The aim of this study was to investigate whether the intracerebral injection of miR-124 in a mouse model of ischemic stroke before or after the peak phase of M1 polarization modifies the M1/M2 balance. For this purpose, a total of 34 adult male C57BL6/N mice (11–12 weeks, 21–25 g; Janvier, Saint Berthevin Cedex, France) were randomly allocated to different experimental groups of miR-124 injection and survival times. Animals were housed under fixed circadian rhythm with ad libitum access to food and water. All surgical and scanning procedures were performed under Isoflurane anesthesia and core body temperature control. Stroke was induced in all mice via the middle cerebral artery occlusion (MCAO) model (cf. below). Successful ischemic stroke lesions were verified by magnetic resonance imaging (MRI) scans 48 h after MCAO, before intracranial miR-124 injection, and again before perfusion fixation at day 6 or day 14, respectively.
In Experiment 1, fifteen mice were subjected to MCAO and perfused 6 days later, divided into the following three groups: i) control mice exposed to stroke only (n = 5), ii) miR-treated mice which received an intracranial injection of liposomated miR-124 48 h after MCAO (n = 5), and iii) mice receiving a liposomated random primer injection 48 h after MCAO (n = 5) as negative control.
In Experiment 2, fourteen mice were subjected to MCAO and perfused 14 days later, divided into the following three groups: i) control mice subjected to stroke only (n = 5), ii) miR-treated mice which received an intracranial injection of liposomated miR-124 48 h after MCAO (n = 6), and iii) the negative control group that was injected with a liposomated random primer 48 h after MCAO (n = 3).
In Experiment 3, stroke was induced in ten mice, and they were perfused 14 days later, divided into the following two groups: control mice subjected to stroke only (n = 5), and the miR-treated mice with liposomated miR-124 injection at day 10 post MCAO (n = 5).
Middle Cerebral Artery Occlusion
Transient occlusion of the right middle cerebral artery (MCAO) was induced with the intraluminal filament model, as described previously (Adamczak et al. 2014). Briefly, mice were initially anesthetized with 2 % Isoflurane (O2:N2O, 30:70 %) and were subcutaneously injected with 4 mg/kg Carprofen (Pfizer, Berlin, Germany). The body core temperature was controlled during surgery via a temperature-regulated heating pad (medres GmbH, Cologne, Germany). The right common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA) were presented. A silicon rubber-coated filament with the length of 20 mm and a tip diameter of 170 μm (7017PK5Re, Doccol Corporation, Sharon, USA) was advanced into the ICA lumen until it blocked the origin of MCA. During 30 min occlusion time, mice were allowed to recover from anesthesia in a controlled heating box. Afterwards, the animals were re-anesthetized, the reperfusion was induced by filament withdrawal and the CCA was ligated. Following surgery, NaCl was subcutaneously injected in all animals twice daily until stabilization of body weight. Only the mice with cortico-striatal lesion, observed by MRI two days after MCAO, were selected for the experiments.
Intracranial Injection of Liposomated miR-124
T2-weighted MRI was used to determine the injection coordinates 48 h after MCAO. At the day of injection the miR-124 (2 μg in 50 μl PBS, PM10691 Applied Biosystems, Carlsbad, CA, USA) or the negative control of miR (2 μg in 50 μl PBS , AM17110; Applied Biosystems) was mixed with the transfection agent Lipofectamine 2000 (Invitrogen, Paisley, UK). Intracranial injection into the right striatum, ipsilateral to the lesion, was performed as described elsewhere (Aswendt et al. 2012). Briefly, mice were fixed in a stereotactic frame (Stoelting, Dublin, Ireland), and 2.0 μl of the suspension containing 100 ng of miR or random primer was injected with a Hamilton syringe (26G needle) at the following coordinates relative to bregma: AP +0.5; L + 1.4; DV −2.4 with an infusion rate of 500 nl/min. The needle was kept in place for another 5 min before removal.
To observe the different aspects of neurological functions, a set of behavioral tests was performed before and every 2 days after MCAO using the modified neurological deficit scores (mNDS), a modification of a previous report (Chen et al. 2001). The modified NDS (mNDS) has already been described recently (Hamzei Taj et al. 2016). In short, it consists of a set of motor tests (muscle status and abnormal movement), sensory tests (tactile, and proprioceptive), and reflex tests on a scale of 0–16. One point was given for the failure of a performed test or for the loss of a tested reflex. Thus, higher scores indicate higher severity of ischemia.
Magnetic Resonance Imaging
MRI was performed using a 9.4 T Biospec animal MRI system with a 20 cm horizontal bore magnet (Bruker BioSpin, Ettlingen, Germany) equipped with actively shielded gradient coils (BGA9S, 750 mT/m, Bruker BioSpin), using ParaVision 5 software. Radio frequency (RF) transmission was achieved with a 112/72 mm od/id mouse resonator (Bruker) while signal was received with a dedicated quadrature mouse head surface coil (Bruker BioSpin). Mice were anesthetized with 2 % Isoflurane in a 30/70 oxygen/air mixture and placed in an animal holder (Bruker) using a tooth bar and ear bars for stable positioning. Respiration rate was monitored using a pressure sensitive pad placed underneath the mice. The physiological status of the animal was monitored with DASYlab software (National Instruments, Austin, TX, USA). The body core temperature was monitored with a rectal probe and was kept constant at 37 ± 1.0 °C using a water blanket connected to a feedback controlled automated temperature control unit (medres, Cologne).
Tripilot gradient-echo scans were used for definite positioning of the mouse head in the magnet. For lesion visualization, T2-weighted images were acquired with a multi-slice multi-echo (MSME) spin echo sequence (TR/TE = 5000 ms/10 ms, 16 echoes, 10 coronal slices, slice thickness 0.5 mm, FOV 2.5 × 2.5 cm2, matrix 256 × 256, resolution 98 × 98 μm2, bandwidth 75 kHz).
Quantitative T2 maps were calculated from the multi-echo trains using the IDL software (Exelis Visual Information Solutions, Boulder, CO, USA), by pixelwise fitting signal intensities to a mono-exponential decay curve. Average T2 relaxation of the healthy cortex and striatum was determined. For determination of the lesion volume, all pixels on the ipsilateral hemisphere were counted with T2 values above the threshold of normal T2 + 2 standard deviations.
Animals were allowed to survive for 6 or 14 days after MCAO and were subsequently transcardially perfused under deep Isoflurane anesthesia with 20 ml cold phosphate buffered saline (PBS), followed by 20 ml 4 % paraformaldehyde (PFA). Afterwards, brains were post-fixed in 4 % PFA overnight and then cryo-protected by immersion in 30 % sucrose solution for the next 2 days at 4 °C. Then, brains were frozen in −40 °C methylbutane (Sigma-Aldrich, Taufkirchen, Germany) and subsequently stored at −80 °C. The brains were cut into 10 μm coronal slices using a cryostat (Leica Microsystems, Wetzlar, Germany), directly mounted, and stored at −20 °C.
At the day of immunostaining the cryosections were kept at room temperature (RT) for 30 min, and for stainings of ionized calcium-binding adapter molecule 1(Iba-1) acetone pre-treatment at −20 °C for 20 min was performed. To prevent non-specific binding of antibodies, the sections were pre-incubated in 5 % normal serum and 0.25 % Triton X-100, in potassium phosphate buffered saline (KPBS) for 60 min at RT. The treated sections were incubated overnight at 4 °C with subsequent primary antibodies: rabbit polyclonal anti Iba-1 (1:1000, 019–19,741; Wako Chem, Osaka, Japan), mouse anti-mannose receptor, MMR/CD206, (1:200, AF2535; R & D Systems, Minneapolis, MN, USA) and rat anti CD16/32 (1:200, 101,301, Biolegend, San Diego, CA), followed by Cy5 and Cy3 conjugated secondary antibodies (1:200, Jackson Immuno Research, West Grove, PA, USA) for 2 h at RT. For nuclear staining Hoechst 33,342 (1:1000; Invitrogen, Carlsbad, USA) was added together with secondary antibodies. Negatively stained control sections were included with equal preparation, excluding primary antibodies.
Three sections per mouse were imaged with a fluorescent microscope (BZ-9000 Keyence, Osaka, Japan) with 4×, 20× and 40× magnification objectives. 6 different regions of interest (ROIs) in each brain section were selected with the exposure time kept constant: the border and the core region of the ischemic hemisphere in the cortex and the striatum, further, two ROIs in the cortex and striatum of the intact hemisphere.
Quantitative Immunohistochemical Analysis
To observe the polarization of microglia/macrophages, immunofluorescence images were analyzed using NIH ImageJ analysis software (ImageJ) and TissueQuest 4.0 (TissueGnostics, Vienna, Austria), a specific image cytometry analysis software. Cells were identified on the tissue section based on the nuclei staining (nuclei size, staining intensity and discrimination by area was optimized manually) followed by the analysis of specific staining. To achieve optimal cell detection the background threshold was defined and due to the coverage of Iba-1 cell ramification with total nuclei intensity, the cut-offs were defined for each ROI to discriminate false signals. Scattergrams were created to visualize the corresponding positive cells in the source ROI through the real-time back gating component. Mean intensity and the relative number of co-expressed Iba-1 and CD206 or CD16/32 were obtained, and mean values were estimated from analyses of at least three brain sections per mouse. To observe M1 and M2 representative markers in double staining with Iba-1, the above-defined 6 different ROIs provided the following number of cells/mm3: i) total cell density according to the nuclei staining, ii) Iba-1+ microglial/macrophage cell density, iii) CD206+/Iba-1+ cell density, iv) CD16/32+/Iba-1+ cell density. Then, a mean was calculated for each region for each animal.
Data were analyzed by SPSS version 22 (IBM SPSS statistics, Ehningen, Germany). The Normality test and homogeneity of variances were evaluated for all data. For behavioral scores (mNDS) the nonparametric analysis approach, Kruskal-Wallis H, was performed. IHC data from early time point of liposomated miR-124 injection at day 2 were tested for significant changes between the 3 groups using one-way analysis of variance (ANOVA) with Bonferroni corrected posthoc comparisons. Using an independent one-tailed Student’s t-test, we tested the data of late liposomated miR-124 injection to observe the significant changes between the stroke-only group and the group treated with miR-124 at day 10.
The IHC numbers are represented by box-and-whisker plots (Figs. 3, 4 and 6) wherein each box shows the central 50 % of the data points, the interquartile range (IQR), a horizontal line in each box indicates the median, and the vertical bars speak for the spread of 1.5 × IQR. Dots display outliers, which were included in calculations of significance. The box-and-whisker plots were generated by using SPSS version 22.
Bivariate correlation analysis between M1/M2 expression by Iba-1+ cells and mNDS were measured with Spearman’s correlation coefficient. Regression was done with M1/M2 expression by Iba-1+ cells as the independent variable and mNDS as the dependent variable. A p-value ≤0.05 was regarded statistically significant.