21206 Invitrogen Cy2 Mouse 1:500 PA42003 GE Healthcare Lifescience Streptavidin Alexa Fluor-555 1:500 S32355 Molecular Probes Streptavidin Alexa Fluor-488 1:500 S11223 Molecular Probes Streptavidin-HRP 1:500 SA-5004 Vector Laboratories DAPI 1:10000 D9542 Sigma Aldrich Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 4 of 22 containing 1% Triton-X100 (TBST, pH 7.4). In the case sections were incubated overnight at 4 °C and 1 h at RT of Pu.1 staining, sections were exposed to antigen re- with anti-CD28 and anti-CTLA-4 followed by anti- trieval by treatment with sodium citrate buffer (pH 8.5) rabbit Alexa-Fluor 555-conjugated secondary antibody for 40 min at 80 °C. Endogenous peroxidase was blocked (Table 2). by incubating the sections for 10 min with 2% H2O2 in To analyze the localization of MHCII+ cells, double 70% methanol. For BrdU detection, DNA was denatured immunofluorescence combining MHCII with laminin by first incubating in 0.082 N HCl for 10 min at 4 °C was performed (Table 2). Sections were firstly processed and then for 30 min in 0.82 N HCl at 37 °C. Sections for MHCII as described above, but using Alexa-Fluor were then rinsed with borate buffer (pH 8.5) and 0.5% 555-conjugated streptavidin instead of horseradish Triton X-100 in TBS. Afterwards, all sections were incu- streptavidin-peroxidase (Table 2). Then, sections were bated for 1 h at room temperature (RT) in blocking buf- incubated overnight at 4 °C and 1 h at RT with anti- fer solution (BB) containing 10% fetal bovine serum in laminin followed by anti-rabbit Alexa-Fluor 488- TBST. After that, sections were incubated overnight at 4 conjugated secondary antibody (Table 2). °C followed by 1h at RT with primary antibodies diluted To study the phenotype of MHCII+ cells, double im- in BB, as specified in Table 2. Sections from spleen and munofluorescence combining MHCII with Tmem119, a gut were used as positive control whereas sections incu- defined microglia-specific marker [10], was performed bated in BB lacking the primary antibody were used as (Table 2). Sections were firstly processed for Tmem119 as negative control. After washes with TBST, sections were described above, but using Alexa-Fluor 488-conjugated incubated for 1 h at RT with biotinylated anti-rabbit sec- streptavidin instead of horseradish streptavidin-peroxidase ondary antibody biotinylated anti-hamster secondary (Table 2). After that, sections were incubated overnight at antibody or biotinylated anti-rat secondary antibody 4 °C and 1 h at RT with anti-MHCII followed by anti-rat (Table 2) diluted in BB. After 1 h at RT in streptavidin- Alexa-Fluor 555-conjugated secondary antibody (Table 2). peroxidase, the reaction was visualized by incubating the Sections incubated in BB lacking the primary antibody sections in the 3,3-diaminobenzidine (DAB) kit (SK- were used as negative control and sections from spleen 4100; Vector Laboratories, USA) following the manufac- and spinal cord from EAE-induced mice were used as turer’s instructions. Finally, sections were mounted on positive control for the different immunostains. Finally, slides, counterstained with 1% toluidine blue, dehy- all double-immunostained sections were mounted on drated in graded alcohols, and, after xylene treatment, slides, coverslipped with Fluoromount GTM (0100-01; coverslipped with DPX. Sections were analyzed with a SouthernBiotech, Birmingham, AL), and analyzed using Nikon Eclipse 80i brightfield microscope and photo- Nikon Eclipse E600 fluorescence microscope and a Zeiss graphed with a DXM 1200F Nikon digital camera. LSM 700 confocal microscope using a × 40 objective. Double stain immunohistochemistry Terminal dUTP nick end labeling Double immunolabeling for BrdU with either Iba1 or In addition to caspase-3 immunohistochemistry, the study GFAP was carried out to identify proliferating microglia/ of microglial cell death was performed on sections double macrophages and astrocytes. Sections were firstly proc- stained for terminal dUTP nick end labeling (TUNEL) essed for BrdU as described above, with anti-rat Alexa- and Iba1. Briefly, sections were mounted on slides and Fluor 555-conjugated antibody as secondary antibody treated for 5 min with 100% methanol for endogenous (Table 2). After several washes with TBST and 1 h incuba- peroxidase blocking. Then, sections were rinsed in 10 mM tion in BB at RT, sections were incubated overnight at 4 Tris buffer (pH 8) and 5 mM EDTA followed by incuba- °C followed by 1 h at RT with rabbit anti-Iba1 and mouse tion for 15 min at RT in the same buffer plus Proteinase K anti-GFAP antibody diluted in BB (Table 2). After washes (20 μg/mL). After several washes with 5 mM EDTA, sec- in TBST, sections were incubated with either anti-rabbit tions were incubated for 10 min at RT in TdT buffer con- Alexa-Fluor 488-conjugated or anti-mouse Cy2- taining 30 mM Tris, 140 mM sodium cacodilate, and 1 conjugated secondary antibodies (Table 2). Finally, sec- mM cobalt chloride (pH 7.7). After that, sections were in- tions were washed with TBST, followed by TBS and TB, cubated for 20 min at 37 °C in TdT buffer plus 0.161 U/ and the nuclei stained with 4,9,6-diamidino-2-phenylin- μL TdT enzyme (Terminal Transferase, 3333566 Roche, dole (DAPI) for 10 min (Table 2). Manheim, Germany) and 0.0161 nmol/μL of biotin-16- To study the expression of co-stimulatory molecules dUTP (1093070, Roche, Manheim, Germany). The reac- on T cells double immunolabeling combining CD28 or tion was stopped by submerging sections twice in citrate CTLA-4 with CD3 were performed (Table 2). Sections buffer (300 mM sodium chloride, 30 mM sodium citrate, were firstly processed for CD3 as described above, but 5 mM EDTA) for 5 min. After several washes with TBS, using Alexa-Fluor 488-conjugated streptavidin instead of sections were incubated for 1 h at RT with HRP- horseradish streptavidin-peroxidase (Table 2). Then, conjugated streptavidin (Table 2) and the peroxidase Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 5 of 22 reaction visualized by incubation in a 3,3-DAB kit plus 1% In order to quantify microglial cell density, sections im- of cobalt (SK-4100; Vector Laboratories, USA), following munostained for the transcription factor Pu.1 from a the manufacturer’s instructions. After that, sections were minimum of three NL and three lesioned WT and GFAP- incubated with Iba1 antibody (1:1000) (Table 2) using the IL6Tg at 2, 3, 7, 14, and 21 dpl were analyzed. In this case, same protocol described in the single IHC section. Sec- a total of 6 photographs from 3 different sections per ani- tions from gut were used as positive control. Finally, sec- mal were captured with the × 10 objective, using the same tions were dehydrated in graded alcohols and, after xylene device and software referred above. The number of Pu.1+ treatment, coverslipped in DPX. cells in the ML of the DG was obtained using the “Auto- matic Cell Counter” (ITCN) plug-in from NIH Image J® software (Wayne Rasband, National Institutes of Health, Sulfide-silver staining USA). Data were expressed in cells/mm2. Collateral sprouting in non-lesioned and lesioned ani- Microglial cell proliferation was quantified on sections mals was analyzed at 14 and 21 dpl, using a variation of immunostained for the mitotic marker pH3 as well as the sulfide-silver staining technique, as described by BrdU in NL and PPT-lesioned animals at 2, 3, and 7dpl Danscher [24]. Briefly, animals were injected intraperito- in the case of pH3 and at 7dpl in the case of BrdU. For neally with sodium selenite (10 mg/kg) and 1 h later, an- pH3, a minimum of three WT and three GFAP-IL6Tg imals were perfused and postfixed as described above. animals per group were analyzed, whereas in the case of After that, brains were removed and cryoprotected with BrdU, five WT and six GFAP-IL6Tg animals were used. 30% sucrose solution in 0.1 M phosphate buffer at 4 °C The number of both pH3+ and BrdU+ cells in the ML of for 48 h and frozen as described above. Frozen parallel DG was manually counted on 20 (for pH3) and 10 dif- transversal sections (30 μm thick) were obtained using a ferent sections (for BrdU) per animal using a × 20 ob- cryostat (Leica, CM 3050S) and mounted on slides. Sec- jective. Data were averaged and represented as pH3+ tions were incubated to visualize metal precipitates. cells/section or BrdU+ cells/section. Briefly, slides were rinsed in 95% EtOH for 15 min, then To evaluate T lymphocyte infiltration, sections stained 2 min in 70% alcohol, 50% alcohol, and rinsed again with for CD3 were used. At least three WT and three GFAP- dH2O for 30 min. Following rehydration, sections were IL-6Tg NL animals and at 2, 3, 7, and 14dpl were used. incubated with the developer solution that contains 60 All CD3+ cells in the ML of DG were manually counted mL of arabic gum (0.5 kg/L), 10 mL of sodium citrate on 20 sections per animal using a × 20 objective. Data buffer 0.2 M, 15 mL hydroquinone (0.85 g diluted in 15 were averaged and represented as CD3+ cells/section. mL of dH2O), and 15 mL silver lactate (0.12 g diluted in To analyze axonal sprouting, sections stained with the 15 mL of dH2O) in a water bath at 26 °C protected from sulfide-silver staining were used. At least three WT and five light for 60–80 min [24]. After that, sections were rinsed GFAP-IL6Tg NL animals and at 14 and 21 dpl were ana- in 5% sodium thiosulfate for 12 min and rinsed again lyzed. A total of 24 photographs from 8 different hippocam- with dH2O. Sections were postfixated in 70% alcohol for pal sections per animal were captured at × 20 magnification 30 min, dehydrated in alcohol, rinsed in xylene and cov- using the same device and software referred above. The per- erslipped with DPX. centage of area occupied was obtained using the analySIS® software. Brightfield microscopy densitometric analysis Densitometric analysis in NL and lesioned animals was Morphometric analysis performed on sections immunolabeled for Iba1 to assess Morphometric analysis of microglial cells was done on microglia state. A minimum of three animals per genotype sections immunolabeled for Iba1. At least three WT and and experimental time-point were analyzed. A total of 9 three GFAP-IL6Tg animals were analyzed. For each ani- photographs from 3 different sections per animal contain- mal, a total of 30 representative microglial cells were ing the left molecular layer (ML) of the dentate gyrus randomly chosen from 9 different photographs from the (DG) were captured using the × 20 objective with a DXM ML of the DG of the hippocampus and photographed at 1200F Nikon digital camera joined to a brightfield Nikon × 40 magnification. Using the analySIS® software, indi- Eclipse 80i microscope, using the software ACT-1 2.20 vidual cells were isolated and different parameters in- (Nikon corporation) (Suppl. Fig. 1). By means of analySIS® cluding the area occupied, the sphericity (value equal to software, both the percentage of area occupied by the 1 indicates spherical shape and low values increased immunolabeling as well as the intensity of the immuno- elongation), the shape factor (high values indicate round stain (Mean Grey Value Mean) was recorded for each shape and low values ramified morphology), and the photograph. The AI index [2] was calculated as function elongation (value equal to 1 indicates round morph- of the percentage of the immunostained area and the ology and high values increased elongation) were re- Mean Grey Value Mean. corded for each cell. Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 6 of 22 Flow cytometry analysis Foxp3-PE-Cy7, and anti-Gata3-PE (Table 3). In parallel, The phenotype of the microglia/macrophage populations isotype-matched control antibodies for the different and the T cell infiltration in NL and lesioned animals (at fluorochromes (BD Pharmingen) were used as negative 3 and 7 dpl) were analyzed using flow cytometry as pre- control and a cell suspension of splenocytes as positive viously described [1]. control. Data were extrapolated as number of cells using Briefly, animals were anaesthetized and intracardially the Cyto Count™ fluorescent beads following the manu- perfused for 1 min with 0.1 M phosphate buffer solution facturer’s instructions (S2366, Dako Cytomation). Fi- (PBS), brains removed, and the entire ipsilateral hippo- nally, cells were acquired using a FACS Canto flow campus was quickly dissected out. In order to obtain a cytometer (Becton Dickinson, San Jose, CA) and results cell suspension, samples were dissociated through 140 analyzed using the FlowJo® software. The analysis was μm and 70 μm meshes and digested for 30 min at 37 °C performed separately for each animal without any using collagenase type IV (17104-019, Life Technologies) pooling. and DNAsa I (D5025, Sigma). After that, each cellular suspension was centrifuged at RT for 20 min at 2400 Tissue processing for protein analysis rpm in a discontinuous density Percoll gradient (17- Animals used for protein analysis were i.p anaesthetized 0891-02, Amersham-Pharmacia) between 1.03 and 1.08 (as described above) and perfused for 1 min with cold g/mL. Myelin in the upper layer was removed. Cells in 0.1 M PBS (pH 7.4). Subsequently, the entire ipsilateral the interphase and in the clear upper-phase were col- hippocampus was dissected out, snap frozen individually lected, washed in PBS + 2% serum, and the Fc receptors in liquid nitrogen, and stored at − 80 °C. Total protein were blocked by incubation for 10 min in a solution of was extracted by solubilization of samples on lysis buffer, purified CD16/32 diluted in PBS + 2% serum. After- containing 25 mM HEPES, 2% Igepal, 5 mM MgCl2, 1.3 wards, cells were labeled for 30 min at 4 °C with the fol- mM EDTA, 1 mM EGTA, 0.1 M PMSF, and protease (1: lowing four combinations of cell surface antibodies: (1) 100, P8340, Sigma Aldrich) and phosphatase inhibitor anti-CD11b-APC-Cy7, anti-CD45-PerCPCy5, anti- cocktails (1:100, P0044, Sigma Aldrich), for 2 h at 4 °C. CD11c-PE, anti-MHCII-FITC, anti-CD86-PE-Cy7, anti- After solubilization, samples were centrifuged at 13,000 CD80-APC; (2) anti-CD11b-APC-Cy7, anti-CD45- rpm for 5 min at 4 °C and the supernatants collected. PerCPCy5, anti-Ly6C-FITC, anti-F480-APC, and anti- The hippocampus from each animal was analyzed separ- ICOSL-PE; (3) anti-CD3-FITC, anti-CD4-APC-Cy7, and ately. Total protein concentration was determined with a anti-CD8-PerCPC; and (4) anti-CD3-FITC, anti-CD4- commercial Pierce® BCA Protein Assay kit (#23225, APC-Cy7, anti-Tbet-PerCPCy5.5, anti-RORγt-APC, anti- Thermo Scientific) according to manufacturer’s protocol. Table 3 Antibodies used in flow cytometry Target antigen Format Dilution Cat Number Manufacturer Fc blocker CD16/32 Purified 1:250 553142 BD Pharmingen Primary antibodies CD11b APC-Cy7 1:400 557657 BD Pharmingen CD45 PerCPCy5 1:400 557235 BD Biosciences MHCII FITC 1:400 553623 BD Pharmingen CD11c PE 1:400 557401 BD Pharmingen Ly6C FITC 1:400 553104 BD Pharmingen ICOSL PE 1:400 12-5985-82 BD Bioscience CD86 PE-Cy7 1:400 560582 BD Pharmingen CD80 APC 1:400 560016 BD Pharmingen CD11c PE 1:400 557401 BD Pharmingen F4/80 APC 1:400 17-4801-82 eBioscience CD3e FITC 1:400 553062 BD Pharmingen CD4 APC-Cy7 1:400 552051 BD Pharmingen CD8 PerCP 1:400 553036 BD Pharmingen T-bet PerCPCy5.5 1:400 45-5825 eBioscience RORγt APC 1:400 17-6988-82 eBioscience Foxp3 PECy7 1:400 25-5773-80 eBioscience Gata3 PE 1:400 560074 BD Pharmingen Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 7 of 22 Protein lysates were stored at − 80 °C until used for the Results protein microarray analysis. Astrocyte-targeted IL-6 production modifies the number and morphology of microglia/macrophage cell Cytokine analysis populations The cytokines IL-2, IL-6, IFN-γ, IL-1β, IL12p70, IL-17, Analysis of microglia/macrophage cell distribution and IL-10, IL-13, IL-9, IL-5, and IL-4 and the chemokines morphology CXCL10 and CCL2 were analyzed using a Milliplex® To investigate possible changes in the brain cyto- MAP Mouse Cytokine/Chemokine kit (#MCYTOMAG- architecture and cell distribution, by astrocyte-targeted 70K, Merck Millipore) according to the manufacturer’s IL-6 production in the CNS, a microscopic study was instructions. Briefly, 25 μL of each hippocampus extract performed on toluidine blue sections (Suppl Fig. 2). In with a final total protein concentration of 2.5 μg/μL was NL conditions, both WT and GFAP-IL6Tg animals added to the plate, along with the standards in separate showed the same distribution of cells through the ML. wells, containing 25 μL of custom fluorescent beads and However, GFAP-IL6Tg showed an increased number of 25 μL of matrix solution and incubated overnight at 4 °C cells. After PPT, the presence of cells increased in the in a plate-shaker (750 rpm). After two washes with wash outer molecular layer (OML) and medial molecular layer buffer (WB), the plate was incubated with 25 μL of de- (MML) in both WT and GFAP-IL6Tg animals from 3 to tection antibodies for 30 min at RT followed by incuba- 7 dpl. In parallel to changes found in the OML and the tion with 25 μL of Streptavidin-Phycoerythrin for 30 MML, both WT and GFAP-IL6Tg animals showed a min at RT in a plate-shaker (750 rpm). Finally, the plate progressive reduction of cells in the inner molecular was washed two times with WB and 150 μL of Drive layer (IML) along the different time-points (from 3 to 14 fluid was added. Luminex® MAGPIX® device with the dpl) (Suppl Fig. 2). xPONENT® 4.2 software was used to read the plate. Data The possible changes in distribution and morphology were analyzed using the Milliplex® Analyst 5.1 software of the microglia/macrophage cell population evoked by and expressed as pg/mL of protein. the transgenic production of IL-6 were analyzed using Iba1 IHC. Tissue processing for serum samples In NL WT animals, microglial cells showed the char- Animals used for serum samples were i.p anaesthetized acteristic ramified morphology throughout the ML of (as described above) and blood extraction were per- the DG (Fig. 1a, f), whereas in the GFAP-IL6Tg mice, formed by cardiac puncture. After blood extraction, microglia had a significant increase of Iba1 immunoreac- samples were centrifuged at 10,000 rpm for 5 min at 4 tivity (Fig. 1b) as well as morphological changes mainly °C and the supernatants collected. Finally, samples were characterized by a greater number and thickness of ram- stored at − 80 °C until used for the ELISA analysis. ifications and an increase in the number of processes (Fig. 1a). These qualitative morphological changes were IL-6 ELISA quantified by a morphometric analysis of different pa- The IL-6 levels in serum were analyzed using a Mouse rameters such as the area, shape factor, the elongation, IL-6 Uncoated ELISA kit (88-7064, Invitrogen) accord- and sphericity of individual Iba1+ microglial cells. Con- ing to the manufacturer’s instructions. Briefly, the plate firming the qualitative findings, GFAP-IL6Tg animals was coated overnight at 4 °C and after washes; wells showed a significant increase in the area occupied by in- were blocked for 1 h at RT. After several washes, 100 μL dividual microglial cells (Suppl Fig. 3A). Moreover, of each serum extraction and standards, in separated microglial cells in transgenic animals showed a higher wells, were added and incubated overnight at 4 °C. After ramified morphology, as indicated by less values of shape reagents incubations, and the addition of TMB solution factor (Suppl Fig. 3D), and a rounded morphology and Stop solution, results were read at 450 nm in the (Suppl Fig. 3B-C). microplate reader VarioskanTM Lux (ThermoFisher Sci- After PPT, Iba1 immunoreactivity in WT animals in- entific) and data were expressed as pg/mL of protein. creased at 3 dpl and decreased at 21 dpl, whereas in GFAP-IL6Tg animals Iba1 showed a later increase at 7 Statistical analysis dpl and remained stable until 21 dpl (Fig. 1a, b). Notice- Statistics were performed using the Graph Pad Prism 5.0 ably, at 2 and 14 dpl (Fig. 1b), GFAP-IL6Tg mice ® software. To study the differences between NL WT showed higher levels of microglial cell Iba1 compared to and NL GFAP-IL6Tg animals unpaired Student’s T test WT (1651 ± 376.6 vs. 3145 ± 202.3 AI, WT/2dpl vs. Tg/ was used, while two-way ANOVA followed by Tukey’s 2dpl, p = 0.014 and 4160 ± 939.0 vs. 2374 ± 63.5 AI, post hoc analysis was used to study the effect of the le- WT/14dpl vs. Tg/14dpl, p = 0.032). To quantify the dy- sion in both genotypes. All experimental values were namics of microglial activation following PPT lesion in expressed as mean values ± SD. both WT and GFAP-IL6Tg animals, the fold-changes Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 8 of 22 Fig. 1 Microglial cell activation. Representative images from WT and GFAP-IL6Tg mice showing Iba1 staining in the granular (GL) as well as the inner, medial, and outer molecular layer (IML, MML, and OML, respectively) of the dentate gyrus (DG) in non-lesioned (NL) and PPT-lesioned hippocampus from 3 to 21dpl. Scale bar = 20 μm (a) and 10 μm (f). b Graph showing the time course of Iba1 immunostaining, expressed as AI (area x intensity). In comparison with WT, GFAP-IL6Tg animals showed increased levels of Iba1 at 2 and 14dpl. c, d Graphs showing the fold change increase of Iba1 expression in comparison with the corresponding NL animals in both WT and GFAP-IL6Tg mice. e Graph showing the area occupied by Iba1 in the IML in both WT and GFAP-IL6Tg at 3 and 7 dpl. A minimum of 3 animals per genotype and experimental time-point were analyzed. A total of 9 photographs from 3 different hippocampal sections were used. Data are represented as mean ± SD. The significances are represented as #p ≤ 0.05 vs. NL of respective group; $p ≤ 0.05 vs. 7dpl of respective group. Significant differences between genotypes are represented as *p ≤ 0.05 with respect to the corresponding NL animals were cal- early time-points (3 and 7dpl), microglial cells in both culated (Fig. 1c, d). This revealed that, although the genotypes showed the typical “bushy” shapes character- levels of Iba1 in GFAP-IL6Tg animals were higher than ized by short and stubby ramifications (Suppl Fig. 3). At in WT, the upregulation of Iba1 with respect to its basal later time-points (14 and 21 dpl), microglial cells in both levels was less pronounced in the transgenic animals WT and GFAP-IL6Tg animals returned to ramified than in WT, at specific time-points along the progres- morphologies, although comparatively, microglia in the sion of the PPT lesion. Thus, while WT animals showed transgenic animals displayed a greater number and around a 3-fold increase of Iba1 at 7dpl in comparison thickness of ramifications than WT at these time-points with their basal levels (Fig. 1c), in GFAP-IL6Tg mice, (Fig. 1f). this increase remained less than 2-fold at any time-point analyzed (Fig. 1d). Finally, no differences in the area oc- Analysis of microglia/macrophage cell density cupied by Iba1 were observed in the IML of both WT To evaluate possible changes in the number of micro- and GFAP-IL6Tg animals after PPT (Fig. 1e). glia/macrophages IHC for the transcription factor Pu.1, In addition to changes in Iba1 levels, after PPT, both a specific myeloid marker [40], was used. WT and GFAP-IL6Tg mice showed alterations in micro- In parallel to changes in microglia/macrophage distri- glial cell morphology with some differences noted. At bution and morphology, our observations revealed Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 9 of 22 around a 2-fold increase in the number of Pu.1+ cells in differences between genotypes. After PPT, although the the ML of the DG in NL GFAP-IL6Tg animals in com- peak of proliferation was at 2 dpl in both genotypes, in parison with NL WT (Suppl. Fig. 4 and Suppl. Fig. 11B). WT animals the number of pH3+decreased at 3 dpl, After PPT, the number of Pu.1+ cells in WT animals whereas in GFAP-IL6Tg animals the number of pH3+ increased progressively from 2 to 3 dpl, whereas in cells decreased later on, at 7 dpl (Fig. 2a, c). Further- GFAP-IL6Tg animals, this increase was observed only at more, to explore whether, additionally to the dynamics 3 dpl (Suppl. Fig. 4A). At later time points, in WT ani- of proliferation, IL-6 overproduction produced modifica- mals, Pu.1+ cells exhibited a marked decrease at 14 dpl, tions in the total amount of proliferating cells, a daily in- whereas in GFAP-IL6Tg mice remained stable until 21 jection of BrdU was performed during 7 days. This study dpl (Suppl. Fig. 4A). Consequently, in transgenic ani- demonstrated that the total number of accumulated mals, the number of Pu.1+ cells at 14 dpl was signifi- BrdU+/Iba1+ cells until 7dpl was significantly higher in cantly higher than WT (759 ± 292.8 vs. 1384 ± 305.2 GFAP-IL6Tg animals than in WT (23.06 ± 7.39 vs. 36.19 number of cells, WT/14dpl vs. Tg/14dpl, p = 0.010) ± 6.16 number of cells, WT/7dpl vs. Tg/7dpl, p = 0.049) (Suppl. Fig. 4B). Notably, when dynamics of Pu.1 upreg- (Fig. 2b). Using double stain immunohistochemistry, it ulation were compared with the corresponding NL, it was determined that parenchymal BrdU+ cells in both was found again that WT animals had a nearly 5-fold in- WT and GFAP-IL6Tg animals corresponded mainly to crease in the number of Pu.1+ cells in the ML of the DG Iba1+ microglia/macrophages, although some scattered (Suppl. Fig. 4C), whereas GFAP-IL6Tg animals showed proliferating astrocytes (BrdU+/GFAP+) were also found only a 2-fold increase in Pu.1+ cells compared with NL (Fig. 2d, e). GFAP-IL6Tg animals (Suppl. Fig. 4D). Regarding microglial cell death, no staining for either active caspase-3 or TUNEL was detected in the denner- Astrocyte-targeted IL-6 production increases microglial vated ML of the DG in either WT or GFAP-IL6Tg ani- cell proliferation after PPT but has no effects on cell mals from 2 to 21 dpl (data not shown). death In order to study whether changes observed in micro- Astrocyte-targeted IL-6 production increases the number glial cell numbers in GFAP-IL6Tg animals were related of monocytes/macrophages to changes in either proliferation and/or microglial cell Flow cytometry was used to assess the number of micro- death, pH3 and BrdU detection was used for the analysis glia/macrophage populations induced by the transgenic of proliferation (Fig. 2) and activated caspase-3 and production of IL-6 (Suppl. Fig. 5). The microglia/macro- TUNEL for apoptosis evaluation. phage population was identified based on the positive In both NL WT and NL GFAP-IL6Tg animals, few CD11b expression in combination with differential ex- pH3+ cells were observed in the ML of the DG, with no pression levels of CD45 (Suppl. Fig. 5A). This helped in Fig. 2 Microglial proliferation. a Graph showing the quantification of pH3+ cells in non-lesioned (NL) and PPT-lesioned animals from 2 to 7 dpl, in WT and GFAP-IL6Tg mice. b Graph showing the quantification of BrdU+/Iba1+ cells in PPT-lesioned animals at 7dpl in WT and GFAP-IL6Tg animals. Note that the total number of BrdU+ proliferating cells is significantly higher in transgenic animals. c Representative images of pH3+ cells in WT and Tg animals at 2 dpl. Scale bar = 20um. d Representative images of double IHC combining BrdU (red) and Iba1 (green). e Representative images of double IHC combining BrdU (red) and GFAP (green). Scale bar = 10 μm. For pH3, a minimum of three WT and three GFAP-IL6Tg animals per group were analyzed, whereas in the case of BrdU five WT and six GFAP-IL6Tg animals were used. Data are represented as mean ± SD. The significances are represented as #p ≤ 0.01 vs. NL of respective group; &p ≤ 0.05 vs. 2dpl of respective group and %p ≤ 0.01 vs. 3dpl of respective group. Significant differences between genotypes are represented as *p ≤ 0.05 Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 10 of 22 differentiating homeostatic from activated microglia as 0.69 vs. 5.07 ± 2.30 number of cells, WT/NL vs. Tg/NL, well as macrophages. Thus, ramified or homeostatic p = 0.026) in transgenic animals than WT (Suppl. Fig. microglia were identified as CD11b+/CD45low and acti- 11F and G). vated microglia as CD11b+/CD45int. In this study, the After PPT, in WT animals, no statistically significant term CD11b+/CD45low/int has been used to refer to the differences in the number of F4/80+ and Ly6C+ cells population of homeostatic and activated microglia jointly. were found after PPT in comparison with the corre- Moreover, the CD11b+/CD45high population was identi- sponding NL controls. However, in GFAP-IL6Tg ani- fied, which may include highly activated microglia, mono- mals, the number of Ly6C+ cells increased a 3 dpl within cytes, macrophages, and dendritic cells. The percentage of the CD11b+/CD45high cell population (Suppl. Fig. 6C). CD11b+/CD45high population, in the GFAP-IL6Tg, was Comparisons between genotypes revealed no significant higher than in WT at all time points analyzed (Suppl. Fig. variations in the number of either F4/80+ or Ly6C+ cells 7). within the CD11b+/CD45low/int cell population (Suppl. Taking in account the total number of cells, in NL ani- Fig. 6A and C). However, a greater number of F4/80+ mals, our results showed a significant increase of both cells in the CD11b+/CD45high cell population was found CD11b+/CD45low/int (318.3 ± 124.7 vs. 524.3 ± 154.5 at 3 and 7dpl in GFAP-IL6Tg animals (Suppl. Fig. 6A). number of cells, WT/NL vs. Tg/NL, p = 0.08) and Furthermore, analysis of the levels of F4/80, calculated CD11b+/CD45high populations (7.50 ± 3.06 vs. 45.57 ± using the mean fluorescence intensity, demonstrated a 14.06 number of cells, WT/NL vs. Tg/NL, p = 0.001) in higher F4/80 in the CD11b+/CD45high cell population of GFAP-IL6Tg animals (Suppl. Fig. 11C and D). GFAP-IL6Tg animals in both NL and at 7 dpl (Suppl. After PPT, GFAP-IL6Tg animals exhibited an increase Fig. 6B). However, in the CD11b+/CD45high cell popula- in the number of both CD11b+CD45low/int and CD11b+/ tion, the levels of Ly6C in NL transgenic animals were CD45high cell population at 7dpl when compared to their less than WT (4273 ± 148.1 vs. 2144 ± 523.9 mean own NL, while in WT animals no changes were ob- fluorescence, WT/NL vs. Tg/NL, p = 0.0001) (Suppl. Fig. served. Interestingly, the levels of CD45 (mean fluores- 6D). cence intensity) were always higher in the CD11b+/ CD45low/int cell population of GFAP-IL6Tg animals than Astrocyte-targeted IL-6 production modifies the in WT (Suppl. Fig. 5B). However, in CD11b+/CD45high phenotype of CD11b+/CD45high cell population cell population, no differences between genotypes were In order to investigate the putative changes that trans- observed (Suppl. Fig. 5B). genic IL-6 production may exert on the phenotype of Using CD45 immunohistochemistry, it was confirmed the microglia and macrophage cell populations, the ex- that GFAP-IL6Tg animals showed a higher activated pression of different cell activation markers mostly re- microglial phenotype (Suppl. Fig. 5C). Also, after PPT, lated to antigen presentation, including MHCII, CD80, scattered CD45+ cells with round morphology were ob- CD86, ICOSL, and CD11c, in both CD11b+/CD45low/int served through the ML of the FD in both experimental and CD11b+/CD45high populations was studied by flow groups that could correspond to T cells and/or infil- cytometry (Fig. 3 and Suppl. Figs. 9 and 12). trated monocytes (Suppl. Fig. 5C black arrows). In NL animals, in both WT and GFAP-IL6Tg animals, Taking into account that the population of CD11b+/ a small population of CD11b+/CD45low/int cells CD45high cells includes monocytes, macrophages, den- expressed MHCII (WT: 1.2 ± 0.6%; Tg: 3.3 ± 0.7%) and dritic cells, and highly activated microglia, we next ad- ICOSL (WT: 0.22 ± 0.1%; Tg: 0.1 ± 0.08%) (Fig. 3a and dressed whether the changes observed in the number of Suppl. Fig. 9A), whereas the number increased in the microglia/macrophages were due specifically to varia- CD11b+/CD45high population (WT: 53.5 ± 9.4%; Tg: tions in the number of monocytes or rather to the pres- 53.23 ± 6%) (WT: 34.8 ± 5.4%; Tg: 22.4 ± 0.7%) (Fig. 3a ence of highly activated microglia. For this purpose, the and Suppl. Fig. 9B). expression of the macrophage marker F4/80 and the The number of cells expressing MHCII+ in both popu- monocytic marker Ly6C were evaluated in the CD11b+/ lations was always significantly higher in NL GFAP- CD45low/int and CD11b+/CD45high cell populations IL6Tg animals than in NL WT (Fig. 3a and Suppl. Fig. (Suppl. Fig. 6). 11H-I). In the case of the ICOSL population, NL GFAP- In NL animals, the CD11b+/CD45low/int cell population IL6Tg animals only showed a significant increase in the in GFAP-IL6Tg animals had a higher number of F4/80+ CD11b+/CD45high population compared with WT cells, but not Ly6C+ cells compared to WT (Suppl. Fig. (Suppl. Fig. 11J). 11E). On the other hand, the CD11b+/CD45high cell After PPT, an increase in the number of both population showed an increased number of cells ex- CD11b+/CD45low/int/MHCII+ and CD11b+/CD45high/ pressing F4/80 (2.69 ± 2.36 vs. 56.44 ± 32.48 number of MHCII+ population was only observed in GFAP-IL6Tg cells, WT/NL vs. Tg/NL, p = 0.046) and Ly6C (1.54 ± animals (Fig. 3a). Moreover, in GFAP-IL6Tg animals, Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 11 of 22 Fig. 3 Phenotypic analysis of microglia and macrophage populations. a Graphs showing the number of MHCII+ cells in the CD11b+/CD45low/int and CD11b+/CD45high cell population. b Representative images from WT and GFAP-IL6Tg mice showing MHCII staining in the ML of the DG in NL and PPT- lesioned hippocampus from 3 to 7 dpl. Scale bar = 20 μm. c Representative images, from WT and GFAP-IL6Tg mice, of double IHC combining MHCII (red) and Tmem119 (green) at 7 dpl. White arrows indicate MHCII+/Tmem119− cells and yellow arrow indicate microglia Tmem119+ Scale bar = 5 μm. d Graphs showing the number of CD80+, CD86+, and CD11c+ cells in the CD11b+/CD45high/MHCII+ and CD11b+/CD45high/MHCII− cell populations. A minimum of three WT and three GFAP-IL6Tg animals per group were used for this study. Data are represented as mean ± SD. The significances are represented as #p ≤ 0.05 vs NL of respective group; %p ≤ 0.1 vs. NL of respective group; &p ≤ 0.001 vs. NL of respective group and $p ≤ 0.05 vs. 3dpl of respective group. Significant differences between genotypes are represented as *p ≤ 0.05 the number of ICOSL+ cells increased at 7 dpl within PPT (Fig. 3b). These cells had an elongated morphology the CD11b+/CD45low/int and CD11b+/CD45high popula- and stained with CD206 a marker commonly used to tion, whereas in WT no changes were observed (Suppl. identify perivascular macrophages (Suppl. Fig. 8A). Fur- Fig. 9A-B). thermore, using double immunofluorescence with lam- Using MHCII immunostaining, MHCII+ cells in the ML inin, the location of these MHCII+ cells in the perivascular of the DG were morphologically identified as perivascular space was established (Suppl. Fig. 8B). Noticeably, at 7 dpl, cells in both WT and transgenic animals in NL and after we observed a subpopulation of MHCII+/Tmem119− Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 12 of 22 ramified microglia in the parenchyma of WT animals but significant difference was found in any T cell population not in GFAP-IL6Tg mice (Fig. 3c). between WT and GFAP-IL6Tg mice (Fig. 4b). Taking into account that the major differences ob- Finally, the different subtypes of CD4+ T-helper lym- served in terms of MHCII was in the CD11b/CD45high phocytes were also determined by flow cytometry using population, we focused on the study of the co- antibodies against lineage-specific transcription factors expression of CD80, CD86, and CD11c with MHCII only (Fig. 4d). After gating in the CD3+CD4+ T-helper cell in this population (Fig. 3d). In NL conditions, in contrast population, the number of Tbet+ (for Th1), Gata 3+ (for to WT animals in which these populations were not de- Th2), RORγ+ (for Th17) and Foxp3+ (for T-regulatory) tected, a small population of CD80+ cells (3 ± 1.1%) and cells was analyzed (Fig. 4c). In both NL conditions and a higher number of CD86+ cells (16.3 ± 4.8%) was ob- after PPT, no significant differences in the number of served within the CD11b+/CD45high/MHCII+ population Th-cell infiltration between the two genotypes were in GFAP-IL6Tg animals. In addition, a population of found (Fig. 4d). CD11b+/CD45high/MHCII−/CD86+ was found in NL The possible changes in the distribution and the pat- transgenic animals (Fig. 3d). tern of recruitment of T cell population, in the ML of In GFAP-IL6Tg animals, it was observed an increase the DG, were analyzed using sections immunostained of cells expressing CD80 or CD86 (Fig. 3d), either with for CD3 (a pan T cell marker) of NL and in PPT- or without MHCII. However, the number of cells co- lesioned WT and GFAP-IL6Tg animals (Fig. 4f). In NL expressing CD80 and CD86 only increased within the conditions, a higher number of CD3+ cells was detected CD11b+/CD45high/MHCII+ population. Furthermore, in GFAP-IL6Tg animals (Suppl. Fig. 11M). After PPT, populations of MHCII+/CD11c+ and MHCII-/CD11c+ WT animals showed two waves of T cell infiltration, the cells, with or without CD86, were observed at 7 dpl in first at 2 dpl and the second at 7 dpl. In GFAP-IL6Tg WT (Fig. 3d). Moreover, the number of CD11c+/CD86+ mice, we only detected an increase in lymphocytes at 7 cells either with or without MHCII was always higher in dpl. Moreover, in both genotypes, the CD3+ cells were GFAP-IL6Tg mice than in WT (Fig. 3d). observed randomly distributed through the ML of the As mentioned above, no CD80+, CD86+, and cells co- DG and the hilus (Fig. 4f). From 7 to 14dpl, the numbers expressing both markers were observed in the CD11b+/ of these cells did not change in either WT or GFAP- CD45low/int/MHCII+ population at any time-point ana- IL6Tg mice (Fig. 4e). lyzed in any genotype (data not shown). Absence of T cell co-stimulatory molecules after PPT Astrocyte-targeted IL-6 production promotes T cell In order to investigate the putative changes that trans- infiltration after PPT genic IL-6 production may exert on the communication Another aspect was whether transgenic production of between T cells and microglia/macrophages, we analyzed IL-6 was able to modify the infiltration and/or differenti- CTLA-4 and CD28, two identified ligands for CD80 and ation of T-cells after PPT. CD86, in PPT-lesioned WT and transgenic mice at 7dpl. In order to study the dynamics of the lymphocyte pop- Our results showed no detectable CTLA-4 and CD28 in ulations, the number of CD4+ and CD8+ cells within the either WT or GFAP-IL6Tg mice at any time-point stud- gated CD3+ T cell population was analyzed in the entire ied (data not shown). ipsilateral hippocampus by flow cytometry, in NL and in PPT-lesioned WT and GFAP-IL6Tg mice at 7dpl (Fig. Astrocyte-targeted IL-6 production induces alterations in 4a and b). the cytokine/chemokine microenvironment after PPT In comparison with NL WT, NL GFAP-IL6Tg animals We also addressed whether the changes observed in showed a higher number of CD4+ T cells (20.44 ± 6.25 transgenic animals correlated with variations in various vs. 92.27 ± 43.29 number of cells, WT/NL vs. Tg/NL, p cytokines involved in microglial cell activation and T cell = 0.046), CD8+ T cells (29.62 ± 1.63 vs. 294.9 ± 139.9 differentiation, including pro-inflammatory cytokines number of cells, WT/NL vs. Tg/NL, p = 0.031), and such as IL-2, IL-6, IFN-γ, IL-1β, IL-12p70, and IL-17; CD3+/CD4−/CD8− cells (8.73 ± 1.59 vs. 15.34 ± 4.27 anti-inflammatory cytokines such as IL-10, IL-13, IL-9, number of cells, WT/NL vs. Tg/NL, p = 0.07) but simi- IL-5, and IL-4; and chemokines, related with leukocyte lar numbers of CD3+/CD4+/CD8+ cells (Fig. 4b and recruitment, like CXCL10 and CCL2 (Suppl. Fig. 10). Suppl. Fig. 11K and L). In NL conditions, when only animals of the two geno- After PPT, at 7 dpl, both experimental groups showed types are compared, significant differences were ob- an increase in CD3+/CD4−/CD8− T cells, whereas in the served in the presence of the pro-inflammatory CD3+/CD4+/CD8+ population, only GFAP-IL6Tg ani- cytokines IL-6 and IL-17 and the anti-inflammatory cy- mals showed an increase, and in the CD3+/CD4+ popu- tokines IL-10 and IL-13 in the nervous parenchyma lation was only observed in WT animals (Fig. 4b). No (Suppl. Fig. 11N-Q). In contrast, no modifications of IL- Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 13 of 22 Fig. 4 T cell infiltration after PPT. a Representative dot plot of SSC/CD3 expression of cells obtained from the lesioned WT animals. The square delimits the CD3+ cell population used in this study. Representative dot plot of CD3/CD4/CD8 cells obtained from the spleen, positive control (top), and hippocampus (bottom) of lesioned WT. b Graphs showing the number of CD4+, CD3+/CD4+/CD8+ cells, CD8+ and CD3+/CD4-/CD8− in NL and at 7dpl. c Representative histogram-plot showing the expression of Tbet, Gata3, RORgt, and Foxp3 in the population of CD3+Th-cells in comparison to the corresponding isotype control, in which the positive staining was defined. d Graphs showing the number of the different subtypes of CD4+ T-helper lymphocytes: Tbet+, Gata 3+, RORg+, and Foxp3+ in NL and at 7dpl. e Graph showing the quantification of CD3+ immunostaining cells in non-lesioned (NL) and PPT-lesioned hippocampus from 2 to 14 dpl, in WT and GFAP-IL6Tg mice. f Representative images showing CD3 immunostaining in WT and GFAP-IL6Tg mice at 7 dpl. Scale bar = 20 μm. A minimum of four WT and four GFAP-IL6Tg animals per group were used for this study. Data are represented as mean ± SD. The significances are represented as #p ≤ 0.05 vs. NL of respective group; $p ≤ 0.1 vs. NL of respective group and %p ≤ 0.05 vs. 3dpl of respective group. Significant differences between genotypes are represented as *p ≤ 0.05 Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 14 of 22 6 levels in the serum were found in transgenic animals Levels of these chemokines were higher in transgenic (Suppl. Fig. 10N). animals than WT at 3dpl (Suppl. Fig. 10L and M). After PPT, WT animals only showed induction of the CXCL10 and CCL2 chemokines, while no induction was Astrocyte-targeted IL-6 production reduces collateral detected in any of the cytokines studied at the different sprouting after PPT time-points analyzed (Suppl. Fig. 10L and M). Finally, it was explored whether the modifications de- PPT-lesioned GFAP-IL6Tg animals showed, in relation tected in the population of microglia/macrophages, as to the WT, differences in the levels of pro-inflammatory well as in the infiltration of immune cells in GFAP- and anti-inflammatory cytokines as well as in chemo- IL6Tg animals, affects the collateral sprouting after per- kines. In concordance with the transgene expression, the forant path sectioning. Sprouting evaluation was done at cytokine with the greatest difference was IL-6, which later time points (14 and 21dpl) using the selenium- was at significantly higher levels at 3 and 7 dpl. In the silver staining technique, which reveals glutamatergic case of pro-inflammatory cytokines, transgenic animals zinc-rich buttons as in the hippocampal mossy fibers had significantly higher IL-1β at 3dpl (Suppl. Fig. 10D), (Fig. 5) and other telencephalic pathways. but lower IFN-γ and IL-12p70 at 14dpl (Suppl. Fig. 10C In NL animals, WT and GFAP-IL6Tg, large labeled but- and E). In the case of anti-inflammatory cytokines, tons were densely packed in the hilus of the dentate gyrus GFAP-IL6Tg animals showed higher levels of IL-10 than and only few crossed the granular layer and reached the WT at 7 dpl but had a significant decrease at 14dpl inner third of the molecular layer. After PPT, at 21 dpl, (Suppl. Fig. 10G). Finally, the GFAP-IL6Tg mice had a WT animals showed an increase in the area occupied by significant increase in the chemokines CXCL10 and large, labeled buttons in the ML, indicative of sprouting, CCL2 at 3dpl that decreased significantly at 7 and 14dpl. whereas in GFAP-IL6Tg animals, the increase was detected Fig. 5 Collateral sprouting. a Graph showing the time course of the selenium-silver staining for zinc (arrows) expressed as proportion of occupied area. Note that GFAP-IL6Tg animals showed a reduction of Zinc+ staining at 21dpl. b Representative images from WT and GFAP-IL6Tg mice, at 21 dpl, showing staining in the ML of the DG (GL indicates granular layer). Scale bar = 20 μm. At least three WT and five GFAP-IL6Tg animals (NL animals and at 14 and 21dpl) were analyzed. A total of 24 photographs from 8 different hippocampal sections per animal were quantified. Data are represented as mean ± SD. The significances are represented as: #p ≤ 0.05 vs. NL of respective group and $p ≤ 0.01 vs. NL of respective group. Significant differences between genotypes are represented as: *p ≤ 0.05 Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 15 of 22 Fig. 6 Summary figure. Schematic representation of the main changes observed in the non-lesioned (NL) (green) and PPT-lesioned hippocampus (red) in both WT and GFAP-IL6Tg animals at 14 dpl (Fig. 5a). Consequently, at 21 dpl, transgenic ani- CD45high/F4/80+ and CD11b+/CD45high/Ly6C+ monocytes mals showed significantly less labeled buttons in the ML and T lymphocytes were found in GFAP-IL6Tg mice. All compared with WT animals (2.32 ± 0.41 vs. 1.29 ± 0.36% these changes are linked to significant modifications in area, WT/21dpl vs. Tg/21dpl, p = 0.049) (Fig. 5b and c). the cytokine/chemokine microenvironment and had a deleterious effect on axonal sprouting. Discussion The present study demonstrates that astrocyte-targeted Astrocyte-targeted production of IL-6 induces alterations IL-6 production causes significant alterations in the in the microglia/macrophage, lymphocyte, and cytokine/ phenotype and density of the microglia/macrophage pop- chemokine microenvironment ulations in the hippocampus of NL and PPT-lesioned ani- The first interesting result of this study is that the chronic mals. Moreover, an increase in the infiltration of CD11b+/ production of IL-6 increased the number of microglia/ Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 16 of 22 macrophages and modified their phenotype in homeosta- that could produce this cytokine in the CNS [69, 93] and sis. Thus, 2.5 times more CD11b+/CD45low/int and could be linked to alterations in the blood-brain barrier CD11b+/CD45high cells were found. Although this in- (BBB) [50], a well-described feature of these transgenic creased number might reflect a higher infiltration of mice [14, 19] and the higher activation phenotype of monocytes, the number of CD11b+/CD45high/F4/80+ and microglia observed in basal conditions. Moreover, an ef- CD11b+/CD45high/Ly6C+ monocytes detected was not fect of IL-17 on the upregulation of microglial activation enough to explain the increase of microglia/macrophages has been reported, especially in demyelinating and auto- in transgenic animals. In this regard, unpublished observa- immune diseases [53, 96] coinciding with the higher ac- tions from our laboratory, demonstrated a higher number tivation profile observed in our transgenic animals in of microglia/macrophages during brain development in basal conditions. On the other hand, the higher levels of GFAP-IL6Tg mice, raising the possibility that the differ- the classical anti-inflammatory cytokines IL-10 and IL- ences found in the number of cells in the adult would be 13 may be related to the increased number of perivascu- linked to modifications from the post-natal period. lar macrophages and/or CD8+ T cells, two cell-type with Regarding the phenotype of the microglia/macrophages, the capacity to produce IL-10 [60, 63]. our study revealed that IL-6 promotes the activation of the microglial/macrophage cell population towards a more react- Increased microglia/macrophage in GFAP-IL6Tg mice after ive macrophage-like cell, with higher expression of CD45 PPT might be due to both proliferation and recruitment and F4/80 and a lower Ly6C. This low level of Ly6C could of bone-marrow derived monocytes be interpreted as a stabilization of the entrance of these In agreement with previous studies in mice [29, 31, 43, monocytes, whose levels decreased once in the parenchyma 58, 66, 94], we found a progressive increase in the num- [38]. In concordance, IL-6 has been described as a potent ac- ber of microglia/macrophages as well as an increase in tivator of microglia/macrophage cells both in vitro and Iba1 expression along with the progression of PPT in in vivo [15, 17, 21, 57, 91]. Also, astrocyte-targeted produc- the molecular layer of the DG in both genotypes. How- tion of IL-6 induces an increase in the number of CD11b+/ ever, in our study, the density of microglia/macrophages CD45high/MHCII+ cells and in cells expressing co- was higher in transgenic animals than in WT. This dif- stimulatory molecules, including ICOSL, CD80 and CD86, ference could be related to three changes after PPT: suggesting a putative transformation into competent antigen microglial proliferation; infiltration of blood-derived presenting cells. Importantly, MHCII+ cells corresponded ex- monocytes [5, 7, 59, 95], which transformed into micro- clusively to perivascular macrophages and not parenchymal glial/macrophage cells [52]; and of microglial cells from microglia, highlighting a necessity to study in more detail the adjacent non-deafferented areas, such as the CA1, the role of these macrophages under homeostatic situations. IML, and the hilus [29, 52, 59, 95]. In agreement with previous studies describing increased Our results clearly showed an increase in the number basal lymphocyte infiltration in specific areas of the CNS of of proliferating pH3+ cells, as well as in the total number GFAP-IL6Tg mice [17, 18], we observed an increase in the of mitotic microglial cells (BrdU+), in GFAP-IL6Tg ani- number of T cells infiltrating the molecular layer of the DG mals. This result was associated with an increase of IL-6, in transgenic animals. The increase was due to the presence a potent inducer of microglial proliferation [56, 62, 80, of CD4+ T-helper, CD8+ T-cytotoxic, and CD3+CD4−CD8− 87]. Beyond proliferation, the recruitment of blood- γδ T cells. Studies in vitro have demonstrated the ability of derived monocytes paralleled this increase of microglia/ IL-6 to induce a unique non-classical effector CD8+T cell macrophages in the transgenic mice. Thus, a higher subpopulation, called Tc17, characterized by a high produc- number of CD11b+/CD45high cells containing a high tion of IL-17 [69], and whose primary function is to number of F4/80+ cells and Ly6C+ cells, widely used contribute to inflammation and the recruitment of lympho- markers for monocyte/macrophages [49], was observed. cytes and myeloid cells [39]. In concordance, we found ele- As we will comment later, this greater presence of vated IL-17 in NL GFAP-IL6Tg animals. monocytes/macrophages correlated with the increased IL-6-induced modifications in the microglia/macro- CCL2 production observed in GFAP-IL6Tg mice, a key phage population and T cell infiltration were linked to a chemokine involved in the leukocyte migration in the specific cytokine/chemokine profile, characterized by PPT-model [5], suggesting that not only proliferation higher IL-6, IL-17, IL-13, and IL-10 production. As ex- but also infiltration is involved in the microglia/macro- pected, major differences between WT and GFAP-IL6Tg phages increase. animals were found for IL-6, which showed around a 5- Another question is why the increased number of fold increase in the GFAP-IL6Tg animals. As already microglia/macrophages in PPT-lesioned transgenic ani- commented, the upregulation of the pro-inflammatory mals remained at later time-points. One possibility is cytokine IL-17 could be related with a higher number of that the effects of chronic IL-6 production were related CD8+ T cells and γδ T cells, one of the two cell-types to the prevention of microglial/macrophage cell number Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 17 of 22 restoration, by affecting either microglial cell death or great reduction [82]. In contrast, myeloid cells re- monocyte exit from the CNS or even both. However, in cruited into the lesions in the course of brain inflam- concordance with previous studies [8, 78], we could not mation in rodents and humans do not express find any conclusive evidence of increased microglia/ TMEM119. However, a recent study has been de- macrophage death, as we were not able to detect any scribed that the regulation of microglial TMEM119 in evidence of apoptosis, monitored by active caspase-3 MS is dependent on their inflammatory environment and TUNEL, at any time-point analyzed in any [92]. In this regard, the lack of expression of genotype. TMEM119 in MHCII+ cells does not provide evidence if these cells were resident microglia or infiltrated After PPT, astrocyte-targeted production of IL-6 macrophages. Similarly, expression of P2Y12R, an- promoted a shift in the phenotype of CD11b+/CD45high other marker commonly used to discriminate between cells and avoided the transformation of any parenchymal macrophages and microglia, by its stable expression microglia into putative MHCII+ antigen presenting cells in microglia, was downregulated in our PPT-model Phenotypically, we demonstrated that, after PPT, part [65] making not possible to use it for this purpose. of the activated microglia/macrophage population in Another interesting result was the observation of a WT animals, expressed CD11c, MHCII, CD86, and population of cells within the CD11b+/CD45high popula- ICOSL, molecules commonly associated with an tion expressing CD86 and CD80 without MHCII in both antigen-presenting cell phenotype. These results are WT and GFAP-IL6Tg mice. Noticeably, these popula- in agreement with already published papers reporting tions were observed in transgenic animals already in the expression of MHCII and CD86, but a lack of homeostatic situations. This cellular phenotype has been CD80 in microglia/macrophages after PPT [9]. The associated with peripheral innate immune responses phenotype observed in lesioned GFAP-IL6Tg animals [74], and makes us hypothesize that transgenic animals was similar to the reported in lesioned WT animals, presented a “primed microglia” with a desensitized pro- and only a slight increase in the number of MHCII+/ file in NL conditions, as suggested by other authors [72], CD80+ and MHCII+/CD86+ cells at 7dpl and in the and retained microglial activation in an innate immune number of and MHCII−/CD86+ cells at 3 dpl were phenotype after lesion. found in the CD11b+/CD45high cell population of transgenic mice. Nevertheless, when we investigated Astrocyte-targeted production of IL-6 increased the cells responsible for the expression of these mole- lymphocyte recruitment after PPT without modifying cules using immunohistochemistry, in the molecular their differentiation layer of the DG, we found major differences between Two different waves of T cell infiltration have been de- WT and GFAP-IL6Tg mice. In WT animals, in scribed in the PPT paradigm, the first wave at early addition to perivascular macrophages, activated rami- time-points (2–3 dpl) and the second, later on, at 7 dpl fied microglia in the parenchyma expressed MHCII [5, 6]. Our results showed that, in contrast to WT, and CD11c and thus were equipped to engage an GFAP-IL6Tg mice presented an increase in T cell infil- adaptive immune response. However, in GFAP-IL6Tg tration in the molecular layer only at 7dpl, coinciding animals, microglia did not express MHCII or CD11c; with the role attributed to IL-6 as a T cell recruitment this property is restricted to perivascular macro- molecule in peritoneal inflammation and in CNS-lesion phages, suggesting that IL6-induced microglia are in- models [3, 67] and a massive decrease in the recruitment adequate to act as antigen presenting cells and thus of T cells after facial nerve injury in IL-6-deficient mice to drive an adaptive immune response, despite the [36]. In most of these studies, the effects of IL-6 on higher presence of MHCII+ and CD11c+ cells. This lymphocyte recruitment have been related to alterations reduction of MHCII observed in microglia might be in the production of adhesion molecules and/or chemo- explained by the higher expression of IL-10 observed kines, including CXCL10 and CCL4, among others [67]. at this same time-point in GFAP-IL6Tg mice and the In the present study, we observed that after PPT GFAP- ability of this cytokine to reduce MHCII on macro- IL6Tg animals had elevated levels of CXCL10 and phages [68, 78]. CCL2, two crucial chemoattractant proteins for both In the present study, in both genotypes, all MHCII+ lymphocytes and monocytes in this paradigm [5], indi- cells were TMEM119−, described as a specific marker cating that IL-6 may be promoting the infiltration of T of microglia that discriminates from blood-derived cells by modifying the chemokine microenvironment. macrophages in the human brain [81]. However, sev- Taking into consideration the role of IL-6 to drive eral studies revealed that in homeostatic conditions, lymphocyte differentiation, we analyzed the phenotype microglia express high levels of TMEM119, while in of infiltrated T cells in the PPT-lesioned hippocampus neurodegeneration disorders microglial cells show a in more detail, quantifying the number of both CD8+ T- Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 18 of 22 cytotoxic and CD4+ T-helper lymphocytes, as well as the lower axonal sprouting at 21 dpl than WT. Although different subtypes of T-helper cells. Until now, the spe- in vitro IL-6 has a beneficial effect on axonal sprouting cific subtypes of T lymphocytes infiltrating the paren- [44], in vivo our results showed a deleterious effect. chyma of the deafferented hippocampus in the PPT- One putative explanation for the decreased axonal paradigm were not characterized and to our knowledge, sprouting observed in transgenic mice may lie in the our study is the first to demonstrate the presence of a distinct cytokine/chemokine microenvironment ob- small population of the four subtypes of T-helper lym- served. As we already commented, the molecules ob- phocytes, Th1 (Tbet+), Th2 (GATA3+), Th17 (RORg+), served in WT animals are more related to the and Treg (Foxp3+), in the PPT-lesioned hippocampus. acquired immune response [9], whereas in GFAP- No differences between WT and GFAP-IL6Tg animals IL6Tg, the environment generated was more associ- were found in any T-helper populations analyzed or ated with a pro-inflammatory profile characteristic of its expression of co-stimulatory molecules, suggesting the innate immune response. that the increased number of lymphocytes found in Also, it is possible that some of the altered cyto- GFAP-IL6Tg animals might be caused by the in- kines, and especially IL-6, act on the cells directly re- creased expression of CXCL10 and CCL2 rather than sponsible for this axonal sprouting phenomenon. by specific action of IL-6 on T cell differentiation in Astrocytes are the brain cells linked to axonal sprout- this paradigm. ing, as they secrete proteoglycans, such as neurocan, tenascin-C, brevican, widely demonstrated with an in- Astrocyte-targeted production of IL-6 promoted hibitory role of axonal sprouting after PPT [26, 42, alterations in the cytokine/chemokine profile associated 85, 90]. Moreover, pro-inflammatory environments es- with PPT pecially containing cytokines like TNF-α and IL-1β Astrocyte-targeted production of IL-6 markedly altered have been linked to the upregulation of these inhibi- the cytokine/chemokine profile not only in the unle- tory proteoglycans, and thus with the inhibition of sioned state, as already discussed (see the first section of axonal outgrowth [41, 85]. More studies analyzing the this discussion) but also after PPT-lesion. As expected, composition of the extracellular matrix associated major differences between WT and GFAP-IL6Tg ani- with PPT in GFAP-IL6Tg animals are needed to con- mals were found for IL-6, which showed around a 5-fold firm this hypothesis. increase in the GFAP-IL6Tg animals in both unlesioned Taking into consideration all the results observed in hippocampus and at early time-points after PPT. In NL GFAP-IL6Tg mice, together with the fact that the addition to modifications in chemokines involved in the net change in microglial/macrophage cell activation recruitment of macrophages and T cells, such as after PPT was less pronounced in transgenic animals CXCL10 and CCL2, transgenic mice showed distur- than in WT mice, lead as to speculate that IL-6 in- bances in the cytokine profile that may drive the type of duced a “primed” microglia/macrophage phenotype. the immune response. In this sense, astrocyte-targeted Accordingly, some of the features observed in the IL-6 production resulted in increased levels at early population of microglia/macrophages of transgenic time-points (3 dpl) of IL-1β, cytokine more related with mice are commonly linked to a primed state includ- innate immune responses [28, 47, 63, 70, 88, 93]. Re- ing higher expression of pro-inflammatory mediators markably, cytokines like IL-17 and IL-13, whose levels as well as MHCII expression [75]. The first descrip- we would expect to see increased given their upregula- tions of primed microglia were done in the context of tion in basal conditions, remained unaltered in lesioned LPS-administration and after IL-1β treatment and animals. This fact suggests they are not crucial to the identified primed microglia as an altered state that, evolution of the PPT lesion and could explain the lack among other features, presented an exaggerated re- of differences in the shift of T lymphocyte subtypes ob- sponse to inflammation. However, nowadays, this served between WT and GFAP-IL6Tg mice. primed stated has been also described in physiological aging conditions and stress [37, 73] in relation to the Astrocyte-targeted production of IL-6 reduced axonal development of chronic neurodegenerative diseases sprouting [76]. In fact, the concept of primed microglia has Finally, we analyzed whether the modification of the im- been revised and regarded as having an ongoing state mune response observed in GFAP-IL6Tg mice had any of activation that may respond differently to a sec- effect on the axonal sprouting of the remaining unle- ondary stimulation [72]. Furthermore, the concept of sioned axons of mossy cells, which typically occurs in microglial “innate immune memory” has been recently the deafferented areas after PPT and is used as a marker demonstrated, with an enhancement or suppression of of lesion outcome at later time-points [4, 25, 27, 33]. a secondary insult, depending on the type, duration Our results indicated that GFAP-IL6Tg animals had and intensity of the initial stimulus [72]. Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 19 of 22 Conclusions NL and PPT-lesioned animals. Graphs showing the mean fluorescence in- Altogether, we propose that chronic exposure to IL6 in- tensity of CD45 levels in the CD11b+/CD45low/int and CD11b+/CD45high duced an environment characteristic of an innate immune populations in NL and PPT-lesioned animals. (C) Representative images from WT and GFAP-IL6Tg mice showing CD45 staining in the GL and ML response, with the presence of “primed” microglia/macro- of the DG in NL and PPT-lesioned hippocampus at 3 and 7 dpl. Note that phages with a desensitized profile, infiltrating monocytes both CD45+ ramified (arrowheads) and CD45+ round cells (arrows) were and T cells as well as increased in IL-6, IL-17, IL10, and observed in both genotypes. Scale bar = 20μm. A minimum of five WT and five GFAP-IL6Tg animals per group were used for this study. Scale IL-13. After PPT, in WT animals, the immune response is bar = 50μm. Data are represented as mean ± SD. The significances are characterized by parenchymal microglia showing MHCII represented as &p≤0.01vs NL of respective group; #p≤0.05 vs NL of re- and co-stimulatory molecule phenotype. In contrast, the spective group and $p≤0.001vs NL of respective group. Significant differ- ences between genotypes are represented as *p≤0.05 and **p≤0.01. desensitized microglia in transgenic animals is not able to Additional file 6: Supplementary Fig. 6. Monocyte infiltration. (A-B) express molecules linked to antigenic presentation follow- Graphs showing the number of cells expressing the F4/80 and the mean ing the PPT. This generates a differential microglial fluorescence. (C-D) Graphs showing the number of cells expressing the phenotype with increased IL-1β that has a negative effect monocyte-related marker Ly6C and the mean fluorescence. A minimum of five WT and five GFAP-IL6Tg animals per group were used for this on the pro-regenerative environment and, maybe through study. Data are represented as mean ± SD. The significances are repre- action on astrocytes, reduce the axonal sprouting associ- sented as #p≤0.05 vs NL of respective group. Significant differences be- ated with PPT (see Fig. 6). tween genotypes are represented as *p≤0.05, **p≤0.01 and ****p≤0.0001. Additional file 7: Supplementary Fig. 7. Percentages of microglia/ macrophage population in NL and after PPT. Representative graph Supplementary Information showing the percentage of CD11b+/CD45low/int (dark grey) and CD11b+/ The online version contains supplementary material available at https://doi. CD45high (light grey) population in both WT and GFAP-IL6Tg animals in org/10.1186/s12974-020-02063-1. NL conditions and after PPT. Additional file 8: Supplementary Fig. 8. CD206 and Laminin Additional file 1: Supplementary Fig. 1. Area selected for expression after PPT. (A) Representative images from WT and GFAP-IL6Tg densitometry study. Representative image from WT mice showing Iba1+ mice showing CD206 staining in the ML of the DG at 7 dpl. Black arrows staining after perforant pathway transection. The selected area (black indicate CD206+ cells. Scale bar = 20μm. (B) Representative images, from squares) represented the area analyzed for densitometry. Scale bar = WT and GFAP-IL6Tg mice, of double IHC combining MHCII (red) and Lam- 50μm. inin (green) at 7 dpl. White arrows indicate MHCII+ cells in the perivascu- lar space. Scale bar = 10μm Additional file 2: Supplementary Fig. 2. Toluidine blue staining. Representative images, from WT and GFAP-IL6Tg mice, showing toluidine Additional file 9: Supplementary Fig. 9. ICOSL expression. (A) Graph blue staining in the ML of the DG in NL conditions and from 3 to 21 dpl showing the number of ICOSL+ cells in non-lesioned (NL) and PPT- after PPT. Scale bar = 20 μm. lesioned hippocampus, from 3 to 7 dpl, within the CD11b+/CD45low/int and CD11b+/CD45high cell populations. A minimum of five WT and five Additional file 3: Supplementary Fig. 3. Morphological GFAP-IL6Tg animals per group were used for this study. Data are repre- characterization of microglia. (A–D) Graphs showing the quantification of sented as mean ± SD. The significances are represented as #p≤0.05 vs NL the area occupied by Iba1+ labeled cells (A), (B) the elongation values of respective group. (value equal to 1 indicates round morphology and high values increased elongation), (C) the sphericity values (value equal to 1 indicates index of Additional file 10: Supplementary Fig. 10. Cytokines and sphericity) and (D) the shape factor (high values indicate round shape chemokines expression. Graphs showing the time course of expression of and low values ramified morphology), calculated for individual cells. The IL-2, IL-6, IFNγ, IL1β, IL12p70, IL-17, IL-10, IL-13, IL-9, IL-5, IL-4, CXCL10 and significances are represented as #p≤0.001 vs NL of respective group and CCL2 in non-lesioned (NL) and PPT-lesioned animals from 3 to 14dpl, in &p≤0.05 vs 3dpl of respective group. Significant differences between both WT and GFAP-IL6Tg animals. (N) Graph showing the IL-6 levels in genotypes are represented as *p≤0.05 and **p≤0.01. serum in both NL WT and NL GFAP-IL6Tg animals. At least four WT and five GFAP-IL6Tg animals for each time point were used. Data are repre- Additional file 4: Supplementary Fig. 4. Microglial cell density. (A) sented as mean ± SD. The significances are represented as #p≤0.05, Graph showing the quantification of Pu.1+ cells in non-lesioned (NL) and %p≤0.01 and &p≤0.001 vs NL of respective group; ^p≤0.1, $p≤0.05, ´´ lesioned animals from 2 to 21dpl after PPT, in WT and GFAP-IL6Tg ani- p≤0.01 and ∞p≤0.001 vs 3dpl of respective group and αp≤0.05 vs 7dpl mals. (B) Representative images from WT and GFAP-IL6Tg mice showing of respective group. Significant differences between genotypes are repre- Pu.1+ staining in the ML of the DG at 14 dpl. Note that transgenic mice sented as ***p≤0.001, **p≤0.01, *p≤0.05. showed an increased number of Pu.1+ cells in NL and at 14 dpl. Scale bar = 20μm. (C and D) Graphs showing the fold changes increase of Pu.1+ Additional file 11: Supplementary Fig. 11. Non-lesioned study in WT cells in WT (C) and GFAP-IL6Tg (D) in comparison to their corresponding and GFAP-IL6Tg animals. Representative graphs showing the differences NL animals. A minimum of three NL and three lesioned WT and GFAP- between WT and GFAP-IL6Tg animals in NL conditions. A minimum of IL6Tg at 2, 3, 7, 14 and 21 dpl were analyzed. A total of 6 photographs three WT and three GFAP-IL6Tg animals per group were used. Data are from 3 different hippocampal sections per animal were used. Data are represented as mean ± SD. The significances are represented represented as mean ± SD. The significances are represented as ***p≤0.0001, **p≤0.01, *p≤0.05. #p≤0.01vs NL of respective group; &p≤0.05 and %p≤0.01 vs 2dpl of re- Additional file 12: Supplementary Fig. 12. Gating strategy for flow spective group and $p≤0.001 vs 7dpl of respective group. Significant dif- cytometry. (A – F) Representative dot plot and histogram plot from ferences between genotypes are represented as *p≤0.05. individual hippocampus from WT and GFAP-IL6Tg animals in NL and after Additional file 5: Supplementary Fig. 5. Microglia/macrophages PPT. First, population/live cells were gated based on SSC-A and FSC-A, populations. (A) Representative dot plot of CD11b/CD45 expression of and then microglia/macrophage cells were gated based on CD45 and cells obtained from the hippocampus of non-lesioned (NL) WT animals. CD11b expression. CD11b+/CD45low/int and CD11b+/CD45high population The square delimits the CD11b+/CD45+ population of cells used in this were discriminated according to the levels of CD45. Microglia/macro- study. Representative histogram where the populations of CD11b+/ phages phenotype were studied by MHCII, CD80, CD86, Ly6C and F4/80 CD45low/int (microglia) and CD11b+/CD45high (macrophages) were de- expression and gated based on CD11b+/CD45+/Igs isotype control fined. (B) Graphs showing the number of cells and the mean fluorescence antibodies. intensity in the CD11b+/CD45low/int and CD11b+/CD45high populations in Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 20 of 22 Abbreviations 3. Almolda B, Villacampa N, Manders P, Hidalgo J, Campbell IL, Gonzalez B, BB: Blocking buffer; BBB: Blood-brain barrier; BrdU: 5′ Bromodeoxyuridine; Castellano B. Effects of astrocyte-targeted production of interleukin-6 in the CNS: Central nervous system; DAB: 3,3-Diaminobenzidine; DAPI: 4,9,6- mouse on the host response to nerve injury. Glia. 2014;62(7):1142–61. Diamidino-2-phenylindole; DG: Dentate gyrus; dpl: Day post lesion; https://doi.org/10.1002/glia.22668. GFAP: Glial fibrillary acidic protein; GFAP-IL6Tg: Transgenic mice with 4. Amaral DG, Witter MP. The three-dimensional organization of the astrocyte-targeted IL6-production; GL: Granular layer; hippocampal formation: a review of anatomical data. Neuroscience. 1989; IHC: Immunohistochemistry; IL: Interleukin; IML: Inner molecular layer; 31(3):571–91. https://doi.org/10.1016/0306-4522(89)90424-7. i.p.: Intraperitoneally; ML: Molecular layer; MML: Medial molecular layer; 5. Babcock AA, Kuziel WA, Rivest S, Owens T. Chemokine expression by glial NL: Non-lesioned; OML: Outer molecular layer; PBS: Phosphate buffer cells directs leukocytes to sites of axonal injury in the CNS. J Neurosci. 2003; solution; PP: Perforant pathway; PPT: Perforant pathway transection; 23(21):7922–30. RT: Room temperature; TBS: Tris-buffered saline; TBST: TBS containing 1% 6. Babcock AA, Wirenfeldt M, Holm T, Nielsen HH, Dissing-Olesen L, Toft- Triton-X100; Tg: Transgenic; WB: Wash buffer; WT: Wild-type Hansen H, et al. Toll-like receptor 2 signaling in response to brain injury: an innate bridge to neuroinflammation. J Neurosci. 2006;26(49):12826–37. Acknowledgements https://doi.org/10.1523/JNEUROSCI.4937-05.2006. The authors would like to thank Miguel A. Martil and Isabella Appiah for their 7. Bechmann I, Goldmann J, Kovac AD, Kwidzinski E, Simburger E, Naftolin F, technical help. et al. Circulating monocytic cells infiltrate layers of anterograde axonal The authors declare that they have no competing interests. degeneration where they transform into microglia. FASEB J. 2005;19(6):647– 9. https://doi.org/10.1096/fj.04-2599fje. Authors’ contributions 8. Bechmann I, Mor G, Nilsen J, Eliza M, Nitsch R, Naftolin F. FasL MR designed and performed the experiments, including acquiring data and (CD95L, Apo1L) is expressed in the normal rat and human brain: analyzing data; and wrote the manuscript. BA designed experiments, evidence for the existence of an immunological brain barrier. Glia. analyzed data, and revised manuscript. JPC contributed to design and 1999;27(1):62–74. perform the sulfide-silver staining. ILC revised the manuscript. BG designed 9. Bechmann I, Peter S, Beyer M, Gimsa U, Nitsch R. Presence of B7--2 (CD86) the research and revised the manuscript and BC provided financing, de- and lack of B7--1 CD(80) on myelin phagocytosing MHC-II-positive rat signed the research, and revised the manuscript. The authors read and ap- microglia is associated with nondestructive immunity in vivo. FASEB J. 2001; proved the final manuscript. 15(6):1086–8. https://doi.org/10.1096/fj.00-0563fje. 10. Bennett ML, Bennett FC, Liddelow SA, Ajami B, Zamanian JL, Fernhoff NB, Funding et al. New tools for studying microglia in the mouse and human CNS. Proc This work was supported by the Spanish Ministry of Science and Innovation Natl Acad Sci U S A. 2016;113(12):E1738–46. https://doi.org/10.1073/pnas. grants BFU2014-55459 and BFU2017-87843-R. 1525528113. 11. Benveniste EN. Cytokine actions in the central nervous system. Cytokine Availability of data and materials Growth Factor Rev. 1998;9(3-4):259–75. All data generated or analyzed during this study are included in this 12. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, et al. Reciprocal published article [and its supplementary information files]. developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature. 2006;441(7090):235–8. https://doi.org/10.1038/ Ethics approval and consent to participate nature04753. All experimental animal work was conducted in accordance to Spanish 13. Boche D, Nicoll JA. Neuroinflammation in ageing and in neurodegenerative regulations (Ley 32/2007, Real Decreto 1201/2005, Ley 9/2003 and Real disease. Neuropathol Appl Neurobiol. 2013;39(1):1–2. https://doi.org/10. Decreto 178/2004) in agreement with European Union directives (86/609/ 1111/nan.12009. CEE, 91/628/CEE and 92/65/CEE) and was approved by the Ethical 14. Brett FM, Mizisin AP, Powell HC, Campbell IL. Evolution of neuropathologic Commission of the Autonomous University of Barcelona. All efforts were abnormalities associated with blood-brain barrier breakdown in transgenic made to minimize the number of animals used to produce reliable scientific mice expressing interleukin-6 in astrocytes. J Neuropathol Exp Neurol. 1995; data, as well as animal suffering. 54(6):766–75. https://doi.org/10.1097/00005072-199511000-00003. 15. Brunello AG, Weissenberger J, Kappeler A, Vallan C, Peters M, Rose-John S, Consent for publication Weis J. Astrocytic alterations in interleukin-6/Soluble interleukin-6 receptor Not applicable. alpha double-transgenic mice. Am J Pathol. 2000;157(5):1485–93. 16. Campbell IL. Transgenic mice and cytokine actions in the brain: bridging Competing interests the gap between structural and functional neuropathology. Brain Res Brain The authors declare that they have no competing interests. Res Rev. 1998;26(2-3):327–36. 17. Campbell IL, Abraham CR, Masliah E, Kemper P, Inglis JD, Oldstone MB, Author details Mucke L. Neurologic disease induced in transgenic mice by cerebral 1 overexpression of interleukin 6. Proc Natl Acad Sci U S A. 1993;90(21): Department of Cell Biology, Physiology and Immunology, Institute of Neuroscience, Universitat Autònoma de Barcelona, 08193, Bellaterra, 10061–5. Barcelona, Spain. 2Department of Cell Biology, Physiology and Immunology, 18. Campbell IL, Hofer MJ, Pagenstecher A. Transgenic models for cytokine- Universitat de Barcelona, 08028 Barcelona, Spain. 3School of Life and induced neurological disease. Biochim Biophys Acta. 2010;1802(10):903–17. Environmental Sciences and Charles Perkins Centre, University of Sydney, https://doi.org/10.1016/j.bbadis.2009.10.004. Sydney, NSW 2006, Australia. 19. Campbell IL, Stalder AK, Akwa Y, Pagenstecher A, Asensio VC. Transgenic models to study the actions of cytokines in the central nervous system. Received: 15 April 2020 Accepted: 16 December 2020 Neuroimmunomodulation. 1998;5(3-4):126–35. https://doi.org/10.1159/ 000026329. 20. Cao Z, Gao Y, Bryson JB, Hou J, Chaudhry N, Siddiq M, et al. The cytokine References interleukin-6 is sufficient but not necessary to mimic the peripheral 1. Almolda B, Costa M, Montoya M, Gonzalez B, Castellano B. CD4 microglial conditioning lesion effect on axonal growth. J Neurosci. 2006;26(20):5565– expression correlates with spontaneous clinical improvement in the acute 73. https://doi.org/10.1523/JNEUROSCI.0815-06.2006. Lewis rat EAE model. J Neuroimmunol. 2009;209(1-2):65–80. https://doi.org/ 21. Chiang CS, Stalder A, Samimi A, Campbell IL. Reactive gliosis as a 10.1016/j.jneuroim.2009.01.026. consequence of interleukin-6 expression in the brain: studies in transgenic 2. Almolda B, de Labra C, Barrera I, Gruart A, Delgado-Garcia JM, Villacampa N, mice. Dev Neurosci. 1994;16(3-4):212–21. https://doi.org/10.1159/000112109. et al. Alterations in microglial phenotype and hippocampal neuronal 22. Conroy SM, Nguyen V, Quina LA, Blakely-Gonzales P, Ur C, Netzeband JG, function in transgenic mice with astrocyte-targeted production of et al. Interleukin-6 produces neuronal loss in developing cerebellar granule interleukin-10. Brain Behav Immun. 2015;45:80–97. https://doi.org/10.1016/j. neuron cultures. J Neuroimmunol. 2004;155(1-2):43–54. https://doi.org/10. bbi.2014.10.015. 1016/j.jneuroim.2004.06.014. Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 21 of 22 23. Cornfield LJ, Sills MA. High affinity interleukin-6 binding sites in bovine 46. Hanisch UK. Microglia as a source and target of cytokines. Glia. 2002;40(2): hypothalamus. Eur J Pharmacol. 1991;202(1):113–5. 140–55. https://doi.org/10.1002/glia.10161. 24. Danscher G. Exogenous selenium in the brain. A histochemical technique 47. Hewett SJ, Jackman NA, Claycomb RJ. Interleukin-1beta in central nervous for light and electron microscopical localization of catalytic selenium bonds. system injury and repair. Eur J Neurodegener Dis. 2012;1(2):195–211. Histochemistry. 1982;76(3):281–93. 48. Hirbec HE, Noristani HN, Perrin FE. Microglia Responses in Acute and 25. Deller T, Del Turco D, Rappert A, Bechmann I. Structural reorganization of Chronic Neurological Diseases: What Microglia-Specific Transcriptomic the dentate gyrus following entorhinal denervation: species differences Studies Taught (and did Not Teach) Us. Front Aging Neurosci. 2017;9:227. between rat and mouse. Prog Brain Res. 2007;163:501–28. https://doi.org/10. https://doi.org/10.3389/fnagi.2017.00227. 1016/S0079-6123(07)63027-1. 49. Hume DA, Ross IL, Himes SR, Sasmono RT, Wells CA, Ravasi T. The 26. Deller T, Frotscher M. Lesion-induced plasticity of central neurons: sprouting mononuclear phagocyte system revisited. J Leukoc Biol. 2002;72(4):621–7. of single fibres in the rat hippocampus after unilateral entorhinal cortex 50. Huppert J, Closhen D, Croxford A, White R, Kulig P, Pietrowski E, et al. lesion. Prog Neurobiol. 1997;53(6):687–727. Cellular mechanisms of IL-17-induced blood-brain barrier disruption. FASEB 27. Deller T, Nitsch R, Frotscher M. Layer-specific sprouting of commissural J. 2010;24(4):1023–34. https://doi.org/10.1096/fj.09-141978. fibres to the rat fascia dentata after unilateral entorhinal cortex lesion: a 51. Ikeda K, Iwasaki Y, Shiojima T, Kinoshita M. Neuroprotective effect of various Phaseolus vulgaris leucoagglutinin tracing study. Neuroscience. 1996;71(3): cytokines on developing spinal motoneurons following axotomy. J Neurol 651–60. https://doi.org/10.1016/0306-4522(95)00475-0. Sci. 1996;135(2):109–13. 28. Dinarello CA. Overview of the IL-1 family in innate inflammation and 52. Kaminski M, Bechmann I, Pohland M, Kiwit J, Nitsch R, Glumm J. Migration acquired immunity. Immunol Rev. 2018;281(1):8–27. https://doi.org/10.1111/ of monocytes after intracerebral injection at entorhinal cortex lesion site. J imr.12621. Leukoc Biol. 2012;92(1):31–9. https://doi.org/10.1189/jlb.0511241. 29. Dissing-Olesen L, Ladeby R, Nielsen HH, Toft-Hansen H, Dalmau I, Finsen B. 53. Kawanokuchi J, Shimizu K, Nitta A, Yamada K, Mizuno T, Takeuchi H, Axonal lesion-induced microglial proliferation and microglial cluster Suzumura A. Production and functions of IL-17 in microglia. J formation in the mouse. Neuroscience. 2007;149(1):112–22. https://doi.org/ Neuroimmunol. 2008;194(1-2):54–61. https://doi.org/10.1016/j.jneuroim.2007. 10.1016/j.neuroscience.2007.06.037. 11.006. 30. Erta M, Quintana A, Hidalgo J. Interleukin-6, a major cytokine in the central nervous 54. Kishimoto T. Interleukin-6: discovery of a pleiotropic cytokine. Arthritis Res system. Int J Biol Sci. 2012;8(9):1254–66. https://doi.org/10.7150/ijbs.4679. Ther. 2006;8(Suppl 2):S2. https://doi.org/10.1186/ar1916. 31. Finsen B, Jensen MB, Lomholt ND, Hegelund IV, Poulsen FR, Owens T. 55. Kishimoto T, Akira S, Narazaki M, Taga T. Interleukin-6 family of cytokines Axotomy-induced glial reactions in normal and cytokine transgenic mice. and gp130. Blood. 1995;86(4):1243–54. Adv Exp Med Biol. 1999;468:157–71. 56. Kloss CU, Kreutzberg GW, Raivich G. Proliferation of ramified microglia on an 32. Finsen B, Owens T. Innate immune responses in central nervous system astrocyte monolayer: characterization of stimulatory and inhibitory inflammation. FEBS Lett. 2011;585(23):3806–12. https://doi.org/10.1016/j. cytokines. J Neurosci Res. 1997;49(2):248–54. febslet.2011.05.030. 57. Krady JK, Lin HW, Liberto CM, Basu A, Kremlev SG, Levison SW. Ciliary 33. Frotscher M, Heimrich B, Deller T. Sprouting in the hippocampus is layer- neurotrophic factor and interleukin-6 differentially activate microglia. J specific. Trends Neurosci. 1997;20(5):218–23. Neurosci Res. 2008;86(7):1538–47. https://doi.org/10.1002/jnr.21620. 34. Gadient RA, Otten U. Expression of interleukin-6 (IL-6) and interleukin-6 58. Ladeby R, Wirenfeldt M, Dalmau I, Gregersen R, Garcia-Ovejero D, Babcock receptor (IL-6R) mRNAs in rat brain during postnatal development. Brain A, et al. Proliferating resident microglia express the stem cell antigen CD34 Res. 1994;637(1-2):10–4. in response to acute neural injury. Glia. 2005a;50(2):121–31. https://doi.org/ 35. Gadient RA, Otten UH. Interleukin-6 (IL-6)--a molecule with both beneficial 10.1002/glia.20159. and destructive potentials. Prog Neurobiol. 1997;52(5):379–90. 59. Ladeby R, Wirenfeldt M, Garcia-Ovejero D, Fenger C, Dissing-Olesen L, 36. Galiano M, Liu ZQ, Kalla R, Bohatschek M, Koppius A, Gschwendtner A, et al. Dalmau I, Finsen B. Microglial cell population dynamics in the injured adult Interleukin-6 (IL6) and cellular response to facial nerve injury: effects on central nervous system. Brain Res Brain Res Rev. 2005b;48(2):196–206. lymphocyte recruitment, early microglial activation and axonal outgrowth in https://doi.org/10.1016/j.brainresrev.2004.12.009. IL6-deficient mice. Eur J Neurosci. 2001;14(2):327–41. 60. Lapenna A, De Palma M, Lewis CE. Perivascular macrophages in health and 37. Garner KM, Amin R, Johnson RW, Scarlett EJ, Burton MD. Microglia priming disease. Nat Rev Immunol. 2018;18(11):689–702. https://doi.org/10.1038/ by interleukin-6 signaling is enhanced in aged mice. J Neuroimmunol. 2018; s41577-018-0056-9. 324:90–9. https://doi.org/10.1016/j.jneuroim.2018.09.002. 61. Li Q, Barres BA. Microglia and macrophages in brain homeostasis and 38. Garre JM, Yang G. Contributions of monocytes to nervous system disorders. disease. Nat Rev Immunol. 2018;18(4):225–42. https://doi.org/10.1038/nri. J Mol Med. 2018;96(9):873–83. https://doi.org/10.1007/s00109-018-1672-3. 2017.125. 39. Gartlan KH, Markey KA, Varelias A, Bunting MD, Koyama M, Kuns RD, et al. 62. Liva SM, de Vellis J. IL-5 induces proliferation and activation of microglia via Tc17 cells are a proinflammatory, plastic lineage of pathogenic CD8+ T cells an unknown receptor. Neurochem Res. 2001;26(6):629–37. that induce GVHD without antileukemic effects. Blood. 2015;126(13):1609– 63. Lobo-Silva D, Carriche GM, Castro AG, Roque S, Saraiva M. Balancing the 20. https://doi.org/10.1182/blood-2015-01-622662. immune response in the brain: IL-10 and its regulation. J 40. Gomez-Nicola D, Fransen NL, Suzzi S, Perry VH. Regulation of microglial Neuroinflammation. 2016;13(1):297. https://doi.org/10.1186/s12974-016- proliferation during chronic neurodegeneration. J Neurosci. 2013;33(6):2481– 0763-8. 93. https://doi.org/10.1523/JNEUROSCI.4440-12.2013. 64. Loddick SA, Turnbull AV, Rothwell NJ. Cerebral interleukin-6 is 41. Gu WL, Fu SL, Wang YX, Li Y, Wang XF, Xu XM, Lu PH. Expression and neuroprotective during permanent focal cerebral ischemia in the rat. J regulation of versican in neural precursor cells and their lineages. Acta Cereb Blood Flow Metab. 1998;18(2):176–9. https://doi.org/10.1097/ Pharmacol Sin. 2007;28(10):1519–30. https://doi.org/10.1111/j.1745-7254. 00004647-199802000-00008. 2007.00659.x. 65. Manich G, Gomez-Lopez RG, Almolda B, Villacampa N, Shrivastava K, 42. Haas CA, Rauch U, Thon N, Merten T, Deller T. Entorhinal cortex lesion in adult Gonzalez B, Castellano B. Differential roles of TREM2+ microglia in rats induces the expression of the neuronal chondroitin sulfate proteoglycan anterograde and retrograde axonal injury models. Front Cell Neurosci. 2020. neurocan in reactive astrocytes. J Neurosci. 1999;19(22):9953–63. https://doi.org/10.3389/fncel.2020.567404. 43. Hailer NP, Grampp A, Nitsch R. Proliferation of microglia and astrocytes in 66. Matthews DA, Cotman C, Lynch G. An electron microscopic study of lesion- the dentate gyrus following entorhinal cortex lesion: a quantitative induced synaptogenesis in the dentate gyrus of the adult rat. II. bromodeoxyuridine-labelling study. Eur J Neurosci. 1999;11(9):3359–64. Reappearance of morphologically normal synaptic contacts. Brain Res. 1976; 44. Hakkoum D, Stoppini L, Muller D. Interleukin-6 promotes sprouting and 115(1):23–41. functional recovery in lesioned organotypic hippocampal slice cultures. J 67. McLoughlin RM, Jenkins BJ, Grail D, Williams AS, Fielding CA, Parker CR, Neurochem. 2007;100(3):747–57. https://doi.org/10.1111/j.1471-4159.2006. et al. IL-6 trans-signaling via STAT3 directs T cell infiltration in acute 04257.x. inflammation. Proc Natl Acad Sci U S A. 2005;102(27):9589–94. https://doi. 45. Hanamsagar R, Bilbo SD. Environment matters: microglia function and org/10.1073/pnas.0501794102. dysfunction in a changing world. Curr Opin Neurobiol. 2017;47:146–55. 68. Mittal SK, Cho KJ, Ishido S, Roche PA. Interleukin 10 (IL-10)-mediated https://doi.org/10.1016/j.conb.2017.10.007. Immunosuppression: march-i induction regulates antigen presentation by Recasens et al. Journal of Neuroinflammation (2021) 18:31 Page 22 of 22 macrophages but not dendritic cells. J Biol Chem. 2015;290(45):27158–67. multiple sclerosis white and grey matter lesions is dependent on their https://doi.org/10.1074/jbc.M115.682708. inflammatory environment. Acta Neuropathol Commun. 2019;7(1):206. 69. Mittrucker HW, Visekruna A, Huber M. Heterogeneity in the differentiation https://doi.org/10.1186/s40478-019-0850-z. and function of CD8(+) T cells. Arch Immunol Ther Exp (Warsz). 2014;62(6): 93. Waisman A, Hauptmann J, Regen T. The role of IL-17 in CNS diseases. Acta 449–58. https://doi.org/10.1007/s00005-014-0293-y. Neuropathol. 2015;129(5):625–37. https://doi.org/10.1007/s00401-015-1402-7. 70. Mori S, Maher P, Conti B. Neuroimmunology of the Interleukins 13 and 4. 94. Wirenfeldt M, Babcock AA, Ladeby R, Lambertsen KL, Dagnaes-Hansen F, Brain Sci. 2016;6(2). https://doi.org/10.3390/brainsci6020018. Leslie RG, et al. Reactive microgliosis engages distinct responses by 71. Nakamura M, Okada S, Toyama Y, Okano H. Role of IL-6 in spinal cord injury microglial subpopulations after minor central nervous system injury. J in a mouse model. Clin Rev Allergy Immunol. 2005;28(3):197–204. https:// Neurosci Res. 2005;82(4):507–14. https://doi.org/10.1002/jnr.20659. doi.org/10.1385/CRIAI:28:3:197. 95. Wirenfeldt M, Dissing-Olesen L, Anne Babcock A, Nielsen M, Meldgaard M, 72. Neher JJ, Cunningham C. Priming microglia for innate immune memory in Zimmer J, et al. Population control of resident and immigrant microglia by the brain. Trends Immunol. 2019;40(4):358–74. https://doi.org/10.1016/j.it. mitosis and apoptosis. Am J Pathol. 2007;171(2):617–31. https://doi.org/10. 2019.02.001. 2353/ajpath.2007.061044. 73. Niraula A, Sheridan JF, Godbout JP. Microglia Priming with Aging and Stress. 96. Zimmermann J, Emrich M, Krauthausen M, Saxe S, Nitsch L, Heneka MT, et al. Neuropsychopharmacology. 2017;42(1):318–33. https://doi.org/10.1038/npp. IL-17A promotes granulocyte infiltration, myelin loss, microglia activation, and 2016.185. behavioral deficits during cuprizone-induced demyelination. Mol Neurobiol. 74. Nolan A, Kobayashi H, Naveed B, Kelly A, Hoshino Y, Hoshino S, et al. 2018;55(2):946–57. https://doi.org/10.1007/s12035-016-0368-3. Differential role for CD80 and CD86 in the regulation of the innate immune response in murine polymicrobial sepsis. PLoS One. 2009;4(8):e6600. https:// doi.org/10.1371/journal.pone.0006600. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in 75. Norden DM, Godbout JP. Review: microglia of the aged brain: primed to be published maps and institutional affiliations. activated and resistant to regulation. Neuropathol Appl Neurobiol. 2013; 39(1):19–34. https://doi.org/10.1111/j.1365-2990.2012.01306.x. 76. Perry VH, Holmes C. Microglial priming in neurodegenerative disease. Nat Rev Neurol. 2014;10(4):217–24. https://doi.org/10.1038/nrneurol.2014.38. 77. Raivich G, Jones LL, Werner A, Bluthmann H, Doetschmann T, Kreutzberg GW. Molecular signals for glial activation: pro- and anti-inflammatory cytokines in the injured brain. Acta Neurochir Suppl. 1999;73:21–30. 78. Recasens M, Shrivastava K, Almolda B, Gonzalez B, Castellano B. Astrocyte- targeted IL-10 production decreases proliferation and induces a downregulation of activated microglia/macrophages after PPT. Glia. 2018. https://doi.org/10.1002/glia.23573. 79. Reyes TM, Fabry Z, Coe CL. Brain endothelial cell production of a neuroprotective cytokine, interleukin-6, in response to noxious stimuli. Brain Res. 1999;851(1-2):215–20. 80. Ringheim GE. Mitogenic effects of interleukin-5 on microglia. Neurosci Lett. 1995;201(2):131–4. 81. Satoh J, Kino Y, Asahina N, Takitani M, Miyoshi J, Ishida T, Saito Y. TMEM119 marks a subset of microglia in the human brain. Neuropathology. 2016; 36(1):39–49. https://doi.org/10.1111/neup.12235. 82. Satoh JI, Kino Y, Yanaizu M, Ishida T, Saito Y. Microglia express TMEM119 in the brains of Nasu-Hakola disease. Intractable Rare Dis Res. 2019;8(4):260–5. https://doi.org/10.5582/irdr.2019.01123. 83. Schobitz B, de Kloet ER, Sutanto W, Holsboer F. Cellular localization of interleukin 6 mRNA and interleukin 6 receptor mRNA in rat brain. Eur J Neurosci. 1993;5(11):1426–35. 84. Schobitz B, Voorhuis DA, De Kloet ER. Localization of interleukin 6 mRNA and interleukin 6 receptor mRNA in rat brain. Neurosci Lett. 1992;136(2): 189–92. 85. Siebert JR, Conta Steencken A, Osterhout DJ. Chondroitin sulfate proteoglycans in the nervous system: inhibitors to repair. Biomed Res Int. 2014;2014:845323. https://doi.org/10.1155/2014/845323. 86. Spooren A, Kolmus K, Laureys G, Clinckers R, De Keyser J, Haegeman G, Gerlo S. Interleukin-6, a mental cytokine. Brain Res Rev. 2011;67(1-2):157–83. https://doi.org/10.1016/j.brainresrev.2011.01.002. 87. Streit WJ. Microglial response to brain injury: a brief synopsis. Toxicol Pathol. 2000;28(1):28–30. https://doi.org/10.1177/019262330002800104. 88. Swardfager W, Winer DA, Herrmann N, Winer S, Lanctot KL. Interleukin-17 in post-stroke neurodegeneration. Neurosci Biobehav Rev. 2013;37(3):436–47. https://doi.org/10.1016/j.neubiorev.2013.01.021. 89. Taga T, Kishimoto T. Gp130 and the interleukin-6 family of cytokines. Annu Rev Immunol. 1997;15:797–819. https://doi.org/10.1146/annurev.immunol.15. 1.797. 90. Thon N, Haas CA, Rauch U, Merten T, Fassler R, Frotscher M, Deller T. The chondroitin sulphate proteoglycan brevican is upregulated by astrocytes after entorhinal cortex lesions in adult rats. Eur J Neurosci. 2000;12(7):2547–58. 91. Tilgner J, Volk B, Kaltschmidt C. Continuous interleukin-6 application in vivo via macroencapsulation of interleukin-6-expressing COS-7 cells induces massive gliosis. Glia. 2001;35(3):234–45. 92. van Wageningen TA, Vlaar E, Kooij G, Jongenelen CAM, Geurts JJG, van Dam AM. Regulation of microglial TMEM119 and P2RY12 immunoreactivity in