Animals
This study was approved by the ethics committee at Karolinska Institutet, Stockholms Norra djurförsöksetiska nämnd (approval number: N94/15 and N126/16) and was conducted according to the relevant guidelines and regulations (Swedish Animal Welfare Act 1988:543), as well as the ARRIVE guidelines for animal experiments. We used Cx3cr1EYFP-CreER mice, which express a Cre-ERT2 fusion protein and enhanced yellow fluorescent protein (EYFP), and Rosa26DTA+/- mice, which carry a loxP-flanked stop cassette associated with an attenuated diphtheria toxin [41]. Crossbreeding of these two strains produced Cx3cr1CreER-EYFP+/Rosa26DTA+/- mice (DTA+ mice, microglia-depleted mice) and Cx3cr1 CreER-EYFP+/Rosa26DTA-/- mice (DTA- mice, control mice) (Figure 1a). All the mice were housed in a humidity-controlled room with a 12-h light-dark cycle and ad libitum access to food and water.
Tamoxifen administration
Tamoxifen was purchased from Sigma-Aldrich (Cat. #T5648), dissolved in corn oil (8 mg tamoxifen/1 ml corn oil) at 70 °C, and intraperitoneally (50 mg/kg) injected to the mice at P8 and P9.
Induction of hypoxic-ischemic encephalopathy
HIE induction was performed without knowing the genotype. Unilateral HI was induced at P10 using the modified Rice-Vannucci model [42-44]. After sedation with isoflurane (4% for induction and 2% for maintenance) and local anesthesia with Marcaine (Astra Zeneca), the left common carotid artery was isolated and occluded through 8-Watt electronic coagulation (Figure S1). The skin incision was sutured with 6-0 silk surgical thread and infiltrated with additional local anesthesia. All mice were subjected to ischemic surgery within < 5 min. Pups were returned to the dam for 1 h with subsequent placement in a hypoxic chamber (BioSpherix, NY, USA) with 10% O2 in 90% N2 for 1 h. For mice in the sham group, the carotid artery was isolated but not cauterized. For injury evaluation and cytokine analysis, 40 and 64 pups underwent the HI insult, respectively. The sham group was included 41 pups.
Immunohistochemistry
The mice were anesthetized using an injection of 50 mg/kg pentobarbital (APL, Sweden) and transcardially perfused with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde (PFA) at P10, P13, P17, and P24 for a repopulation study and at P13 after HI for evaluation of the infarction size. The removed brains were fixed in 4% PFA for 24 h, dehydrated, embedded in a paraffin block, and coronally sectioned to 5-µm thick slices using a sliding microtome. Six levels of each brain were collected as previously reported [18, 45, 46]. The first level was obtained corresponding to the bregma 1.3 mm in the adult mouse brain with every 100th section being collected and deparaffinized. After deparaffinization, antigen retrieval was performed in 10 mM citric buffer (pH 6.0) for 10 min. After blocking, the sections were exposed to primary antibodies in a humidity chamber box overnight at 4 °C. The primary antibodies were those against rabbit Iba-1 obtained from Wako (#1919741) and diluted at 1:1000, mouse microtubule-associated protein-2 (MAP2) obtained from Sigma (#M4403) and diluted at 1:1000, rabbit anti glial fibrillary acidic protein (GFAP) diluted at 1:500 (#ab5804), chicken anti-green fluorescent protein obtained from Abcam (#ab13970) and diluted at 1:500, and mouse anti myelin basic protein (MBP) obtained from Covance and diluted 1:1000. After washing, biotin-conjugated donkey anti-rabbit secondary antibody (Jackson Immuno Research) diluted at 1:1000 was applied for Iba-1 while the VECTOR M.O.M.™ Immunodetection Kit was used for MAP-2. After blocking endogenous peroxidase using 0.3% H2O2, the sections were visualized using the VECTASTAIN Elite ABC-Peroxidase Kit and 3,3’-diaminobenzidine. For GFAP, biotinylated donkey anti-rabbit secondary antibody was applied for 1 h prior to incubation with VECTASTAIN® Elite ABC-Peroxidase Kit. For immunofluorescence analyses, the following secondary antibodies were used: donkey anti-rabbit IgG Alexa Fluor 555 (1:1000) and donkey anti-chicken Alexa Fluor 488 (1:1000). Nucleus staining was performed using Hoechst 33342 (1:1000, ThermoFisher). Coverslips were mounted onto slides using antifade reagent ProLong Gold (Molecular probes/Life technologies).
Iba-1+ cell counting
Sections were examined using Zeiss Axio Imager M2 (Carl Zeiss, Göttingen, Germany). The microglia number was determined by counting the Iba-1-positive cells in the whole hemisphere. The total Iba-1+ cell number was calculated using the fractionator option in Stereo Investigator software (MicroBrightField, Colchester, Vt., USA). Briefly, each section was divided into 200 approximately equal squares (400 × 400 µm, sampling grid). Subsequently, the number of Iba-1+ cells with a complete cell body was counted in each 100 × 100 µm red-green square (counting frame) using a 20x air objective. The software allowed estimation of the total cell number and the whole coronal section area. Slides from the DTA+ mice at P10, P13, and P17 showed few unevenly distributed microglial cells; therefore, all the Iba-1+ cells were directly counted without using a fractionator. Based on these data, the Iba-1+ cell density was calculated.
Fluorescence-Activated Cell Sorting
Mice were sacrificed using an intraperitoneal injection of 50 mg/kg pentobarbital (APL, Sweden) and transcardially perfused with PBS. After 5 minutes of perfusion, the brains were extracted and the olfactory bulbs and cerebellum removed. Subsequently, the right and left hemispheres were separated and individually processed. Next, the tissue was minced with a scalpel and digested for 30 min at 37 °C with Hank’s Balanced Salt Solution (HBSS) containing 0.02% DNase 1 (Sigma-Aldrich) and 0.005% collagenase (Sigma-Aldrich). The reaction was stopped using cold HBSS containing 0.01% ethylenediaminetetraacetic acid (EDTA); subsequently, the samples were strained and centrifuged at 4 °C (300 g, 10 min). The pellets were resuspended in 38% percoll solution and centrifuged at 4 °C (800 g, 10 min). After careful removal of the supernatant, the pellets were rinsed with PBS and transferred to a V-bottom plate for staining. The samples were incubated for 20 min at 4 °C with PBS containing CD11b (1:100, M1-70, BioLegend), CD45 (1:1000, 30-F11, BioLegend), and LIVE/DEAD™ Fixable Near-IR Dead Cell Stain Kit (Invitrogen) and rinsed before analysis. The cell numbers were determined using the Gallios flow cytometer (Beckman Coulter) and analyzed using Kaluza software (Beckman Coulter).
Injury evaluation
At P13, tissue loss was assessed by determining the MAP-2-stained area as previously described.[18, 45] First, the MAP-2-positive area was separately measured in each hemisphere using Stereo Investigator software (MicroBrightField, USA). Second, the difference in the areas was calculated by subtracting the staining area in the contralateral hemisphere from that in the ipsilateral hemisphere. Next, the total tissue loss volume was calculated based on the Cavalieri principle using the following formula: V = ΣA·P·T (where V is the total volume, ΣA is the sum of the areas measured, P is the inverse of the sampling fraction, and T is the section thickness). This evaluation was performed with genotype blinding.
TUNEL and NeuN double staining
After deparaffinization, antigen retrieval was performed in 10 mM citric buffer (pH 6.0) for 15 min. The sections were incubated with 5% normal donkey serum and 0.3% Triton X-100 in PBS for 30 min at room temperature followed by rabbit monoclonal anti-NeuN antibody (1:100, MABN140, Millipore Corpo, CA, USA) at 4 °C overnight. After washing out the primary antibody, donkey anti-rabbit Alexa Fluor 488 (1:300, A-212206, Thermo Fisher Scientific) was applied for 2 hours at room temperature. The sections were rinsed with PBS followed by the terminal deoxynucleotidyl transferase-dUTP nick end labeling (TUNEL) (Click-iT Plus TUNEL Assay Kit, C10618, Thermo Fisher Scientific Europe BV) assay according to the manufacturer’s protocol. Stereological analysis was performed using Stereo Investigator software (MicroBrightField, USA). The number of TUNEL+ cells was counted under x20 magnification [47, 48]. First, each region’s contour (striatum, hippocampus, cortex, and thalamus) was traced using a 5x air objective. The counting frame and grid size were both set to 150 × 150 µm and the number of TUNEL-positive cells was counted using a 20x air objective. Cell counting was performed by an investigator blinded to the genotype. Fluorescent images were obtained using a LSM700 laser scanning confocal microscope (Axio-observer Z1; Carl Zeiss microscopy, Germany) and analyzed using the software ZEN 2.3 (blue edition; Zeiss).
Genotyping
Post-sacrifice genotyping was performed using a piece of tissue that was placed in 75 µl extraction buffer containing 25 mM sodium hydroxide and 0.2 mM EDTA in distilled water and incubated at 98 ℃ for 60 min. Next, 75 µl of 40 mM tris-hydroxymethyl aminomethane-hydrochloric acid (pH 5.0) was added and centrifuged at 6000 revolutions per minute (rpm) for 3 minutes. The supernatant was collected for genotyping. Regarding the polymerase chain reaction (PCR), 2 µl of cDNA was mixed with 0.25 µl of MyTaq™ HS DNA Polymerase, 10 µl of 5x MyTaq™ Reaction Buffer (Bioline meridian, London, UK), 3 µl of each of the corresponding primers (0.5 µM), and 34.75 µl of distilled water (total 50 µl). Table S1 shows each primer sequence. The Cre allele was amplified using 30 cycles of 95 °C for 30 s, 61.5 °C for 30 s, and 72 °C for 60 s. For the DTA allele, there were 35 cycles of 94 °C for 30 s, 60 °C for 60 s, and 72 °C for 60 s. The PCR products were separated on 1% agarose gel and stained with GelRed® Nucleic Acid Gel Stain (Biotium, Inc. Fremont, CA, USA).
Quantitative polymerase chain reaction
Total RNA was isolated from the ipsilateral hemisphere after HI using the RNeasy Kit (QIAGEN) and measured using a NanoDrop (Thermo Scientific). All the samples had a nucleotide ratio (A260: A280) within the range of 1.9-2.1. A total of 1 µg of extracted RNA underwent reverse transcription into cDNA using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s protocol. Quantitative PCR (qPCR) was performed using the Step-One-Plus Real-Time PCR machine (Applied Biosystems) with Power SYBR Green PCR Master Mix (Applied Biosystems) and the primers presented in Table S2. The set-up condition was initial denaturation at 95 °C for 5 min, subsequent denaturation at 95 °C for 5 s, annealing at 60 °C for 10 s, and elongation for 30 s for a total of 40 cycles. The cycle time values were normalized to the β-actin of the same sample. The mRNA expression levels were calculated using the delta-delta CT method as fold changes compared to the DTA- sham samples at P13. For pre-HI insult assessments, DTA+ samples were compared with DTA- samples at P10 after tamoxifen administration at P8 and P9. All samples were examined in triplicate. The aforementioned analyses were performed using the StepOne software program (version 2.3, Applied Biosystems).
Quantitative enzyme-linked immunosorbent assay
Ipsilateral hemispheres were collected from the pups at P13 after administering the HI insult at P10. Tissue samples were homogenized in N-PER™ Neuronal Protein Extraction Reagent (Thermo Fisher) with cOmplete™, Mini, EDTA-free Protease Inhibitor Cocktail (Sigma-Aldrich) using pellet pestles, blue polypropylene (Sigma-Aldrich) on ice. After 1-h agitation at 4 °C, lysates were centrifuged at 14000 rpm for 10 min. The supernatant’s protein concentration was measured using the Pierce BCA Protein Assay kit (Thermo Scientific) according to the manufacturer’s protocol. Lysate aliquots were stored at -80 °C for subsequent analysis. After the preparation of all the samples, the IL-10 and TGF-β levels in the lysates were measured using Mouse IL-10 Quantikine ELISA Kit (R&D system) and Mouse/Rat/Porcine/Canine TGF-beta 1 Quantikine ELISA Kit (R&D system). For TGF-β assessments, the samples were treated with 1N HCL to activate the latent TGF-β before measurement using an ELISA kit. The optical density was determined at 450 nm using a microplate reader (FLUOstar Omega, BMG LABTECH, Germany) within 15 min after stopping the reactions. All measurements were performed in duplicate.
Statistical analysis
Values are presented as mean ± standard deviation. Statistical analysis was performed using GraphPad Prism ver. 7 (GraphPad Software, Inc., San Diego, CA, USA). One-way ANOVA was used for multiple comparisons followed by the Sidak test for determining statistical significance. Moreover, we performed two-way ANOVA with the Bonferroni multiple comparison test as the post hoc test. In non-normally distributed data, Kruskal-Wallis test was used followed by the Dunn’s comparison to be consistent with the results as non-parametric test. Statistical significance was defined at p < 0.05.