Microglial angiotensin type 2 receptors mediate sex-specific expression of inflammatory cytokines independently of circulating estrogen

RESEARCH ARTICLE Microglial angiotensin type 2 receptors mediate sex-specific expression of inflammatory cytokines independently of circulating estrogen Pablo Garrido-Gil 1,2 | Maria A. Pedrosa 1,2 | Maria Garcia-Garrote 1,2 | Ana Pequeño-Valtierra 3 | Jorge Rodríguez-Castro 3 | Daniel García-Souto 3 | Ana I. Rodríguez-Pérez 1,2 | Jose L. Labandeira-Garcia 1,2 1 Laboratory of Cellular and Molecular Neurobiology of Parkinson's disease, Research Center for Research in Molecular Medicine and Chronic Diseases (CIMUS), Department of Morphological Sciences, IDIS, University of Santiago de Compostela, Santiago de Compostela, Spain 2 Networking Research Center on Neurodegenerative Diseases (CiberNed), Madrid, Spain 3 Laboratory of Genomes and Disease, Center for Research in Molecular Medicine and Chronic Diseases (CIMUS), University of Santiago de Compostela, Santiago de Compostela, Spain Correspondence Jose L. Labandeira-Garcia and Ana I. Rodríguez-Pérez, Department of Morphological Sciences, CIMUS, University of Santiago de Compostela, 15782 Santiago de Compostela, Spain. Email: [email protected] and anai. [email protected] Funding information Consellería de Cultura, Educaci on e Ordenaci on Universitaria, Xunta de Galicia, Grant/Award Numbers: XUGA, ED431C 2018/10, ED431G/05; Instituto de Salud Carlos III, Grant/Award Numbers: PI20/00385, RD16/0011/0016, CIBERNED; Secretaría de Estado de Investigaci on, Desarrollo e Innovaci on, Grant/Award Number: RTI2018-098830-B-I00; Regional European Development Fund (FEDER) Abstract There are sex differences in microglia, which can maintain sex-related gene expression and functional differences in the absence of circulating sex steroids. The angiotensin type 2 (AT2) receptors mediate anti-inflammatory actions in different tissues, including brain. In mice, we performed RT-PCR analysis of microglia isolated from adult brains and RNA scope in situ hybridization from males, females, ovariectomized females, orchiectomized males and brain masculinized females. We also compared wild type and AT2 knockout mice. The expression of AT2 receptors in microglial cells showed sex differences with much higher AT2 mRNA expression in females than in males, and this was not dependent on circulating gonadal hormones, as observed using ovariectomized females, brain masculinized females and orchiectomized males. These results suggest genomic reasons, possibly related to sex chromosome complement, for sex differences in AT2 expression in microglia, as the AT2 receptor gene is located in the X chromosome. Furthermore, sex differences in expression of AT2 receptors were associated to sex differences in microglial expression of key anti-inflammatory cytokines such as interleukin-10 and pro-inflammatory cytokines such as interleukin-1βand interleukin-6. In conclusion, sex differences in microglial AT2 receptor expression appear as a major factor contributing to sex differences in the neuroinflammatory responses beyond the effects of circulating steroids. KEYWORDS AT2 receptor, gender, gonadal hormones, interleukins, neuroinflammation, sex chromosome complement, sex dimorphism Pablo Garrido-Gil and Maria A. Pedrosa contributed equally to experiments. Received: 2 April 2022 Revised: 17 July 2022 Accepted: 25 July 2022 DOI: 10.1002/glia.24255 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2022 The Authors. GLIA published by Wiley Periodicals LLC. 2348 Glia. 2022;70:2348–2360. wileyonlinelibrary.com/journal/glia 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1|INTRODUCTION It is now known that sex modifies the outcome of many neurological diseases and particularly those related to neuroinflammation (Mazure & Swendsen, 2016; Villapol et al., 2017). Several studies have shown that microglial cells show sex differences, and this may play a major role in sex differences in disease outcome (Han et al., 2021;Kerretal.,2019). Initially, sex differences were related to effects of circulating gonadal hormones, particularly on microglial inflammasome (Habib & Beyer, 2015). Microglial cells respond to in vitro treatments with gonadal steroids and changes in neuroinflammatory or neurodegenerative responses were observed after ovariectomies (Acosta-Martinez, 2020; Rodriguez-Perez et al., 2011, 2012; VanRyzin et al., 2019). However, a series of recent studies have revealed that other factors, more difficult to identify in classic in vitro studies, may also determine sex differences in microglia. Microglia can maintain sex-related gene expression and functional differences in the absence of circulating sex steroids (Bordt et al., 2020; Villa et al., 2018). It has been suggested that hormones mayactduringveryearlystagesof brain development, leading to permanent sex-related changes in microglia (Hanamsagar et al., 2017; Villa et al., 2016). Recent studies also suggest the primary action of genes located in the sex chromosomes, so that hormones may be considered as the most important secondary factor downstream of primary effects of sex chromosomes (Arnold, 2020). Incomplete X chromosome inactivation may lead to double expression of some genes leading to sex differences in the related functions (Syrett & Anguera, 2019; Tukiainen et al., 2017), including inflammatory responses (Qi et al., 2021). Most of the escape genes lie in the Xq region, which is the older evolutionary region of the chromosome (Ross et al., 2005). The angiotensin type 2 (AT2) receptors mediate anti-inflammatory actions in different tissues, including brain (Bhat et al., 2019; Fatima et al., 2021). In the brain, AT2 receptors have been observed in neurons and glial cells of rodents, monkeys and humans (Garrido-Gil et al., 2013, 2017), and promote neuroprotective and anti-inflammatory effects, inducing a shift from pro-inflammatory to anti-inflammatory microglial phenotypes (Jackson et al., 2020;Rodriguez-Perezetal.,2020). Interestingly, AT2 receptors were found overexpressed in several murine female tissues relative to males (Garrido-Gil et al., 2021;Okumura et al., 2005; Rodriguez-Perez et al., 2011), and the AT2 receptor gene is in the X chromosome, and particularly in the Xq23 region (de Gasparo et al., 2000). Therefore, both hormonal and sex chromosome complement effects may be responsible for primary sex differences in AT2 expression and differences in microglial responses. In the present study, we isolated microglia from adult healthy mouse brains of males, females, ovariectomized (OVX) females, masculinized females and orchiectomized males, and from both wild type (WT) and AT2 knockout (KO) mice. In microglial cells from the different hormonal conditions, we studied the AT2 receptor expression and the expression of several interleukins (IL) that have previously been shown to be related with the AT2 receptor activity (IL-10, IL-6, and IL-1β). The substantia nigra has the highest levels of microglia in healthy mouse brains (Lawson et al., 1990), and was used to further confirm the presence of AT2 receptors in microglial cells using RNA scope in situ hybridization. 2|METHODS 2.1 |Experimental design Young adult (2–4-month-old) male and female wild-type (WT) and AT2 receptor knockout (AT2 KO; generous gift of Dr. Daniel Henrion) C57/BL6 mice were used. Animals were housed in conditions of constant room temperature (21–22C) and a 12:12 h light–dark cycle and given free access to food and water. All experiments were carried out in accordance with the European Directive 2010/63/EU and the Spanish RD/53/2013 and were approved by the corresponding Ethics Committee at the University of Santiago de Compostela. A first series of experiments were performed in WT mice. In a group of mice, adult microglial cells were isolated from brains of male (n=14), female (n=14) and ovariectomized (OVX)-female (n=14) WT mice and were used to study the effect of sex and estrogen on the AT2 receptor and anti-inflammatory (IL-10) and pro-inflammatory (IL1β, IL-6) cytokine mRNA expression. Two mouse brains were used to obtain one sample of microglial cells. A second group of male (n=8) and female (n=8) WT mice were quickly perfused with ice-cold saline and brains extracted and processed for microglial isolation, as controls to exclude any contamination by circulating blood cells. An additional group of male (n=10) and female (n=10) WT mice were included to observe the mRNA expression of AT2 receptors in microglial cells in brain tissue sections containing the substantia nigra by RNA scope in situ hybridization (both double in situ or immuno/in situ labeling). In a second series of experiments, adult microglia were isolated from male AT2 KO mice (n=14), female AT2 KO mice (n=14) and ovariectomized female AT2 KO mice (n=14) to evaluate the effect of AT2 receptor on sexand estrogen-related differences in proand anti-inflammatory cytokine expression. A third series of mice were used to observe the response of above mentioned IL to a pro-inflammatory stimulus in the presence and absence of AT2 receptors: WT female mice (n=4) and AT2 KO female mice (n=4) were treated with LPS or vehicle (n=4) and sacrificed 24 h later for microglia extraction. In additional series of mice, other possible mechanisms involved in sex differences observed in microglial AT2 receptor expression were investigated. First, male mice were bilaterally orchiectomized (n=6) to analyze possible effects of male gonadal hormones. Second, female pups (n=11) were subjected to brain masculinization using subcutaneous injections of estradiol benzoate. Finally, microglial cells isolated from wild type females (n=3) and males (n=3) were used to study the possibility of observing sex differences in the methylation status in the promoter of the AT2 gene (Agtr2). 2.2 |Estrogen depletion by ovariectomy and ELISA analysis Female mice were bilaterally ovariectomized through a dorsal incision as previously described (Garrido-Gil et al., 2021). The anesthetized animal was placed in the prone position with the tail facing the GARRIDO-GIL ET AL.2349 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License surgeon and the dorsal surgical area was shaved and cleaned with povidone iodine. A skin incision was made at the level of the midline between the caudal edge of the rib cage and the base of the tail. Then, an incision (bilateral to the spinal column) was made in the muscle wall. The ovaries were then located and removed outside. A sterile silk ligation was placed around the ovary, and it was removed. Finally, the muscle wall was closed with stitches, and the skin with staples. The animals were injected with atipamezole to promote their recovery, and with buprenorphine as an analgesic. Microglia was isolated 4 weeks after the operation. 17β-estradiol levels were quantified (pg/ml) in female (OVX, and non-OVX) and male mouse serum using a specific enzyme-linked immunosorbent assay (ab108667) from Abcam, following the manufacturer's instructions. Briefly, 17β-estradiol was added to standard, controls, and samples in the respective 96-well microtiter plates supplied by manufacturer and incubated for 2 h at 37C. When incubation was completed, the wells were washed, and TMB substrate solution was added and incubated at room temperature in the dark. After that, stop solution was added and the absorbance of the samples was read at 450 nm in an Infinite M200 multiwell plate reader (TECAN). 2.3 |Brain masculinization, male orchiectomy, and LPS treatment Brain masculinization was carried out by subcutaneous injections of 5mgof17β-estradiol benzoate (E8515, Sigma) dissolved in 50 ml of sterile sunflower oil. Male and female pups (n=11) were injected at days P2, P5, and P8 (Villa et al., 2018). The effectiveness of the intervention was tested by microscopic evaluation of the types of cells present in vaginal smears, a common procedure used to document the stages of the estrous cycle (Cora et al., 2015). The masculinized females were unable to cycle at 8 weeks of age. Male mice were bilaterally orchiectomized (n=6) through a small incision at the tip of the scrotum. Anesthetized animals were immobilized, shaved, and disinfected. The tunic was opened, and the testis and epididymis were exteriorized. After placing a sterile silk ligation around the caudal zone of the spermatic cord, the testis and epididymis were removed, and the remaining tissue returned into the sac. The procedure was repeated with the other testis. Finally, the muscle wall was closed with stitches, and the skin with staples. In sham orchiectomized animals (n=6) scrotal sac was opened without removing the testis. Finally, the incision was closed, and sutured. Animals were allowed to recover for 2 weeks before study. Testosterone levels were quantified in sham (i.e., controls, n=6) and in orchiectomized mice (n=6) serum using a specific competitive enzyme-linked immunosorbent assay (ADI-900-065 from Enzo), following strictly the manufacturer's instructions. Optical density was measured at 405 nm (with correction at 570 nm) using an Infinite M200 multiwell plate reader (TECAN; 2.0 software) and testosterone concentrations were quantified using a specific standard curve (4PL curve fit). Testosterone levels were 4.393 ng/ml ± 0.404 (SEM) in sham-operated mice and 0.513 ng/ml ± 0.0289 (SEM) in orchiectomized mice. To observe the response to the LPS pro-inflammatory stimulus in the presence and absence of AT2 receptors, WT female mice (n=4) and AT2 KO female mice (n=4) were treated with LPS (lipopolysaccharide; 5 mg/kg i.p.; Sigma) and sacrificed 24 h later, as previously described (Rodriguez-Perez et al., 2020). The control group were WT female mice treated with sterile saline solution (n=4). Finally, microglial cells isolated from a group of male (n=3) and female (n=3) mice were used to study possible sex differences in the methylation status in the promoter of the AT2 gene (Agtr2). 2.4 |Adult microglia isolation Isolation of adult microglia from mouse brains was performed as previously described (Rodriguez-Perez et al., 2020) with slight modifications. Two mouse brains were used to obtain one sample of microglial cells. Brains from perfused (ice-cold saline; 0.9% in DEPC water) or non-perfused mice were removed and washed in free-serum DMEM/ F12 with penicillin/streptomycin and glucose. Then, each brain hemisphere was fully minced with a scalpel into small pieces (approximate 1mm 2 ), repeatedly washed to remove blood cells from vessels, transferred to a Dounce homogenizer containing 5 ml of the enzymatic mix (1 mg/ml papain, 6 U/ml dispase and 20 U/ml DNase in a DMEM/F12 solution with 12.5 mM HEPES and 37.5 mM NaCl) and dissociated on ice with 10–12 passes using the loose-fitting pestle. Dissociated tissue was transferred to a 50 ml conical tube and the enzymatic digestion was performed at 37C for 30 min. Then, cell suspension was washed with 10 ml of ice-cold Hanks' balanced salt solution (HBSS), filtered (100 μm cell strainer) and centrifuged (300 g at 4C for 5 min). The pellet was resuspended in 8 ml of 35% Percoll at RT and transferred to a clean conical tube. After overlaying with 5 ml of HBSS, each sample was incubated for 5 min on ice, and centrifuged (800 g at 4C for 20 min) without braking. Then, the cellular pellet was washed with HBSS, filter (30 μm), centrifuged (400 g at 4C for 5 min), and subsequently, resuspended in 90 μl of separation buffer (PBS with 0.5% bovine serum albumin and 2 mM EDTA), and was transferred to a 15 ml conical tube. For microglia isolation, 10 μl of CD11b (Microglia) MicroBeads (130-093-634, MACS Miltenyi Biotec) were added, and then, the cell-bead suspension was incubated (15 min at 4C) under agitation, washed, resuspended in 500 μl of separation buffer and applied onto MS columns (130-042-210, MACS Miltenyi Biotec) according to manufacturer instructions. 2.5 |Flow cytometry analysis Microglial cells isolated by magnetic sorting were incubated (15 min at 4C) under agitation with monoclonal antibodies against CD11b (PE-conjugated: eBioscience 12-0112, 1:170), CD45 (FITC-conjugated: eBioscience 11-0451, 1:200) or isotype controls (eBioscience 12-4331, 1:50, and 11-4331, 1:100). Then, cells were rinsed in 2 ml ice-cold PBS, pelleted at 300 g for 10 min at 4C and resuspended in 300 μl of separation buffer. To discriminate between living and dead 2350 GARRIDO-GIL ET AL. 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License cells, cells were incubated with DAPI 10 min before flow cytometry. Cells were analyzed on an Accuri C6 flow cytometer (BD Biosciences). For each staining condition (cells only, isotype control, and antibody of interest), at least 1 10 4 events were collected and then analyzed using a FlowJo v10 software (BD Biosciences). Microglia purity was defined as the percentage of all living CD11b+cells that showed CD45 low expression. Positive labeled cells were determined by setting threshold on background staining of isotype control. 2.6 |Real time quantitative PCR Firstly, total RNA extraction was performed with the RNeasy minikit (Qiagen) protocol and then, the RNA was reverse transcribed to complementary DNA using the high-capacity RNA-to-cDNA kit (4387406, Applied Biosystems). Subsequently, the real time qPCR analysis was performed with a QuantStudio3 platform (Applied Biosystems) and the EvaGreen qPCR MasterMix (Applied Biological Materials Inc.). β-actin was used as housekeeping gene and was amplified in parallel with the genes of interest. The relative levels of mRNA were evaluated by the delta Ct method (2 ΔΔCt ), where Ct is the cycle threshold. Expression of each gene was obtained as relative to that of the housekeeping transcripts. Forward (F) and reverse (R) primers sequences, designed using Beacon Designer software (Premier Biosoft), were as follows: for β-actin, F 50-TCGTGCGTGACATTAAAGAG-30,R5 0-TGC CACAGGATTCCATACC-30; for AT2, F 50-TGTGGTCTCACTGTTTT GTTGTCA-30,R5 0-TTGCCCAGAGAGGGAGGGTA-30; for IL-1β, F5 0CCGTGTCTTCCTAAAGTATGG-30,R5 0-GTTTCTTGTGACCCTG AGC-30; for IL-6, F 50-GACTGATGCTGGTGACAAC-30,R5 0-GAGT GGTATCCTCTGTGAAGT-30; and for IL-10, F 50-CATACTGCTAACCG ACTC-30,R5 0-AATGCTCCTTGATTTCTGG-30. 2.7 |RNA scope in situ hybridization Brains from adult mice were rapidly removed, embedded in OCT, frozen with liquid nitrogen cooled-isopentane, and stored at 80C. Then, coronal sections (15 μm thick) containing the substantia nigra were cut at 20C with a cryostat (Thermo Scientific), mounted on glass slides (Superfrost Plus, Thermo Fisher Scientific), and stored at 80C until use. Tissue collection and sectioning were performed in RNase free conditions. For in situ hybridization, slides containing nigral tissue sections were processed with the RNAscope Fluorescent Multiplex Assay using the RNAscope ® Fluorescent Multiplex Reagent Kit (320293-USM; Advanced Cell Diagnostics, ACD, Newark, CA), according to the manufacturer's instructions. Tissue sections were fixed with 4% paraformaldehyde for 30 min at 4C, dehydrated in graded ethanol (50%, 70%, and 100% for 5 min each step) and incubated with Protease IV (322340, ACD) for 30 min at RT. Then, hybridization was performed into the ACD HybEZ™II oven (321720, AC) at 40C for 2 h using the following target probes: Mus musculus angiotensin II receptor type 2mRNA(Agtr2 probe; accession number NM_007429.5, target region 397-1697; 406471-C2, ACD) and Mus musculus integrin alpha M (Itgam), transcript variant 1, mRNA (accession number NM_001082960.1, target region 538-1528; 311491 -C3, ACD), as a microglial marker. In addition, RNAscope 3-plex Negative Control Probe_Mm (320871, ACD) and RNAscope 3plex Positive Control Probe_Mm (320881, ACD) were used as control probes. After probes hybridization, the tissue sections were sequentially incubated into the ACD HybEZ™II oven at 40C with amplification reagents (Detection Reagents kit, 320851, ACD): AMP-1 for 30 min, AMP-2 for 15 min, AMP-3 for 30 min and AMP-4 C for 15 min, which assigned the different fluorophores to each channel: Agtr2 probe (channel 2; fluorophore: Atto 647), Itgam (channel 3; Alexa 488). Then, tissue sections were incubated with DAPI (320858, ACD) for 30 s at RT and immediately cover slipped with Fluoromount-G™Mounting Medium (00-4958-02, Invitrogen). 2.8 |RNA scope in situ hybridization combined with immunohistochemistry The combination of the RNA scope in situ hybridization technique and the immunofluorescence technique was performed as previously described (Lanfranco et al., 2017; Villapol et al., 2017), with slight modifications. For tissue preparation, mice were first transcardially perfused with 0.9% saline and then with cold 4% paraformaldehyde in 0.1 M PBS (pH 7.4). The brains were removed, washed and cryoprotected in PBS containing 20% sucrose (24 h at 4C). Brains were embedded in OCT and frozen with liquid nitrogen cooled-isopentane. Then, coronal tissue sections (15 μm thick) were cut with a cryostat (Thermo Scientific), mounted on glass slides (Epredia™Polysine Adhesion Slides, Thermo Fisher Scientific), and stored at 80C until use. Fluorescent in situ hybridization was performed using RNA scope ® Technology 2.5 Red fluorescent kit for fixed tissue (Advanced Cell Diagnostics, ACD; Hayward, CA), according to the manufacturer's instructions. Briefly, tissue sections were incubated (30 min at 60C) in a HybEZTM II oven (321720, AC), pretreated (10 min at RT) with a H 2 O 2 solution (322335, ACD) and boiled (5 min) with RNA scope Target Retrieval (322001, ACD). Then tissue sections were washed in 100% ethanol and incubated with Protease Plus (322331, ACD) for 30 min at 40C. In situ hybridization was performed using a Mus musculus angiotensin II receptor type 2 mRNA probe (Agtr2; accession number NM_007429.5, target region 397-1697; 403991-C1, ACD) for 2 h at 40C. Then, sequential amplification steps were performed at 40C (Detection Reagent-RED kit, 322360, ACD). Finally, tissue sections were incubated with a detection solution containing a mixture ratio of Red-A to Red-B of 1:60. Positive hybridization consisted of a punctate signal in the far-red spectrum (Excitation: 647 nm; Emission: 657–727 nm) representing a single mRNA target molecule. Following in situ hybridization, tissue sections were processed for immunofluorescence for the microglial marker Iba1. In short, posthybridized sections were first blocked with 5% normal donkey serum (Sigma) in PBS-tween for 1 h at RT and then incubated overnight at 4C with a rabbit polyclonal antibody against Iba-1 (1:100; 019-19741, Wako). The immunoreaction was visualized with an Alexa GARRIDO-GIL ET AL.2351 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License fluor 488-conjugated donkey anti-rabbit IgG (1:200, 2 h at RT; Molecular Probes, Eugene, OR). Finally, tissue sections were incubated with DAPI (320858, ACD) for 30 s at RT and immediately cover slipped with Fluoromount-G™Mounting Medium (00-4958-02, Invitrogen). 2.9 |Images acquisition and quantitative analysis Tissue sections processed for RNA scope in situ hybridization alone or in combination with immunofluorescence technique were visualized using a confocal laser scanning microscope (AOBS-SP5X; Leica Microsystems Heidelberg GmbH, Mannheim, Germany). Fluorescent signals of AT2 mRNA and CD11b mRNA hybridization and immunofluorescence for Iba1 were imaged with a 40and 63 objective lens. For quantitative analysis, three substantia nigra sections from each animal (n=5 per group), selected throughout its anteroposterior axis (anterior, central, and posterior levels) were analyzed by investigators blinded to experimental conditions. Images from six microscopic fields (400.15 mm 2 by field) per each tissue section were obtained within the substantia nigra by laser scanning microscopy and using constant microscope parameters and similar laser intensity. The total number of microglial cells and the total number of microglial cells that express AT2 mRNAs were quantified using the ImageJ64 software (National Institute of Health. Bethesda, MD, USA). Only cell profiles showing a labeled nucleus and microglial morphology were included. The negative probe used as a control did not contain any stained cells. Data were presented as the mean number of cells per mm 2 . The percentage of microglial cells expressing AT2 mRNA was calculated from the corresponding absolute values. In addition, the number of AT2 mRNA positive dots per each cell was quantified in these microscopic fields to assess the intracellular mRNA levels of AT2. A total of 60 microglial cells per animal were analyzed using the ImageJ64 software. 2.10 |Methylation status on the Agtr2 gene promotor DNA was isolated from microglia from brains of male (n=3) and female (n=3) wild type mice and converted with sodium bisulfite using the EZ DNA Methylation-Gold Kit (Zymo Research; D5005) following the manufacturer recommendations. A set of PCR primers (MMU_AGTR2_met1L: 50-TTAGAATTTTGTAGGTTGAAGGTTT-30; MMU_AGTR2_met1R: 50-CAAAATACAATTAAAATACAAAAAAAA-30) was designed with Methprimer (Li & Dahiya, 2002) to amplify a 250 bp fragment encompassing the entire promoter of the AT2 receptor gene (Agtr2), as recovered from the Mus musculus reference genome (NC_000086.8). Next, we carried out PCR followed by Sanger and single-molecule sequencing of the resulting amplicons using a Seqstudio (Applied Biosystems) and a MinION (Oxford Nanopore Technologies Ltd, ONT), respectively. PCR conditions included 35 cycles of 95C for 20 s, 50C for 20 s and 72C for 20 s, in a mix including 40 ng transformed DNA and 1DreamTaq Hot Start Green FIGURE 1 Flow cytometry analysis of the CD11b+cells. Cells isolated from non-perfused [(a)–(e)] and perfused [(f)–(j)] brains did not show significant differences. [(a), (f)] Histograms showing two CD45-expressing cell populations: CD45 low (microglia) and CD 45 high (possible macrophages). [(b), (g)] Microglia, identified as CD11b+and CD45 low (Q3), comprised the highest proportion of cells (91.7%–94.5%), whereas a low percentage of potential macrophages (CD11b+/CD45 high, Q2) were observed. [(c), (h)] Forward scatter (FSC) analysis showing a slight smaller size of the CD45 low cells (red) respect to CD45 high (blue) cells. [(d), (i)] CD11b analysis showed a slight lower expression within CD45 low cells (red) respect to CD45 high (blue) cells. [(e), (j)] Forward scatter (FSC) and side scatter (SSC) characteristics of CD11b+/CD45 low (red) and CD11b+/CD45 high (blue) are shown 2352 GARRIDO-GIL ET AL. 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PCR Master Mix (Thermo Scientific; MAN0015977). PCR amplicons were purified with ExoSAP-IT Express (Applied Biosystems; 75001.1. ML), amplified using the BigDye Terminator v3.1 kit (ThermoFisher; 4337455) and cleaned with BigDye XTerminator kit (ThermoFisher; 4376486) prior to Sanger sequencing in a SeqStudio genetic analyzer controlled by the Sequencing Analysis Software v6.0 (Applied Biosystems; 4474950). Electropherograms were trimmed, visualized and mapped to the reference sequence with Geneious. ONT libraries were constructed with the Sequencing 1D ligation kit (Oxford Nanopore Technologies Ltd; SQK-LSK109), including DNA repair with NEBNext FFPE DNA Repair Mix (New England BioLabs) and NEBNext Ultra II Ligation Module (New England BioLabs) and loaded onto MinION R9.4 flowcells (Oxford Nanopore Technologies Ltd; FLO-MIN106 revD), controlled by MinKNOW v19.12.5 and sequenced to 1000coverage. High accuracy base-calling was performed with Guppy v2.3.1 to generate fastq files and these were further mapped to the Mus musculus reference genome (NC_000086.8) with minimap2 v2.14-r88344. The resulting SAM files were processed with Samtools v1.744 and visualized with the Integrative Genome Viewer (IGV) to assess cytosine methylation status in the promotor and adjacent regions. 2.11 |Statistical analysis All data were obtained from at least three independent experiments and were expressed as means ± SEM. Two group comparisons were performed with Student's t-test. Multiple comparisons were analyzed by one-way ANOVA (when one factor was studied) or two-way ANOVA (when two factors were studied), followed by Student–Newman–Keuls post-hoc test or the corresponding nonparametric test. The normality of populations and homogeneity of variances were tested before each ANOVA. Differences were considered statistically significant at p≤.05. Statistical analyses were carried out with SigmaPlot 11.0 (Systat Software, Inc., San José, CA/USA). Statistics are detailed in Supplemental file 2: Table S1. 3|RESULTS 3.1 |Characterization of models First, we characterized our model of microglial isolation to confirm that the isolated CD11b+cells were microglial cells and that the analyzed population was not constituted by other types of brain cells or blood cells. The efficacy of microglial isolation, confirmed by flow cytometry analysis of these CD11b+cells, showed similar results when perfused (Figure 1a–e) or non-perfused (Figure 1f–j) brains were used for isolation. Flow cytometry analysis showed two CD45-expressing cell subpopulations: CD45 low (microglia) and CD 45 high (possible macrophages) (Figure 1a,f). Microglia, identified as CD11b+and CD45 low, comprised the highest proportion of cells (91.7%–94.5%), whereas a low percentage of potential macrophages (CD11b+/CD45 high) were observed (Figure 1b,g). However, brain-resident microglia can upregulate CD45 expression in response to insults (Sedgwick et al., 1991), indicating difficulty to stablish that high CD45 are not microglia. The presence of these two cell populations was also confirmed by analysis of size (Figure 1c,h), CD11b expression levels (Figure 1d,i) and internal complexity (i.e., granularity) (Figure 1e,j). Thus, the microglia (CD11b+/CD45 low) showed a smaller size, a lower expression of CD11b and a lower internal complexity than CD11b+/CD45 high cells. FIGURE 2 Estradiol concentrations (pg/ml) in different mouse groups and absence of AT2 mRNA in AT2 KO microglia. Estradiol (E2) serum concentrations in wild type (wt) males, wt females, ovariectomized (OVX) wt females, AT2 KO females and OVX AT2 KO females are shown in (a). The lack of expression of AT2 mRNA in adult microglia isolated AT2 KO mice and the presence in WT mice was demonstrated by RT-PCR [(b)–(d)]. Specific AT2 amplification by RT-PCR in adult microglia of WT mice was demonstrated by the presence of a unique peak at the corresponding melting temperature (T m ) in the melt curves, and no peak was observed in the melt curve plot of AT2 KO mice (d). Representative bands of AT2 expression by RT-PCR in isolated mouse WT and AT2 KO adult microglia [(b), (c)]. Data are mean ± SEMs. *p< .05 compared to male group, # p< .05 compared to wt female group, & p< .05 compared to AT2 KO females. Multiple comparisons were analyzed by Kruskal-Wallis one way analysis of variance on ranks followed by Dunn's method post-hoc test. KO, knockout; OVX, ovariectomized; wt, wild type GARRIDO-GIL ET AL.2353 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License The levels of circulating estrogens in the different mouse models used in this study and efficacy of ovariectomies were estimated by determination of levels of Estradiol (E2) concentrations in wt males (4.658 ± 0.579 pg/ml), wt females (72.123 ± 5.526 pg/ml), ovariectomized (OVX) wt females (4.869 ± 0.551 pg/ml), AT2 KO females (72.702 ± 4.866 pg/ml) and OVX AT2 KO females (4.385 ± 0.645 pg/ml) (Figure 2a). These levels of E2 were similar to those obtained in our previous studies in mice (Garrido-Gil et al., 2021). Then, we confirmed the specific AT2 amplification by RT-PCR in adult microglia. The specificity was demonstrated by the presence of a unique peak at the corresponding melting temperature (T m ) in the melt curves, while no peak was observed in the melt curve plot of the negative controls. The lack of AT2 receptors in the microglia isolated from AT2 KO mice was confirmed using primers designed against the mRNA region (second putative transmembrane domain) that is codified by the deleted sequence (exon 3) used for disrupting the Agtr2 gene in these KO mice, which clearly showed that AT2 mRNA was expressed in microglia isolated from WT mice but was absent in microglia isolated from AT2 KO mice (Figure 2b–d). 3.2 |Levels of AT2 mRNA and inflammationrelated cytokine expression in microglia from WT males and females We analyzed the expression of AT2 mRNA and inflammation-related cytokines mRNA in microglial cells isolated from non-perfused (Figure 3a–d) and perfused (Figure 3e–h) brains of adult WT males, WT females and WT OVX females. AT2 mRNA expression was much higher in microglia from females and OVX females than in microglia from males (Figure 3a,e). There was a decrease in microglial AT2 expression in OVX females as compared with control females that was not statistically significant. Microglia isolated from male brains showed lower levels of mRNA expression for the anti-inflammatory cytokine IL-10 than microglia from females and OVX females, and no significant difference was observed between females and OVX females (Figure 3b,f). However, mRNA expression for the pro-inflammatory cytokines IL-1βand IL-6 was significantly higher in microglia isolated from males than in microglia isolated from females or OVX females, and no significant differences in IL-1βand IL-6 expression were detected between females and OVX females (Figure 3c,d,g,h). These results were confirmed both FIGURE 3 mRNA expression of AT2 [(a), (e)], IL-10 [(b), (f)], IL-1β[(c), (g)] and IL-6 [(d), (h)] in microglia. Microglia was isolated from perfused [(a)-(d)] and non-perfused [(e)–(h)] brains of wild type (WT) males, WT females and ovariectomized (OVX) WT females using RT-PCR. For mRNA the comparative cycle threshold values method (2 ΔΔCt ) was used. Data are mean ± SEMs. *p< .05 compared to wild type males. One-way ANOVA followed by Student's Newman–Keuls post-hoc test [(a)–(d)] and Student's ttests [(e)–(h)] 2354 GARRIDO-GIL ET AL. 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License in perfused and non-perfused brains of WT mice indicating that these effects are not related to circulating macrophages. Altogether suggests that differences in expression of AT2 receptors, the antiinflammatory cytokine IL-10 and the pro-inflammatory cytokines IL1βand IL-6 in microglia of adult healthy WT mice are primary related to sex of mice. Finally, we confirmed the expression of AT2 mRNA in microglia from healthy brains of WT male and female mice using in situ RNA scope in tissue sections from the nigral region, as well as in situ hybridization combined with immunohistochemistry (Figure 4). This again confirmed the presence of AT2 expression in cells labeled by microglial markers and showing clear microglial morphology (i.e., microglia) in control brains. As observed in isolated microglia, AT2 mRNA labeling was more marked in female than in male microglia (Figure 4a–l). In situ procedures also showed that females had significantly higher number of AT2-labeled microglial cells, and significantly higher number of AT2 mRNA dots by cell than males (Figure 4m–p). 3.3 |Levels of inflammation-related cytokine expression in microglia from WT and AT2 KO males and females In a second series of experiments, we used wild-type and AT2 KO males, females and OVX females to know whether differences in the microglial expression of IL-10, IL-1βand IL-6 between males and females (control or OVX females) and not dependent of circulating FIGURE 4 AT2 receptor mRNA expression using in situ RNA scope probes (Agtr2) and double immune/in situ RNA scope. Representative photomicrographs showing microglia from the nigral region in male [(a)–(d)] and female [(e)–(h)] wild-type (wt) mice using double in situ RNA scope (Agtr2 in red and integrin alpha M, Itgam in blue), and counterstained with DAPI (white). [(i)–(l)] Double immuno/in situ showing Iba-1 (microglial marker; green) and Agtr2 (red, arrows). [(m)–(p)] Agtr2 was more abundant in microglia from females both in number (m), percentage (o) of labeled cells and dots by cell (p), without significant difference in the total number of microglial cells (n). Data are mean ± SEMs. *p< .05 compared to wt males. Student's ttests [(m)–(p)]. Scale bar 12.5 μm GARRIDO-GIL ET AL.2355 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License gonadal steroids could be related to the observed differences in AT2 expression between males and females. Changes in IL-10, IL-1β, and IL-6 expression induced by AT2 deletion were more marked in females than in males, which is consistent with the higher levels of AT2 expression observed in females relative to males. Interestingly, no significant differences were observed in microglial expression of IL-10, IL-1β, and IL-6 between males, females and OVX females of AT2 KO mice, suggesting that differences in AT2 expression play a major role in sex differences in regulation of microglial IL-10, IL-1β, and IL-6 expression in the control brains (Figure 5a–c). Under intense pro-inflammatory stimulus with LPS, there was a marked increase in levels of mRNA expression for IL10 in WT mice that was much higher than that observed in AT2 KO mice, while mRNA expression of the pro-inflammatory cytokine IL-6 was significantly higher in the AT2 KO mice relative to theWTmice(Figure6a,b). FIGURE 5 IL-10 (a), IL-1β(b), and IL-6 (c) mRNA in microglia isolated from different mouse groups. Wild-type (wt) or AT2 knockout (KO) male, female and ovariectomized (OVX) female brains were analyzed using RT-PCR. For mRNA, the comparative cycle threshold values method (2 ΔΔCt ) was used. Data are mean ± SEMs. *p< .05 compared to the corresponding wt control, $ p< .05 compared to wt males. Multiple comparisons were analyzed by two-way ANOVA followed by Student–Newman–Keuls post-hoc test [(b), (c)] and twoway ANOVA on ranks Kruskal Wallis, (a) FIGURE 6 Effects of LPS injection on microglial interleukin mRNA expression in the presence or the absence of AT2 receptors (a, b), and effect of female brain masculinization (c) and male orchiectomy (d) on AT2 mRNA expression using RT-PCR. In vivo injection of LPS induced a marked increase in levels of IL-10 mRNA expression in microglia from WT mice that was much higher than in microglia from AT2 KO mice (a), while IL-6 mRNA expression was significantly higher in microglia from AT2 KO mice than in microglia from WT mice (b). Microglia from masculinized female brains showed higher AT2 mRNA expression than microglia from WT males (c). Microglia from orchiectomized WT males showed AT2 mRNA expression similar than microglia from control WT males (d). For mRNA, the comparative cycle threshold values method (2 ΔΔCt ) was used. Data are mean ± SEMs. *p< .05 compared to the corresponding WT control, # p< .05 compared to WT +LPS. Oneway ANOVA followed by Student's Newman–Keuls post-hoc test (a); Kruskal-Wallis one way analysis of variance on ranks followed by Student's Newman–Keuls post-hoc test (b) and Student's ttests [(c), (d)] 2356 GARRIDO-GIL ET AL. 10981136, 2022, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/glia.24255 by Consorcio Interuniversitario Do Sistema Universitario De Galicia (Cisug), Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License